==================================================================================================== import os import sys with open(sys.argv[0]) as f: code = f.read() # read the code of this file ASAP, for logging import uuid import glob import time from dataclasses import dataclass import numpy as np import torch from torch import nn import torch.nn.functional as F import torch.distributed as dist import torch._inductor.config as config from torch.nn.parallel import DistributedDataParallel as DDP # ----------------------------------------------------------------------------- # Muon optimizer def zeropower_via_svd(G, steps=None): U, S, V = G.svd() return U @ V.T @torch.compile def zeropower_via_newtonschulz5(G, steps=10, eps=1e-7): """ Newton-Schulz iteration to compute the zeroth power / orthogonalization of G. We opt to use a quintic iteration whose coefficients are selected to maximize the slope at zero. For the purpose of minimizing steps, it turns out to be empirically effective to keep increasing the slope at zero even beyond the point where the iteration no longer converges all the way to one everywhere on the interval. This iteration therefore does not produce UV^T but rather something like US'V^T where S' is diagonal with S_{ii}' \sim Uniform(0.5, 1.5), which turns out not to hurt model performance at all relative to UV^T, where USV^T = G is the SVD. """ assert len(G.shape) == 2 a, b, c = (3.4445, -4.7750, 2.0315) X = G.bfloat16() X /= (X.norm() + eps) # ensure top singular value <= 1 if G.size(0) > G.size(1): X = X.T for _ in range(steps): A = X @ X.T B = A @ X X = a * X + b * B + c * A @ B if G.size(0) > G.size(1): X = X.T return X zeropower_backends = dict(svd=zeropower_via_svd, newtonschulz5=zeropower_via_newtonschulz5) class Muon(torch.optim.Optimizer): """ Muon - MomentUm Orthogonalized by Newton-schulz Muon internally runs standard SGD-momentum, and then performs an orthogonalization post- processing step, in which each 2D parameter's update is replaced with the nearest orthogonal matrix. To efficiently orthogonalize each update, we use a Newton-Schulz iteration, which has the advantage that it can be stably run in bfloat16 on the GPU. Some warnings: - This optimizer assumes that all parameters passed in are 2D. - It should not be used for the embedding layer, the final fully connected layer, or any {0,1}-D parameters; those should all be optimized by a standard method (e.g., AdamW). - To use it with 4D convolutional filters, it works well to just flatten their last 3 dimensions. - We believe it is unlikely to work well for training with small batch size. - We believe it may not work well for finetuning pretrained models, but we haven't tested this. - We have not yet tried this optimizer for training scenarios larger than NanoGPT (124M). Arguments: lr: The learning rate used by the internal SGD. momentum: The momentum used by the internal SGD. nesterov: Whether to use Nesterov-style momentum in the internal SGD. (recommended) backend: The chosen backend for the orthogonalization step. (recommended: 'newtonschulz5') backend_steps: The number of iteration steps to use in the backend, if it is iterative. """ def __init__(self, params, lr=3e-4, momentum=0.95, nesterov=True, backend='newtonschulz5', backend_steps=5, rank=0, world_size=1): defaults = dict(lr=lr, momentum=momentum, nesterov=nesterov, backend=backend, backend_steps=backend_steps) super().__init__(params, defaults) self.rank = rank self.world_size = world_size def step(self): for group in self.param_groups: lr = group['lr'] momentum = group['momentum'] zeropower_backend = zeropower_backends[group['backend']] # generate weight updates in distributed fashion total_params = sum(p.numel() for p in group['params']) updates_flat = torch.zeros(total_params, device='cuda', dtype=torch.bfloat16) curr_idx = 0 for i, p in enumerate(group['params']): # luckily this will perfectly distribute a transformer with multiple of 4 layers to 8 GPUs if i % self.world_size == self.rank: g = p.grad if g is None: continue state = self.state[p] if 'momentum_buffer' not in state: state['momentum_buffer'] = torch.zeros_like(g) buf = state['momentum_buffer'] buf.mul_(momentum).add_(g) if group['nesterov']: g = g.add(buf, alpha=momentum) g = zeropower_backend(g, steps=group['backend_steps']) g *= max(g.size(0), g.size(1))**0.5 # scale to have update.square().mean() == 1 updates_flat[curr_idx:curr_idx+p.numel()] = g.flatten() curr_idx += p.numel() # sync updates across devices. we are not memory-constrained so can do this simple deserialization dist.all_reduce(updates_flat, op=dist.ReduceOp.SUM) # deserialize and apply updates curr_idx = 0 for p in group['params']: g = updates_flat[curr_idx:curr_idx+p.numel()].view_as(p.data).type_as(p.data) p.data.add_(g, alpha=-lr) curr_idx += p.numel() # ----------------------------------------------------------------------------- # PyTorch nn.Module definitions for the GPT-2 model class Rotary(torch.nn.Module): def __init__(self, dim, base=10000): super().__init__() self.inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float() / dim)) self.seq_len_cached = None self.cos_cached = None self.sin_cached = None def forward(self, x): seq_len = x.shape[1] if seq_len != self.seq_len_cached: self.seq_len_cached = seq_len t = torch.arange(seq_len, device=x.device).type_as(self.inv_freq) freqs = torch.outer(t, self.inv_freq).to(x.device) self.cos_cached = freqs.cos().bfloat16() self.sin_cached = freqs.sin().bfloat16() return self.cos_cached[None, :, None, :], self.sin_cached[None, :, None, :] def apply_rotary_emb(x, cos, sin): assert x.ndim == 4 # multihead attention d = x.shape[3]//2 x1 = x[..., :d] x2 = x[..., d:] y1 = x1 * cos + x2 * sin y2 = x1 * (-sin) + x2 * cos return torch.cat([y1, y2], 3).type_as(x) class CausalSelfAttention(nn.Module): def __init__(self, config): super().__init__() self.n_head = config.n_head self.n_embd = config.n_embd self.head_dim = self.n_embd // self.n_head assert self.n_embd % self.n_head == 0 self.c_q = nn.Linear(self.n_embd, self.n_embd, bias=False) self.c_k = nn.Linear(self.n_embd, self.n_embd, bias=False) self.c_v = nn.Linear(self.n_embd, self.n_embd, bias=False) # output projection self.c_proj = nn.Linear(self.n_embd, self.n_embd, bias=False) self.c_proj.weight.data.zero_() # zero init suggested by @Grad62304977 self.rotary = Rotary(self.head_dim) def forward(self, x): B, T, C = x.size() # batch size, sequence length, embedding dimensionality (n_embd) q = self.c_q(x).view(B, T, self.n_head, self.head_dim) k = self.c_k(x).view(B, T, self.n_head, self.head_dim) v = self.c_v(x).view(B, T, self.n_head, self.head_dim) cos, sin = self.rotary(q) q, k = F.rms_norm(q, (q.size(-1),)), F.rms_norm(k, (k.size(-1),)) # QK norm suggested by @Grad62304977 q, k = apply_rotary_emb(q, cos, sin), apply_rotary_emb(k, cos, sin) y = F.scaled_dot_product_attention(q.transpose(1, 2), k.transpose(1, 2), v.transpose(1, 2), is_causal=True) y = y.transpose(1, 2).contiguous().view_as(x) # re-assemble all head outputs side by side y = self.c_proj(y) return y class MLP(nn.Module): def __init__(self, config): super().__init__() self.c_fc = nn.Linear(config.n_embd, 4 * config.n_embd, bias=False) self.c_proj = nn.Linear(4 * config.n_embd, config.n_embd, bias=False) self.c_proj.weight.data.zero_() # zero init suggested by @Grad62304977 def forward(self, x): x = self.c_fc(x) x = F.relu(x).square() # https://arxiv.org/abs/2109.08668v2; ~1-2% better than GELU; suggested by @SKYLINEZ007 and @Grad62304977 x = self.c_proj(x) return x class Block(nn.Module): def __init__(self, config): super().__init__() self.attn = CausalSelfAttention(config) self.mlp = MLP(config) def forward(self, x): x = x + self.attn(F.rms_norm(x, (x.size(-1),))) x = x + self.mlp(F.rms_norm(x, (x.size(-1),))) return x # ----------------------------------------------------------------------------- # The main GPT-2 model @dataclass class GPTConfig: vocab_size : int = 50304 n_layer : int = 12 n_head : int = 6 # head dim 128 suggested by @Grad62304977 n_embd : int = 768 class GPT(nn.Module): def __init__(self, config): super().__init__() self.config = config self.transformer = nn.ModuleDict(dict( wte = nn.Embedding(config.vocab_size, config.n_embd), h = nn.ModuleList([Block(config) for _ in range(config.n_layer)]), )) self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False) self.transformer.wte.weight = self.lm_head.weight # https://paperswithcode.com/method/weight-tying def forward(self, idx, targets=None, return_logits=True): # forward the GPT model itself x = self.transformer.wte(idx) # token embeddings of shape (b, t, n_embd) for block in self.transformer.h: x = block(x) x = F.rms_norm(x, (x.size(-1),)) if targets is not None: # if we are given some desired targets also calculate the loss logits = self.lm_head(x) logits = logits.float() # use tf32/fp32 for logits loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1), ignore_index=-1) else: # inference-time mini-optimization: only forward the lm_head on the very last position logits = self.lm_head(x[:, [-1], :]) # note: using list [-1] to preserve the time dim logits = logits.float() # use tf32/fp32 for logits loss = None # there are performance reasons why not returning logits is prudent, if not needed if not return_logits: logits = None return logits, loss # ----------------------------------------------------------------------------- # Our own simple Distributed Data Loader def _peek_data_shard(filename): # only reads the header, returns header data with open(filename, "rb") as f: # first read the header, which is 256 int32 integers (4 bytes each) header = np.frombuffer(f.read(256*4), dtype=np.int32) if header[0] != 20240520: print("ERROR: magic number mismatch in the data .bin file!") print("---> HINT: Are you passing in a correct file with --input_bin?") print("---> HINT: Dataset encoding changed recently, re-run data prepro or refer again to README") print("---> HINT: For example re-run: `python dev/data/tinyshakespeare.py`, then re-try") exit(1) assert header[1] == 1, "unsupported version" ntok = header[2] # number of tokens (claimed) return ntok # for now just return the number of tokens def _load_data_shard(filename): with open(filename, "rb") as f: # first read the header, which is 256 int32 integers (4 bytes each) header = np.frombuffer(f.read(256*4), dtype=np.int32) assert header[0] == 20240520, "magic number mismatch in the data .bin file" assert header[1] == 1, "unsupported version" ntok = header[2] # number of tokens (claimed) # the rest of it are tokens, stored as uint16 tokens = np.frombuffer(f.read(), dtype=np.uint16) assert len(tokens) == ntok, "number of tokens read does not match header?" return tokens class DistributedDataLoader: def __init__(self, filename_pattern, B, T, process_rank, num_processes): self.process_rank = process_rank self.num_processes = num_processes self.B = B self.T = T # glob files that match the pattern self.files = sorted(glob.glob(filename_pattern)) assert len(self.files) > 0, f"did not find any files that match the pattern {filename_pattern}" # load and validate all data shards, count number of tokens in total ntok_total = 0 for fname in self.files: shard_ntok = _peek_data_shard(fname) assert shard_ntok >= num_processes * B * T + 1 ntok_total += int(shard_ntok) self.ntok_total = ntok_total # kick things off self.reset() def reset(self): self.current_shard = 0 self.current_position = self.process_rank * self.B * self.T self.tokens = _load_data_shard(self.files[self.current_shard]) def advance(self): # advance to next data shard self.current_shard = (self.current_shard + 1) % len(self.files) self.current_position = self.process_rank * self.B * self.T self.tokens = _load_data_shard(self.files[self.current_shard]) def next_batch(self): B = self.B T = self.T buf = self.tokens[self.current_position : self.current_position+B*T+1] buf = torch.tensor(buf.astype(np.int32), dtype=torch.long) x = (buf[:-1]).view(B, T) # inputs y = (buf[1:]).view(B, T) # targets # advance current position and load next shard if necessary self.current_position += B * T * self.num_processes if self.current_position + (B * T * self.num_processes + 1) > len(self.tokens): self.advance() return x.cuda(), y.cuda() # ----------------------------------------------------------------------------- # int main @dataclass class Hyperparameters: # data hyperparams input_bin : str = 'data/fineweb10B/fineweb_train_*.bin' # input .bin to train on input_val_bin : str = 'data/fineweb10B/fineweb_val_*.bin' # input .bin to eval validation loss on # optimization hyperparams batch_size : int = 8*64 # batch size, in sequences, across all devices device_batch_size : int = 64 # batch size, in sequences, per device sequence_length : int = 1024 # sequence length, in tokens num_iterations : int = 5100 # number of iterations to run learning_rate : float = 0.0036 warmup_iters : int = 0 warmdown_iters : int = 1450 # number of iterations of linear warmup/warmdown for triangular or trapezoidal schedule weight_decay : float = 0 # evaluation and logging hyperparams val_loss_every : int = 125 # every how many steps to evaluate val loss? 0 for only at the end val_tokens : int = 10485760 # how many tokens of validation data? it's important to keep this fixed for consistent comparisons save_every : int = 0 # every how many steps to save the checkpoint? 0 for only at the end args = Hyperparameters() # set up DDP (distributed data parallel). torchrun sets this env variable assert torch.cuda.is_available() dist.init_process_group(backend='nccl') ddp_rank = int(os.environ['RANK']) ddp_local_rank = int(os.environ['LOCAL_RANK']) ddp_world_size = int(os.environ['WORLD_SIZE']) device = f'cuda:{ddp_local_rank}' torch.cuda.set_device(device) print(f"using device: {device}") master_process = (ddp_rank == 0) # this process will do logging, checkpointing etc. # convenience variables B, T = args.device_batch_size, args.sequence_length # calculate the number of steps to take in the val loop. assert args.val_tokens % (B * T * ddp_world_size) == 0 val_steps = args.val_tokens // (B * T * ddp_world_size) # calculate the steps of gradient accumulation required to attain the desired global batch size. assert args.batch_size % (B * ddp_world_size) == 0 train_accumulation_steps = args.batch_size // (B * ddp_world_size) # load tokens train_loader = DistributedDataLoader(args.input_bin, B, T, ddp_rank, ddp_world_size) val_loader = DistributedDataLoader(args.input_val_bin, B, T, ddp_rank, ddp_world_size) if master_process: print(f"Training DataLoader: total number of tokens: {train_loader.ntok_total} across {len(train_loader.files)} files") print(f"Validation DataLoader: total number of tokens: {val_loader.ntok_total} across {len(val_loader.files)} files") x, y = train_loader.next_batch() # there are only 50257 unique GPT-2 tokens; we extend to nearest multiple of 128 for efficiency. suggested to me by @Grad62304977. # this originates from Karpathy's experiments. num_vocab = 50304 model = GPT(GPTConfig(vocab_size=num_vocab, n_layer=12, n_head=6, n_embd=768)) model = model.cuda() if hasattr(config, "coordinate_descent_tuning"): config.coordinate_descent_tuning = True # suggested by @Chillee model = torch.compile(model) # here we wrap model into DDP container model = DDP(model, device_ids=[ddp_local_rank]) raw_model = model.module # always contains the "raw" unwrapped model ctx = torch.amp.autocast(device_type='cuda', dtype=torch.bfloat16) # init the optimizer(s) optimizer1 = torch.optim.AdamW(raw_model.lm_head.parameters(), lr=args.learning_rate, betas=(0.9, 0.95), weight_decay=args.weight_decay, fused=True) optimizer2 = Muon(raw_model.transformer.h.parameters(), lr=0.1*args.learning_rate, momentum=0.95, rank=ddp_rank, world_size=ddp_world_size) optimizers = [optimizer1, optimizer2] # learning rate decay scheduler (linear warmup and warmdown) def get_lr(it): assert it <= args.num_iterations # 1) linear warmup for warmup_iters steps if it < args.warmup_iters: return (it+1) / args.warmup_iters # 2) constant lr for a while elif it < args.num_iterations - args.warmdown_iters: return 1.0 # 3) linear warmdown else: decay_ratio = (args.num_iterations - it) / args.warmdown_iters return decay_ratio schedulers = [torch.optim.lr_scheduler.LambdaLR(opt, get_lr) for opt in optimizers] # begin logging if master_process: run_id = str(uuid.uuid4()) logdir = 'logs/%s/' % run_id os.makedirs(logdir, exist_ok=True) logfile = 'logs/%s.txt' % run_id # create the log file with open(logfile, "w") as f: # begin the log by printing this file (the Python code) f.write('='*100 + '\n') f.write(code) f.write('='*100 + '\n') # log information about the hardware/software environment this is running on # and print the full `nvidia-smi` to file f.write(f"Running pytorch {torch.version.__version__} compiled for CUDA {torch.version.cuda}\nnvidia-smi:\n") import subprocess result = subprocess.run(['nvidia-smi'], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True) f.write(f'{result.stdout}\n') f.write('='*100 + '\n') training_time_ms = 0 # start the clock torch.cuda.synchronize() t0 = time.time() # begin training train_loader.reset() for step in range(args.num_iterations + 1): last_step = (step == args.num_iterations) # This effectively ignores timing first 10 steps, which are slower for weird reasons. # Alternately, and slightly more correctly in terms of benchmarking, we could do 10 # steps with dummy data first, and then re-initialize the model and reset the loader. if step == 10: training_time_ms = 0 t0 = time.time() timed_steps = float('nan') if step <= 11 else (step - 10) + 1 # <= 11 to avoid bug in val # once in a while evaluate the validation dataset if (last_step or (args.val_loss_every > 0 and step % args.val_loss_every == 0)): # stop the clock torch.cuda.synchronize() training_time_ms += 1000 * (time.time() - t0) # run validation batches model.eval() val_loader.reset() val_loss = 0.0 for _ in range(val_steps): x_val, y_val = val_loader.next_batch() with ctx: # of course, we'd like to use no_grad() here too, but that creates a torch.compile error for some reason _, loss = model(x_val, y_val, return_logits=False) val_loss += loss.detach() del loss dist.all_reduce(val_loss, op=dist.ReduceOp.AVG) val_loss /= val_steps # log val loss to console and to logfile if master_process: print(f'step:{step}/{args.num_iterations} val_loss:{val_loss:.4f} train_time:{training_time_ms:.0f}ms step_avg:{training_time_ms/(timed_steps-1):.2f}ms') with open(logfile, "a") as f: f.write(f'step:{step}/{args.num_iterations} val_loss:{val_loss:.4f} train_time:{training_time_ms:.0f}ms step_avg:{training_time_ms/(timed_steps-1):.2f}ms\n') # start the clock again torch.cuda.synchronize() t0 = time.time() if master_process and (last_step or (args.save_every > 0 and step % args.save_every == 0)): # stop the clock torch.cuda.synchronize() training_time_ms += 1000 * (time.time() - t0) # save the state of the training process log = dict(step=step, code=code, model=raw_model.state_dict(), optimizers=[opt.state_dict() for opt in optimizers]) torch.save(log, 'logs/%s/state_step%06d.pt' % (run_id, step)) # start the clock again torch.cuda.synchronize() t0 = time.time() # bit confusing: we want to make sure to eval on 0th iteration # but also after the very last iteration. so we loop for step <= num_iterations # instead of just < num_iterations (one extra due to <=), only to do # the validation/sampling one last time, and then we break right here as we're done. if last_step: break # --------------- TRAINING SECTION BEGIN ----------------- model.train() for i in range(1, train_accumulation_steps+1): # forward pass with ctx: _, loss = model(x, y, return_logits=False) train_loss = loss.detach() # advance the dataset for the next batch x, y = train_loader.next_batch() # backward pass if i < train_accumulation_steps: with model.no_sync(): # there's no need to sync gradients every accumulation step loss.backward() else: loss.backward() # just sync on the last step for p in model.parameters(): p.grad /= train_accumulation_steps # step the optimizers and schedulers for opt, sched in zip(optimizers, schedulers): opt.step() sched.step() # null the gradients model.zero_grad(set_to_none=True) # --------------- TRAINING SECTION END ------------------- # everything that follows now is just diagnostics, prints, logging, etc. #dist.all_reduce(train_loss, op=dist.ReduceOp.AVG) # all-reducing the training loss would be more correct in terms of logging, but slower if master_process: approx_time = training_time_ms + 1000 * (time.time() - t0) print(f"step:{step+1}/{args.num_iterations} train_loss:{train_loss.item():.4f} train_time:{approx_time:.0f}ms step_avg:{approx_time/timed_steps:.2f}ms") with open(logfile, "a") as f: f.write(f"step:{step+1}/{args.num_iterations} train_loss:{train_loss.item():.4f} train_time:{approx_time:.0f}ms step_avg:{approx_time/timed_steps:.2f}ms\n") if master_process: print(f"peak memory consumption: {torch.cuda.max_memory_allocated() // 1024 // 1024} MiB") # ------------------------------------------------------------------------- # clean up nice dist.destroy_process_group() ==================================================================================================== Running pytorch 2.4.1+cu121 compiled for CUDA 12.1 nvidia-smi: Wed Oct 16 13:25:45 2024 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 555.42.06 Driver Version: 555.42.06 CUDA Version: 12.5 | |-----------------------------------------+------------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+======================| | 0 NVIDIA H100 80GB HBM3 Off | 00000000:18:00.0 Off | 0 | | N/A 28C P0 111W / 700W | 5786MiB / 81559MiB | 9% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ | 1 NVIDIA H100 80GB HBM3 Off | 00000000:2A:00.0 Off | 0 | | N/A 30C P0 116W / 700W | 5834MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ | 2 NVIDIA H100 80GB HBM3 Off | 00000000:3A:00.0 Off | 0 | | N/A 31C P0 113W / 700W | 5834MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ | 3 NVIDIA H100 80GB HBM3 Off | 00000000:5D:00.0 Off | 0 | | N/A 28C P0 115W / 700W | 5834MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ | 4 NVIDIA H100 80GB HBM3 Off | 00000000:84:00.0 Off | 0 | | N/A 28C P0 113W / 700W | 5834MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ | 5 NVIDIA H100 80GB HBM3 Off | 00000000:8B:00.0 Off | 0 | | N/A 30C P0 115W / 700W | 5834MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ | 6 NVIDIA H100 80GB HBM3 Off | 00000000:91:00.0 Off | 0 | | N/A 29C P0 112W / 700W | 5834MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ | 7 NVIDIA H100 80GB HBM3 Off | 00000000:E4:00.0 Off | 0 | | N/A 28C P0 116W / 700W | 5594MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+------------------------+----------------------+ +-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=========================================================================================| | 0 N/A N/A 20785 C /usr/bin/python3 0MiB | | 1 N/A N/A 20786 C /usr/bin/python3 0MiB | | 2 N/A N/A 20787 C /usr/bin/python3 0MiB | | 3 N/A N/A 20788 C /usr/bin/python3 0MiB | | 4 N/A N/A 20789 C /usr/bin/python3 0MiB | | 5 N/A N/A 20790 C /usr/bin/python3 0MiB | | 6 N/A N/A 20791 C /usr/bin/python3 0MiB | | 7 N/A N/A 20792 C /usr/bin/python3 0MiB | +-----------------------------------------------------------------------------------------+ ==================================================================================================== step:0/5100 val_loss:16.0098 train_time:286ms step_avg:nanms step:1/5100 train_loss:16.0051 train_time:60444ms step_avg:nanms step:2/5100 train_loss:9.5200 train_time:60535ms step_avg:nanms step:3/5100 train_loss:8.7468 train_time:60688ms step_avg:nanms step:4/5100 train_loss:8.0303 train_time:60841ms step_avg:nanms step:5/5100 train_loss:7.4920 train_time:60993ms step_avg:nanms step:6/5100 train_loss:7.4504 train_time:61145ms step_avg:nanms step:7/5100 train_loss:7.3315 train_time:61298ms step_avg:nanms step:8/5100 train_loss:7.5719 train_time:61450ms step_avg:nanms step:9/5100 train_loss:7.4521 train_time:61604ms step_avg:nanms step:10/5100 train_loss:7.0940 train_time:61759ms step_avg:nanms step:11/5100 train_loss:6.9980 train_time:90ms step_avg:nanms step:12/5100 train_loss:6.8961 train_time:242ms step_avg:nanms step:13/5100 train_loss:6.7010 train_time:395ms step_avg:131.59ms step:14/5100 train_loss:6.6766 train_time:548ms step_avg:136.89ms step:15/5100 train_loss:6.6384 train_time:700ms step_avg:140.03ms step:16/5100 train_loss:6.5587 train_time:853ms step_avg:142.19ms step:17/5100 train_loss:6.5682 train_time:1009ms step_avg:144.12ms step:18/5100 train_loss:6.5854 train_time:1163ms step_avg:145.32ms step:19/5100 train_loss:6.4148 train_time:1315ms step_avg:146.14ms step:20/5100 train_loss:6.4305 train_time:1469ms step_avg:146.92ms step:21/5100 train_loss:6.0939 train_time:1622ms step_avg:147.48ms step:22/5100 train_loss:6.4600 train_time:1777ms step_avg:148.05ms step:23/5100 train_loss:6.6648 train_time:1929ms step_avg:148.36ms step:24/5100 train_loss:6.3462 train_time:2082ms step_avg:148.74ms step:25/5100 train_loss:6.4746 train_time:2236ms step_avg:149.06ms step:26/5100 train_loss:6.1830 train_time:2388ms step_avg:149.27ms step:27/5100 train_loss:6.1035 train_time:2542ms step_avg:149.55ms step:28/5100 train_loss:6.2450 train_time:2694ms step_avg:149.68ms step:29/5100 train_loss:5.9279 train_time:2848ms step_avg:149.89ms step:30/5100 train_loss:6.2054 train_time:3001ms step_avg:150.03ms step:31/5100 train_loss:6.0422 train_time:3154ms step_avg:150.20ms step:32/5100 train_loss:6.0101 train_time:3308ms step_avg:150.35ms step:33/5100 train_loss:5.8360 train_time:3461ms step_avg:150.48ms step:34/5100 train_loss:6.1218 train_time:3613ms step_avg:150.56ms step:35/5100 train_loss:6.0536 train_time:3766ms step_avg:150.66ms step:36/5100 train_loss:6.1885 train_time:3919ms step_avg:150.74ms step:37/5100 train_loss:6.1319 train_time:4073ms step_avg:150.86ms step:38/5100 train_loss:6.0204 train_time:4226ms step_avg:150.91ms step:39/5100 train_loss:5.9126 train_time:4381ms step_avg:151.06ms step:40/5100 train_loss:5.9277 train_time:4533ms step_avg:151.11ms step:41/5100 train_loss:5.8419 train_time:4686ms step_avg:151.15ms step:42/5100 train_loss:5.8712 train_time:4839ms step_avg:151.20ms step:43/5100 train_loss:5.7509 train_time:4992ms step_avg:151.26ms step:44/5100 train_loss:5.8549 train_time:5144ms step_avg:151.28ms step:45/5100 train_loss:5.8095 train_time:5297ms step_avg:151.34ms step:46/5100 train_loss:5.9779 train_time:5449ms step_avg:151.37ms step:47/5100 train_loss:5.7638 train_time:5603ms step_avg:151.42ms step:48/5100 train_loss:5.6291 train_time:5757ms step_avg:151.50ms step:49/5100 train_loss:5.8346 train_time:5911ms step_avg:151.55ms step:50/5100 train_loss:5.7183 train_time:6063ms step_avg:151.58ms step:51/5100 train_loss:5.8558 train_time:6216ms step_avg:151.62ms step:52/5100 train_loss:5.7231 train_time:6369ms step_avg:151.65ms step:53/5100 train_loss:5.5794 train_time:6521ms step_avg:151.65ms step:54/5100 train_loss:5.7221 train_time:6675ms step_avg:151.71ms step:55/5100 train_loss:5.6005 train_time:6828ms step_avg:151.73ms step:56/5100 train_loss:5.9398 train_time:6982ms step_avg:151.77ms step:57/5100 train_loss:5.5927 train_time:7136ms step_avg:151.83ms step:58/5100 train_loss:5.4627 train_time:7289ms step_avg:151.86ms step:59/5100 train_loss:5.6116 train_time:7442ms step_avg:151.88ms step:60/5100 train_loss:5.5761 train_time:7596ms step_avg:151.91ms step:61/5100 train_loss:5.6906 train_time:7749ms step_avg:151.93ms step:62/5100 train_loss:5.4456 train_time:7901ms step_avg:151.95ms step:63/5100 train_loss:5.5463 train_time:8055ms step_avg:151.98ms step:64/5100 train_loss:5.5327 train_time:8208ms step_avg:152.01ms step:65/5100 train_loss:5.2273 train_time:8361ms step_avg:152.02ms step:66/5100 train_loss:5.3407 train_time:8514ms step_avg:152.04ms step:67/5100 train_loss:5.4952 train_time:8667ms step_avg:152.05ms step:68/5100 train_loss:5.3736 train_time:8819ms step_avg:152.05ms step:69/5100 train_loss:5.6261 train_time:8973ms step_avg:152.09ms step:70/5100 train_loss:5.2765 train_time:9126ms step_avg:152.11ms step:71/5100 train_loss:5.2919 train_time:9280ms step_avg:152.14ms step:72/5100 train_loss:5.5050 train_time:9434ms step_avg:152.16ms step:73/5100 train_loss:5.4406 train_time:9587ms step_avg:152.17ms step:74/5100 train_loss:5.3141 train_time:9740ms step_avg:152.18ms step:75/5100 train_loss:5.4444 train_time:9892ms step_avg:152.19ms step:76/5100 train_loss:5.4133 train_time:10046ms step_avg:152.21ms step:77/5100 train_loss:5.3648 train_time:10199ms step_avg:152.22ms step:78/5100 train_loss:5.4578 train_time:10352ms step_avg:152.24ms step:79/5100 train_loss:5.5200 train_time:10506ms step_avg:152.26ms step:80/5100 train_loss:5.3211 train_time:10660ms step_avg:152.28ms step:81/5100 train_loss:5.4260 train_time:10811ms step_avg:152.27ms step:82/5100 train_loss:5.1893 train_time:10965ms step_avg:152.29ms step:83/5100 train_loss:5.3617 train_time:11118ms step_avg:152.30ms step:84/5100 train_loss:5.3141 train_time:11272ms step_avg:152.32ms step:85/5100 train_loss:5.2966 train_time:11424ms step_avg:152.32ms step:86/5100 train_loss:5.1611 train_time:11580ms step_avg:152.37ms step:87/5100 train_loss:5.3763 train_time:11733ms step_avg:152.38ms step:88/5100 train_loss:5.2845 train_time:11885ms step_avg:152.38ms step:89/5100 train_loss:5.3295 train_time:12039ms step_avg:152.40ms step:90/5100 train_loss:5.2873 train_time:12193ms step_avg:152.41ms step:91/5100 train_loss:5.2253 train_time:12345ms step_avg:152.41ms step:92/5100 train_loss:5.2012 train_time:12499ms step_avg:152.42ms step:93/5100 train_loss:5.3525 train_time:12652ms step_avg:152.43ms step:94/5100 train_loss:5.1548 train_time:12806ms step_avg:152.45ms step:95/5100 train_loss:5.1634 train_time:12959ms step_avg:152.46ms step:96/5100 train_loss:5.2021 train_time:13113ms step_avg:152.47ms step:97/5100 train_loss:5.1079 train_time:13266ms step_avg:152.48ms step:98/5100 train_loss:5.1923 train_time:13419ms step_avg:152.49ms step:99/5100 train_loss:5.1158 train_time:13573ms step_avg:152.50ms step:100/5100 train_loss:5.2396 train_time:13726ms step_avg:152.52ms step:101/5100 train_loss:5.2138 train_time:13880ms step_avg:152.53ms step:102/5100 train_loss:5.1164 train_time:14034ms step_avg:152.54ms step:103/5100 train_loss:5.2165 train_time:14187ms step_avg:152.55ms step:104/5100 train_loss:5.1526 train_time:14340ms step_avg:152.56ms step:105/5100 train_loss:5.0218 train_time:14494ms step_avg:152.57ms step:106/5100 train_loss:5.1203 train_time:14648ms step_avg:152.58ms step:107/5100 train_loss:5.3267 train_time:14800ms step_avg:152.58ms step:108/5100 train_loss:5.0854 train_time:14954ms step_avg:152.60ms step:109/5100 train_loss:4.8868 train_time:15108ms step_avg:152.60ms step:110/5100 train_loss:5.0813 train_time:15262ms step_avg:152.62ms step:111/5100 train_loss:5.0444 train_time:15415ms step_avg:152.63ms step:112/5100 train_loss:5.0087 train_time:15568ms step_avg:152.63ms step:113/5100 train_loss:5.1094 train_time:15721ms step_avg:152.63ms step:114/5100 train_loss:5.0430 train_time:15875ms step_avg:152.64ms step:115/5100 train_loss:4.9000 train_time:16028ms step_avg:152.65ms step:116/5100 train_loss:5.0611 train_time:16183ms step_avg:152.67ms step:117/5100 train_loss:4.9731 train_time:16337ms step_avg:152.68ms step:118/5100 train_loss:4.9182 train_time:16490ms step_avg:152.68ms step:119/5100 train_loss:5.0629 train_time:16642ms step_avg:152.68ms step:120/5100 train_loss:5.0233 train_time:16796ms step_avg:152.69ms step:121/5100 train_loss:4.9470 train_time:16951ms step_avg:152.72ms step:122/5100 train_loss:4.8487 train_time:17106ms step_avg:152.73ms step:123/5100 train_loss:4.9770 train_time:17260ms step_avg:152.74ms step:124/5100 train_loss:4.8247 train_time:17414ms step_avg:152.75ms step:125/5100 train_loss:5.1450 train_time:17568ms step_avg:152.76ms step:125/5100 val_loss:4.9669 train_time:17631ms step_avg:153.32ms step:126/5100 train_loss:5.0153 train_time:17724ms step_avg:152.79ms step:127/5100 train_loss:4.9507 train_time:17883ms step_avg:152.85ms step:128/5100 train_loss:5.0129 train_time:18036ms step_avg:152.85ms step:129/5100 train_loss:4.8857 train_time:18188ms step_avg:152.84ms step:130/5100 train_loss:5.1978 train_time:18341ms step_avg:152.84ms step:131/5100 train_loss:4.9433 train_time:18493ms step_avg:152.83ms step:132/5100 train_loss:4.9528 train_time:18645ms step_avg:152.83ms step:133/5100 train_loss:4.9133 train_time:18801ms step_avg:152.85ms step:134/5100 train_loss:4.9434 train_time:18955ms step_avg:152.86ms step:135/5100 train_loss:4.8356 train_time:19108ms step_avg:152.87ms step:136/5100 train_loss:4.9590 train_time:19262ms step_avg:152.87ms step:137/5100 train_loss:4.7379 train_time:19415ms step_avg:152.87ms step:138/5100 train_loss:4.9000 train_time:19566ms step_avg:152.86ms step:139/5100 train_loss:4.8433 train_time:19720ms step_avg:152.87ms step:140/5100 train_loss:4.8878 train_time:19874ms step_avg:152.88ms step:141/5100 train_loss:4.9478 train_time:20027ms step_avg:152.88ms step:142/5100 train_loss:4.8241 train_time:20181ms step_avg:152.89ms step:143/5100 train_loss:4.8760 train_time:20335ms step_avg:152.89ms step:144/5100 train_loss:4.7360 train_time:20489ms step_avg:152.91ms step:145/5100 train_loss:4.8745 train_time:20642ms step_avg:152.90ms step:146/5100 train_loss:4.8265 train_time:20796ms step_avg:152.91ms step:147/5100 train_loss:4.7005 train_time:20948ms step_avg:152.91ms step:148/5100 train_loss:4.8448 train_time:21102ms step_avg:152.91ms step:149/5100 train_loss:4.8398 train_time:21255ms step_avg:152.92ms step:150/5100 train_loss:4.8679 train_time:21409ms step_avg:152.92ms step:151/5100 train_loss:4.9167 train_time:21563ms step_avg:152.93ms step:152/5100 train_loss:4.7985 train_time:21716ms step_avg:152.93ms step:153/5100 train_loss:4.7928 train_time:21869ms step_avg:152.93ms step:154/5100 train_loss:4.8811 train_time:22022ms step_avg:152.93ms step:155/5100 train_loss:4.8371 train_time:22177ms step_avg:152.94ms step:156/5100 train_loss:4.7923 train_time:22329ms step_avg:152.94ms step:157/5100 train_loss:4.8188 train_time:22483ms step_avg:152.94ms step:158/5100 train_loss:4.9365 train_time:22636ms step_avg:152.95ms step:159/5100 train_loss:4.7256 train_time:22789ms step_avg:152.95ms step:160/5100 train_loss:4.8005 train_time:22944ms step_avg:152.96ms step:161/5100 train_loss:4.6270 train_time:23096ms step_avg:152.96ms step:162/5100 train_loss:4.8134 train_time:23249ms step_avg:152.96ms step:163/5100 train_loss:4.8456 train_time:23402ms step_avg:152.95ms step:164/5100 train_loss:4.8269 train_time:23555ms step_avg:152.96ms step:165/5100 train_loss:4.6422 train_time:23709ms step_avg:152.96ms step:166/5100 train_loss:4.7655 train_time:23863ms step_avg:152.97ms step:167/5100 train_loss:4.9010 train_time:24016ms step_avg:152.97ms step:168/5100 train_loss:4.6904 train_time:24170ms step_avg:152.98ms step:169/5100 train_loss:4.7889 train_time:24322ms step_avg:152.97ms step:170/5100 train_loss:4.6299 train_time:24477ms step_avg:152.98ms step:171/5100 train_loss:4.5340 train_time:24630ms step_avg:152.98ms step:172/5100 train_loss:4.6949 train_time:24783ms step_avg:152.98ms step:173/5100 train_loss:4.6711 train_time:24937ms step_avg:152.99ms step:174/5100 train_loss:4.7388 train_time:25090ms step_avg:152.99ms step:175/5100 train_loss:4.8856 train_time:25243ms step_avg:152.99ms step:176/5100 train_loss:4.7307 train_time:25395ms step_avg:152.98ms step:177/5100 train_loss:4.5852 train_time:25549ms step_avg:152.99ms step:178/5100 train_loss:4.5581 train_time:25701ms step_avg:152.98ms step:179/5100 train_loss:4.6316 train_time:25855ms step_avg:152.99ms step:180/5100 train_loss:4.6384 train_time:26009ms step_avg:152.99ms step:181/5100 train_loss:4.6272 train_time:26163ms step_avg:153.00ms step:182/5100 train_loss:4.7598 train_time:26316ms step_avg:153.00ms step:183/5100 train_loss:4.6281 train_time:26469ms step_avg:153.00ms step:184/5100 train_loss:4.5748 train_time:26621ms step_avg:153.00ms step:185/5100 train_loss:4.5907 train_time:26775ms step_avg:153.00ms step:186/5100 train_loss:4.7187 train_time:26928ms step_avg:153.00ms step:187/5100 train_loss:4.6292 train_time:27082ms step_avg:153.01ms step:188/5100 train_loss:4.8232 train_time:27235ms step_avg:153.01ms step:189/5100 train_loss:4.6354 train_time:27487ms step_avg:153.56ms step:190/5100 train_loss:4.5647 train_time:27775ms step_avg:154.31ms step:191/5100 train_loss:4.7053 train_time:27925ms step_avg:154.28ms step:192/5100 train_loss:4.5519 train_time:28077ms step_avg:154.27ms step:193/5100 train_loss:4.4761 train_time:28230ms step_avg:154.26ms step:194/5100 train_loss:4.6939 train_time:28383ms step_avg:154.25ms step:195/5100 train_loss:4.6267 train_time:28535ms step_avg:154.24ms step:196/5100 train_loss:4.8096 train_time:28688ms step_avg:154.23ms step:197/5100 train_loss:4.6832 train_time:28841ms step_avg:154.23ms step:198/5100 train_loss:4.5200 train_time:28996ms step_avg:154.24ms step:199/5100 train_loss:4.5972 train_time:29150ms step_avg:154.23ms step:200/5100 train_loss:4.4624 train_time:29302ms step_avg:154.22ms step:201/5100 train_loss:4.5590 train_time:29455ms step_avg:154.22ms step:202/5100 train_loss:4.4535 train_time:29608ms step_avg:154.21ms step:203/5100 train_loss:4.7074 train_time:29761ms step_avg:154.20ms step:204/5100 train_loss:4.5699 train_time:29915ms step_avg:154.20ms step:205/5100 train_loss:4.5984 train_time:30069ms step_avg:154.20ms step:206/5100 train_loss:4.7220 train_time:30222ms step_avg:154.19ms step:207/5100 train_loss:4.3753 train_time:30376ms step_avg:154.19ms step:208/5100 train_loss:4.5346 train_time:30529ms step_avg:154.19ms step:209/5100 train_loss:4.5112 train_time:30682ms step_avg:154.18ms step:210/5100 train_loss:4.6699 train_time:30835ms step_avg:154.18ms step:211/5100 train_loss:4.5814 train_time:30988ms step_avg:154.17ms step:212/5100 train_loss:4.4721 train_time:31140ms step_avg:154.16ms step:213/5100 train_loss:4.5835 train_time:31294ms step_avg:154.16ms step:214/5100 train_loss:4.4429 train_time:31448ms step_avg:154.16ms step:215/5100 train_loss:4.5102 train_time:31600ms step_avg:154.15ms step:216/5100 train_loss:4.3772 train_time:31754ms step_avg:154.14ms step:217/5100 train_loss:4.4848 train_time:31906ms step_avg:154.14ms step:218/5100 train_loss:4.4521 train_time:32060ms step_avg:154.14ms step:219/5100 train_loss:4.4727 train_time:32214ms step_avg:154.13ms step:220/5100 train_loss:4.4671 train_time:32367ms step_avg:154.13ms step:221/5100 train_loss:4.5000 train_time:32520ms step_avg:154.12ms step:222/5100 train_loss:4.5143 train_time:32673ms step_avg:154.12ms step:223/5100 train_loss:4.4404 train_time:32826ms step_avg:154.11ms step:224/5100 train_loss:4.4540 train_time:32980ms step_avg:154.11ms step:225/5100 train_loss:4.6417 train_time:33134ms step_avg:154.11ms step:226/5100 train_loss:4.3033 train_time:33287ms step_avg:154.11ms step:227/5100 train_loss:4.3602 train_time:33442ms step_avg:154.11ms step:228/5100 train_loss:4.3672 train_time:33595ms step_avg:154.11ms step:229/5100 train_loss:4.5179 train_time:33748ms step_avg:154.10ms step:230/5100 train_loss:4.3141 train_time:33901ms step_avg:154.10ms step:231/5100 train_loss:4.4583 train_time:34055ms step_avg:154.10ms step:232/5100 train_loss:4.3263 train_time:34208ms step_avg:154.09ms step:233/5100 train_loss:4.3273 train_time:34362ms step_avg:154.09ms step:234/5100 train_loss:4.4929 train_time:34515ms step_avg:154.09ms step:235/5100 train_loss:4.3778 train_time:34668ms step_avg:154.08ms step:236/5100 train_loss:4.2735 train_time:34820ms step_avg:154.07ms step:237/5100 train_loss:4.4817 train_time:34974ms step_avg:154.07ms step:238/5100 train_loss:4.4401 train_time:35127ms step_avg:154.06ms step:239/5100 train_loss:4.3106 train_time:35280ms step_avg:154.06ms step:240/5100 train_loss:4.4647 train_time:35434ms step_avg:154.06ms step:241/5100 train_loss:4.4568 train_time:35588ms step_avg:154.06ms step:242/5100 train_loss:4.3369 train_time:35742ms step_avg:154.06ms step:243/5100 train_loss:4.5240 train_time:35896ms step_avg:154.06ms step:244/5100 train_loss:4.3587 train_time:36048ms step_avg:154.05ms step:245/5100 train_loss:4.3904 train_time:36201ms step_avg:154.05ms step:246/5100 train_loss:4.4742 train_time:36354ms step_avg:154.04ms step:247/5100 train_loss:4.4093 train_time:36508ms step_avg:154.04ms step:248/5100 train_loss:4.3459 train_time:36660ms step_avg:154.04ms step:249/5100 train_loss:4.4756 train_time:36814ms step_avg:154.03ms step:250/5100 train_loss:4.2456 train_time:36967ms step_avg:154.03ms step:250/5100 val_loss:4.3420 train_time:37031ms step_avg:154.30ms step:251/5100 train_loss:4.2975 train_time:37122ms step_avg:154.03ms step:252/5100 train_loss:4.4059 train_time:37282ms step_avg:154.06ms step:253/5100 train_loss:4.4499 train_time:37436ms step_avg:154.06ms step:254/5100 train_loss:4.2720 train_time:37587ms step_avg:154.04ms step:255/5100 train_loss:4.2149 train_time:37740ms step_avg:154.04ms step:256/5100 train_loss:4.3911 train_time:37892ms step_avg:154.03ms step:257/5100 train_loss:4.3103 train_time:38045ms step_avg:154.03ms step:258/5100 train_loss:4.3216 train_time:38199ms step_avg:154.03ms step:259/5100 train_loss:4.2828 train_time:38353ms step_avg:154.03ms step:260/5100 train_loss:4.3226 train_time:38506ms step_avg:154.03ms step:261/5100 train_loss:4.3673 train_time:38659ms step_avg:154.02ms step:262/5100 train_loss:4.3305 train_time:38811ms step_avg:154.01ms step:263/5100 train_loss:4.2978 train_time:38964ms step_avg:154.01ms step:264/5100 train_loss:4.2126 train_time:39117ms step_avg:154.00ms step:265/5100 train_loss:4.2962 train_time:39271ms step_avg:154.00ms step:266/5100 train_loss:4.1531 train_time:39424ms step_avg:154.00ms step:267/5100 train_loss:4.2122 train_time:39577ms step_avg:154.00ms step:268/5100 train_loss:4.2295 train_time:39731ms step_avg:153.99ms step:269/5100 train_loss:4.2357 train_time:39883ms step_avg:153.99ms step:270/5100 train_loss:4.1582 train_time:40036ms step_avg:153.98ms step:271/5100 train_loss:4.3940 train_time:40188ms step_avg:153.98ms step:272/5100 train_loss:4.2917 train_time:40341ms step_avg:153.97ms step:273/5100 train_loss:4.1918 train_time:40495ms step_avg:153.97ms step:274/5100 train_loss:4.2445 train_time:40649ms step_avg:153.97ms step:275/5100 train_loss:4.3226 train_time:40803ms step_avg:153.97ms step:276/5100 train_loss:4.3393 train_time:40956ms step_avg:153.97ms step:277/5100 train_loss:4.5242 train_time:41109ms step_avg:153.97ms step:278/5100 train_loss:4.3126 train_time:41261ms step_avg:153.96ms step:279/5100 train_loss:4.3895 train_time:41414ms step_avg:153.96ms step:280/5100 train_loss:4.2820 train_time:41567ms step_avg:153.95ms step:281/5100 train_loss:4.3993 train_time:41719ms step_avg:153.95ms step:282/5100 train_loss:4.2367 train_time:41874ms step_avg:153.95ms step:283/5100 train_loss:4.2551 train_time:42027ms step_avg:153.95ms step:284/5100 train_loss:4.1814 train_time:42180ms step_avg:153.94ms step:285/5100 train_loss:4.3337 train_time:42334ms step_avg:153.94ms step:286/5100 train_loss:4.3377 train_time:42486ms step_avg:153.94ms step:287/5100 train_loss:4.3617 train_time:42638ms step_avg:153.93ms step:288/5100 train_loss:4.1948 train_time:42791ms step_avg:153.93ms step:289/5100 train_loss:4.2898 train_time:42944ms step_avg:153.92ms step:290/5100 train_loss:4.1467 train_time:43097ms step_avg:153.92ms step:291/5100 train_loss:4.1432 train_time:43251ms step_avg:153.92ms step:292/5100 train_loss:4.2244 train_time:43404ms step_avg:153.92ms step:293/5100 train_loss:4.1343 train_time:43557ms step_avg:153.91ms step:294/5100 train_loss:4.1883 train_time:43711ms step_avg:153.91ms step:295/5100 train_loss:4.2213 train_time:43863ms step_avg:153.91ms step:296/5100 train_loss:4.1006 train_time:44015ms step_avg:153.90ms step:297/5100 train_loss:4.1221 train_time:44169ms step_avg:153.90ms step:298/5100 train_loss:4.1272 train_time:44321ms step_avg:153.89ms step:299/5100 train_loss:4.2334 train_time:44475ms step_avg:153.89ms step:300/5100 train_loss:4.0936 train_time:44629ms step_avg:153.89ms step:301/5100 train_loss:4.2395 train_time:44782ms step_avg:153.89ms step:302/5100 train_loss:4.2444 train_time:44936ms step_avg:153.89ms step:303/5100 train_loss:4.1797 train_time:45088ms step_avg:153.88ms step:304/5100 train_loss:4.2444 train_time:45241ms step_avg:153.88ms step:305/5100 train_loss:4.2217 train_time:45394ms step_avg:153.88ms step:306/5100 train_loss:4.6954 train_time:45547ms step_avg:153.88ms step:307/5100 train_loss:4.1937 train_time:45699ms step_avg:153.87ms step:308/5100 train_loss:4.0995 train_time:45853ms step_avg:153.87ms step:309/5100 train_loss:4.2599 train_time:46005ms step_avg:153.86ms step:310/5100 train_loss:4.1046 train_time:46158ms step_avg:153.86ms step:311/5100 train_loss:4.3338 train_time:46311ms step_avg:153.86ms step:312/5100 train_loss:4.1882 train_time:46465ms step_avg:153.86ms step:313/5100 train_loss:4.1218 train_time:46616ms step_avg:153.85ms step:314/5100 train_loss:4.2396 train_time:46771ms step_avg:153.85ms step:315/5100 train_loss:4.3367 train_time:46923ms step_avg:153.85ms step:316/5100 train_loss:4.2161 train_time:47077ms step_avg:153.85ms step:317/5100 train_loss:4.0477 train_time:47231ms step_avg:153.85ms step:318/5100 train_loss:4.1242 train_time:47384ms step_avg:153.84ms step:319/5100 train_loss:4.1575 train_time:47536ms step_avg:153.84ms step:320/5100 train_loss:4.1365 train_time:47689ms step_avg:153.84ms step:321/5100 train_loss:4.2392 train_time:47842ms step_avg:153.83ms step:322/5100 train_loss:4.1963 train_time:47995ms step_avg:153.83ms step:323/5100 train_loss:4.1661 train_time:48149ms step_avg:153.83ms step:324/5100 train_loss:4.2576 train_time:48303ms step_avg:153.83ms step:325/5100 train_loss:4.2126 train_time:48456ms step_avg:153.83ms step:326/5100 train_loss:4.2788 train_time:48610ms step_avg:153.83ms step:327/5100 train_loss:4.1298 train_time:48762ms step_avg:153.82ms step:328/5100 train_loss:4.6276 train_time:48915ms step_avg:153.82ms step:329/5100 train_loss:4.3128 train_time:49068ms step_avg:153.82ms step:330/5100 train_loss:4.0488 train_time:49220ms step_avg:153.81ms step:331/5100 train_loss:4.0034 train_time:49375ms step_avg:153.82ms step:332/5100 train_loss:4.2231 train_time:49528ms step_avg:153.81ms step:333/5100 train_loss:4.1383 train_time:49681ms step_avg:153.81ms step:334/5100 train_loss:4.1230 train_time:49835ms step_avg:153.81ms step:335/5100 train_loss:4.0844 train_time:49986ms step_avg:153.80ms step:336/5100 train_loss:4.2570 train_time:50138ms step_avg:153.80ms step:337/5100 train_loss:4.1924 train_time:50291ms step_avg:153.79ms step:338/5100 train_loss:4.6707 train_time:50444ms step_avg:153.79ms step:339/5100 train_loss:4.1788 train_time:50597ms step_avg:153.79ms step:340/5100 train_loss:4.1239 train_time:50751ms step_avg:153.79ms step:341/5100 train_loss:4.1599 train_time:50904ms step_avg:153.79ms step:342/5100 train_loss:4.0778 train_time:51057ms step_avg:153.79ms step:343/5100 train_loss:4.0494 train_time:51209ms step_avg:153.78ms step:344/5100 train_loss:4.0997 train_time:51361ms step_avg:153.78ms step:345/5100 train_loss:4.2330 train_time:51514ms step_avg:153.77ms step:346/5100 train_loss:4.0750 train_time:51666ms step_avg:153.77ms step:347/5100 train_loss:4.0057 train_time:51818ms step_avg:153.76ms step:348/5100 train_loss:4.0549 train_time:51973ms step_avg:153.77ms step:349/5100 train_loss:4.0957 train_time:52125ms step_avg:153.76ms step:350/5100 train_loss:4.0547 train_time:52278ms step_avg:153.76ms step:351/5100 train_loss:3.7710 train_time:52431ms step_avg:153.76ms step:352/5100 train_loss:4.0464 train_time:52584ms step_avg:153.75ms step:353/5100 train_loss:4.3910 train_time:52736ms step_avg:153.75ms step:354/5100 train_loss:3.8918 train_time:52889ms step_avg:153.75ms step:355/5100 train_loss:4.1504 train_time:53042ms step_avg:153.75ms step:356/5100 train_loss:4.0241 train_time:53195ms step_avg:153.74ms step:357/5100 train_loss:4.1256 train_time:53348ms step_avg:153.74ms step:358/5100 train_loss:4.0636 train_time:53502ms step_avg:153.74ms step:359/5100 train_loss:4.0779 train_time:53655ms step_avg:153.74ms step:360/5100 train_loss:4.1309 train_time:53808ms step_avg:153.74ms step:361/5100 train_loss:3.6846 train_time:53960ms step_avg:153.73ms step:362/5100 train_loss:4.2538 train_time:54114ms step_avg:153.73ms step:363/5100 train_loss:4.1440 train_time:54266ms step_avg:153.73ms step:364/5100 train_loss:4.0703 train_time:54418ms step_avg:153.72ms step:365/5100 train_loss:3.9769 train_time:54572ms step_avg:153.72ms step:366/5100 train_loss:4.1383 train_time:54725ms step_avg:153.72ms step:367/5100 train_loss:4.1000 train_time:54879ms step_avg:153.72ms step:368/5100 train_loss:4.0840 train_time:55033ms step_avg:153.72ms step:369/5100 train_loss:4.0698 train_time:55185ms step_avg:153.72ms step:370/5100 train_loss:3.9639 train_time:55337ms step_avg:153.71ms step:371/5100 train_loss:4.1196 train_time:55490ms step_avg:153.71ms step:372/5100 train_loss:3.9940 train_time:55643ms step_avg:153.71ms step:373/5100 train_loss:3.9145 train_time:55795ms step_avg:153.70ms step:374/5100 train_loss:4.1327 train_time:55948ms step_avg:153.70ms step:375/5100 train_loss:4.0621 train_time:56101ms step_avg:153.70ms step:375/5100 val_loss:4.0605 train_time:56167ms step_avg:153.88ms step:376/5100 train_loss:4.0358 train_time:56261ms step_avg:153.72ms step:377/5100 train_loss:4.0955 train_time:56417ms step_avg:153.72ms step:378/5100 train_loss:4.0143 train_time:56664ms step_avg:153.98ms step:379/5100 train_loss:4.0625 train_time:56824ms step_avg:154.00ms step:380/5100 train_loss:4.1093 train_time:57099ms step_avg:154.32ms step:381/5100 train_loss:4.1681 train_time:57249ms step_avg:154.31ms step:382/5100 train_loss:4.0795 train_time:57401ms step_avg:154.30ms step:383/5100 train_loss:4.0486 train_time:57553ms step_avg:154.30ms step:384/5100 train_loss:4.0117 train_time:57704ms step_avg:154.29ms step:385/5100 train_loss:4.0987 train_time:57857ms step_avg:154.28ms step:386/5100 train_loss:4.0053 train_time:58009ms step_avg:154.28ms step:387/5100 train_loss:4.1190 train_time:58163ms step_avg:154.28ms step:388/5100 train_loss:4.3082 train_time:58317ms step_avg:154.28ms step:389/5100 train_loss:4.0209 train_time:58469ms step_avg:154.27ms step:390/5100 train_loss:4.0149 train_time:58621ms step_avg:154.26ms step:391/5100 train_loss:4.1150 train_time:58773ms step_avg:154.26ms step:392/5100 train_loss:4.0298 train_time:58924ms step_avg:154.25ms step:393/5100 train_loss:4.1390 train_time:59077ms step_avg:154.25ms step:394/5100 train_loss:3.9758 train_time:59230ms step_avg:154.25ms step:395/5100 train_loss:4.1108 train_time:59384ms step_avg:154.24ms step:396/5100 train_loss:3.8561 train_time:59537ms step_avg:154.24ms step:397/5100 train_loss:4.0557 train_time:59689ms step_avg:154.24ms step:398/5100 train_loss:4.1114 train_time:59842ms step_avg:154.23ms step:399/5100 train_loss:4.1186 train_time:59995ms step_avg:154.23ms step:400/5100 train_loss:4.0074 train_time:60147ms step_avg:154.22ms step:401/5100 train_loss:4.0659 train_time:60300ms step_avg:154.22ms step:402/5100 train_loss:4.1272 train_time:60452ms step_avg:154.21ms step:403/5100 train_loss:4.0661 train_time:60604ms step_avg:154.21ms step:404/5100 train_loss:4.1703 train_time:60757ms step_avg:154.20ms step:405/5100 train_loss:3.9245 train_time:60910ms step_avg:154.20ms step:406/5100 train_loss:4.0160 train_time:61062ms step_avg:154.20ms step:407/5100 train_loss:4.3058 train_time:61214ms step_avg:154.19ms step:408/5100 train_loss:4.0165 train_time:61368ms step_avg:154.19ms step:409/5100 train_loss:4.0389 train_time:61521ms step_avg:154.19ms step:410/5100 train_loss:4.0845 train_time:61674ms step_avg:154.18ms step:411/5100 train_loss:3.9669 train_time:61825ms step_avg:154.18ms step:412/5100 train_loss:3.9798 train_time:61980ms step_avg:154.18ms step:413/5100 train_loss:4.4027 train_time:62131ms step_avg:154.17ms step:414/5100 train_loss:3.9010 train_time:62284ms step_avg:154.17ms step:415/5100 train_loss:4.2339 train_time:62437ms step_avg:154.17ms step:416/5100 train_loss:3.9805 train_time:62590ms step_avg:154.16ms step:417/5100 train_loss:3.9805 train_time:62743ms step_avg:154.16ms step:418/5100 train_loss:4.1775 train_time:62896ms step_avg:154.16ms step:419/5100 train_loss:3.9101 train_time:63048ms step_avg:154.15ms step:420/5100 train_loss:4.0172 train_time:63201ms step_avg:154.15ms step:421/5100 train_loss:3.9475 train_time:63353ms step_avg:154.14ms step:422/5100 train_loss:3.8606 train_time:63504ms step_avg:154.14ms step:423/5100 train_loss:4.0012 train_time:63657ms step_avg:154.13ms step:424/5100 train_loss:4.0865 train_time:63810ms step_avg:154.13ms step:425/5100 train_loss:3.8474 train_time:63964ms step_avg:154.13ms step:426/5100 train_loss:4.0296 train_time:64117ms step_avg:154.13ms step:427/5100 train_loss:3.9098 train_time:64270ms step_avg:154.12ms step:428/5100 train_loss:4.1203 train_time:64422ms step_avg:154.12ms step:429/5100 train_loss:4.0361 train_time:64574ms step_avg:154.11ms step:430/5100 train_loss:3.9692 train_time:64727ms step_avg:154.11ms step:431/5100 train_loss:3.9398 train_time:64881ms step_avg:154.11ms step:432/5100 train_loss:3.8442 train_time:65033ms step_avg:154.11ms step:433/5100 train_loss:3.9791 train_time:65185ms step_avg:154.10ms step:434/5100 train_loss:4.0415 train_time:65339ms step_avg:154.10ms step:435/5100 train_loss:3.9856 train_time:65491ms step_avg:154.10ms step:436/5100 train_loss:4.0374 train_time:65643ms step_avg:154.09ms step:437/5100 train_loss:4.0445 train_time:65795ms step_avg:154.09ms step:438/5100 train_loss:3.9223 train_time:65948ms step_avg:154.08ms step:439/5100 train_loss:3.9369 train_time:66100ms step_avg:154.08ms step:440/5100 train_loss:3.9161 train_time:66253ms step_avg:154.08ms step:441/5100 train_loss:4.0959 train_time:66405ms step_avg:154.07ms step:442/5100 train_loss:3.9856 train_time:66558ms step_avg:154.07ms step:443/5100 train_loss:3.9691 train_time:66711ms step_avg:154.07ms step:444/5100 train_loss:3.8572 train_time:66863ms step_avg:154.06ms step:445/5100 train_loss:4.1291 train_time:67015ms step_avg:154.06ms step:446/5100 train_loss:4.0580 train_time:67167ms step_avg:154.05ms step:447/5100 train_loss:4.0527 train_time:67320ms step_avg:154.05ms step:448/5100 train_loss:3.9666 train_time:67472ms step_avg:154.05ms step:449/5100 train_loss:4.0696 train_time:67624ms step_avg:154.04ms step:450/5100 train_loss:3.8919 train_time:67777ms step_avg:154.04ms step:451/5100 train_loss:3.9378 train_time:67930ms step_avg:154.04ms step:452/5100 train_loss:3.7968 train_time:68083ms step_avg:154.03ms step:453/5100 train_loss:3.9182 train_time:68235ms step_avg:154.03ms step:454/5100 train_loss:3.8897 train_time:68387ms step_avg:154.03ms step:455/5100 train_loss:3.8440 train_time:68541ms step_avg:154.03ms step:456/5100 train_loss:4.0581 train_time:68693ms step_avg:154.02ms step:457/5100 train_loss:3.9337 train_time:68845ms step_avg:154.02ms step:458/5100 train_loss:4.0065 train_time:68997ms step_avg:154.01ms step:459/5100 train_loss:4.0492 train_time:69150ms step_avg:154.01ms step:460/5100 train_loss:3.8527 train_time:69302ms step_avg:154.00ms step:461/5100 train_loss:4.0139 train_time:69454ms step_avg:154.00ms step:462/5100 train_loss:3.9158 train_time:69607ms step_avg:154.00ms step:463/5100 train_loss:3.9277 train_time:69760ms step_avg:154.00ms step:464/5100 train_loss:3.9938 train_time:69913ms step_avg:153.99ms step:465/5100 train_loss:3.9279 train_time:70066ms step_avg:153.99ms step:466/5100 train_loss:3.9349 train_time:70219ms step_avg:153.99ms step:467/5100 train_loss:4.0280 train_time:70371ms step_avg:153.98ms step:468/5100 train_loss:4.0424 train_time:70522ms step_avg:153.98ms step:469/5100 train_loss:4.0130 train_time:70675ms step_avg:153.98ms step:470/5100 train_loss:3.9030 train_time:70828ms step_avg:153.97ms step:471/5100 train_loss:3.9867 train_time:70982ms step_avg:153.97ms step:472/5100 train_loss:4.0408 train_time:71135ms step_avg:153.97ms step:473/5100 train_loss:3.9811 train_time:71288ms step_avg:153.97ms step:474/5100 train_loss:3.9444 train_time:71441ms step_avg:153.97ms step:475/5100 train_loss:3.7950 train_time:71592ms step_avg:153.96ms step:476/5100 train_loss:4.2213 train_time:71745ms step_avg:153.96ms step:477/5100 train_loss:3.9856 train_time:71897ms step_avg:153.95ms step:478/5100 train_loss:3.7981 train_time:72050ms step_avg:153.95ms step:479/5100 train_loss:4.0325 train_time:72202ms step_avg:153.95ms step:480/5100 train_loss:3.9844 train_time:72355ms step_avg:153.95ms step:481/5100 train_loss:4.1200 train_time:72508ms step_avg:153.94ms step:482/5100 train_loss:3.9348 train_time:72660ms step_avg:153.94ms step:483/5100 train_loss:3.7371 train_time:72812ms step_avg:153.94ms step:484/5100 train_loss:4.0251 train_time:72965ms step_avg:153.93ms step:485/5100 train_loss:3.8776 train_time:73117ms step_avg:153.93ms step:486/5100 train_loss:3.8893 train_time:73269ms step_avg:153.93ms step:487/5100 train_loss:3.8172 train_time:73421ms step_avg:153.92ms step:488/5100 train_loss:3.8799 train_time:73574ms step_avg:153.92ms step:489/5100 train_loss:4.0861 train_time:73727ms step_avg:153.92ms step:490/5100 train_loss:3.9214 train_time:73880ms step_avg:153.92ms step:491/5100 train_loss:3.8161 train_time:74032ms step_avg:153.91ms step:492/5100 train_loss:3.8331 train_time:74184ms step_avg:153.91ms step:493/5100 train_loss:3.9507 train_time:74337ms step_avg:153.91ms step:494/5100 train_loss:3.7949 train_time:74489ms step_avg:153.90ms step:495/5100 train_loss:3.9257 train_time:74642ms step_avg:153.90ms step:496/5100 train_loss:3.8603 train_time:74794ms step_avg:153.90ms step:497/5100 train_loss:3.7517 train_time:74947ms step_avg:153.89ms step:498/5100 train_loss:3.9420 train_time:75100ms step_avg:153.89ms step:499/5100 train_loss:4.0135 train_time:75253ms step_avg:153.89ms step:500/5100 train_loss:4.0457 train_time:75405ms step_avg:153.89ms step:500/5100 val_loss:3.9223 train_time:75469ms step_avg:154.02ms step:501/5100 train_loss:3.9545 train_time:75563ms step_avg:153.90ms step:502/5100 train_loss:4.0124 train_time:75719ms step_avg:153.90ms step:503/5100 train_loss:3.9554 train_time:75872ms step_avg:153.90ms step:504/5100 train_loss:3.9958 train_time:76024ms step_avg:153.89ms step:505/5100 train_loss:3.9469 train_time:76175ms step_avg:153.89ms step:506/5100 train_loss:4.0246 train_time:76326ms step_avg:153.88ms step:507/5100 train_loss:3.8496 train_time:76478ms step_avg:153.88ms step:508/5100 train_loss:3.9744 train_time:76631ms step_avg:153.88ms step:509/5100 train_loss:4.0491 train_time:76786ms step_avg:153.88ms step:510/5100 train_loss:3.9922 train_time:76939ms step_avg:153.88ms step:511/5100 train_loss:3.7937 train_time:77092ms step_avg:153.88ms step:512/5100 train_loss:3.9895 train_time:77244ms step_avg:153.87ms step:513/5100 train_loss:3.9329 train_time:77396ms step_avg:153.87ms step:514/5100 train_loss:3.8939 train_time:77548ms step_avg:153.86ms step:515/5100 train_loss:3.9840 train_time:77700ms step_avg:153.86ms step:516/5100 train_loss:3.9598 train_time:77853ms step_avg:153.86ms step:517/5100 train_loss:4.2963 train_time:78006ms step_avg:153.86ms step:518/5100 train_loss:3.8910 train_time:78158ms step_avg:153.85ms step:519/5100 train_loss:3.9963 train_time:78311ms step_avg:153.85ms step:520/5100 train_loss:3.8985 train_time:78463ms step_avg:153.85ms step:521/5100 train_loss:3.8989 train_time:78615ms step_avg:153.85ms step:522/5100 train_loss:3.8496 train_time:78767ms step_avg:153.84ms step:523/5100 train_loss:3.8650 train_time:78921ms step_avg:153.84ms step:524/5100 train_loss:4.5024 train_time:79073ms step_avg:153.84ms step:525/5100 train_loss:3.9569 train_time:79225ms step_avg:153.83ms step:526/5100 train_loss:3.8964 train_time:79377ms step_avg:153.83ms step:527/5100 train_loss:3.9043 train_time:79529ms step_avg:153.83ms step:528/5100 train_loss:3.8614 train_time:79682ms step_avg:153.83ms step:529/5100 train_loss:3.8385 train_time:79835ms step_avg:153.82ms step:530/5100 train_loss:4.0564 train_time:79987ms step_avg:153.82ms step:531/5100 train_loss:3.8500 train_time:80140ms step_avg:153.82ms step:532/5100 train_loss:4.1298 train_time:80293ms step_avg:153.82ms step:533/5100 train_loss:3.9445 train_time:80444ms step_avg:153.81ms step:534/5100 train_loss:3.8711 train_time:80597ms step_avg:153.81ms step:535/5100 train_loss:3.8925 train_time:80750ms step_avg:153.81ms step:536/5100 train_loss:3.8238 train_time:80903ms step_avg:153.81ms step:537/5100 train_loss:3.9536 train_time:81055ms step_avg:153.80ms step:538/5100 train_loss:3.9431 train_time:81207ms step_avg:153.80ms step:539/5100 train_loss:3.8394 train_time:81359ms step_avg:153.80ms step:540/5100 train_loss:4.3366 train_time:81512ms step_avg:153.80ms step:541/5100 train_loss:3.8852 train_time:81664ms step_avg:153.79ms step:542/5100 train_loss:3.9939 train_time:81817ms step_avg:153.79ms step:543/5100 train_loss:3.8149 train_time:81969ms step_avg:153.79ms step:544/5100 train_loss:3.7955 train_time:82122ms step_avg:153.79ms step:545/5100 train_loss:3.8767 train_time:82274ms step_avg:153.78ms step:546/5100 train_loss:3.8051 train_time:82426ms step_avg:153.78ms step:547/5100 train_loss:3.8477 train_time:82579ms step_avg:153.78ms step:548/5100 train_loss:3.8688 train_time:82731ms step_avg:153.78ms step:549/5100 train_loss:3.8400 train_time:82884ms step_avg:153.77ms step:550/5100 train_loss:3.9326 train_time:83036ms step_avg:153.77ms step:551/5100 train_loss:3.8177 train_time:83188ms step_avg:153.77ms step:552/5100 train_loss:3.8347 train_time:83340ms step_avg:153.76ms step:553/5100 train_loss:4.1639 train_time:83493ms step_avg:153.76ms step:554/5100 train_loss:3.9584 train_time:83646ms step_avg:153.76ms step:555/5100 train_loss:3.9208 train_time:83798ms step_avg:153.76ms step:556/5100 train_loss:3.8653 train_time:83952ms step_avg:153.76ms step:557/5100 train_loss:3.9042 train_time:84104ms step_avg:153.75ms step:558/5100 train_loss:3.5674 train_time:84256ms step_avg:153.75ms step:559/5100 train_loss:3.8168 train_time:84407ms step_avg:153.75ms step:560/5100 train_loss:3.8671 train_time:84560ms step_avg:153.74ms step:561/5100 train_loss:3.9088 train_time:84712ms step_avg:153.74ms step:562/5100 train_loss:3.8191 train_time:84864ms step_avg:153.74ms step:563/5100 train_loss:3.7601 train_time:85018ms step_avg:153.74ms step:564/5100 train_loss:3.9682 train_time:85170ms step_avg:153.74ms step:565/5100 train_loss:3.7806 train_time:85323ms step_avg:153.73ms step:566/5100 train_loss:3.9003 train_time:85474ms step_avg:153.73ms step:567/5100 train_loss:3.8406 train_time:85720ms step_avg:153.90ms step:568/5100 train_loss:3.7959 train_time:85882ms step_avg:153.91ms step:569/5100 train_loss:3.8915 train_time:86034ms step_avg:153.91ms step:570/5100 train_loss:3.8609 train_time:86308ms step_avg:154.12ms step:571/5100 train_loss:3.8902 train_time:86459ms step_avg:154.12ms step:572/5100 train_loss:3.9767 train_time:86611ms step_avg:154.11ms step:573/5100 train_loss:3.9273 train_time:86764ms step_avg:154.11ms step:574/5100 train_loss:3.9301 train_time:86915ms step_avg:154.10ms step:575/5100 train_loss:3.9820 train_time:87069ms step_avg:154.10ms step:576/5100 train_loss:3.9391 train_time:87221ms step_avg:154.10ms step:577/5100 train_loss:3.9610 train_time:87375ms step_avg:154.10ms step:578/5100 train_loss:3.8985 train_time:87528ms step_avg:154.10ms step:579/5100 train_loss:3.8805 train_time:87681ms step_avg:154.10ms step:580/5100 train_loss:3.8670 train_time:87833ms step_avg:154.09ms step:581/5100 train_loss:3.8086 train_time:87988ms step_avg:154.09ms step:582/5100 train_loss:3.8361 train_time:88143ms step_avg:154.10ms step:583/5100 train_loss:4.0641 train_time:88295ms step_avg:154.09ms step:584/5100 train_loss:3.8361 train_time:88447ms step_avg:154.09ms step:585/5100 train_loss:3.7947 train_time:88599ms step_avg:154.09ms step:586/5100 train_loss:3.9901 train_time:88753ms step_avg:154.08ms step:587/5100 train_loss:3.7377 train_time:88905ms step_avg:154.08ms step:588/5100 train_loss:3.8753 train_time:89059ms step_avg:154.08ms step:589/5100 train_loss:3.8558 train_time:89212ms step_avg:154.08ms step:590/5100 train_loss:4.2102 train_time:89366ms step_avg:154.08ms step:591/5100 train_loss:3.9851 train_time:89519ms step_avg:154.08ms step:592/5100 train_loss:3.7246 train_time:89671ms step_avg:154.07ms step:593/5100 train_loss:3.7434 train_time:89824ms step_avg:154.07ms step:594/5100 train_loss:3.7318 train_time:89976ms step_avg:154.07ms step:595/5100 train_loss:3.7749 train_time:90128ms step_avg:154.06ms step:596/5100 train_loss:4.1295 train_time:90282ms step_avg:154.07ms step:597/5100 train_loss:3.8534 train_time:90435ms step_avg:154.06ms step:598/5100 train_loss:3.7917 train_time:90587ms step_avg:154.06ms step:599/5100 train_loss:3.8684 train_time:90741ms step_avg:154.06ms step:600/5100 train_loss:3.6855 train_time:90895ms step_avg:154.06ms step:601/5100 train_loss:3.8052 train_time:91047ms step_avg:154.06ms step:602/5100 train_loss:3.8401 train_time:91199ms step_avg:154.05ms step:603/5100 train_loss:3.8581 train_time:91352ms step_avg:154.05ms step:604/5100 train_loss:3.9854 train_time:91505ms step_avg:154.05ms step:605/5100 train_loss:3.8414 train_time:91658ms step_avg:154.05ms step:606/5100 train_loss:3.8255 train_time:91810ms step_avg:154.04ms step:607/5100 train_loss:3.7750 train_time:91963ms step_avg:154.04ms step:608/5100 train_loss:4.0208 train_time:92116ms step_avg:154.04ms step:609/5100 train_loss:3.8582 train_time:92267ms step_avg:154.03ms step:610/5100 train_loss:3.8258 train_time:92422ms step_avg:154.04ms step:611/5100 train_loss:3.9201 train_time:92574ms step_avg:154.03ms step:612/5100 train_loss:3.8263 train_time:92726ms step_avg:154.03ms step:613/5100 train_loss:3.8066 train_time:92879ms step_avg:154.03ms step:614/5100 train_loss:3.9705 train_time:93033ms step_avg:154.03ms step:615/5100 train_loss:3.9331 train_time:93186ms step_avg:154.03ms step:616/5100 train_loss:3.8954 train_time:93339ms step_avg:154.02ms step:617/5100 train_loss:3.8259 train_time:93492ms step_avg:154.02ms step:618/5100 train_loss:3.7808 train_time:93645ms step_avg:154.02ms step:619/5100 train_loss:3.8829 train_time:93796ms step_avg:154.02ms step:620/5100 train_loss:3.7824 train_time:93948ms step_avg:154.01ms step:621/5100 train_loss:3.7950 train_time:94100ms step_avg:154.01ms step:622/5100 train_loss:4.1115 train_time:94254ms step_avg:154.01ms step:623/5100 train_loss:3.7989 train_time:94405ms step_avg:154.01ms step:624/5100 train_loss:3.8281 train_time:94558ms step_avg:154.00ms step:625/5100 train_loss:3.8974 train_time:94709ms step_avg:154.00ms step:625/5100 val_loss:3.8349 train_time:94774ms step_avg:154.10ms step:626/5100 train_loss:3.9289 train_time:94868ms step_avg:154.01ms step:627/5100 train_loss:3.9600 train_time:95024ms step_avg:154.01ms step:628/5100 train_loss:3.9377 train_time:95177ms step_avg:154.01ms step:629/5100 train_loss:3.9813 train_time:95327ms step_avg:154.00ms step:630/5100 train_loss:3.7929 train_time:95480ms step_avg:154.00ms step:631/5100 train_loss:3.9270 train_time:95632ms step_avg:154.00ms step:632/5100 train_loss:3.9583 train_time:95785ms step_avg:154.00ms step:633/5100 train_loss:3.8671 train_time:95940ms step_avg:154.00ms step:634/5100 train_loss:3.7922 train_time:96093ms step_avg:153.99ms step:635/5100 train_loss:3.8892 train_time:96245ms step_avg:153.99ms step:636/5100 train_loss:4.1504 train_time:96397ms step_avg:153.99ms step:637/5100 train_loss:3.7350 train_time:96549ms step_avg:153.99ms step:638/5100 train_loss:3.5588 train_time:96701ms step_avg:153.98ms step:639/5100 train_loss:3.7865 train_time:96853ms step_avg:153.98ms step:640/5100 train_loss:3.8232 train_time:97005ms step_avg:153.98ms step:641/5100 train_loss:3.7789 train_time:97159ms step_avg:153.98ms step:642/5100 train_loss:3.7766 train_time:97312ms step_avg:153.97ms step:643/5100 train_loss:3.8260 train_time:97464ms step_avg:153.97ms step:644/5100 train_loss:3.8381 train_time:97617ms step_avg:153.97ms step:645/5100 train_loss:3.7621 train_time:97771ms step_avg:153.97ms step:646/5100 train_loss:3.9827 train_time:97922ms step_avg:153.97ms step:647/5100 train_loss:3.8833 train_time:98075ms step_avg:153.96ms step:648/5100 train_loss:3.8763 train_time:98226ms step_avg:153.96ms step:649/5100 train_loss:3.9045 train_time:98379ms step_avg:153.96ms step:650/5100 train_loss:3.9650 train_time:98531ms step_avg:153.95ms step:651/5100 train_loss:3.8229 train_time:98683ms step_avg:153.95ms step:652/5100 train_loss:3.9720 train_time:98836ms step_avg:153.95ms step:653/5100 train_loss:3.7950 train_time:98989ms step_avg:153.95ms step:654/5100 train_loss:3.8690 train_time:99141ms step_avg:153.95ms step:655/5100 train_loss:3.6274 train_time:99293ms step_avg:153.94ms step:656/5100 train_loss:3.7787 train_time:99446ms step_avg:153.94ms step:657/5100 train_loss:3.7832 train_time:99598ms step_avg:153.94ms step:658/5100 train_loss:3.7195 train_time:99751ms step_avg:153.94ms step:659/5100 train_loss:3.8938 train_time:99903ms step_avg:153.93ms step:660/5100 train_loss:3.7940 train_time:100055ms step_avg:153.93ms step:661/5100 train_loss:3.8842 train_time:100208ms step_avg:153.93ms step:662/5100 train_loss:3.9577 train_time:100360ms step_avg:153.93ms step:663/5100 train_loss:3.8720 train_time:100513ms step_avg:153.92ms step:664/5100 train_loss:3.7527 train_time:100664ms step_avg:153.92ms step:665/5100 train_loss:3.8394 train_time:100816ms step_avg:153.92ms step:666/5100 train_loss:3.7040 train_time:100969ms step_avg:153.92ms step:667/5100 train_loss:3.9935 train_time:101122ms step_avg:153.91ms step:668/5100 train_loss:3.8284 train_time:101274ms step_avg:153.91ms step:669/5100 train_loss:3.8349 train_time:101426ms step_avg:153.91ms step:670/5100 train_loss:3.6860 train_time:101581ms step_avg:153.91ms step:671/5100 train_loss:3.8019 train_time:101732ms step_avg:153.91ms step:672/5100 train_loss:3.7624 train_time:101884ms step_avg:153.90ms step:673/5100 train_loss:3.7745 train_time:102037ms step_avg:153.90ms step:674/5100 train_loss:4.0546 train_time:102191ms step_avg:153.90ms step:675/5100 train_loss:3.8483 train_time:102343ms step_avg:153.90ms step:676/5100 train_loss:3.9224 train_time:102496ms step_avg:153.90ms step:677/5100 train_loss:3.6948 train_time:102649ms step_avg:153.90ms step:678/5100 train_loss:3.7957 train_time:102801ms step_avg:153.89ms step:679/5100 train_loss:3.7404 train_time:102953ms step_avg:153.89ms step:680/5100 train_loss:3.8915 train_time:103107ms step_avg:153.89ms step:681/5100 train_loss:3.7909 train_time:103259ms step_avg:153.89ms step:682/5100 train_loss:3.8228 train_time:103412ms step_avg:153.89ms step:683/5100 train_loss:3.8833 train_time:103565ms step_avg:153.88ms step:684/5100 train_loss:3.9359 train_time:103717ms step_avg:153.88ms step:685/5100 train_loss:3.8341 train_time:103869ms step_avg:153.88ms step:686/5100 train_loss:3.9055 train_time:104021ms step_avg:153.88ms step:687/5100 train_loss:3.8300 train_time:104173ms step_avg:153.87ms step:688/5100 train_loss:3.8797 train_time:104326ms step_avg:153.87ms step:689/5100 train_loss:3.5017 train_time:104479ms step_avg:153.87ms step:690/5100 train_loss:3.6153 train_time:104631ms step_avg:153.87ms step:691/5100 train_loss:3.7621 train_time:104783ms step_avg:153.87ms step:692/5100 train_loss:3.6358 train_time:104936ms step_avg:153.86ms step:693/5100 train_loss:3.8506 train_time:105089ms step_avg:153.86ms step:694/5100 train_loss:3.8703 train_time:105241ms step_avg:153.86ms step:695/5100 train_loss:3.7562 train_time:105393ms step_avg:153.86ms step:696/5100 train_loss:3.7428 train_time:105545ms step_avg:153.86ms step:697/5100 train_loss:4.0574 train_time:105698ms step_avg:153.85ms step:698/5100 train_loss:3.8016 train_time:105850ms step_avg:153.85ms step:699/5100 train_loss:3.8452 train_time:106002ms step_avg:153.85ms step:700/5100 train_loss:3.9977 train_time:106155ms step_avg:153.85ms step:701/5100 train_loss:3.7819 train_time:106307ms step_avg:153.85ms step:702/5100 train_loss:3.7432 train_time:106459ms step_avg:153.84ms step:703/5100 train_loss:3.7280 train_time:106613ms step_avg:153.84ms step:704/5100 train_loss:3.6748 train_time:106765ms step_avg:153.84ms step:705/5100 train_loss:3.7681 train_time:106917ms step_avg:153.84ms step:706/5100 train_loss:3.7613 train_time:107070ms step_avg:153.84ms step:707/5100 train_loss:3.7780 train_time:107222ms step_avg:153.83ms step:708/5100 train_loss:3.8522 train_time:107375ms step_avg:153.83ms step:709/5100 train_loss:3.7965 train_time:107526ms step_avg:153.83ms step:710/5100 train_loss:3.7821 train_time:107679ms step_avg:153.83ms step:711/5100 train_loss:3.7510 train_time:107832ms step_avg:153.83ms step:712/5100 train_loss:3.7882 train_time:107984ms step_avg:153.82ms step:713/5100 train_loss:3.8523 train_time:108137ms step_avg:153.82ms step:714/5100 train_loss:3.8517 train_time:108289ms step_avg:153.82ms step:715/5100 train_loss:3.7719 train_time:108441ms step_avg:153.82ms step:716/5100 train_loss:3.7745 train_time:108594ms step_avg:153.82ms step:717/5100 train_loss:3.7891 train_time:108747ms step_avg:153.82ms step:718/5100 train_loss:3.9368 train_time:108899ms step_avg:153.81ms step:719/5100 train_loss:3.7933 train_time:109052ms step_avg:153.81ms step:720/5100 train_loss:3.8709 train_time:109204ms step_avg:153.81ms step:721/5100 train_loss:4.0322 train_time:109356ms step_avg:153.81ms step:722/5100 train_loss:3.6629 train_time:109509ms step_avg:153.81ms step:723/5100 train_loss:3.9255 train_time:109662ms step_avg:153.80ms step:724/5100 train_loss:3.9819 train_time:109814ms step_avg:153.80ms step:725/5100 train_loss:3.7698 train_time:109965ms step_avg:153.80ms step:726/5100 train_loss:3.8473 train_time:110117ms step_avg:153.79ms step:727/5100 train_loss:3.7453 train_time:110270ms step_avg:153.79ms step:728/5100 train_loss:3.7577 train_time:110422ms step_avg:153.79ms step:729/5100 train_loss:3.9400 train_time:110575ms step_avg:153.79ms step:730/5100 train_loss:3.8899 train_time:110728ms step_avg:153.79ms step:731/5100 train_loss:3.8869 train_time:110880ms step_avg:153.79ms step:732/5100 train_loss:3.7652 train_time:111032ms step_avg:153.78ms step:733/5100 train_loss:3.7931 train_time:111184ms step_avg:153.78ms step:734/5100 train_loss:4.0344 train_time:111337ms step_avg:153.78ms step:735/5100 train_loss:3.7594 train_time:111490ms step_avg:153.78ms step:736/5100 train_loss:3.8268 train_time:111642ms step_avg:153.78ms step:737/5100 train_loss:3.9500 train_time:111795ms step_avg:153.78ms step:738/5100 train_loss:3.8610 train_time:111948ms step_avg:153.77ms step:739/5100 train_loss:3.8033 train_time:112099ms step_avg:153.77ms step:740/5100 train_loss:3.6951 train_time:112251ms step_avg:153.77ms step:741/5100 train_loss:4.3429 train_time:112404ms step_avg:153.77ms step:742/5100 train_loss:3.6997 train_time:112555ms step_avg:153.76ms step:743/5100 train_loss:3.7840 train_time:112708ms step_avg:153.76ms step:744/5100 train_loss:3.7832 train_time:112861ms step_avg:153.76ms step:745/5100 train_loss:3.8452 train_time:113014ms step_avg:153.76ms step:746/5100 train_loss:3.8240 train_time:113166ms step_avg:153.76ms step:747/5100 train_loss:3.7983 train_time:113318ms step_avg:153.76ms step:748/5100 train_loss:3.8343 train_time:113470ms step_avg:153.75ms step:749/5100 train_loss:3.7610 train_time:113622ms step_avg:153.75ms step:750/5100 train_loss:3.7724 train_time:113774ms step_avg:153.75ms step:750/5100 val_loss:3.7745 train_time:113838ms step_avg:153.84ms step:751/5100 train_loss:3.8027 train_time:113932ms step_avg:153.75ms step:752/5100 train_loss:3.7658 train_time:114088ms step_avg:153.76ms step:753/5100 train_loss:3.7988 train_time:114240ms step_avg:153.76ms step:754/5100 train_loss:3.8203 train_time:114392ms step_avg:153.75ms step:755/5100 train_loss:3.7954 train_time:114544ms step_avg:153.75ms step:756/5100 train_loss:3.8654 train_time:114793ms step_avg:153.88ms step:757/5100 train_loss:3.6922 train_time:114954ms step_avg:153.89ms step:758/5100 train_loss:3.9381 train_time:115104ms step_avg:153.88ms step:759/5100 train_loss:3.8526 train_time:115257ms step_avg:153.88ms step:760/5100 train_loss:3.7829 train_time:115540ms step_avg:154.05ms step:761/5100 train_loss:3.8927 train_time:115706ms step_avg:154.07ms step:762/5100 train_loss:3.6004 train_time:115859ms step_avg:154.07ms step:763/5100 train_loss:3.7539 train_time:116010ms step_avg:154.06ms step:764/5100 train_loss:3.8676 train_time:116162ms step_avg:154.06ms step:765/5100 train_loss:3.5194 train_time:116314ms step_avg:154.06ms step:766/5100 train_loss:3.9417 train_time:116466ms step_avg:154.06ms step:767/5100 train_loss:3.7963 train_time:116619ms step_avg:154.05ms step:768/5100 train_loss:3.7578 train_time:116772ms step_avg:154.05ms step:769/5100 train_loss:3.7724 train_time:116924ms step_avg:154.05ms step:770/5100 train_loss:3.7962 train_time:117078ms step_avg:154.05ms step:771/5100 train_loss:3.8550 train_time:117229ms step_avg:154.05ms step:772/5100 train_loss:4.0807 train_time:117382ms step_avg:154.04ms step:773/5100 train_loss:3.6603 train_time:117534ms step_avg:154.04ms step:774/5100 train_loss:3.8566 train_time:117686ms step_avg:154.04ms step:775/5100 train_loss:3.8395 train_time:117839ms step_avg:154.04ms step:776/5100 train_loss:3.8094 train_time:117991ms step_avg:154.04ms step:777/5100 train_loss:3.6113 train_time:118143ms step_avg:154.03ms step:778/5100 train_loss:3.6073 train_time:118296ms step_avg:154.03ms step:779/5100 train_loss:3.6774 train_time:118448ms step_avg:154.03ms step:780/5100 train_loss:3.7684 train_time:118602ms step_avg:154.03ms step:781/5100 train_loss:3.8067 train_time:118756ms step_avg:154.03ms step:782/5100 train_loss:3.8612 train_time:118909ms step_avg:154.03ms step:783/5100 train_loss:3.7773 train_time:119062ms step_avg:154.03ms step:784/5100 train_loss:3.7728 train_time:119215ms step_avg:154.02ms step:785/5100 train_loss:3.7769 train_time:119366ms step_avg:154.02ms step:786/5100 train_loss:3.7557 train_time:119518ms step_avg:154.02ms step:787/5100 train_loss:3.6554 train_time:119671ms step_avg:154.02ms step:788/5100 train_loss:3.9053 train_time:119823ms step_avg:154.01ms step:789/5100 train_loss:3.6982 train_time:119976ms step_avg:154.01ms step:790/5100 train_loss:3.7615 train_time:120129ms step_avg:154.01ms step:791/5100 train_loss:3.8271 train_time:120283ms step_avg:154.01ms step:792/5100 train_loss:3.9595 train_time:120435ms step_avg:154.01ms step:793/5100 train_loss:3.9705 train_time:120587ms step_avg:154.01ms step:794/5100 train_loss:3.6825 train_time:120740ms step_avg:154.01ms step:795/5100 train_loss:3.8030 train_time:120891ms step_avg:154.00ms step:796/5100 train_loss:3.8572 train_time:121043ms step_avg:154.00ms step:797/5100 train_loss:3.9614 train_time:121197ms step_avg:154.00ms step:798/5100 train_loss:3.7142 train_time:121350ms step_avg:154.00ms step:799/5100 train_loss:3.8613 train_time:121501ms step_avg:153.99ms step:800/5100 train_loss:3.7518 train_time:121654ms step_avg:153.99ms step:801/5100 train_loss:3.7435 train_time:121806ms step_avg:153.99ms step:802/5100 train_loss:3.8309 train_time:121959ms step_avg:153.99ms step:803/5100 train_loss:3.6853 train_time:122112ms step_avg:153.99ms step:804/5100 train_loss:3.7255 train_time:122264ms step_avg:153.98ms step:805/5100 train_loss:3.8341 train_time:122418ms step_avg:153.98ms step:806/5100 train_loss:3.7367 train_time:122570ms step_avg:153.98ms step:807/5100 train_loss:3.7441 train_time:122722ms step_avg:153.98ms step:808/5100 train_loss:3.8415 train_time:122875ms step_avg:153.98ms step:809/5100 train_loss:3.7592 train_time:123028ms step_avg:153.98ms step:810/5100 train_loss:3.6862 train_time:123181ms step_avg:153.98ms step:811/5100 train_loss:3.7705 train_time:123333ms step_avg:153.97ms step:812/5100 train_loss:3.7973 train_time:123486ms step_avg:153.97ms step:813/5100 train_loss:3.7899 train_time:123639ms step_avg:153.97ms step:814/5100 train_loss:3.8252 train_time:123791ms step_avg:153.97ms step:815/5100 train_loss:3.7752 train_time:123944ms step_avg:153.97ms step:816/5100 train_loss:3.7556 train_time:124096ms step_avg:153.97ms step:817/5100 train_loss:3.8586 train_time:124250ms step_avg:153.96ms step:818/5100 train_loss:3.9643 train_time:124402ms step_avg:153.96ms step:819/5100 train_loss:3.7284 train_time:124555ms step_avg:153.96ms step:820/5100 train_loss:3.9251 train_time:124708ms step_avg:153.96ms step:821/5100 train_loss:3.7009 train_time:124861ms step_avg:153.96ms step:822/5100 train_loss:3.7474 train_time:125012ms step_avg:153.96ms step:823/5100 train_loss:3.8635 train_time:125164ms step_avg:153.95ms step:824/5100 train_loss:3.7851 train_time:125317ms step_avg:153.95ms step:825/5100 train_loss:3.7123 train_time:125469ms step_avg:153.95ms step:826/5100 train_loss:3.8153 train_time:125621ms step_avg:153.95ms step:827/5100 train_loss:3.7016 train_time:125774ms step_avg:153.95ms step:828/5100 train_loss:3.9345 train_time:125926ms step_avg:153.94ms step:829/5100 train_loss:3.8215 train_time:126081ms step_avg:153.95ms step:830/5100 train_loss:3.8810 train_time:126233ms step_avg:153.94ms step:831/5100 train_loss:3.7345 train_time:126386ms step_avg:153.94ms step:832/5100 train_loss:3.7815 train_time:126539ms step_avg:153.94ms step:833/5100 train_loss:3.7119 train_time:126693ms step_avg:153.94ms step:834/5100 train_loss:3.8395 train_time:126844ms step_avg:153.94ms step:835/5100 train_loss:3.6840 train_time:126997ms step_avg:153.94ms step:836/5100 train_loss:3.6577 train_time:127149ms step_avg:153.93ms step:837/5100 train_loss:3.9288 train_time:127302ms step_avg:153.93ms step:838/5100 train_loss:3.6163 train_time:127454ms step_avg:153.93ms step:839/5100 train_loss:3.7864 train_time:127606ms step_avg:153.93ms step:840/5100 train_loss:3.6348 train_time:127758ms step_avg:153.93ms step:841/5100 train_loss:3.6762 train_time:127911ms step_avg:153.92ms step:842/5100 train_loss:3.7573 train_time:128064ms step_avg:153.92ms step:843/5100 train_loss:3.7758 train_time:128217ms step_avg:153.92ms step:844/5100 train_loss:3.7857 train_time:128369ms step_avg:153.92ms step:845/5100 train_loss:3.6275 train_time:128522ms step_avg:153.92ms step:846/5100 train_loss:3.8682 train_time:128674ms step_avg:153.92ms step:847/5100 train_loss:3.7256 train_time:128826ms step_avg:153.91ms step:848/5100 train_loss:3.6875 train_time:128980ms step_avg:153.91ms step:849/5100 train_loss:3.8306 train_time:129133ms step_avg:153.91ms step:850/5100 train_loss:3.6962 train_time:129285ms step_avg:153.91ms step:851/5100 train_loss:3.6463 train_time:129439ms step_avg:153.91ms step:852/5100 train_loss:3.9410 train_time:129591ms step_avg:153.91ms step:853/5100 train_loss:3.6490 train_time:129743ms step_avg:153.91ms step:854/5100 train_loss:3.7638 train_time:129895ms step_avg:153.90ms step:855/5100 train_loss:3.8490 train_time:130048ms step_avg:153.90ms step:856/5100 train_loss:3.7279 train_time:130200ms step_avg:153.90ms step:857/5100 train_loss:3.7475 train_time:130352ms step_avg:153.90ms step:858/5100 train_loss:3.8023 train_time:130504ms step_avg:153.90ms step:859/5100 train_loss:3.6789 train_time:130657ms step_avg:153.89ms step:860/5100 train_loss:3.7558 train_time:130812ms step_avg:153.90ms step:861/5100 train_loss:3.7988 train_time:130963ms step_avg:153.89ms step:862/5100 train_loss:3.8392 train_time:131115ms step_avg:153.89ms step:863/5100 train_loss:3.7918 train_time:131268ms step_avg:153.89ms step:864/5100 train_loss:3.7688 train_time:131420ms step_avg:153.89ms step:865/5100 train_loss:3.5913 train_time:131572ms step_avg:153.89ms step:866/5100 train_loss:3.7880 train_time:131724ms step_avg:153.88ms step:867/5100 train_loss:4.0571 train_time:131877ms step_avg:153.88ms step:868/5100 train_loss:3.6451 train_time:132030ms step_avg:153.88ms step:869/5100 train_loss:3.8319 train_time:132182ms step_avg:153.88ms step:870/5100 train_loss:3.8084 train_time:132334ms step_avg:153.88ms step:871/5100 train_loss:3.6480 train_time:132488ms step_avg:153.88ms step:872/5100 train_loss:3.6213 train_time:132640ms step_avg:153.87ms step:873/5100 train_loss:3.8601 train_time:132792ms step_avg:153.87ms step:874/5100 train_loss:3.6543 train_time:132944ms step_avg:153.87ms step:875/5100 train_loss:3.3726 train_time:133097ms step_avg:153.87ms step:875/5100 val_loss:3.7258 train_time:133161ms step_avg:153.94ms step:876/5100 train_loss:3.8387 train_time:133257ms step_avg:153.88ms step:877/5100 train_loss:3.6476 train_time:133414ms step_avg:153.88ms step:878/5100 train_loss:3.8273 train_time:133566ms step_avg:153.88ms step:879/5100 train_loss:3.6822 train_time:133718ms step_avg:153.88ms step:880/5100 train_loss:3.8520 train_time:133870ms step_avg:153.87ms step:881/5100 train_loss:3.5219 train_time:134022ms step_avg:153.87ms step:882/5100 train_loss:3.6976 train_time:134175ms step_avg:153.87ms step:883/5100 train_loss:3.8867 train_time:134327ms step_avg:153.87ms step:884/5100 train_loss:4.0445 train_time:134482ms step_avg:153.87ms step:885/5100 train_loss:3.7715 train_time:134635ms step_avg:153.87ms step:886/5100 train_loss:3.6858 train_time:134788ms step_avg:153.87ms step:887/5100 train_loss:3.7784 train_time:134941ms step_avg:153.87ms step:888/5100 train_loss:4.2854 train_time:135092ms step_avg:153.86ms step:889/5100 train_loss:4.0469 train_time:135246ms step_avg:153.86ms step:890/5100 train_loss:3.7227 train_time:135397ms step_avg:153.86ms step:891/5100 train_loss:3.7380 train_time:135551ms step_avg:153.86ms step:892/5100 train_loss:3.5625 train_time:135704ms step_avg:153.86ms step:893/5100 train_loss:3.9008 train_time:135857ms step_avg:153.86ms step:894/5100 train_loss:3.6243 train_time:136010ms step_avg:153.86ms step:895/5100 train_loss:3.8764 train_time:136163ms step_avg:153.86ms step:896/5100 train_loss:3.8902 train_time:136316ms step_avg:153.86ms step:897/5100 train_loss:3.7006 train_time:136468ms step_avg:153.85ms step:898/5100 train_loss:3.7343 train_time:136622ms step_avg:153.85ms step:899/5100 train_loss:3.7898 train_time:136774ms step_avg:153.85ms step:900/5100 train_loss:3.6790 train_time:136926ms step_avg:153.85ms step:901/5100 train_loss:3.6176 train_time:137080ms step_avg:153.85ms step:902/5100 train_loss:3.8306 train_time:137233ms step_avg:153.85ms step:903/5100 train_loss:3.8357 train_time:137385ms step_avg:153.85ms step:904/5100 train_loss:3.7336 train_time:137538ms step_avg:153.85ms step:905/5100 train_loss:3.6954 train_time:137691ms step_avg:153.85ms step:906/5100 train_loss:3.6952 train_time:137844ms step_avg:153.84ms step:907/5100 train_loss:3.9209 train_time:137997ms step_avg:153.84ms step:908/5100 train_loss:3.7093 train_time:138150ms step_avg:153.84ms step:909/5100 train_loss:3.7522 train_time:138302ms step_avg:153.84ms step:910/5100 train_loss:3.6584 train_time:138457ms step_avg:153.84ms step:911/5100 train_loss:3.7455 train_time:138610ms step_avg:153.84ms step:912/5100 train_loss:3.8131 train_time:138764ms step_avg:153.84ms step:913/5100 train_loss:3.8008 train_time:138917ms step_avg:153.84ms step:914/5100 train_loss:3.6805 train_time:139069ms step_avg:153.84ms step:915/5100 train_loss:3.9323 train_time:139222ms step_avg:153.84ms step:916/5100 train_loss:3.7320 train_time:139375ms step_avg:153.84ms step:917/5100 train_loss:3.8247 train_time:139527ms step_avg:153.83ms step:918/5100 train_loss:3.8004 train_time:139680ms step_avg:153.83ms step:919/5100 train_loss:5.0237 train_time:139833ms step_avg:153.83ms step:920/5100 train_loss:3.7184 train_time:139986ms step_avg:153.83ms step:921/5100 train_loss:3.7735 train_time:140141ms step_avg:153.83ms step:922/5100 train_loss:3.7354 train_time:140293ms step_avg:153.83ms step:923/5100 train_loss:3.7902 train_time:140446ms step_avg:153.83ms step:924/5100 train_loss:3.7966 train_time:140598ms step_avg:153.83ms step:925/5100 train_loss:3.8839 train_time:140751ms step_avg:153.83ms step:926/5100 train_loss:3.8644 train_time:140903ms step_avg:153.82ms step:927/5100 train_loss:3.7527 train_time:141056ms step_avg:153.82ms step:928/5100 train_loss:3.7431 train_time:141210ms step_avg:153.82ms step:929/5100 train_loss:3.9716 train_time:141363ms step_avg:153.82ms step:930/5100 train_loss:3.8107 train_time:141516ms step_avg:153.82ms step:931/5100 train_loss:3.5936 train_time:141669ms step_avg:153.82ms step:932/5100 train_loss:3.6880 train_time:141822ms step_avg:153.82ms step:933/5100 train_loss:3.8720 train_time:141974ms step_avg:153.82ms step:934/5100 train_loss:3.5962 train_time:142127ms step_avg:153.82ms step:935/5100 train_loss:3.7743 train_time:142281ms step_avg:153.82ms step:936/5100 train_loss:3.6506 train_time:142434ms step_avg:153.82ms step:937/5100 train_loss:3.7068 train_time:142586ms step_avg:153.81ms step:938/5100 train_loss:3.8065 train_time:142739ms step_avg:153.81ms step:939/5100 train_loss:3.7312 train_time:142892ms step_avg:153.81ms step:940/5100 train_loss:3.9027 train_time:143046ms step_avg:153.81ms step:941/5100 train_loss:3.6779 train_time:143198ms step_avg:153.81ms step:942/5100 train_loss:3.7481 train_time:143351ms step_avg:153.81ms step:943/5100 train_loss:3.5500 train_time:143504ms step_avg:153.81ms step:944/5100 train_loss:3.8966 train_time:143657ms step_avg:153.81ms step:945/5100 train_loss:3.6084 train_time:143907ms step_avg:153.91ms step:946/5100 train_loss:3.6217 train_time:144070ms step_avg:153.92ms step:947/5100 train_loss:5.2585 train_time:144222ms step_avg:153.92ms step:948/5100 train_loss:3.7953 train_time:144375ms step_avg:153.92ms step:949/5100 train_loss:3.6925 train_time:144527ms step_avg:153.92ms step:950/5100 train_loss:3.5941 train_time:144804ms step_avg:154.05ms step:951/5100 train_loss:3.6467 train_time:144955ms step_avg:154.04ms step:952/5100 train_loss:3.6090 train_time:145106ms step_avg:154.04ms step:953/5100 train_loss:3.6724 train_time:145258ms step_avg:154.04ms step:954/5100 train_loss:3.7548 train_time:145410ms step_avg:154.04ms step:955/5100 train_loss:3.6305 train_time:145562ms step_avg:154.03ms step:956/5100 train_loss:3.6769 train_time:145715ms step_avg:154.03ms step:957/5100 train_loss:3.6417 train_time:145867ms step_avg:154.03ms step:958/5100 train_loss:3.7008 train_time:146021ms step_avg:154.03ms step:959/5100 train_loss:3.6917 train_time:146173ms step_avg:154.03ms step:960/5100 train_loss:3.6973 train_time:146324ms step_avg:154.03ms step:961/5100 train_loss:3.5877 train_time:146476ms step_avg:154.02ms step:962/5100 train_loss:3.8559 train_time:146628ms step_avg:154.02ms step:963/5100 train_loss:3.8051 train_time:146782ms step_avg:154.02ms step:964/5100 train_loss:3.6786 train_time:146934ms step_avg:154.02ms step:965/5100 train_loss:3.6411 train_time:147087ms step_avg:154.02ms step:966/5100 train_loss:3.6840 train_time:147239ms step_avg:154.02ms step:967/5100 train_loss:3.9082 train_time:147391ms step_avg:154.01ms step:968/5100 train_loss:3.7320 train_time:147543ms step_avg:154.01ms step:969/5100 train_loss:3.7192 train_time:147696ms step_avg:154.01ms step:970/5100 train_loss:3.7707 train_time:147849ms step_avg:154.01ms step:971/5100 train_loss:3.5889 train_time:148002ms step_avg:154.01ms step:972/5100 train_loss:3.7415 train_time:148155ms step_avg:154.01ms step:973/5100 train_loss:3.6954 train_time:148306ms step_avg:154.00ms step:974/5100 train_loss:3.7432 train_time:148461ms step_avg:154.01ms step:975/5100 train_loss:3.8110 train_time:148615ms step_avg:154.01ms step:976/5100 train_loss:3.6827 train_time:148767ms step_avg:154.00ms step:977/5100 train_loss:3.8830 train_time:148920ms step_avg:154.00ms step:978/5100 train_loss:3.7639 train_time:149071ms step_avg:154.00ms step:979/5100 train_loss:3.5966 train_time:149224ms step_avg:154.00ms step:980/5100 train_loss:3.8885 train_time:149377ms step_avg:154.00ms step:981/5100 train_loss:3.6212 train_time:149530ms step_avg:154.00ms step:982/5100 train_loss:3.7802 train_time:149683ms step_avg:153.99ms step:983/5100 train_loss:3.7581 train_time:149835ms step_avg:153.99ms step:984/5100 train_loss:3.7719 train_time:149987ms step_avg:153.99ms step:985/5100 train_loss:3.7119 train_time:150139ms step_avg:153.99ms step:986/5100 train_loss:3.7974 train_time:150292ms step_avg:153.99ms step:987/5100 train_loss:3.6163 train_time:150443ms step_avg:153.99ms step:988/5100 train_loss:3.6949 train_time:150596ms step_avg:153.98ms step:989/5100 train_loss:3.7067 train_time:150748ms step_avg:153.98ms step:990/5100 train_loss:3.6311 train_time:150902ms step_avg:153.98ms step:991/5100 train_loss:3.8495 train_time:151053ms step_avg:153.98ms step:992/5100 train_loss:3.6724 train_time:151205ms step_avg:153.98ms step:993/5100 train_loss:3.6425 train_time:151356ms step_avg:153.97ms step:994/5100 train_loss:3.7142 train_time:151509ms step_avg:153.97ms step:995/5100 train_loss:3.7991 train_time:151662ms step_avg:153.97ms step:996/5100 train_loss:3.7372 train_time:151816ms step_avg:153.97ms step:997/5100 train_loss:3.6557 train_time:151968ms step_avg:153.97ms step:998/5100 train_loss:4.0065 train_time:152121ms step_avg:153.97ms step:999/5100 train_loss:3.6632 train_time:152272ms step_avg:153.97ms step:1000/5100 train_loss:3.7897 train_time:152423ms step_avg:153.96ms step:1000/5100 val_loss:3.6852 train_time:152487ms step_avg:154.03ms step:1001/5100 train_loss:3.6581 train_time:152585ms step_avg:153.97ms step:1002/5100 train_loss:3.7123 train_time:152742ms step_avg:153.97ms step:1003/5100 train_loss:3.5956 train_time:152894ms step_avg:153.97ms step:1004/5100 train_loss:3.7763 train_time:153046ms step_avg:153.97ms step:1005/5100 train_loss:3.8272 train_time:153198ms step_avg:153.97ms step:1006/5100 train_loss:3.5954 train_time:153350ms step_avg:153.97ms step:1007/5100 train_loss:3.6796 train_time:153502ms step_avg:153.96ms step:1008/5100 train_loss:3.6512 train_time:153655ms step_avg:153.96ms step:1009/5100 train_loss:3.7688 train_time:153807ms step_avg:153.96ms step:1010/5100 train_loss:3.8698 train_time:153961ms step_avg:153.96ms step:1011/5100 train_loss:3.7722 train_time:154114ms step_avg:153.96ms step:1012/5100 train_loss:3.7353 train_time:154266ms step_avg:153.96ms step:1013/5100 train_loss:3.5909 train_time:154420ms step_avg:153.96ms step:1014/5100 train_loss:3.7354 train_time:154572ms step_avg:153.96ms step:1015/5100 train_loss:3.8484 train_time:154724ms step_avg:153.95ms step:1016/5100 train_loss:3.5547 train_time:154877ms step_avg:153.95ms step:1017/5100 train_loss:3.6441 train_time:155030ms step_avg:153.95ms step:1018/5100 train_loss:3.6524 train_time:155184ms step_avg:153.95ms step:1019/5100 train_loss:3.5928 train_time:155337ms step_avg:153.95ms step:1020/5100 train_loss:3.7336 train_time:155491ms step_avg:153.95ms step:1021/5100 train_loss:3.6442 train_time:155643ms step_avg:153.95ms step:1022/5100 train_loss:3.5803 train_time:155796ms step_avg:153.95ms step:1023/5100 train_loss:3.6894 train_time:155949ms step_avg:153.95ms step:1024/5100 train_loss:3.7150 train_time:156102ms step_avg:153.95ms step:1025/5100 train_loss:3.6923 train_time:156255ms step_avg:153.95ms step:1026/5100 train_loss:3.7018 train_time:156408ms step_avg:153.94ms step:1027/5100 train_loss:3.8682 train_time:156562ms step_avg:153.94ms step:1028/5100 train_loss:3.5430 train_time:156715ms step_avg:153.94ms step:1029/5100 train_loss:3.6083 train_time:156867ms step_avg:153.94ms step:1030/5100 train_loss:3.5620 train_time:157020ms step_avg:153.94ms step:1031/5100 train_loss:3.7267 train_time:157173ms step_avg:153.94ms step:1032/5100 train_loss:3.7163 train_time:157325ms step_avg:153.94ms step:1033/5100 train_loss:3.8985 train_time:157479ms step_avg:153.94ms step:1034/5100 train_loss:3.7085 train_time:157631ms step_avg:153.94ms step:1035/5100 train_loss:3.6318 train_time:157785ms step_avg:153.94ms step:1036/5100 train_loss:3.6472 train_time:157937ms step_avg:153.94ms step:1037/5100 train_loss:3.7087 train_time:158091ms step_avg:153.93ms step:1038/5100 train_loss:4.0167 train_time:158244ms step_avg:153.93ms step:1039/5100 train_loss:3.8413 train_time:158396ms step_avg:153.93ms step:1040/5100 train_loss:3.7315 train_time:158549ms step_avg:153.93ms step:1041/5100 train_loss:3.6272 train_time:158702ms step_avg:153.93ms step:1042/5100 train_loss:3.6945 train_time:158855ms step_avg:153.93ms step:1043/5100 train_loss:3.7370 train_time:159007ms step_avg:153.93ms step:1044/5100 train_loss:3.6602 train_time:159159ms step_avg:153.93ms step:1045/5100 train_loss:3.6715 train_time:159312ms step_avg:153.92ms step:1046/5100 train_loss:3.7534 train_time:159464ms step_avg:153.92ms step:1047/5100 train_loss:3.6544 train_time:159618ms step_avg:153.92ms step:1048/5100 train_loss:3.8636 train_time:159772ms step_avg:153.92ms step:1049/5100 train_loss:3.7186 train_time:159924ms step_avg:153.92ms step:1050/5100 train_loss:3.6429 train_time:160077ms step_avg:153.92ms step:1051/5100 train_loss:3.6040 train_time:160229ms step_avg:153.92ms step:1052/5100 train_loss:3.7298 train_time:160383ms step_avg:153.92ms step:1053/5100 train_loss:3.6026 train_time:160536ms step_avg:153.92ms step:1054/5100 train_loss:3.9269 train_time:160688ms step_avg:153.92ms step:1055/5100 train_loss:3.7619 train_time:160842ms step_avg:153.92ms step:1056/5100 train_loss:3.6200 train_time:160994ms step_avg:153.91ms step:1057/5100 train_loss:3.7285 train_time:161146ms step_avg:153.91ms step:1058/5100 train_loss:3.8018 train_time:161299ms step_avg:153.91ms step:1059/5100 train_loss:3.5233 train_time:161452ms step_avg:153.91ms step:1060/5100 train_loss:3.6444 train_time:161606ms step_avg:153.91ms step:1061/5100 train_loss:3.6729 train_time:161759ms step_avg:153.91ms step:1062/5100 train_loss:3.6379 train_time:161913ms step_avg:153.91ms step:1063/5100 train_loss:3.6110 train_time:162065ms step_avg:153.91ms step:1064/5100 train_loss:3.7102 train_time:162218ms step_avg:153.91ms step:1065/5100 train_loss:3.6122 train_time:162371ms step_avg:153.91ms step:1066/5100 train_loss:3.5916 train_time:162523ms step_avg:153.90ms step:1067/5100 train_loss:3.6191 train_time:162677ms step_avg:153.90ms step:1068/5100 train_loss:3.5282 train_time:162829ms step_avg:153.90ms step:1069/5100 train_loss:3.6449 train_time:162983ms step_avg:153.90ms step:1070/5100 train_loss:3.5172 train_time:163137ms step_avg:153.90ms step:1071/5100 train_loss:3.7742 train_time:163289ms step_avg:153.90ms step:1072/5100 train_loss:3.7232 train_time:163442ms step_avg:153.90ms step:1073/5100 train_loss:3.6715 train_time:163595ms step_avg:153.90ms step:1074/5100 train_loss:3.7417 train_time:163747ms step_avg:153.90ms step:1075/5100 train_loss:3.6888 train_time:163901ms step_avg:153.90ms step:1076/5100 train_loss:3.6208 train_time:164054ms step_avg:153.90ms step:1077/5100 train_loss:4.0139 train_time:164206ms step_avg:153.90ms step:1078/5100 train_loss:3.6928 train_time:164360ms step_avg:153.89ms step:1079/5100 train_loss:3.3573 train_time:164513ms step_avg:153.89ms step:1080/5100 train_loss:3.7553 train_time:164665ms step_avg:153.89ms step:1081/5100 train_loss:3.6745 train_time:164818ms step_avg:153.89ms step:1082/5100 train_loss:3.7399 train_time:164970ms step_avg:153.89ms step:1083/5100 train_loss:3.8339 train_time:165123ms step_avg:153.89ms step:1084/5100 train_loss:3.7275 train_time:165277ms step_avg:153.89ms step:1085/5100 train_loss:3.7021 train_time:165429ms step_avg:153.89ms step:1086/5100 train_loss:3.6691 train_time:165583ms step_avg:153.89ms step:1087/5100 train_loss:3.8622 train_time:165736ms step_avg:153.89ms step:1088/5100 train_loss:3.7630 train_time:165889ms step_avg:153.89ms step:1089/5100 train_loss:3.5868 train_time:166041ms step_avg:153.88ms step:1090/5100 train_loss:3.6085 train_time:166193ms step_avg:153.88ms step:1091/5100 train_loss:3.7281 train_time:166347ms step_avg:153.88ms step:1092/5100 train_loss:3.5230 train_time:166499ms step_avg:153.88ms step:1093/5100 train_loss:3.7212 train_time:166652ms step_avg:153.88ms step:1094/5100 train_loss:3.8571 train_time:166804ms step_avg:153.88ms step:1095/5100 train_loss:3.6959 train_time:166956ms step_avg:153.88ms step:1096/5100 train_loss:3.6466 train_time:167109ms step_avg:153.88ms step:1097/5100 train_loss:3.6699 train_time:167262ms step_avg:153.87ms step:1098/5100 train_loss:3.7133 train_time:167414ms step_avg:153.87ms step:1099/5100 train_loss:3.7883 train_time:167566ms step_avg:153.87ms step:1100/5100 train_loss:3.7402 train_time:167719ms step_avg:153.87ms step:1101/5100 train_loss:3.6769 train_time:167872ms step_avg:153.87ms step:1102/5100 train_loss:3.5248 train_time:168023ms step_avg:153.87ms step:1103/5100 train_loss:3.5926 train_time:168176ms step_avg:153.87ms step:1104/5100 train_loss:3.6785 train_time:168328ms step_avg:153.86ms step:1105/5100 train_loss:3.5573 train_time:168483ms step_avg:153.87ms step:1106/5100 train_loss:4.3115 train_time:168635ms step_avg:153.86ms step:1107/5100 train_loss:3.4631 train_time:168789ms step_avg:153.86ms step:1108/5100 train_loss:3.8030 train_time:168941ms step_avg:153.86ms step:1109/5100 train_loss:3.5862 train_time:169093ms step_avg:153.86ms step:1110/5100 train_loss:3.7330 train_time:169245ms step_avg:153.86ms step:1111/5100 train_loss:3.6669 train_time:169398ms step_avg:153.86ms step:1112/5100 train_loss:3.7096 train_time:169550ms step_avg:153.86ms step:1113/5100 train_loss:3.8003 train_time:169702ms step_avg:153.86ms step:1114/5100 train_loss:3.6609 train_time:169855ms step_avg:153.85ms step:1115/5100 train_loss:3.5939 train_time:170007ms step_avg:153.85ms step:1116/5100 train_loss:3.5038 train_time:170160ms step_avg:153.85ms step:1117/5100 train_loss:3.6723 train_time:170314ms step_avg:153.85ms step:1118/5100 train_loss:3.8282 train_time:170465ms step_avg:153.85ms step:1119/5100 train_loss:3.8650 train_time:170618ms step_avg:153.85ms step:1120/5100 train_loss:3.7005 train_time:170770ms step_avg:153.85ms step:1121/5100 train_loss:3.7295 train_time:170922ms step_avg:153.85ms step:1122/5100 train_loss:3.6308 train_time:171074ms step_avg:153.84ms step:1123/5100 train_loss:3.6817 train_time:171227ms step_avg:153.84ms step:1124/5100 train_loss:3.8277 train_time:171379ms step_avg:153.84ms step:1125/5100 train_loss:3.5929 train_time:171532ms step_avg:153.84ms step:1125/5100 val_loss:3.6553 train_time:171596ms step_avg:153.90ms step:1126/5100 train_loss:3.4858 train_time:171691ms step_avg:153.84ms step:1127/5100 train_loss:3.7167 train_time:171848ms step_avg:153.85ms step:1128/5100 train_loss:3.9272 train_time:172000ms step_avg:153.85ms step:1129/5100 train_loss:3.4693 train_time:172152ms step_avg:153.84ms step:1130/5100 train_loss:3.7907 train_time:172304ms step_avg:153.84ms step:1131/5100 train_loss:3.6228 train_time:172456ms step_avg:153.84ms step:1132/5100 train_loss:3.6518 train_time:172608ms step_avg:153.84ms step:1133/5100 train_loss:3.6110 train_time:172760ms step_avg:153.84ms step:1134/5100 train_loss:3.7697 train_time:173016ms step_avg:153.93ms step:1135/5100 train_loss:3.7012 train_time:173176ms step_avg:153.93ms step:1136/5100 train_loss:3.7462 train_time:173328ms step_avg:153.93ms step:1137/5100 train_loss:3.7850 train_time:173481ms step_avg:153.93ms step:1138/5100 train_loss:3.7016 train_time:173632ms step_avg:153.93ms step:1139/5100 train_loss:3.5931 train_time:173785ms step_avg:153.93ms step:1140/5100 train_loss:3.9029 train_time:174060ms step_avg:154.04ms step:1141/5100 train_loss:3.7084 train_time:174210ms step_avg:154.03ms step:1142/5100 train_loss:3.8110 train_time:174362ms step_avg:154.03ms step:1143/5100 train_loss:3.6985 train_time:174514ms step_avg:154.03ms step:1144/5100 train_loss:3.6032 train_time:174666ms step_avg:154.03ms step:1145/5100 train_loss:3.7090 train_time:174817ms step_avg:154.02ms step:1146/5100 train_loss:3.8349 train_time:174969ms step_avg:154.02ms step:1147/5100 train_loss:3.7954 train_time:175122ms step_avg:154.02ms step:1148/5100 train_loss:3.7196 train_time:175276ms step_avg:154.02ms step:1149/5100 train_loss:3.7374 train_time:175430ms step_avg:154.02ms step:1150/5100 train_loss:3.5900 train_time:175582ms step_avg:154.02ms step:1151/5100 train_loss:3.6060 train_time:175734ms step_avg:154.02ms step:1152/5100 train_loss:3.5773 train_time:175886ms step_avg:154.02ms step:1153/5100 train_loss:3.7298 train_time:176038ms step_avg:154.01ms step:1154/5100 train_loss:3.6997 train_time:176190ms step_avg:154.01ms step:1155/5100 train_loss:3.7538 train_time:176343ms step_avg:154.01ms step:1156/5100 train_loss:3.6010 train_time:176496ms step_avg:154.01ms step:1157/5100 train_loss:3.7763 train_time:176648ms step_avg:154.01ms step:1158/5100 train_loss:3.7374 train_time:176800ms step_avg:154.01ms step:1159/5100 train_loss:3.5414 train_time:176952ms step_avg:154.01ms step:1160/5100 train_loss:3.5771 train_time:177106ms step_avg:154.00ms step:1161/5100 train_loss:3.5712 train_time:177259ms step_avg:154.00ms step:1162/5100 train_loss:3.3907 train_time:177411ms step_avg:154.00ms step:1163/5100 train_loss:3.6812 train_time:177563ms step_avg:154.00ms step:1164/5100 train_loss:3.6580 train_time:177715ms step_avg:154.00ms step:1165/5100 train_loss:3.5229 train_time:177868ms step_avg:154.00ms step:1166/5100 train_loss:3.5165 train_time:178020ms step_avg:154.00ms step:1167/5100 train_loss:3.6222 train_time:178173ms step_avg:154.00ms step:1168/5100 train_loss:3.6305 train_time:178325ms step_avg:153.99ms step:1169/5100 train_loss:3.9521 train_time:178478ms step_avg:153.99ms step:1170/5100 train_loss:3.6377 train_time:178630ms step_avg:153.99ms step:1171/5100 train_loss:3.6493 train_time:178783ms step_avg:153.99ms step:1172/5100 train_loss:3.5648 train_time:178935ms step_avg:153.99ms step:1173/5100 train_loss:3.6517 train_time:179087ms step_avg:153.99ms step:1174/5100 train_loss:3.7893 train_time:179239ms step_avg:153.99ms step:1175/5100 train_loss:3.6260 train_time:179391ms step_avg:153.98ms step:1176/5100 train_loss:3.6504 train_time:179544ms step_avg:153.98ms step:1177/5100 train_loss:3.6969 train_time:179696ms step_avg:153.98ms step:1178/5100 train_loss:3.6818 train_time:179848ms step_avg:153.98ms step:1179/5100 train_loss:3.7397 train_time:180001ms step_avg:153.98ms step:1180/5100 train_loss:3.6496 train_time:180154ms step_avg:153.98ms step:1181/5100 train_loss:3.6563 train_time:180307ms step_avg:153.98ms step:1182/5100 train_loss:3.5931 train_time:180459ms step_avg:153.98ms step:1183/5100 train_loss:3.6539 train_time:180613ms step_avg:153.98ms step:1184/5100 train_loss:3.5818 train_time:180765ms step_avg:153.97ms step:1185/5100 train_loss:3.7465 train_time:180917ms step_avg:153.97ms step:1186/5100 train_loss:3.8087 train_time:181069ms step_avg:153.97ms step:1187/5100 train_loss:3.6053 train_time:181222ms step_avg:153.97ms step:1188/5100 train_loss:3.6597 train_time:181376ms step_avg:153.97ms step:1189/5100 train_loss:3.6824 train_time:181527ms step_avg:153.97ms step:1190/5100 train_loss:3.5246 train_time:181680ms step_avg:153.97ms step:1191/5100 train_loss:3.7017 train_time:181833ms step_avg:153.97ms step:1192/5100 train_loss:3.8458 train_time:181986ms step_avg:153.96ms step:1193/5100 train_loss:3.6435 train_time:182138ms step_avg:153.96ms step:1194/5100 train_loss:3.5303 train_time:182290ms step_avg:153.96ms step:1195/5100 train_loss:3.8209 train_time:182443ms step_avg:153.96ms step:1196/5100 train_loss:3.6206 train_time:182596ms step_avg:153.96ms step:1197/5100 train_loss:3.6324 train_time:182748ms step_avg:153.96ms step:1198/5100 train_loss:3.5327 train_time:182901ms step_avg:153.96ms step:1199/5100 train_loss:3.5425 train_time:183053ms step_avg:153.96ms step:1200/5100 train_loss:3.5951 train_time:183207ms step_avg:153.96ms step:1201/5100 train_loss:3.6771 train_time:183358ms step_avg:153.95ms step:1202/5100 train_loss:3.7520 train_time:183511ms step_avg:153.95ms step:1203/5100 train_loss:3.7678 train_time:183663ms step_avg:153.95ms step:1204/5100 train_loss:3.6627 train_time:183816ms step_avg:153.95ms step:1205/5100 train_loss:3.5833 train_time:183968ms step_avg:153.95ms step:1206/5100 train_loss:3.6691 train_time:184121ms step_avg:153.95ms step:1207/5100 train_loss:3.7239 train_time:184274ms step_avg:153.95ms step:1208/5100 train_loss:3.7750 train_time:184426ms step_avg:153.94ms step:1209/5100 train_loss:3.6488 train_time:184578ms step_avg:153.94ms step:1210/5100 train_loss:3.5042 train_time:184731ms step_avg:153.94ms step:1211/5100 train_loss:3.5602 train_time:184883ms step_avg:153.94ms step:1212/5100 train_loss:3.6546 train_time:185036ms step_avg:153.94ms step:1213/5100 train_loss:3.6663 train_time:185188ms step_avg:153.94ms step:1214/5100 train_loss:3.7003 train_time:185341ms step_avg:153.94ms step:1215/5100 train_loss:3.5884 train_time:185494ms step_avg:153.94ms step:1216/5100 train_loss:3.6481 train_time:185646ms step_avg:153.94ms step:1217/5100 train_loss:3.5907 train_time:185798ms step_avg:153.93ms step:1218/5100 train_loss:3.5845 train_time:185952ms step_avg:153.93ms step:1219/5100 train_loss:3.6815 train_time:186105ms step_avg:153.93ms step:1220/5100 train_loss:3.5319 train_time:186257ms step_avg:153.93ms step:1221/5100 train_loss:3.7464 train_time:186410ms step_avg:153.93ms step:1222/5100 train_loss:3.7689 train_time:186562ms step_avg:153.93ms step:1223/5100 train_loss:3.7134 train_time:186714ms step_avg:153.93ms step:1224/5100 train_loss:3.5472 train_time:186867ms step_avg:153.93ms step:1225/5100 train_loss:3.5407 train_time:187019ms step_avg:153.93ms step:1226/5100 train_loss:3.6197 train_time:187172ms step_avg:153.92ms step:1227/5100 train_loss:3.6069 train_time:187324ms step_avg:153.92ms step:1228/5100 train_loss:3.5398 train_time:187479ms step_avg:153.92ms step:1229/5100 train_loss:3.7142 train_time:187631ms step_avg:153.92ms step:1230/5100 train_loss:3.6343 train_time:187783ms step_avg:153.92ms step:1231/5100 train_loss:3.6896 train_time:187936ms step_avg:153.92ms step:1232/5100 train_loss:3.8495 train_time:188088ms step_avg:153.92ms step:1233/5100 train_loss:3.7494 train_time:188242ms step_avg:153.92ms step:1234/5100 train_loss:3.6886 train_time:188395ms step_avg:153.92ms step:1235/5100 train_loss:3.8405 train_time:188548ms step_avg:153.92ms step:1236/5100 train_loss:3.6008 train_time:188700ms step_avg:153.92ms step:1237/5100 train_loss:3.5661 train_time:188853ms step_avg:153.91ms step:1238/5100 train_loss:3.5163 train_time:189005ms step_avg:153.91ms step:1239/5100 train_loss:3.5903 train_time:189158ms step_avg:153.91ms step:1240/5100 train_loss:3.5976 train_time:189311ms step_avg:153.91ms step:1241/5100 train_loss:3.6441 train_time:189463ms step_avg:153.91ms step:1242/5100 train_loss:3.6914 train_time:189615ms step_avg:153.91ms step:1243/5100 train_loss:3.5609 train_time:189768ms step_avg:153.91ms step:1244/5100 train_loss:3.6588 train_time:189920ms step_avg:153.91ms step:1245/5100 train_loss:3.6749 train_time:190073ms step_avg:153.91ms step:1246/5100 train_loss:3.6781 train_time:190226ms step_avg:153.90ms step:1247/5100 train_loss:3.5046 train_time:190379ms step_avg:153.90ms step:1248/5100 train_loss:3.6447 train_time:190532ms step_avg:153.90ms step:1249/5100 train_loss:3.7114 train_time:190684ms step_avg:153.90ms step:1250/5100 train_loss:3.6708 train_time:190836ms step_avg:153.90ms step:1250/5100 val_loss:3.6258 train_time:190899ms step_avg:153.95ms step:1251/5100 train_loss:3.5697 train_time:190994ms step_avg:153.90ms step:1252/5100 train_loss:3.7782 train_time:191150ms step_avg:153.91ms step:1253/5100 train_loss:3.6464 train_time:191302ms step_avg:153.90ms step:1254/5100 train_loss:3.5740 train_time:191454ms step_avg:153.90ms step:1255/5100 train_loss:3.7048 train_time:191606ms step_avg:153.90ms step:1256/5100 train_loss:3.7743 train_time:191758ms step_avg:153.90ms step:1257/5100 train_loss:3.5774 train_time:191909ms step_avg:153.90ms step:1258/5100 train_loss:3.6082 train_time:192064ms step_avg:153.90ms step:1259/5100 train_loss:3.6408 train_time:192216ms step_avg:153.90ms step:1260/5100 train_loss:3.6098 train_time:192369ms step_avg:153.90ms step:1261/5100 train_loss:3.4706 train_time:192521ms step_avg:153.89ms step:1262/5100 train_loss:3.5693 train_time:192674ms step_avg:153.89ms step:1263/5100 train_loss:3.6403 train_time:192826ms step_avg:153.89ms step:1264/5100 train_loss:3.4816 train_time:192978ms step_avg:153.89ms step:1265/5100 train_loss:3.7059 train_time:193132ms step_avg:153.89ms step:1266/5100 train_loss:3.6938 train_time:193284ms step_avg:153.89ms step:1267/5100 train_loss:3.6921 train_time:193437ms step_avg:153.89ms step:1268/5100 train_loss:3.6323 train_time:193589ms step_avg:153.89ms step:1269/5100 train_loss:3.6759 train_time:193741ms step_avg:153.88ms step:1270/5100 train_loss:3.5243 train_time:193893ms step_avg:153.88ms step:1271/5100 train_loss:3.3741 train_time:194045ms step_avg:153.88ms step:1272/5100 train_loss:3.6532 train_time:194198ms step_avg:153.88ms step:1273/5100 train_loss:3.6227 train_time:194351ms step_avg:153.88ms step:1274/5100 train_loss:3.6791 train_time:194504ms step_avg:153.88ms step:1275/5100 train_loss:3.6158 train_time:194656ms step_avg:153.88ms step:1276/5100 train_loss:3.7154 train_time:194809ms step_avg:153.88ms step:1277/5100 train_loss:3.7375 train_time:194961ms step_avg:153.88ms step:1278/5100 train_loss:3.6975 train_time:195112ms step_avg:153.87ms step:1279/5100 train_loss:3.6845 train_time:195265ms step_avg:153.87ms step:1280/5100 train_loss:3.5156 train_time:195420ms step_avg:153.87ms step:1281/5100 train_loss:3.6328 train_time:195571ms step_avg:153.87ms step:1282/5100 train_loss:3.6985 train_time:195724ms step_avg:153.87ms step:1283/5100 train_loss:3.7363 train_time:195877ms step_avg:153.87ms step:1284/5100 train_loss:3.6246 train_time:196029ms step_avg:153.87ms step:1285/5100 train_loss:3.6434 train_time:196181ms step_avg:153.87ms step:1286/5100 train_loss:3.6307 train_time:196334ms step_avg:153.87ms step:1287/5100 train_loss:3.6068 train_time:196486ms step_avg:153.87ms step:1288/5100 train_loss:3.7431 train_time:196639ms step_avg:153.86ms step:1289/5100 train_loss:3.5835 train_time:196792ms step_avg:153.86ms step:1290/5100 train_loss:3.6548 train_time:196945ms step_avg:153.86ms step:1291/5100 train_loss:3.7303 train_time:197097ms step_avg:153.86ms step:1292/5100 train_loss:3.6548 train_time:197251ms step_avg:153.86ms step:1293/5100 train_loss:3.7580 train_time:197403ms step_avg:153.86ms step:1294/5100 train_loss:3.7775 train_time:197556ms step_avg:153.86ms step:1295/5100 train_loss:3.7435 train_time:197709ms step_avg:153.86ms step:1296/5100 train_loss:3.5577 train_time:197862ms step_avg:153.86ms step:1297/5100 train_loss:3.6307 train_time:198013ms step_avg:153.86ms step:1298/5100 train_loss:3.5304 train_time:198167ms step_avg:153.86ms step:1299/5100 train_loss:3.6068 train_time:198319ms step_avg:153.86ms step:1300/5100 train_loss:3.6718 train_time:198472ms step_avg:153.85ms step:1301/5100 train_loss:3.6766 train_time:198624ms step_avg:153.85ms step:1302/5100 train_loss:3.6795 train_time:198777ms step_avg:153.85ms step:1303/5100 train_loss:3.8312 train_time:198929ms step_avg:153.85ms step:1304/5100 train_loss:3.6099 train_time:199081ms step_avg:153.85ms step:1305/5100 train_loss:3.8307 train_time:199234ms step_avg:153.85ms step:1306/5100 train_loss:3.5462 train_time:199387ms step_avg:153.85ms step:1307/5100 train_loss:3.7342 train_time:199539ms step_avg:153.85ms step:1308/5100 train_loss:3.7311 train_time:199691ms step_avg:153.85ms step:1309/5100 train_loss:3.5913 train_time:199844ms step_avg:153.84ms step:1310/5100 train_loss:3.5616 train_time:199997ms step_avg:153.84ms step:1311/5100 train_loss:3.5831 train_time:200150ms step_avg:153.84ms step:1312/5100 train_loss:3.5581 train_time:200304ms step_avg:153.84ms step:1313/5100 train_loss:3.6836 train_time:200456ms step_avg:153.84ms step:1314/5100 train_loss:3.6254 train_time:200608ms step_avg:153.84ms step:1315/5100 train_loss:3.3409 train_time:200760ms step_avg:153.84ms step:1316/5100 train_loss:3.5744 train_time:200913ms step_avg:153.84ms step:1317/5100 train_loss:3.6523 train_time:201066ms step_avg:153.84ms step:1318/5100 train_loss:3.6821 train_time:201218ms step_avg:153.84ms step:1319/5100 train_loss:3.5471 train_time:201371ms step_avg:153.84ms step:1320/5100 train_loss:3.6951 train_time:201524ms step_avg:153.83ms step:1321/5100 train_loss:3.7445 train_time:201677ms step_avg:153.83ms step:1322/5100 train_loss:3.6365 train_time:201829ms step_avg:153.83ms step:1323/5100 train_loss:3.5809 train_time:202079ms step_avg:153.91ms step:1324/5100 train_loss:3.6093 train_time:202241ms step_avg:153.91ms step:1325/5100 train_loss:3.7118 train_time:202392ms step_avg:153.91ms step:1326/5100 train_loss:3.7625 train_time:202545ms step_avg:153.91ms step:1327/5100 train_loss:3.5257 train_time:202697ms step_avg:153.91ms step:1328/5100 train_loss:3.4394 train_time:202848ms step_avg:153.91ms step:1329/5100 train_loss:3.7467 train_time:203034ms step_avg:153.93ms step:1330/5100 train_loss:3.6010 train_time:203279ms step_avg:154.00ms step:1331/5100 train_loss:3.7181 train_time:203430ms step_avg:154.00ms step:1332/5100 train_loss:3.6322 train_time:203581ms step_avg:154.00ms step:1333/5100 train_loss:4.0363 train_time:203733ms step_avg:153.99ms step:1334/5100 train_loss:3.7223 train_time:203885ms step_avg:153.99ms step:1335/5100 train_loss:3.6383 train_time:204037ms step_avg:153.99ms step:1336/5100 train_loss:3.5861 train_time:204189ms step_avg:153.99ms step:1337/5100 train_loss:3.5744 train_time:204345ms step_avg:153.99ms step:1338/5100 train_loss:3.8343 train_time:204499ms step_avg:153.99ms step:1339/5100 train_loss:3.7755 train_time:204652ms step_avg:153.99ms step:1340/5100 train_loss:3.6164 train_time:204805ms step_avg:153.99ms step:1341/5100 train_loss:3.5699 train_time:204957ms step_avg:153.99ms step:1342/5100 train_loss:3.8776 train_time:205109ms step_avg:153.99ms step:1343/5100 train_loss:3.6461 train_time:205262ms step_avg:153.99ms step:1344/5100 train_loss:3.6445 train_time:205415ms step_avg:153.98ms step:1345/5100 train_loss:3.7024 train_time:205569ms step_avg:153.98ms step:1346/5100 train_loss:3.6692 train_time:205723ms step_avg:153.98ms step:1347/5100 train_loss:3.5639 train_time:205876ms step_avg:153.98ms step:1348/5100 train_loss:3.5151 train_time:206029ms step_avg:153.98ms step:1349/5100 train_loss:3.6122 train_time:206181ms step_avg:153.98ms step:1350/5100 train_loss:3.5451 train_time:206333ms step_avg:153.98ms step:1351/5100 train_loss:3.6810 train_time:206487ms step_avg:153.98ms step:1352/5100 train_loss:3.5269 train_time:206640ms step_avg:153.98ms step:1353/5100 train_loss:3.5878 train_time:206793ms step_avg:153.98ms step:1354/5100 train_loss:3.7026 train_time:206947ms step_avg:153.98ms step:1355/5100 train_loss:3.5334 train_time:207100ms step_avg:153.98ms step:1356/5100 train_loss:3.4639 train_time:207252ms step_avg:153.98ms step:1357/5100 train_loss:3.8083 train_time:207405ms step_avg:153.98ms step:1358/5100 train_loss:3.7389 train_time:207558ms step_avg:153.97ms step:1359/5100 train_loss:3.4521 train_time:207710ms step_avg:153.97ms step:1360/5100 train_loss:3.7375 train_time:207864ms step_avg:153.97ms step:1361/5100 train_loss:3.6219 train_time:208017ms step_avg:153.97ms step:1362/5100 train_loss:3.4861 train_time:208171ms step_avg:153.97ms step:1363/5100 train_loss:3.6605 train_time:208324ms step_avg:153.97ms step:1364/5100 train_loss:3.5507 train_time:208477ms step_avg:153.97ms step:1365/5100 train_loss:3.5805 train_time:208630ms step_avg:153.97ms step:1366/5100 train_loss:3.5996 train_time:208782ms step_avg:153.97ms step:1367/5100 train_loss:3.7079 train_time:208936ms step_avg:153.97ms step:1368/5100 train_loss:3.6822 train_time:209089ms step_avg:153.97ms step:1369/5100 train_loss:3.6367 train_time:209242ms step_avg:153.97ms step:1370/5100 train_loss:3.5426 train_time:209395ms step_avg:153.97ms step:1371/5100 train_loss:3.8691 train_time:209550ms step_avg:153.97ms step:1372/5100 train_loss:3.6153 train_time:209702ms step_avg:153.97ms step:1373/5100 train_loss:3.6441 train_time:209854ms step_avg:153.96ms step:1374/5100 train_loss:3.6457 train_time:210008ms step_avg:153.96ms step:1375/5100 train_loss:3.4445 train_time:210161ms step_avg:153.96ms step:1375/5100 val_loss:3.6055 train_time:210225ms step_avg:154.01ms step:1376/5100 train_loss:3.8478 train_time:210321ms step_avg:153.97ms step:1377/5100 train_loss:3.6222 train_time:210478ms step_avg:153.97ms step:1378/5100 train_loss:3.7697 train_time:210630ms step_avg:153.97ms step:1379/5100 train_loss:3.8111 train_time:210782ms step_avg:153.97ms step:1380/5100 train_loss:3.4874 train_time:210934ms step_avg:153.97ms step:1381/5100 train_loss:3.6245 train_time:211086ms step_avg:153.97ms step:1382/5100 train_loss:4.0579 train_time:211238ms step_avg:153.96ms step:1383/5100 train_loss:3.5226 train_time:211391ms step_avg:153.96ms step:1384/5100 train_loss:3.6832 train_time:211545ms step_avg:153.96ms step:1385/5100 train_loss:3.7572 train_time:211698ms step_avg:153.96ms step:1386/5100 train_loss:3.6648 train_time:211850ms step_avg:153.96ms step:1387/5100 train_loss:3.6786 train_time:212002ms step_avg:153.96ms step:1388/5100 train_loss:3.4885 train_time:212154ms step_avg:153.96ms step:1389/5100 train_loss:3.6316 train_time:212306ms step_avg:153.96ms step:1390/5100 train_loss:3.6025 train_time:212458ms step_avg:153.96ms step:1391/5100 train_loss:3.8691 train_time:212611ms step_avg:153.95ms step:1392/5100 train_loss:3.5801 train_time:212765ms step_avg:153.95ms step:1393/5100 train_loss:3.5722 train_time:212917ms step_avg:153.95ms step:1394/5100 train_loss:3.5414 train_time:213069ms step_avg:153.95ms step:1395/5100 train_loss:3.8284 train_time:213222ms step_avg:153.95ms step:1396/5100 train_loss:3.7150 train_time:213374ms step_avg:153.95ms step:1397/5100 train_loss:3.7222 train_time:213526ms step_avg:153.95ms step:1398/5100 train_loss:3.5887 train_time:213678ms step_avg:153.95ms step:1399/5100 train_loss:3.5593 train_time:213831ms step_avg:153.95ms step:1400/5100 train_loss:3.6212 train_time:213983ms step_avg:153.94ms step:1401/5100 train_loss:3.6003 train_time:214136ms step_avg:153.94ms step:1402/5100 train_loss:3.6235 train_time:214289ms step_avg:153.94ms step:1403/5100 train_loss:3.5876 train_time:214442ms step_avg:153.94ms step:1404/5100 train_loss:3.8187 train_time:214595ms step_avg:153.94ms step:1405/5100 train_loss:3.5602 train_time:214746ms step_avg:153.94ms step:1406/5100 train_loss:3.6049 train_time:214900ms step_avg:153.94ms step:1407/5100 train_loss:3.6021 train_time:215052ms step_avg:153.94ms step:1408/5100 train_loss:3.4759 train_time:215205ms step_avg:153.94ms step:1409/5100 train_loss:3.5854 train_time:215358ms step_avg:153.94ms step:1410/5100 train_loss:3.5684 train_time:215510ms step_avg:153.94ms step:1411/5100 train_loss:3.5677 train_time:215665ms step_avg:153.94ms step:1412/5100 train_loss:3.6602 train_time:215817ms step_avg:153.94ms step:1413/5100 train_loss:3.5905 train_time:215970ms step_avg:153.93ms step:1414/5100 train_loss:3.6396 train_time:216122ms step_avg:153.93ms step:1415/5100 train_loss:3.6390 train_time:216275ms step_avg:153.93ms step:1416/5100 train_loss:3.7187 train_time:216427ms step_avg:153.93ms step:1417/5100 train_loss:3.5151 train_time:216579ms step_avg:153.93ms step:1418/5100 train_loss:3.5788 train_time:216732ms step_avg:153.93ms step:1419/5100 train_loss:3.6686 train_time:216884ms step_avg:153.93ms step:1420/5100 train_loss:3.7031 train_time:217038ms step_avg:153.93ms step:1421/5100 train_loss:3.6727 train_time:217189ms step_avg:153.93ms step:1422/5100 train_loss:3.6620 train_time:217342ms step_avg:153.93ms step:1423/5100 train_loss:3.6450 train_time:217495ms step_avg:153.92ms step:1424/5100 train_loss:3.6362 train_time:217647ms step_avg:153.92ms step:1425/5100 train_loss:3.6191 train_time:217800ms step_avg:153.92ms step:1426/5100 train_loss:3.4970 train_time:217952ms step_avg:153.92ms step:1427/5100 train_loss:3.6148 train_time:218105ms step_avg:153.92ms step:1428/5100 train_loss:3.5561 train_time:218259ms step_avg:153.92ms step:1429/5100 train_loss:3.6665 train_time:218410ms step_avg:153.92ms step:1430/5100 train_loss:3.6388 train_time:218564ms step_avg:153.92ms step:1431/5100 train_loss:3.5669 train_time:218717ms step_avg:153.92ms step:1432/5100 train_loss:3.6117 train_time:218870ms step_avg:153.92ms step:1433/5100 train_loss:3.6418 train_time:219022ms step_avg:153.92ms step:1434/5100 train_loss:3.5264 train_time:219175ms step_avg:153.91ms step:1435/5100 train_loss:3.6152 train_time:219327ms step_avg:153.91ms step:1436/5100 train_loss:3.4392 train_time:219479ms step_avg:153.91ms step:1437/5100 train_loss:3.5102 train_time:219632ms step_avg:153.91ms step:1438/5100 train_loss:3.7030 train_time:219784ms step_avg:153.91ms step:1439/5100 train_loss:3.6640 train_time:219937ms step_avg:153.91ms step:1440/5100 train_loss:3.6104 train_time:220090ms step_avg:153.91ms step:1441/5100 train_loss:3.4625 train_time:220243ms step_avg:153.91ms step:1442/5100 train_loss:3.6371 train_time:220396ms step_avg:153.91ms step:1443/5100 train_loss:3.6959 train_time:220547ms step_avg:153.91ms step:1444/5100 train_loss:3.7690 train_time:220701ms step_avg:153.91ms step:1445/5100 train_loss:3.7395 train_time:220852ms step_avg:153.90ms step:1446/5100 train_loss:3.6261 train_time:221005ms step_avg:153.90ms step:1447/5100 train_loss:3.4946 train_time:221158ms step_avg:153.90ms step:1448/5100 train_loss:3.5686 train_time:221310ms step_avg:153.90ms step:1449/5100 train_loss:3.5855 train_time:221464ms step_avg:153.90ms step:1450/5100 train_loss:3.7118 train_time:221617ms step_avg:153.90ms step:1451/5100 train_loss:3.7003 train_time:221769ms step_avg:153.90ms step:1452/5100 train_loss:3.5169 train_time:221921ms step_avg:153.90ms step:1453/5100 train_loss:3.6248 train_time:222073ms step_avg:153.90ms step:1454/5100 train_loss:3.5402 train_time:222226ms step_avg:153.90ms step:1455/5100 train_loss:3.5747 train_time:222378ms step_avg:153.89ms step:1456/5100 train_loss:3.6176 train_time:222530ms step_avg:153.89ms step:1457/5100 train_loss:3.5526 train_time:222683ms step_avg:153.89ms step:1458/5100 train_loss:3.4525 train_time:222835ms step_avg:153.89ms step:1459/5100 train_loss:3.6977 train_time:222988ms step_avg:153.89ms step:1460/5100 train_loss:3.5650 train_time:223143ms step_avg:153.89ms step:1461/5100 train_loss:3.6150 train_time:223295ms step_avg:153.89ms step:1462/5100 train_loss:3.7437 train_time:223447ms step_avg:153.89ms step:1463/5100 train_loss:3.5619 train_time:223600ms step_avg:153.89ms step:1464/5100 train_loss:3.7507 train_time:223752ms step_avg:153.89ms step:1465/5100 train_loss:3.6450 train_time:223904ms step_avg:153.89ms step:1466/5100 train_loss:3.6540 train_time:224058ms step_avg:153.89ms step:1467/5100 train_loss:3.5637 train_time:224210ms step_avg:153.88ms step:1468/5100 train_loss:3.7220 train_time:224365ms step_avg:153.89ms step:1469/5100 train_loss:3.5855 train_time:224518ms step_avg:153.88ms step:1470/5100 train_loss:3.5571 train_time:224671ms step_avg:153.88ms step:1471/5100 train_loss:3.6121 train_time:224824ms step_avg:153.88ms step:1472/5100 train_loss:3.5372 train_time:224975ms step_avg:153.88ms step:1473/5100 train_loss:3.6215 train_time:225127ms step_avg:153.88ms step:1474/5100 train_loss:3.7225 train_time:225280ms step_avg:153.88ms step:1475/5100 train_loss:3.6007 train_time:225432ms step_avg:153.88ms step:1476/5100 train_loss:3.4238 train_time:225585ms step_avg:153.88ms step:1477/5100 train_loss:3.5539 train_time:225738ms step_avg:153.88ms step:1478/5100 train_loss:3.5227 train_time:225891ms step_avg:153.88ms step:1479/5100 train_loss:3.6121 train_time:226044ms step_avg:153.88ms step:1480/5100 train_loss:3.6900 train_time:226196ms step_avg:153.88ms step:1481/5100 train_loss:3.5649 train_time:226348ms step_avg:153.87ms step:1482/5100 train_loss:3.7385 train_time:226502ms step_avg:153.87ms step:1483/5100 train_loss:3.6667 train_time:226653ms step_avg:153.87ms step:1484/5100 train_loss:3.5669 train_time:226805ms step_avg:153.87ms step:1485/5100 train_loss:3.5564 train_time:226958ms step_avg:153.87ms step:1486/5100 train_loss:3.5535 train_time:227111ms step_avg:153.87ms step:1487/5100 train_loss:3.5317 train_time:227264ms step_avg:153.87ms step:1488/5100 train_loss:3.6167 train_time:227417ms step_avg:153.87ms step:1489/5100 train_loss:3.5247 train_time:227569ms step_avg:153.87ms step:1490/5100 train_loss:3.6156 train_time:227723ms step_avg:153.87ms step:1491/5100 train_loss:3.5543 train_time:227875ms step_avg:153.87ms step:1492/5100 train_loss:3.4756 train_time:228026ms step_avg:153.86ms step:1493/5100 train_loss:3.5431 train_time:228180ms step_avg:153.86ms step:1494/5100 train_loss:3.7250 train_time:228333ms step_avg:153.86ms step:1495/5100 train_loss:3.5795 train_time:228484ms step_avg:153.86ms step:1496/5100 train_loss:3.3430 train_time:228637ms step_avg:153.86ms step:1497/5100 train_loss:3.6400 train_time:228789ms step_avg:153.86ms step:1498/5100 train_loss:3.6092 train_time:228942ms step_avg:153.86ms step:1499/5100 train_loss:3.6518 train_time:229095ms step_avg:153.86ms step:1500/5100 train_loss:3.6032 train_time:229246ms step_avg:153.86ms step:1500/5100 val_loss:3.5799 train_time:229311ms step_avg:153.90ms step:1501/5100 train_loss:3.5825 train_time:229402ms step_avg:153.86ms step:1502/5100 train_loss:3.3755 train_time:229560ms step_avg:153.86ms step:1503/5100 train_loss:3.6581 train_time:229714ms step_avg:153.86ms step:1504/5100 train_loss:3.5342 train_time:229866ms step_avg:153.86ms step:1505/5100 train_loss:3.5406 train_time:230017ms step_avg:153.86ms step:1506/5100 train_loss:3.4972 train_time:230170ms step_avg:153.86ms step:1507/5100 train_loss:3.5792 train_time:230321ms step_avg:153.86ms step:1508/5100 train_loss:3.4983 train_time:230474ms step_avg:153.85ms step:1509/5100 train_loss:3.8152 train_time:230628ms step_avg:153.85ms step:1510/5100 train_loss:3.5498 train_time:230780ms step_avg:153.85ms step:1511/5100 train_loss:3.5564 train_time:230933ms step_avg:153.85ms step:1512/5100 train_loss:3.6827 train_time:231188ms step_avg:153.92ms step:1513/5100 train_loss:3.7101 train_time:231347ms step_avg:153.92ms step:1514/5100 train_loss:3.5680 train_time:231499ms step_avg:153.92ms step:1515/5100 train_loss:3.4038 train_time:231651ms step_avg:153.92ms step:1516/5100 train_loss:3.5336 train_time:231802ms step_avg:153.92ms step:1517/5100 train_loss:3.5394 train_time:231954ms step_avg:153.92ms step:1518/5100 train_loss:3.6222 train_time:232106ms step_avg:153.92ms step:1519/5100 train_loss:3.5087 train_time:232262ms step_avg:153.92ms step:1520/5100 train_loss:3.7948 train_time:232537ms step_avg:154.00ms step:1521/5100 train_loss:3.4612 train_time:232687ms step_avg:154.00ms step:1522/5100 train_loss:3.5220 train_time:232839ms step_avg:153.99ms step:1523/5100 train_loss:3.6638 train_time:232992ms step_avg:153.99ms step:1524/5100 train_loss:3.5205 train_time:233144ms step_avg:153.99ms step:1525/5100 train_loss:3.6160 train_time:233297ms step_avg:153.99ms step:1526/5100 train_loss:3.6163 train_time:233448ms step_avg:153.99ms step:1527/5100 train_loss:3.5808 train_time:233603ms step_avg:153.99ms step:1528/5100 train_loss:3.5734 train_time:233757ms step_avg:153.99ms step:1529/5100 train_loss:3.7243 train_time:233911ms step_avg:153.99ms step:1530/5100 train_loss:3.6955 train_time:234063ms step_avg:153.99ms step:1531/5100 train_loss:3.5305 train_time:234217ms step_avg:153.99ms step:1532/5100 train_loss:3.4887 train_time:234369ms step_avg:153.99ms step:1533/5100 train_loss:3.6443 train_time:234521ms step_avg:153.99ms step:1534/5100 train_loss:3.5938 train_time:234675ms step_avg:153.99ms step:1535/5100 train_loss:3.5801 train_time:234828ms step_avg:153.99ms step:1536/5100 train_loss:3.5746 train_time:234980ms step_avg:153.98ms step:1537/5100 train_loss:3.5215 train_time:235132ms step_avg:153.98ms step:1538/5100 train_loss:3.5765 train_time:235285ms step_avg:153.98ms step:1539/5100 train_loss:3.7547 train_time:235437ms step_avg:153.98ms step:1540/5100 train_loss:3.6825 train_time:235591ms step_avg:153.98ms step:1541/5100 train_loss:3.5878 train_time:235742ms step_avg:153.98ms step:1542/5100 train_loss:3.5432 train_time:235895ms step_avg:153.98ms step:1543/5100 train_loss:3.5426 train_time:236048ms step_avg:153.98ms step:1544/5100 train_loss:3.5011 train_time:236200ms step_avg:153.98ms step:1545/5100 train_loss:3.5976 train_time:236353ms step_avg:153.98ms step:1546/5100 train_loss:3.5663 train_time:236505ms step_avg:153.97ms step:1547/5100 train_loss:3.5384 train_time:236659ms step_avg:153.97ms step:1548/5100 train_loss:3.5054 train_time:236811ms step_avg:153.97ms step:1549/5100 train_loss:3.5365 train_time:236963ms step_avg:153.97ms step:1550/5100 train_loss:3.6544 train_time:237116ms step_avg:153.97ms step:1551/5100 train_loss:3.5754 train_time:237268ms step_avg:153.97ms step:1552/5100 train_loss:3.5161 train_time:237420ms step_avg:153.97ms step:1553/5100 train_loss:3.5097 train_time:237573ms step_avg:153.97ms step:1554/5100 train_loss:3.5035 train_time:237726ms step_avg:153.97ms step:1555/5100 train_loss:3.6344 train_time:237879ms step_avg:153.97ms step:1556/5100 train_loss:3.6301 train_time:238032ms step_avg:153.97ms step:1557/5100 train_loss:3.5680 train_time:238184ms step_avg:153.97ms step:1558/5100 train_loss:3.6234 train_time:238337ms step_avg:153.96ms step:1559/5100 train_loss:3.5431 train_time:238490ms step_avg:153.96ms step:1560/5100 train_loss:3.4651 train_time:238642ms step_avg:153.96ms step:1561/5100 train_loss:3.7069 train_time:238794ms step_avg:153.96ms step:1562/5100 train_loss:3.5257 train_time:238947ms step_avg:153.96ms step:1563/5100 train_loss:3.5024 train_time:239099ms step_avg:153.96ms step:1564/5100 train_loss:3.6323 train_time:239251ms step_avg:153.96ms step:1565/5100 train_loss:3.4608 train_time:239402ms step_avg:153.96ms step:1566/5100 train_loss:3.5184 train_time:239557ms step_avg:153.96ms step:1567/5100 train_loss:3.6605 train_time:239708ms step_avg:153.96ms step:1568/5100 train_loss:3.5399 train_time:239860ms step_avg:153.95ms step:1569/5100 train_loss:3.5245 train_time:240013ms step_avg:153.95ms step:1570/5100 train_loss:3.6239 train_time:240166ms step_avg:153.95ms step:1571/5100 train_loss:3.6324 train_time:240318ms step_avg:153.95ms step:1572/5100 train_loss:3.4612 train_time:240471ms step_avg:153.95ms step:1573/5100 train_loss:3.4859 train_time:240623ms step_avg:153.95ms step:1574/5100 train_loss:3.6072 train_time:240775ms step_avg:153.95ms step:1575/5100 train_loss:3.4807 train_time:240928ms step_avg:153.95ms step:1576/5100 train_loss:3.6248 train_time:241079ms step_avg:153.95ms step:1577/5100 train_loss:3.5290 train_time:241232ms step_avg:153.94ms step:1578/5100 train_loss:3.5807 train_time:241384ms step_avg:153.94ms step:1579/5100 train_loss:3.5568 train_time:241537ms step_avg:153.94ms step:1580/5100 train_loss:3.5232 train_time:241689ms step_avg:153.94ms step:1581/5100 train_loss:3.4958 train_time:241841ms step_avg:153.94ms step:1582/5100 train_loss:3.7353 train_time:241994ms step_avg:153.94ms step:1583/5100 train_loss:3.5089 train_time:242146ms step_avg:153.94ms step:1584/5100 train_loss:3.6647 train_time:242298ms step_avg:153.94ms step:1585/5100 train_loss:3.4973 train_time:242451ms step_avg:153.94ms step:1586/5100 train_loss:3.6556 train_time:242603ms step_avg:153.94ms step:1587/5100 train_loss:3.4373 train_time:242757ms step_avg:153.94ms step:1588/5100 train_loss:3.6359 train_time:242909ms step_avg:153.93ms step:1589/5100 train_loss:3.5535 train_time:243061ms step_avg:153.93ms step:1590/5100 train_loss:3.7106 train_time:243214ms step_avg:153.93ms step:1591/5100 train_loss:3.5275 train_time:243367ms step_avg:153.93ms step:1592/5100 train_loss:3.5371 train_time:243519ms step_avg:153.93ms step:1593/5100 train_loss:3.6093 train_time:243672ms step_avg:153.93ms step:1594/5100 train_loss:3.5882 train_time:243825ms step_avg:153.93ms step:1595/5100 train_loss:3.5572 train_time:243977ms step_avg:153.93ms step:1596/5100 train_loss:3.7003 train_time:244130ms step_avg:153.93ms step:1597/5100 train_loss:3.4334 train_time:244282ms step_avg:153.93ms step:1598/5100 train_loss:3.5899 train_time:244435ms step_avg:153.93ms step:1599/5100 train_loss:3.6343 train_time:244588ms step_avg:153.93ms step:1600/5100 train_loss:3.6823 train_time:244741ms step_avg:153.92ms step:1601/5100 train_loss:3.5337 train_time:244893ms step_avg:153.92ms step:1602/5100 train_loss:3.8171 train_time:245046ms step_avg:153.92ms step:1603/5100 train_loss:3.7205 train_time:245198ms step_avg:153.92ms step:1604/5100 train_loss:3.4937 train_time:245351ms step_avg:153.92ms step:1605/5100 train_loss:3.5330 train_time:245503ms step_avg:153.92ms step:1606/5100 train_loss:3.4146 train_time:245657ms step_avg:153.92ms step:1607/5100 train_loss:3.7352 train_time:245808ms step_avg:153.92ms step:1608/5100 train_loss:3.5461 train_time:245961ms step_avg:153.92ms step:1609/5100 train_loss:3.5628 train_time:246113ms step_avg:153.92ms step:1610/5100 train_loss:3.5133 train_time:246266ms step_avg:153.92ms step:1611/5100 train_loss:4.1225 train_time:246418ms step_avg:153.92ms step:1612/5100 train_loss:3.7501 train_time:246570ms step_avg:153.91ms step:1613/5100 train_loss:3.6679 train_time:246723ms step_avg:153.91ms step:1614/5100 train_loss:3.5309 train_time:246876ms step_avg:153.91ms step:1615/5100 train_loss:3.5824 train_time:247028ms step_avg:153.91ms step:1616/5100 train_loss:3.5633 train_time:247181ms step_avg:153.91ms step:1617/5100 train_loss:3.5246 train_time:247333ms step_avg:153.91ms step:1618/5100 train_loss:3.6018 train_time:247486ms step_avg:153.91ms step:1619/5100 train_loss:3.5619 train_time:247639ms step_avg:153.91ms step:1620/5100 train_loss:3.4530 train_time:247792ms step_avg:153.91ms step:1621/5100 train_loss:3.7284 train_time:247944ms step_avg:153.91ms step:1622/5100 train_loss:3.6238 train_time:248097ms step_avg:153.91ms step:1623/5100 train_loss:3.4178 train_time:248249ms step_avg:153.90ms step:1624/5100 train_loss:3.5387 train_time:248401ms step_avg:153.90ms step:1625/5100 train_loss:3.4936 train_time:248555ms step_avg:153.90ms step:1625/5100 val_loss:3.5641 train_time:248618ms step_avg:153.94ms step:1626/5100 train_loss:3.5710 train_time:248712ms step_avg:153.91ms step:1627/5100 train_loss:3.5382 train_time:248871ms step_avg:153.91ms step:1628/5100 train_loss:3.4992 train_time:249025ms step_avg:153.91ms step:1629/5100 train_loss:3.6152 train_time:249177ms step_avg:153.91ms step:1630/5100 train_loss:3.5054 train_time:249329ms step_avg:153.91ms step:1631/5100 train_loss:3.5675 train_time:249481ms step_avg:153.91ms step:1632/5100 train_loss:3.4443 train_time:249634ms step_avg:153.90ms step:1633/5100 train_loss:3.4089 train_time:249786ms step_avg:153.90ms step:1634/5100 train_loss:3.5761 train_time:249941ms step_avg:153.90ms step:1635/5100 train_loss:3.5616 train_time:250095ms step_avg:153.90ms step:1636/5100 train_loss:3.4988 train_time:250248ms step_avg:153.90ms step:1637/5100 train_loss:3.5893 train_time:250401ms step_avg:153.90ms step:1638/5100 train_loss:3.6407 train_time:250553ms step_avg:153.90ms step:1639/5100 train_loss:3.6714 train_time:250706ms step_avg:153.90ms step:1640/5100 train_loss:3.8341 train_time:250858ms step_avg:153.90ms step:1641/5100 train_loss:3.6544 train_time:251010ms step_avg:153.90ms step:1642/5100 train_loss:3.5616 train_time:251165ms step_avg:153.90ms step:1643/5100 train_loss:3.6511 train_time:251318ms step_avg:153.90ms step:1644/5100 train_loss:3.5432 train_time:251471ms step_avg:153.90ms step:1645/5100 train_loss:3.5666 train_time:251625ms step_avg:153.90ms step:1646/5100 train_loss:3.5653 train_time:251777ms step_avg:153.90ms step:1647/5100 train_loss:3.3403 train_time:251930ms step_avg:153.90ms step:1648/5100 train_loss:3.5992 train_time:252083ms step_avg:153.90ms step:1649/5100 train_loss:3.4687 train_time:252236ms step_avg:153.90ms step:1650/5100 train_loss:3.5420 train_time:252389ms step_avg:153.90ms step:1651/5100 train_loss:3.5177 train_time:252543ms step_avg:153.90ms step:1652/5100 train_loss:3.5928 train_time:252696ms step_avg:153.90ms step:1653/5100 train_loss:3.5223 train_time:252848ms step_avg:153.89ms step:1654/5100 train_loss:3.6475 train_time:253000ms step_avg:153.89ms step:1655/5100 train_loss:3.6387 train_time:253153ms step_avg:153.89ms step:1656/5100 train_loss:3.4529 train_time:253305ms step_avg:153.89ms step:1657/5100 train_loss:3.6116 train_time:253457ms step_avg:153.89ms step:1658/5100 train_loss:3.5050 train_time:253609ms step_avg:153.89ms step:1659/5100 train_loss:3.4959 train_time:253763ms step_avg:153.89ms step:1660/5100 train_loss:3.5666 train_time:253915ms step_avg:153.89ms step:1661/5100 train_loss:3.5991 train_time:254068ms step_avg:153.89ms step:1662/5100 train_loss:3.5135 train_time:254221ms step_avg:153.89ms step:1663/5100 train_loss:3.6044 train_time:254374ms step_avg:153.89ms step:1664/5100 train_loss:3.6197 train_time:254526ms step_avg:153.89ms step:1665/5100 train_loss:3.6389 train_time:254678ms step_avg:153.88ms step:1666/5100 train_loss:3.6182 train_time:254830ms step_avg:153.88ms step:1667/5100 train_loss:3.7594 train_time:254983ms step_avg:153.88ms step:1668/5100 train_loss:3.4663 train_time:255136ms step_avg:153.88ms step:1669/5100 train_loss:3.5485 train_time:255287ms step_avg:153.88ms step:1670/5100 train_loss:3.4736 train_time:255440ms step_avg:153.88ms step:1671/5100 train_loss:3.4771 train_time:255592ms step_avg:153.88ms step:1672/5100 train_loss:3.6408 train_time:255745ms step_avg:153.88ms step:1673/5100 train_loss:3.8158 train_time:255898ms step_avg:153.88ms step:1674/5100 train_loss:3.5336 train_time:256050ms step_avg:153.88ms step:1675/5100 train_loss:3.5182 train_time:256202ms step_avg:153.88ms step:1676/5100 train_loss:3.4014 train_time:256355ms step_avg:153.87ms step:1677/5100 train_loss:3.6105 train_time:256506ms step_avg:153.87ms step:1678/5100 train_loss:3.5282 train_time:256659ms step_avg:153.87ms step:1679/5100 train_loss:3.5513 train_time:256811ms step_avg:153.87ms step:1680/5100 train_loss:3.5387 train_time:256965ms step_avg:153.87ms step:1681/5100 train_loss:3.3607 train_time:257116ms step_avg:153.87ms step:1682/5100 train_loss:3.5438 train_time:257269ms step_avg:153.87ms step:1683/5100 train_loss:3.5574 train_time:257422ms step_avg:153.87ms step:1684/5100 train_loss:3.5977 train_time:257575ms step_avg:153.87ms step:1685/5100 train_loss:3.6025 train_time:257728ms step_avg:153.87ms step:1686/5100 train_loss:3.5122 train_time:257880ms step_avg:153.87ms step:1687/5100 train_loss:3.6164 train_time:258033ms step_avg:153.87ms step:1688/5100 train_loss:3.4959 train_time:258185ms step_avg:153.86ms step:1689/5100 train_loss:3.5825 train_time:258338ms step_avg:153.86ms step:1690/5100 train_loss:3.4937 train_time:258490ms step_avg:153.86ms step:1691/5100 train_loss:3.3984 train_time:258644ms step_avg:153.86ms step:1692/5100 train_loss:3.5499 train_time:258796ms step_avg:153.86ms step:1693/5100 train_loss:3.5436 train_time:258948ms step_avg:153.86ms step:1694/5100 train_loss:3.4637 train_time:259100ms step_avg:153.86ms step:1695/5100 train_loss:3.9030 train_time:259252ms step_avg:153.86ms step:1696/5100 train_loss:3.6142 train_time:259405ms step_avg:153.86ms step:1697/5100 train_loss:3.5990 train_time:259557ms step_avg:153.86ms step:1698/5100 train_loss:3.5035 train_time:259709ms step_avg:153.86ms step:1699/5100 train_loss:3.4240 train_time:259862ms step_avg:153.86ms step:1700/5100 train_loss:3.5133 train_time:260015ms step_avg:153.86ms step:1701/5100 train_loss:3.5019 train_time:260262ms step_avg:153.91ms step:1702/5100 train_loss:3.5760 train_time:260421ms step_avg:153.91ms step:1703/5100 train_loss:3.5018 train_time:260575ms step_avg:153.91ms step:1704/5100 train_loss:3.7026 train_time:260727ms step_avg:153.91ms step:1705/5100 train_loss:3.4675 train_time:260878ms step_avg:153.91ms step:1706/5100 train_loss:3.6998 train_time:261030ms step_avg:153.91ms step:1707/5100 train_loss:3.5401 train_time:261182ms step_avg:153.91ms step:1708/5100 train_loss:3.3143 train_time:261336ms step_avg:153.91ms step:1709/5100 train_loss:3.6499 train_time:261489ms step_avg:153.91ms step:1710/5100 train_loss:3.5645 train_time:261762ms step_avg:153.98ms step:1711/5100 train_loss:3.5518 train_time:261911ms step_avg:153.97ms step:1712/5100 train_loss:3.5411 train_time:262064ms step_avg:153.97ms step:1713/5100 train_loss:3.5833 train_time:262217ms step_avg:153.97ms step:1714/5100 train_loss:3.6066 train_time:262370ms step_avg:153.97ms step:1715/5100 train_loss:3.5297 train_time:262522ms step_avg:153.97ms step:1716/5100 train_loss:3.5270 train_time:262674ms step_avg:153.97ms step:1717/5100 train_loss:3.3727 train_time:262828ms step_avg:153.97ms step:1718/5100 train_loss:3.5007 train_time:262982ms step_avg:153.97ms step:1719/5100 train_loss:3.5280 train_time:263135ms step_avg:153.97ms step:1720/5100 train_loss:3.4793 train_time:263288ms step_avg:153.97ms step:1721/5100 train_loss:3.6275 train_time:263440ms step_avg:153.97ms step:1722/5100 train_loss:3.4394 train_time:263593ms step_avg:153.97ms step:1723/5100 train_loss:3.5730 train_time:263745ms step_avg:153.97ms step:1724/5100 train_loss:3.6710 train_time:263897ms step_avg:153.97ms step:1725/5100 train_loss:3.5105 train_time:264049ms step_avg:153.96ms step:1726/5100 train_loss:3.7452 train_time:264202ms step_avg:153.96ms step:1727/5100 train_loss:3.5312 train_time:264354ms step_avg:153.96ms step:1728/5100 train_loss:3.5867 train_time:264506ms step_avg:153.96ms step:1729/5100 train_loss:3.5644 train_time:264659ms step_avg:153.96ms step:1730/5100 train_loss:3.5742 train_time:264811ms step_avg:153.96ms step:1731/5100 train_loss:3.9425 train_time:264965ms step_avg:153.96ms step:1732/5100 train_loss:3.5511 train_time:265117ms step_avg:153.96ms step:1733/5100 train_loss:3.6832 train_time:265270ms step_avg:153.96ms step:1734/5100 train_loss:3.4641 train_time:265422ms step_avg:153.96ms step:1735/5100 train_loss:3.5019 train_time:265575ms step_avg:153.96ms step:1736/5100 train_loss:3.5238 train_time:265727ms step_avg:153.96ms step:1737/5100 train_loss:3.5081 train_time:265879ms step_avg:153.95ms step:1738/5100 train_loss:3.6484 train_time:266032ms step_avg:153.95ms step:1739/5100 train_loss:3.5149 train_time:266184ms step_avg:153.95ms step:1740/5100 train_loss:3.5813 train_time:266337ms step_avg:153.95ms step:1741/5100 train_loss:3.6318 train_time:266489ms step_avg:153.95ms step:1742/5100 train_loss:3.4327 train_time:266643ms step_avg:153.95ms step:1743/5100 train_loss:3.3182 train_time:266796ms step_avg:153.95ms step:1744/5100 train_loss:3.2559 train_time:266948ms step_avg:153.95ms step:1745/5100 train_loss:3.5534 train_time:267101ms step_avg:153.95ms step:1746/5100 train_loss:3.5629 train_time:267253ms step_avg:153.95ms step:1747/5100 train_loss:3.5320 train_time:267405ms step_avg:153.95ms step:1748/5100 train_loss:3.5372 train_time:267558ms step_avg:153.95ms step:1749/5100 train_loss:3.7904 train_time:267711ms step_avg:153.95ms step:1750/5100 train_loss:3.4928 train_time:267864ms step_avg:153.94ms step:1750/5100 val_loss:3.5460 train_time:267927ms step_avg:153.98ms step:1751/5100 train_loss:3.5666 train_time:268022ms step_avg:153.95ms step:1752/5100 train_loss:3.5513 train_time:268177ms step_avg:153.95ms step:1753/5100 train_loss:3.1953 train_time:268329ms step_avg:153.95ms step:1754/5100 train_loss:3.3122 train_time:268481ms step_avg:153.95ms step:1755/5100 train_loss:3.4081 train_time:268633ms step_avg:153.94ms step:1756/5100 train_loss:3.3587 train_time:268785ms step_avg:153.94ms step:1757/5100 train_loss:3.5186 train_time:268937ms step_avg:153.94ms step:1758/5100 train_loss:3.4006 train_time:269090ms step_avg:153.94ms step:1759/5100 train_loss:3.3932 train_time:269243ms step_avg:153.94ms step:1760/5100 train_loss:4.4544 train_time:269395ms step_avg:153.94ms step:1761/5100 train_loss:3.5252 train_time:269548ms step_avg:153.94ms step:1762/5100 train_loss:3.5644 train_time:269700ms step_avg:153.94ms step:1763/5100 train_loss:3.5635 train_time:269852ms step_avg:153.94ms step:1764/5100 train_loss:3.5872 train_time:270004ms step_avg:153.94ms step:1765/5100 train_loss:3.4975 train_time:270157ms step_avg:153.94ms step:1766/5100 train_loss:3.5358 train_time:270310ms step_avg:153.93ms step:1767/5100 train_loss:3.5552 train_time:270462ms step_avg:153.93ms step:1768/5100 train_loss:3.8036 train_time:270615ms step_avg:153.93ms step:1769/5100 train_loss:3.5357 train_time:270768ms step_avg:153.93ms step:1770/5100 train_loss:3.5994 train_time:270922ms step_avg:153.93ms step:1771/5100 train_loss:3.9688 train_time:271073ms step_avg:153.93ms step:1772/5100 train_loss:3.5332 train_time:271225ms step_avg:153.93ms step:1773/5100 train_loss:3.4388 train_time:271378ms step_avg:153.93ms step:1774/5100 train_loss:3.6935 train_time:271530ms step_avg:153.93ms step:1775/5100 train_loss:3.4275 train_time:271682ms step_avg:153.93ms step:1776/5100 train_loss:3.5867 train_time:271835ms step_avg:153.93ms step:1777/5100 train_loss:3.6448 train_time:271986ms step_avg:153.93ms step:1778/5100 train_loss:3.7363 train_time:272140ms step_avg:153.93ms step:1779/5100 train_loss:3.5441 train_time:272293ms step_avg:153.92ms step:1780/5100 train_loss:3.8379 train_time:272444ms step_avg:153.92ms step:1781/5100 train_loss:3.6096 train_time:272597ms step_avg:153.92ms step:1782/5100 train_loss:3.6302 train_time:272749ms step_avg:153.92ms step:1783/5100 train_loss:3.4143 train_time:272902ms step_avg:153.92ms step:1784/5100 train_loss:3.4893 train_time:273055ms step_avg:153.92ms step:1785/5100 train_loss:3.6331 train_time:273206ms step_avg:153.92ms step:1786/5100 train_loss:3.5287 train_time:273360ms step_avg:153.92ms step:1787/5100 train_loss:3.6917 train_time:273511ms step_avg:153.92ms step:1788/5100 train_loss:3.5097 train_time:273664ms step_avg:153.92ms step:1789/5100 train_loss:3.4863 train_time:273816ms step_avg:153.92ms step:1790/5100 train_loss:3.6338 train_time:273969ms step_avg:153.92ms step:1791/5100 train_loss:3.5290 train_time:274121ms step_avg:153.91ms step:1792/5100 train_loss:3.4704 train_time:274274ms step_avg:153.91ms step:1793/5100 train_loss:3.6084 train_time:274426ms step_avg:153.91ms step:1794/5100 train_loss:3.4914 train_time:274578ms step_avg:153.91ms step:1795/5100 train_loss:3.4759 train_time:274731ms step_avg:153.91ms step:1796/5100 train_loss:3.5359 train_time:274883ms step_avg:153.91ms step:1797/5100 train_loss:3.4970 train_time:275035ms step_avg:153.91ms step:1798/5100 train_loss:3.6375 train_time:275188ms step_avg:153.91ms step:1799/5100 train_loss:3.5200 train_time:275340ms step_avg:153.91ms step:1800/5100 train_loss:3.5982 train_time:275494ms step_avg:153.91ms step:1801/5100 train_loss:3.5277 train_time:275645ms step_avg:153.91ms step:1802/5100 train_loss:3.5628 train_time:275798ms step_avg:153.90ms step:1803/5100 train_loss:3.4888 train_time:275949ms step_avg:153.90ms step:1804/5100 train_loss:3.4068 train_time:276101ms step_avg:153.90ms step:1805/5100 train_loss:3.6500 train_time:276254ms step_avg:153.90ms step:1806/5100 train_loss:3.5750 train_time:276406ms step_avg:153.90ms step:1807/5100 train_loss:3.5860 train_time:276560ms step_avg:153.90ms step:1808/5100 train_loss:3.6963 train_time:276712ms step_avg:153.90ms step:1809/5100 train_loss:3.4892 train_time:276865ms step_avg:153.90ms step:1810/5100 train_loss:3.5925 train_time:277018ms step_avg:153.90ms step:1811/5100 train_loss:3.7303 train_time:277170ms step_avg:153.90ms step:1812/5100 train_loss:3.5814 train_time:277322ms step_avg:153.90ms step:1813/5100 train_loss:3.6221 train_time:277474ms step_avg:153.90ms step:1814/5100 train_loss:3.6462 train_time:277628ms step_avg:153.90ms step:1815/5100 train_loss:3.5887 train_time:277779ms step_avg:153.89ms step:1816/5100 train_loss:3.6227 train_time:277932ms step_avg:153.89ms step:1817/5100 train_loss:3.5792 train_time:278083ms step_avg:153.89ms step:1818/5100 train_loss:3.6337 train_time:278235ms step_avg:153.89ms step:1819/5100 train_loss:3.5605 train_time:278388ms step_avg:153.89ms step:1820/5100 train_loss:3.5463 train_time:278540ms step_avg:153.89ms step:1821/5100 train_loss:3.5021 train_time:278693ms step_avg:153.89ms step:1822/5100 train_loss:3.4804 train_time:278845ms step_avg:153.89ms step:1823/5100 train_loss:3.4045 train_time:278997ms step_avg:153.89ms step:1824/5100 train_loss:3.5599 train_time:279150ms step_avg:153.89ms step:1825/5100 train_loss:3.6821 train_time:279302ms step_avg:153.89ms step:1826/5100 train_loss:3.6379 train_time:279454ms step_avg:153.88ms step:1827/5100 train_loss:3.6169 train_time:279606ms step_avg:153.88ms step:1828/5100 train_loss:3.4879 train_time:279759ms step_avg:153.88ms step:1829/5100 train_loss:3.5159 train_time:279912ms step_avg:153.88ms step:1830/5100 train_loss:3.6462 train_time:280064ms step_avg:153.88ms step:1831/5100 train_loss:3.4234 train_time:280217ms step_avg:153.88ms step:1832/5100 train_loss:3.5698 train_time:280369ms step_avg:153.88ms step:1833/5100 train_loss:3.4471 train_time:280522ms step_avg:153.88ms step:1834/5100 train_loss:3.7701 train_time:280674ms step_avg:153.88ms step:1835/5100 train_loss:3.6040 train_time:280826ms step_avg:153.88ms step:1836/5100 train_loss:3.5871 train_time:280979ms step_avg:153.88ms step:1837/5100 train_loss:3.7138 train_time:281131ms step_avg:153.88ms step:1838/5100 train_loss:3.5736 train_time:281283ms step_avg:153.87ms step:1839/5100 train_loss:3.4488 train_time:281436ms step_avg:153.87ms step:1840/5100 train_loss:3.5679 train_time:281589ms step_avg:153.87ms step:1841/5100 train_loss:3.4555 train_time:281741ms step_avg:153.87ms step:1842/5100 train_loss:3.5627 train_time:281893ms step_avg:153.87ms step:1843/5100 train_loss:3.6251 train_time:282046ms step_avg:153.87ms step:1844/5100 train_loss:3.3728 train_time:282200ms step_avg:153.87ms step:1845/5100 train_loss:3.4950 train_time:282351ms step_avg:153.87ms step:1846/5100 train_loss:3.5546 train_time:282503ms step_avg:153.87ms step:1847/5100 train_loss:3.5019 train_time:282657ms step_avg:153.87ms step:1848/5100 train_loss:3.3959 train_time:282809ms step_avg:153.87ms step:1849/5100 train_loss:3.6609 train_time:282961ms step_avg:153.87ms step:1850/5100 train_loss:3.4258 train_time:283113ms step_avg:153.87ms step:1851/5100 train_loss:3.5105 train_time:283265ms step_avg:153.87ms step:1852/5100 train_loss:3.4724 train_time:283419ms step_avg:153.86ms step:1853/5100 train_loss:3.6769 train_time:283571ms step_avg:153.86ms step:1854/5100 train_loss:3.6470 train_time:283723ms step_avg:153.86ms step:1855/5100 train_loss:3.5242 train_time:283875ms step_avg:153.86ms step:1856/5100 train_loss:3.4778 train_time:284028ms step_avg:153.86ms step:1857/5100 train_loss:3.5139 train_time:284180ms step_avg:153.86ms step:1858/5100 train_loss:3.7579 train_time:284332ms step_avg:153.86ms step:1859/5100 train_loss:3.5953 train_time:284484ms step_avg:153.86ms step:1860/5100 train_loss:3.5315 train_time:284637ms step_avg:153.86ms step:1861/5100 train_loss:3.5732 train_time:284790ms step_avg:153.86ms step:1862/5100 train_loss:3.4680 train_time:284942ms step_avg:153.86ms step:1863/5100 train_loss:3.4572 train_time:285095ms step_avg:153.86ms step:1864/5100 train_loss:3.5327 train_time:285248ms step_avg:153.86ms step:1865/5100 train_loss:3.5730 train_time:285401ms step_avg:153.85ms step:1866/5100 train_loss:3.3317 train_time:285554ms step_avg:153.85ms step:1867/5100 train_loss:3.4653 train_time:285705ms step_avg:153.85ms step:1868/5100 train_loss:3.4208 train_time:285860ms step_avg:153.85ms step:1869/5100 train_loss:3.4242 train_time:286011ms step_avg:153.85ms step:1870/5100 train_loss:3.5806 train_time:286164ms step_avg:153.85ms step:1871/5100 train_loss:3.5682 train_time:286316ms step_avg:153.85ms step:1872/5100 train_loss:3.5109 train_time:286469ms step_avg:153.85ms step:1873/5100 train_loss:3.5245 train_time:286621ms step_avg:153.85ms step:1874/5100 train_loss:3.4548 train_time:286774ms step_avg:153.85ms step:1875/5100 train_loss:3.5552 train_time:286926ms step_avg:153.85ms step:1875/5100 val_loss:3.5306 train_time:286989ms step_avg:153.88ms step:1876/5100 train_loss:3.5559 train_time:287084ms step_avg:153.85ms step:1877/5100 train_loss:3.4842 train_time:287241ms step_avg:153.85ms step:1878/5100 train_loss:3.5258 train_time:287394ms step_avg:153.85ms step:1879/5100 train_loss:3.6343 train_time:287545ms step_avg:153.85ms step:1880/5100 train_loss:3.5153 train_time:287698ms step_avg:153.85ms step:1881/5100 train_loss:3.5729 train_time:287850ms step_avg:153.85ms step:1882/5100 train_loss:3.4795 train_time:288002ms step_avg:153.85ms step:1883/5100 train_loss:3.5548 train_time:288155ms step_avg:153.85ms step:1884/5100 train_loss:3.5506 train_time:288307ms step_avg:153.85ms step:1885/5100 train_loss:3.3058 train_time:288460ms step_avg:153.85ms step:1886/5100 train_loss:3.7065 train_time:288613ms step_avg:153.84ms step:1887/5100 train_loss:3.4430 train_time:288765ms step_avg:153.84ms step:1888/5100 train_loss:3.4626 train_time:288918ms step_avg:153.84ms step:1889/5100 train_loss:3.5297 train_time:289070ms step_avg:153.84ms step:1890/5100 train_loss:3.5772 train_time:289327ms step_avg:153.90ms step:1891/5100 train_loss:3.3943 train_time:289483ms step_avg:153.90ms step:1892/5100 train_loss:3.6770 train_time:289635ms step_avg:153.90ms step:1893/5100 train_loss:3.4280 train_time:289787ms step_avg:153.90ms step:1894/5100 train_loss:3.5539 train_time:289939ms step_avg:153.90ms step:1895/5100 train_loss:3.6008 train_time:290091ms step_avg:153.89ms step:1896/5100 train_loss:3.3991 train_time:290242ms step_avg:153.89ms step:1897/5100 train_loss:3.5609 train_time:290395ms step_avg:153.89ms step:1898/5100 train_loss:3.5238 train_time:290547ms step_avg:153.89ms step:1899/5100 train_loss:3.5968 train_time:290700ms step_avg:153.89ms step:1900/5100 train_loss:3.3796 train_time:290974ms step_avg:153.95ms step:1901/5100 train_loss:3.6208 train_time:291124ms step_avg:153.95ms step:1902/5100 train_loss:3.5097 train_time:291277ms step_avg:153.95ms step:1903/5100 train_loss:3.6668 train_time:291429ms step_avg:153.95ms step:1904/5100 train_loss:3.4685 train_time:291580ms step_avg:153.95ms step:1905/5100 train_loss:3.7521 train_time:291733ms step_avg:153.95ms step:1906/5100 train_loss:3.4835 train_time:291886ms step_avg:153.95ms step:1907/5100 train_loss:3.4782 train_time:292040ms step_avg:153.95ms step:1908/5100 train_loss:3.5535 train_time:292193ms step_avg:153.95ms step:1909/5100 train_loss:3.4372 train_time:292345ms step_avg:153.95ms step:1910/5100 train_loss:3.5012 train_time:292497ms step_avg:153.95ms step:1911/5100 train_loss:3.5980 train_time:292649ms step_avg:153.94ms step:1912/5100 train_loss:3.5205 train_time:292801ms step_avg:153.94ms step:1913/5100 train_loss:3.3995 train_time:292954ms step_avg:153.94ms step:1914/5100 train_loss:3.2762 train_time:293106ms step_avg:153.94ms step:1915/5100 train_loss:3.4743 train_time:293260ms step_avg:153.94ms step:1916/5100 train_loss:3.6925 train_time:293413ms step_avg:153.94ms step:1917/5100 train_loss:3.6819 train_time:293565ms step_avg:153.94ms step:1918/5100 train_loss:3.6390 train_time:293718ms step_avg:153.94ms step:1919/5100 train_loss:3.4683 train_time:293870ms step_avg:153.94ms step:1920/5100 train_loss:3.7132 train_time:294023ms step_avg:153.94ms step:1921/5100 train_loss:3.5282 train_time:294175ms step_avg:153.94ms step:1922/5100 train_loss:3.4657 train_time:294328ms step_avg:153.94ms step:1923/5100 train_loss:3.6410 train_time:294481ms step_avg:153.94ms step:1924/5100 train_loss:3.6061 train_time:294634ms step_avg:153.94ms step:1925/5100 train_loss:3.4437 train_time:294786ms step_avg:153.94ms step:1926/5100 train_loss:3.4796 train_time:294939ms step_avg:153.93ms step:1927/5100 train_loss:3.3894 train_time:295091ms step_avg:153.93ms step:1928/5100 train_loss:3.4938 train_time:295243ms step_avg:153.93ms step:1929/5100 train_loss:3.3562 train_time:295396ms step_avg:153.93ms step:1930/5100 train_loss:3.4675 train_time:295548ms step_avg:153.93ms step:1931/5100 train_loss:3.6013 train_time:295701ms step_avg:153.93ms step:1932/5100 train_loss:3.4728 train_time:295854ms step_avg:153.93ms step:1933/5100 train_loss:3.6098 train_time:296006ms step_avg:153.93ms step:1934/5100 train_loss:3.4892 train_time:296158ms step_avg:153.93ms step:1935/5100 train_loss:3.5360 train_time:296311ms step_avg:153.93ms step:1936/5100 train_loss:3.5736 train_time:296463ms step_avg:153.93ms step:1937/5100 train_loss:3.5225 train_time:296618ms step_avg:153.93ms step:1938/5100 train_loss:3.5521 train_time:296770ms step_avg:153.93ms step:1939/5100 train_loss:3.4798 train_time:296923ms step_avg:153.93ms step:1940/5100 train_loss:3.5736 train_time:297075ms step_avg:153.93ms step:1941/5100 train_loss:3.6105 train_time:297228ms step_avg:153.92ms step:1942/5100 train_loss:3.4467 train_time:297380ms step_avg:153.92ms step:1943/5100 train_loss:3.4812 train_time:297532ms step_avg:153.92ms step:1944/5100 train_loss:3.5514 train_time:297685ms step_avg:153.92ms step:1945/5100 train_loss:3.3926 train_time:297838ms step_avg:153.92ms step:1946/5100 train_loss:3.6651 train_time:297990ms step_avg:153.92ms step:1947/5100 train_loss:3.5321 train_time:298142ms step_avg:153.92ms step:1948/5100 train_loss:3.5064 train_time:298296ms step_avg:153.92ms step:1949/5100 train_loss:3.5191 train_time:298448ms step_avg:153.92ms step:1950/5100 train_loss:3.4004 train_time:298600ms step_avg:153.92ms step:1951/5100 train_loss:3.5176 train_time:298753ms step_avg:153.92ms step:1952/5100 train_loss:3.3582 train_time:298905ms step_avg:153.92ms step:1953/5100 train_loss:3.5730 train_time:299058ms step_avg:153.92ms step:1954/5100 train_loss:3.5645 train_time:299211ms step_avg:153.92ms step:1955/5100 train_loss:3.5150 train_time:299364ms step_avg:153.91ms step:1956/5100 train_loss:3.4104 train_time:299517ms step_avg:153.91ms step:1957/5100 train_loss:3.4979 train_time:299670ms step_avg:153.91ms step:1958/5100 train_loss:3.6869 train_time:299822ms step_avg:153.91ms step:1959/5100 train_loss:3.6022 train_time:299975ms step_avg:153.91ms step:1960/5100 train_loss:3.6227 train_time:300128ms step_avg:153.91ms step:1961/5100 train_loss:3.4227 train_time:300280ms step_avg:153.91ms step:1962/5100 train_loss:3.5421 train_time:300433ms step_avg:153.91ms step:1963/5100 train_loss:3.5890 train_time:300587ms step_avg:153.91ms step:1964/5100 train_loss:3.5389 train_time:300739ms step_avg:153.91ms step:1965/5100 train_loss:3.4429 train_time:300891ms step_avg:153.91ms step:1966/5100 train_loss:3.8592 train_time:301043ms step_avg:153.91ms step:1967/5100 train_loss:3.4671 train_time:301196ms step_avg:153.91ms step:1968/5100 train_loss:3.4955 train_time:301349ms step_avg:153.91ms step:1969/5100 train_loss:3.5538 train_time:301500ms step_avg:153.91ms step:1970/5100 train_loss:3.5031 train_time:301655ms step_avg:153.91ms step:1971/5100 train_loss:3.4020 train_time:301807ms step_avg:153.90ms step:1972/5100 train_loss:3.3855 train_time:301959ms step_avg:153.90ms step:1973/5100 train_loss:3.5020 train_time:302111ms step_avg:153.90ms step:1974/5100 train_loss:3.4761 train_time:302264ms step_avg:153.90ms step:1975/5100 train_loss:3.4485 train_time:302418ms step_avg:153.90ms step:1976/5100 train_loss:3.6040 train_time:302570ms step_avg:153.90ms step:1977/5100 train_loss:3.4740 train_time:302723ms step_avg:153.90ms step:1978/5100 train_loss:3.8419 train_time:302876ms step_avg:153.90ms step:1979/5100 train_loss:3.5245 train_time:303029ms step_avg:153.90ms step:1980/5100 train_loss:3.5221 train_time:303183ms step_avg:153.90ms step:1981/5100 train_loss:3.5308 train_time:303335ms step_avg:153.90ms step:1982/5100 train_loss:3.5521 train_time:303488ms step_avg:153.90ms step:1983/5100 train_loss:3.4792 train_time:303640ms step_avg:153.90ms step:1984/5100 train_loss:3.4490 train_time:303793ms step_avg:153.90ms step:1985/5100 train_loss:3.5043 train_time:303946ms step_avg:153.90ms step:1986/5100 train_loss:3.5674 train_time:304098ms step_avg:153.90ms step:1987/5100 train_loss:3.5441 train_time:304251ms step_avg:153.90ms step:1988/5100 train_loss:3.5042 train_time:304403ms step_avg:153.89ms step:1989/5100 train_loss:3.5922 train_time:304558ms step_avg:153.89ms step:1990/5100 train_loss:3.6272 train_time:304709ms step_avg:153.89ms step:1991/5100 train_loss:3.4020 train_time:304862ms step_avg:153.89ms step:1992/5100 train_loss:3.4012 train_time:305015ms step_avg:153.89ms step:1993/5100 train_loss:3.5896 train_time:305168ms step_avg:153.89ms step:1994/5100 train_loss:3.4131 train_time:305320ms step_avg:153.89ms step:1995/5100 train_loss:3.4990 train_time:305472ms step_avg:153.89ms step:1996/5100 train_loss:3.5717 train_time:305625ms step_avg:153.89ms step:1997/5100 train_loss:3.4325 train_time:305778ms step_avg:153.89ms step:1998/5100 train_loss:3.5398 train_time:305930ms step_avg:153.89ms step:1999/5100 train_loss:3.5373 train_time:306083ms step_avg:153.89ms step:2000/5100 train_loss:3.4638 train_time:306236ms step_avg:153.89ms step:2000/5100 val_loss:3.5166 train_time:306300ms step_avg:153.92ms step:2001/5100 train_loss:3.6094 train_time:306392ms step_avg:153.89ms step:2002/5100 train_loss:3.5470 train_time:306549ms step_avg:153.89ms step:2003/5100 train_loss:3.6427 train_time:306701ms step_avg:153.89ms step:2004/5100 train_loss:3.5576 train_time:306852ms step_avg:153.89ms step:2005/5100 train_loss:3.5689 train_time:307004ms step_avg:153.89ms step:2006/5100 train_loss:3.4487 train_time:307156ms step_avg:153.89ms step:2007/5100 train_loss:3.4837 train_time:307309ms step_avg:153.89ms step:2008/5100 train_loss:3.5263 train_time:307461ms step_avg:153.88ms step:2009/5100 train_loss:3.5688 train_time:307615ms step_avg:153.88ms step:2010/5100 train_loss:3.4642 train_time:307767ms step_avg:153.88ms step:2011/5100 train_loss:3.5506 train_time:307920ms step_avg:153.88ms step:2012/5100 train_loss:3.5281 train_time:308073ms step_avg:153.88ms step:2013/5100 train_loss:3.5271 train_time:308225ms step_avg:153.88ms step:2014/5100 train_loss:3.4520 train_time:308378ms step_avg:153.88ms step:2015/5100 train_loss:3.4912 train_time:308530ms step_avg:153.88ms step:2016/5100 train_loss:3.5115 train_time:308682ms step_avg:153.88ms step:2017/5100 train_loss:3.6441 train_time:308834ms step_avg:153.88ms step:2018/5100 train_loss:3.4873 train_time:308987ms step_avg:153.88ms step:2019/5100 train_loss:3.6405 train_time:309139ms step_avg:153.88ms step:2020/5100 train_loss:3.6506 train_time:309291ms step_avg:153.88ms step:2021/5100 train_loss:3.3637 train_time:309444ms step_avg:153.88ms step:2022/5100 train_loss:3.5958 train_time:309596ms step_avg:153.87ms step:2023/5100 train_loss:3.5137 train_time:309748ms step_avg:153.87ms step:2024/5100 train_loss:3.6153 train_time:309900ms step_avg:153.87ms step:2025/5100 train_loss:3.6552 train_time:310053ms step_avg:153.87ms step:2026/5100 train_loss:3.4406 train_time:310206ms step_avg:153.87ms step:2027/5100 train_loss:3.4902 train_time:310358ms step_avg:153.87ms step:2028/5100 train_loss:3.3859 train_time:310511ms step_avg:153.87ms step:2029/5100 train_loss:3.4923 train_time:310662ms step_avg:153.87ms step:2030/5100 train_loss:3.4211 train_time:310815ms step_avg:153.87ms step:2031/5100 train_loss:3.5049 train_time:310967ms step_avg:153.87ms step:2032/5100 train_loss:3.5046 train_time:311120ms step_avg:153.87ms step:2033/5100 train_loss:3.5151 train_time:311272ms step_avg:153.87ms step:2034/5100 train_loss:3.4131 train_time:311425ms step_avg:153.87ms step:2035/5100 train_loss:3.5751 train_time:311577ms step_avg:153.87ms step:2036/5100 train_loss:3.5709 train_time:311728ms step_avg:153.86ms step:2037/5100 train_loss:3.5618 train_time:311881ms step_avg:153.86ms step:2038/5100 train_loss:3.4280 train_time:312033ms step_avg:153.86ms step:2039/5100 train_loss:3.6911 train_time:312186ms step_avg:153.86ms step:2040/5100 train_loss:3.5252 train_time:312339ms step_avg:153.86ms step:2041/5100 train_loss:3.5470 train_time:312492ms step_avg:153.86ms step:2042/5100 train_loss:3.4930 train_time:312645ms step_avg:153.86ms step:2043/5100 train_loss:3.3892 train_time:312797ms step_avg:153.86ms step:2044/5100 train_loss:3.5226 train_time:312949ms step_avg:153.86ms step:2045/5100 train_loss:3.5203 train_time:313102ms step_avg:153.86ms step:2046/5100 train_loss:3.3816 train_time:313255ms step_avg:153.86ms step:2047/5100 train_loss:3.4587 train_time:313408ms step_avg:153.86ms step:2048/5100 train_loss:3.5273 train_time:313560ms step_avg:153.86ms step:2049/5100 train_loss:3.4781 train_time:313713ms step_avg:153.86ms step:2050/5100 train_loss:3.5281 train_time:313866ms step_avg:153.86ms step:2051/5100 train_loss:3.6747 train_time:314017ms step_avg:153.85ms step:2052/5100 train_loss:3.5428 train_time:314169ms step_avg:153.85ms step:2053/5100 train_loss:3.4923 train_time:314322ms step_avg:153.85ms step:2054/5100 train_loss:3.4754 train_time:314476ms step_avg:153.85ms step:2055/5100 train_loss:3.3415 train_time:314629ms step_avg:153.85ms step:2056/5100 train_loss:3.4528 train_time:314782ms step_avg:153.85ms step:2057/5100 train_loss:3.6329 train_time:314934ms step_avg:153.85ms step:2058/5100 train_loss:3.6498 train_time:315085ms step_avg:153.85ms step:2059/5100 train_loss:3.5117 train_time:315238ms step_avg:153.85ms step:2060/5100 train_loss:3.5534 train_time:315390ms step_avg:153.85ms step:2061/5100 train_loss:3.5426 train_time:315542ms step_avg:153.85ms step:2062/5100 train_loss:3.4947 train_time:315694ms step_avg:153.85ms step:2063/5100 train_loss:3.4068 train_time:315848ms step_avg:153.85ms step:2064/5100 train_loss:3.7132 train_time:316000ms step_avg:153.85ms step:2065/5100 train_loss:3.5683 train_time:316153ms step_avg:153.85ms step:2066/5100 train_loss:3.5238 train_time:316307ms step_avg:153.85ms step:2067/5100 train_loss:3.5685 train_time:316459ms step_avg:153.85ms step:2068/5100 train_loss:3.4695 train_time:316612ms step_avg:153.84ms step:2069/5100 train_loss:3.5300 train_time:316764ms step_avg:153.84ms step:2070/5100 train_loss:3.6588 train_time:316915ms step_avg:153.84ms step:2071/5100 train_loss:3.6618 train_time:317068ms step_avg:153.84ms step:2072/5100 train_loss:3.5097 train_time:317220ms step_avg:153.84ms step:2073/5100 train_loss:3.5406 train_time:317373ms step_avg:153.84ms step:2074/5100 train_loss:3.4367 train_time:317526ms step_avg:153.84ms step:2075/5100 train_loss:3.9600 train_time:317678ms step_avg:153.84ms step:2076/5100 train_loss:3.3919 train_time:317830ms step_avg:153.84ms step:2077/5100 train_loss:3.5507 train_time:317982ms step_avg:153.84ms step:2078/5100 train_loss:3.4427 train_time:318134ms step_avg:153.84ms step:2079/5100 train_loss:3.4263 train_time:318387ms step_avg:153.88ms step:2080/5100 train_loss:3.5060 train_time:318550ms step_avg:153.89ms step:2081/5100 train_loss:3.7643 train_time:318702ms step_avg:153.89ms step:2082/5100 train_loss:3.3925 train_time:318854ms step_avg:153.89ms step:2083/5100 train_loss:3.7320 train_time:319006ms step_avg:153.89ms step:2084/5100 train_loss:3.4363 train_time:319158ms step_avg:153.89ms step:2085/5100 train_loss:3.4213 train_time:319309ms step_avg:153.88ms step:2086/5100 train_loss:3.6683 train_time:319463ms step_avg:153.88ms step:2087/5100 train_loss:3.5982 train_time:319616ms step_avg:153.88ms step:2088/5100 train_loss:3.5800 train_time:319769ms step_avg:153.88ms step:2089/5100 train_loss:3.6382 train_time:319920ms step_avg:153.88ms step:2090/5100 train_loss:3.5571 train_time:320195ms step_avg:153.94ms step:2091/5100 train_loss:3.5495 train_time:320346ms step_avg:153.94ms step:2092/5100 train_loss:3.5027 train_time:320497ms step_avg:153.94ms step:2093/5100 train_loss:3.5782 train_time:320650ms step_avg:153.94ms step:2094/5100 train_loss:3.4683 train_time:320802ms step_avg:153.94ms step:2095/5100 train_loss:3.2680 train_time:320955ms step_avg:153.94ms step:2096/5100 train_loss:3.4935 train_time:321108ms step_avg:153.93ms step:2097/5100 train_loss:3.6729 train_time:321260ms step_avg:153.93ms step:2098/5100 train_loss:3.4979 train_time:321415ms step_avg:153.93ms step:2099/5100 train_loss:3.3864 train_time:321567ms step_avg:153.93ms step:2100/5100 train_loss:3.4894 train_time:321720ms step_avg:153.93ms step:2101/5100 train_loss:3.4485 train_time:321872ms step_avg:153.93ms step:2102/5100 train_loss:3.5856 train_time:322025ms step_avg:153.93ms step:2103/5100 train_loss:3.4218 train_time:322178ms step_avg:153.93ms step:2104/5100 train_loss:3.3818 train_time:322330ms step_avg:153.93ms step:2105/5100 train_loss:3.6429 train_time:322483ms step_avg:153.93ms step:2106/5100 train_loss:3.3839 train_time:322636ms step_avg:153.93ms step:2107/5100 train_loss:3.7742 train_time:322789ms step_avg:153.93ms step:2108/5100 train_loss:3.6075 train_time:322942ms step_avg:153.93ms step:2109/5100 train_loss:3.5208 train_time:323094ms step_avg:153.93ms step:2110/5100 train_loss:3.5339 train_time:323247ms step_avg:153.93ms step:2111/5100 train_loss:3.3521 train_time:323400ms step_avg:153.93ms step:2112/5100 train_loss:3.8425 train_time:323552ms step_avg:153.93ms step:2113/5100 train_loss:3.5277 train_time:323706ms step_avg:153.93ms step:2114/5100 train_loss:3.4573 train_time:323858ms step_avg:153.92ms step:2115/5100 train_loss:3.5729 train_time:324010ms step_avg:153.92ms step:2116/5100 train_loss:3.5311 train_time:324163ms step_avg:153.92ms step:2117/5100 train_loss:3.5115 train_time:324316ms step_avg:153.92ms step:2118/5100 train_loss:3.5697 train_time:324468ms step_avg:153.92ms step:2119/5100 train_loss:3.4319 train_time:324620ms step_avg:153.92ms step:2120/5100 train_loss:3.4883 train_time:324773ms step_avg:153.92ms step:2121/5100 train_loss:3.1989 train_time:324926ms step_avg:153.92ms step:2122/5100 train_loss:3.4035 train_time:325078ms step_avg:153.92ms step:2123/5100 train_loss:3.5566 train_time:325231ms step_avg:153.92ms step:2124/5100 train_loss:3.4732 train_time:325384ms step_avg:153.92ms step:2125/5100 train_loss:3.6258 train_time:325536ms step_avg:153.92ms step:2125/5100 val_loss:3.5045 train_time:325599ms step_avg:153.95ms step:2126/5100 train_loss:3.4861 train_time:325692ms step_avg:153.92ms step:2127/5100 train_loss:3.6008 train_time:325852ms step_avg:153.92ms step:2128/5100 train_loss:3.5860 train_time:326003ms step_avg:153.92ms step:2129/5100 train_loss:3.4432 train_time:326155ms step_avg:153.92ms step:2130/5100 train_loss:3.4288 train_time:326308ms step_avg:153.92ms step:2131/5100 train_loss:3.4545 train_time:326460ms step_avg:153.92ms step:2132/5100 train_loss:3.6108 train_time:326612ms step_avg:153.92ms step:2133/5100 train_loss:3.4880 train_time:326766ms step_avg:153.92ms step:2134/5100 train_loss:3.4003 train_time:326920ms step_avg:153.92ms step:2135/5100 train_loss:3.4611 train_time:327073ms step_avg:153.92ms step:2136/5100 train_loss:3.5790 train_time:327224ms step_avg:153.92ms step:2137/5100 train_loss:3.5970 train_time:327377ms step_avg:153.91ms step:2138/5100 train_loss:3.5421 train_time:327529ms step_avg:153.91ms step:2139/5100 train_loss:3.5323 train_time:327681ms step_avg:153.91ms step:2140/5100 train_loss:3.5157 train_time:327835ms step_avg:153.91ms step:2141/5100 train_loss:3.6017 train_time:327985ms step_avg:153.91ms step:2142/5100 train_loss:3.9078 train_time:328138ms step_avg:153.91ms step:2143/5100 train_loss:3.4307 train_time:328290ms step_avg:153.91ms step:2144/5100 train_loss:3.4657 train_time:328442ms step_avg:153.91ms step:2145/5100 train_loss:3.5032 train_time:328595ms step_avg:153.91ms step:2146/5100 train_loss:3.6317 train_time:328748ms step_avg:153.91ms step:2147/5100 train_loss:3.5581 train_time:328901ms step_avg:153.91ms step:2148/5100 train_loss:3.9697 train_time:329053ms step_avg:153.91ms step:2149/5100 train_loss:3.4882 train_time:329205ms step_avg:153.91ms step:2150/5100 train_loss:3.4595 train_time:329359ms step_avg:153.91ms step:2151/5100 train_loss:3.5222 train_time:329511ms step_avg:153.91ms step:2152/5100 train_loss:3.5553 train_time:329664ms step_avg:153.90ms step:2153/5100 train_loss:3.5150 train_time:329816ms step_avg:153.90ms step:2154/5100 train_loss:3.4460 train_time:329969ms step_avg:153.90ms step:2155/5100 train_loss:3.6617 train_time:330121ms step_avg:153.90ms step:2156/5100 train_loss:3.2798 train_time:330274ms step_avg:153.90ms step:2157/5100 train_loss:3.4419 train_time:330426ms step_avg:153.90ms step:2158/5100 train_loss:3.5759 train_time:330579ms step_avg:153.90ms step:2159/5100 train_loss:3.5135 train_time:330730ms step_avg:153.90ms step:2160/5100 train_loss:3.6684 train_time:330883ms step_avg:153.90ms step:2161/5100 train_loss:3.5851 train_time:331036ms step_avg:153.90ms step:2162/5100 train_loss:3.5162 train_time:331188ms step_avg:153.90ms step:2163/5100 train_loss:3.4833 train_time:331341ms step_avg:153.90ms step:2164/5100 train_loss:3.4785 train_time:331493ms step_avg:153.90ms step:2165/5100 train_loss:3.5657 train_time:331646ms step_avg:153.90ms step:2166/5100 train_loss:3.5855 train_time:331799ms step_avg:153.90ms step:2167/5100 train_loss:3.5201 train_time:331950ms step_avg:153.89ms step:2168/5100 train_loss:3.4094 train_time:332103ms step_avg:153.89ms step:2169/5100 train_loss:3.4899 train_time:332255ms step_avg:153.89ms step:2170/5100 train_loss:3.5396 train_time:332408ms step_avg:153.89ms step:2171/5100 train_loss:3.6602 train_time:332560ms step_avg:153.89ms step:2172/5100 train_loss:3.4593 train_time:332712ms step_avg:153.89ms step:2173/5100 train_loss:3.4443 train_time:332864ms step_avg:153.89ms step:2174/5100 train_loss:3.4514 train_time:333017ms step_avg:153.89ms step:2175/5100 train_loss:3.5062 train_time:333169ms step_avg:153.89ms step:2176/5100 train_loss:3.4672 train_time:333322ms step_avg:153.89ms step:2177/5100 train_loss:3.4393 train_time:333474ms step_avg:153.89ms step:2178/5100 train_loss:3.6633 train_time:333626ms step_avg:153.89ms step:2179/5100 train_loss:3.4840 train_time:333779ms step_avg:153.89ms step:2180/5100 train_loss:3.4905 train_time:333931ms step_avg:153.89ms step:2181/5100 train_loss:3.5482 train_time:334083ms step_avg:153.88ms step:2182/5100 train_loss:3.5269 train_time:334236ms step_avg:153.88ms step:2183/5100 train_loss:3.4978 train_time:334388ms step_avg:153.88ms step:2184/5100 train_loss:3.3998 train_time:334541ms step_avg:153.88ms step:2185/5100 train_loss:3.5690 train_time:334693ms step_avg:153.88ms step:2186/5100 train_loss:3.7414 train_time:334846ms step_avg:153.88ms step:2187/5100 train_loss:3.3742 train_time:334998ms step_avg:153.88ms step:2188/5100 train_loss:3.4193 train_time:335150ms step_avg:153.88ms step:2189/5100 train_loss:3.2763 train_time:335302ms step_avg:153.88ms step:2190/5100 train_loss:3.4225 train_time:335454ms step_avg:153.88ms step:2191/5100 train_loss:3.5690 train_time:335606ms step_avg:153.88ms step:2192/5100 train_loss:3.5066 train_time:335758ms step_avg:153.88ms step:2193/5100 train_loss:3.7378 train_time:335911ms step_avg:153.88ms step:2194/5100 train_loss:3.5023 train_time:336063ms step_avg:153.87ms step:2195/5100 train_loss:3.5746 train_time:336215ms step_avg:153.87ms step:2196/5100 train_loss:3.5085 train_time:336368ms step_avg:153.87ms step:2197/5100 train_loss:3.4275 train_time:336520ms step_avg:153.87ms step:2198/5100 train_loss:3.5120 train_time:336673ms step_avg:153.87ms step:2199/5100 train_loss:3.4526 train_time:336825ms step_avg:153.87ms step:2200/5100 train_loss:3.4569 train_time:336977ms step_avg:153.87ms step:2201/5100 train_loss:3.5128 train_time:337129ms step_avg:153.87ms step:2202/5100 train_loss:3.4881 train_time:337282ms step_avg:153.87ms step:2203/5100 train_loss:3.4719 train_time:337435ms step_avg:153.87ms step:2204/5100 train_loss:3.9744 train_time:337586ms step_avg:153.87ms step:2205/5100 train_loss:3.3830 train_time:337740ms step_avg:153.87ms step:2206/5100 train_loss:3.5031 train_time:337893ms step_avg:153.87ms step:2207/5100 train_loss:3.5197 train_time:338046ms step_avg:153.87ms step:2208/5100 train_loss:3.5418 train_time:338199ms step_avg:153.87ms step:2209/5100 train_loss:3.4346 train_time:338352ms step_avg:153.87ms step:2210/5100 train_loss:3.5131 train_time:338504ms step_avg:153.87ms step:2211/5100 train_loss:3.5232 train_time:338656ms step_avg:153.86ms step:2212/5100 train_loss:3.5173 train_time:338809ms step_avg:153.86ms step:2213/5100 train_loss:3.5435 train_time:338961ms step_avg:153.86ms step:2214/5100 train_loss:3.4001 train_time:339113ms step_avg:153.86ms step:2215/5100 train_loss:3.4663 train_time:339265ms step_avg:153.86ms step:2216/5100 train_loss:3.6074 train_time:339417ms step_avg:153.86ms step:2217/5100 train_loss:3.5628 train_time:339570ms step_avg:153.86ms step:2218/5100 train_loss:3.5153 train_time:339722ms step_avg:153.86ms step:2219/5100 train_loss:3.5228 train_time:339876ms step_avg:153.86ms step:2220/5100 train_loss:3.4350 train_time:340028ms step_avg:153.86ms step:2221/5100 train_loss:3.6918 train_time:340181ms step_avg:153.86ms step:2222/5100 train_loss:3.5748 train_time:340333ms step_avg:153.86ms step:2223/5100 train_loss:3.5994 train_time:340485ms step_avg:153.86ms step:2224/5100 train_loss:3.4823 train_time:340639ms step_avg:153.86ms step:2225/5100 train_loss:3.6056 train_time:340791ms step_avg:153.86ms step:2226/5100 train_loss:3.3605 train_time:340944ms step_avg:153.86ms step:2227/5100 train_loss:3.6321 train_time:341098ms step_avg:153.86ms step:2228/5100 train_loss:3.5643 train_time:341250ms step_avg:153.85ms step:2229/5100 train_loss:3.3705 train_time:341402ms step_avg:153.85ms step:2230/5100 train_loss:3.7213 train_time:341553ms step_avg:153.85ms step:2231/5100 train_loss:3.4047 train_time:341706ms step_avg:153.85ms step:2232/5100 train_loss:3.8756 train_time:341860ms step_avg:153.85ms step:2233/5100 train_loss:3.5672 train_time:342012ms step_avg:153.85ms step:2234/5100 train_loss:3.5076 train_time:342165ms step_avg:153.85ms step:2235/5100 train_loss:3.5374 train_time:342318ms step_avg:153.85ms step:2236/5100 train_loss:3.3247 train_time:342470ms step_avg:153.85ms step:2237/5100 train_loss:3.3225 train_time:342623ms step_avg:153.85ms step:2238/5100 train_loss:3.5473 train_time:342775ms step_avg:153.85ms step:2239/5100 train_loss:3.6446 train_time:342927ms step_avg:153.85ms step:2240/5100 train_loss:3.3607 train_time:343081ms step_avg:153.85ms step:2241/5100 train_loss:3.4349 train_time:343232ms step_avg:153.85ms step:2242/5100 train_loss:3.6136 train_time:343384ms step_avg:153.85ms step:2243/5100 train_loss:3.5868 train_time:343536ms step_avg:153.85ms step:2244/5100 train_loss:3.4422 train_time:343688ms step_avg:153.84ms step:2245/5100 train_loss:3.5067 train_time:343842ms step_avg:153.84ms step:2246/5100 train_loss:3.5344 train_time:343994ms step_avg:153.84ms step:2247/5100 train_loss:3.3673 train_time:344146ms step_avg:153.84ms step:2248/5100 train_loss:3.3841 train_time:344298ms step_avg:153.84ms step:2249/5100 train_loss:3.6404 train_time:344450ms step_avg:153.84ms step:2250/5100 train_loss:3.3816 train_time:344602ms step_avg:153.84ms step:2250/5100 val_loss:3.4934 train_time:344666ms step_avg:153.87ms step:2251/5100 train_loss:3.3643 train_time:344764ms step_avg:153.84ms step:2252/5100 train_loss:3.4460 train_time:344917ms step_avg:153.84ms step:2253/5100 train_loss:3.4180 train_time:345071ms step_avg:153.84ms step:2254/5100 train_loss:3.4608 train_time:345222ms step_avg:153.84ms step:2255/5100 train_loss:3.5228 train_time:345375ms step_avg:153.84ms step:2256/5100 train_loss:3.4038 train_time:345527ms step_avg:153.84ms step:2257/5100 train_loss:3.6864 train_time:345680ms step_avg:153.84ms step:2258/5100 train_loss:3.5644 train_time:345834ms step_avg:153.84ms step:2259/5100 train_loss:3.8776 train_time:345986ms step_avg:153.84ms step:2260/5100 train_loss:3.5596 train_time:346141ms step_avg:153.84ms step:2261/5100 train_loss:3.6131 train_time:346293ms step_avg:153.84ms step:2262/5100 train_loss:3.5240 train_time:346446ms step_avg:153.84ms step:2263/5100 train_loss:3.5222 train_time:346600ms step_avg:153.84ms step:2264/5100 train_loss:3.2809 train_time:346752ms step_avg:153.84ms step:2265/5100 train_loss:3.4093 train_time:346905ms step_avg:153.84ms step:2266/5100 train_loss:3.6214 train_time:347058ms step_avg:153.84ms step:2267/5100 train_loss:3.3557 train_time:347211ms step_avg:153.84ms step:2268/5100 train_loss:3.4250 train_time:347460ms step_avg:153.88ms step:2269/5100 train_loss:3.4013 train_time:347622ms step_avg:153.88ms step:2270/5100 train_loss:3.3685 train_time:347775ms step_avg:153.88ms step:2271/5100 train_loss:3.7710 train_time:347927ms step_avg:153.88ms step:2272/5100 train_loss:3.4211 train_time:348079ms step_avg:153.88ms step:2273/5100 train_loss:3.4339 train_time:348230ms step_avg:153.88ms step:2274/5100 train_loss:3.5130 train_time:348382ms step_avg:153.88ms step:2275/5100 train_loss:3.4658 train_time:348533ms step_avg:153.88ms step:2276/5100 train_loss:3.4789 train_time:348688ms step_avg:153.88ms step:2277/5100 train_loss:3.3616 train_time:348842ms step_avg:153.88ms step:2278/5100 train_loss:3.4679 train_time:348994ms step_avg:153.88ms step:2279/5100 train_loss:3.5912 train_time:349147ms step_avg:153.88ms step:2280/5100 train_loss:3.3949 train_time:349421ms step_avg:153.93ms step:2281/5100 train_loss:3.4506 train_time:349572ms step_avg:153.93ms step:2282/5100 train_loss:3.4653 train_time:349724ms step_avg:153.93ms step:2283/5100 train_loss:3.6080 train_time:349875ms step_avg:153.93ms step:2284/5100 train_loss:3.4835 train_time:350027ms step_avg:153.93ms step:2285/5100 train_loss:3.5028 train_time:350179ms step_avg:153.92ms step:2286/5100 train_loss:3.5068 train_time:350331ms step_avg:153.92ms step:2287/5100 train_loss:3.5022 train_time:350484ms step_avg:153.92ms step:2288/5100 train_loss:3.4575 train_time:350638ms step_avg:153.92ms step:2289/5100 train_loss:3.5893 train_time:350790ms step_avg:153.92ms step:2290/5100 train_loss:3.5644 train_time:350942ms step_avg:153.92ms step:2291/5100 train_loss:3.4473 train_time:351095ms step_avg:153.92ms step:2292/5100 train_loss:3.7876 train_time:351247ms step_avg:153.92ms step:2293/5100 train_loss:3.4432 train_time:351400ms step_avg:153.92ms step:2294/5100 train_loss:3.3952 train_time:351552ms step_avg:153.92ms step:2295/5100 train_loss:3.5749 train_time:351705ms step_avg:153.92ms step:2296/5100 train_loss:3.5273 train_time:351858ms step_avg:153.92ms step:2297/5100 train_loss:3.4860 train_time:352010ms step_avg:153.92ms step:2298/5100 train_loss:3.8719 train_time:352162ms step_avg:153.92ms step:2299/5100 train_loss:3.3970 train_time:352315ms step_avg:153.92ms step:2300/5100 train_loss:3.3968 train_time:352467ms step_avg:153.92ms step:2301/5100 train_loss:3.7372 train_time:352619ms step_avg:153.92ms step:2302/5100 train_loss:3.4549 train_time:352773ms step_avg:153.91ms step:2303/5100 train_loss:3.4711 train_time:352926ms step_avg:153.91ms step:2304/5100 train_loss:3.4634 train_time:353078ms step_avg:153.91ms step:2305/5100 train_loss:3.3974 train_time:353231ms step_avg:153.91ms step:2306/5100 train_loss:3.5574 train_time:353383ms step_avg:153.91ms step:2307/5100 train_loss:3.4209 train_time:353535ms step_avg:153.91ms step:2308/5100 train_loss:3.4344 train_time:353688ms step_avg:153.91ms step:2309/5100 train_loss:3.5694 train_time:353842ms step_avg:153.91ms step:2310/5100 train_loss:3.5222 train_time:353995ms step_avg:153.91ms step:2311/5100 train_loss:3.3979 train_time:354147ms step_avg:153.91ms step:2312/5100 train_loss:3.5088 train_time:354300ms step_avg:153.91ms step:2313/5100 train_loss:3.6368 train_time:354452ms step_avg:153.91ms step:2314/5100 train_loss:3.4535 train_time:354604ms step_avg:153.91ms step:2315/5100 train_loss:3.3804 train_time:354757ms step_avg:153.91ms step:2316/5100 train_loss:3.4646 train_time:354910ms step_avg:153.91ms step:2317/5100 train_loss:3.3512 train_time:355062ms step_avg:153.91ms step:2318/5100 train_loss:3.4542 train_time:355216ms step_avg:153.91ms step:2319/5100 train_loss:3.4782 train_time:355368ms step_avg:153.91ms step:2320/5100 train_loss:3.3246 train_time:355522ms step_avg:153.91ms step:2321/5100 train_loss:3.4612 train_time:355674ms step_avg:153.90ms step:2322/5100 train_loss:3.5140 train_time:355827ms step_avg:153.90ms step:2323/5100 train_loss:3.4215 train_time:355980ms step_avg:153.90ms step:2324/5100 train_loss:3.4710 train_time:356134ms step_avg:153.90ms step:2325/5100 train_loss:3.3984 train_time:356286ms step_avg:153.90ms step:2326/5100 train_loss:3.5354 train_time:356440ms step_avg:153.90ms step:2327/5100 train_loss:3.5452 train_time:356592ms step_avg:153.90ms step:2328/5100 train_loss:3.3164 train_time:356745ms step_avg:153.90ms step:2329/5100 train_loss:3.4285 train_time:356899ms step_avg:153.90ms step:2330/5100 train_loss:3.4589 train_time:357052ms step_avg:153.90ms step:2331/5100 train_loss:3.4214 train_time:357205ms step_avg:153.90ms step:2332/5100 train_loss:3.6178 train_time:357358ms step_avg:153.90ms step:2333/5100 train_loss:3.4925 train_time:357510ms step_avg:153.90ms step:2334/5100 train_loss:3.4685 train_time:357662ms step_avg:153.90ms step:2335/5100 train_loss:3.5495 train_time:357814ms step_avg:153.90ms step:2336/5100 train_loss:3.3923 train_time:357967ms step_avg:153.90ms step:2337/5100 train_loss:3.5401 train_time:358120ms step_avg:153.90ms step:2338/5100 train_loss:3.5009 train_time:358273ms step_avg:153.90ms step:2339/5100 train_loss:3.4473 train_time:358425ms step_avg:153.90ms step:2340/5100 train_loss:3.5168 train_time:358578ms step_avg:153.90ms step:2341/5100 train_loss:3.5759 train_time:358730ms step_avg:153.90ms step:2342/5100 train_loss:3.4418 train_time:358882ms step_avg:153.89ms step:2343/5100 train_loss:3.4529 train_time:359035ms step_avg:153.89ms step:2344/5100 train_loss:3.5162 train_time:359187ms step_avg:153.89ms step:2345/5100 train_loss:3.4643 train_time:359341ms step_avg:153.89ms step:2346/5100 train_loss:3.5834 train_time:359493ms step_avg:153.89ms step:2347/5100 train_loss:3.4898 train_time:359645ms step_avg:153.89ms step:2348/5100 train_loss:3.5964 train_time:359798ms step_avg:153.89ms step:2349/5100 train_loss:3.5513 train_time:359950ms step_avg:153.89ms step:2350/5100 train_loss:3.5976 train_time:360103ms step_avg:153.89ms step:2351/5100 train_loss:3.2863 train_time:360255ms step_avg:153.89ms step:2352/5100 train_loss:3.4123 train_time:360408ms step_avg:153.89ms step:2353/5100 train_loss:3.4053 train_time:360560ms step_avg:153.89ms step:2354/5100 train_loss:3.6223 train_time:360714ms step_avg:153.89ms step:2355/5100 train_loss:3.4126 train_time:360866ms step_avg:153.89ms step:2356/5100 train_loss:3.4124 train_time:361019ms step_avg:153.89ms step:2357/5100 train_loss:3.5595 train_time:361171ms step_avg:153.89ms step:2358/5100 train_loss:3.4153 train_time:361323ms step_avg:153.89ms step:2359/5100 train_loss:3.5189 train_time:361476ms step_avg:153.89ms step:2360/5100 train_loss:3.4159 train_time:361628ms step_avg:153.88ms step:2361/5100 train_loss:3.4281 train_time:361781ms step_avg:153.88ms step:2362/5100 train_loss:3.4669 train_time:361935ms step_avg:153.88ms step:2363/5100 train_loss:3.5404 train_time:362086ms step_avg:153.88ms step:2364/5100 train_loss:3.4741 train_time:362241ms step_avg:153.88ms step:2365/5100 train_loss:3.9108 train_time:362393ms step_avg:153.88ms step:2366/5100 train_loss:3.5330 train_time:362546ms step_avg:153.88ms step:2367/5100 train_loss:3.6868 train_time:362699ms step_avg:153.88ms step:2368/5100 train_loss:3.5009 train_time:362851ms step_avg:153.88ms step:2369/5100 train_loss:3.5050 train_time:363003ms step_avg:153.88ms step:2370/5100 train_loss:3.5416 train_time:363156ms step_avg:153.88ms step:2371/5100 train_loss:3.4162 train_time:363308ms step_avg:153.88ms step:2372/5100 train_loss:3.6539 train_time:363462ms step_avg:153.88ms step:2373/5100 train_loss:3.4976 train_time:363614ms step_avg:153.88ms step:2374/5100 train_loss:4.0561 train_time:363768ms step_avg:153.88ms step:2375/5100 train_loss:3.4724 train_time:363920ms step_avg:153.88ms step:2375/5100 val_loss:3.4846 train_time:363984ms step_avg:153.90ms step:2376/5100 train_loss:3.3837 train_time:364077ms step_avg:153.88ms step:2377/5100 train_loss:3.5437 train_time:364235ms step_avg:153.88ms step:2378/5100 train_loss:3.5113 train_time:364387ms step_avg:153.88ms step:2379/5100 train_loss:3.5261 train_time:364539ms step_avg:153.88ms step:2380/5100 train_loss:3.5094 train_time:364691ms step_avg:153.88ms step:2381/5100 train_loss:3.4058 train_time:364843ms step_avg:153.88ms step:2382/5100 train_loss:3.5117 train_time:364994ms step_avg:153.88ms step:2383/5100 train_loss:3.5260 train_time:365148ms step_avg:153.88ms step:2384/5100 train_loss:3.4742 train_time:365301ms step_avg:153.88ms step:2385/5100 train_loss:3.4030 train_time:365454ms step_avg:153.88ms step:2386/5100 train_loss:3.5125 train_time:365607ms step_avg:153.87ms step:2387/5100 train_loss:3.4714 train_time:365758ms step_avg:153.87ms step:2388/5100 train_loss:3.4759 train_time:365910ms step_avg:153.87ms step:2389/5100 train_loss:3.5055 train_time:366063ms step_avg:153.87ms step:2390/5100 train_loss:3.4922 train_time:366216ms step_avg:153.87ms step:2391/5100 train_loss:3.4903 train_time:366369ms step_avg:153.87ms step:2392/5100 train_loss:3.3671 train_time:366521ms step_avg:153.87ms step:2393/5100 train_loss:3.5855 train_time:366673ms step_avg:153.87ms step:2394/5100 train_loss:3.4270 train_time:366825ms step_avg:153.87ms step:2395/5100 train_loss:3.5297 train_time:366976ms step_avg:153.87ms step:2396/5100 train_loss:3.6425 train_time:367129ms step_avg:153.87ms step:2397/5100 train_loss:3.6526 train_time:367282ms step_avg:153.87ms step:2398/5100 train_loss:3.6048 train_time:367434ms step_avg:153.87ms step:2399/5100 train_loss:3.5740 train_time:367586ms step_avg:153.87ms step:2400/5100 train_loss:3.4447 train_time:367741ms step_avg:153.87ms step:2401/5100 train_loss:3.4460 train_time:367892ms step_avg:153.87ms step:2402/5100 train_loss:3.5524 train_time:368044ms step_avg:153.86ms step:2403/5100 train_loss:3.3909 train_time:368196ms step_avg:153.86ms step:2404/5100 train_loss:3.5248 train_time:368348ms step_avg:153.86ms step:2405/5100 train_loss:3.7381 train_time:368500ms step_avg:153.86ms step:2406/5100 train_loss:3.4623 train_time:368652ms step_avg:153.86ms step:2407/5100 train_loss:3.6087 train_time:368805ms step_avg:153.86ms step:2408/5100 train_loss:3.4707 train_time:368957ms step_avg:153.86ms step:2409/5100 train_loss:3.4027 train_time:369110ms step_avg:153.86ms step:2410/5100 train_loss:3.5365 train_time:369263ms step_avg:153.86ms step:2411/5100 train_loss:3.3284 train_time:369415ms step_avg:153.86ms step:2412/5100 train_loss:3.7616 train_time:369567ms step_avg:153.86ms step:2413/5100 train_loss:3.4473 train_time:369720ms step_avg:153.86ms step:2414/5100 train_loss:3.5206 train_time:369872ms step_avg:153.86ms step:2415/5100 train_loss:3.4450 train_time:370025ms step_avg:153.86ms step:2416/5100 train_loss:3.5117 train_time:370177ms step_avg:153.86ms step:2417/5100 train_loss:3.3244 train_time:370330ms step_avg:153.86ms step:2418/5100 train_loss:3.2612 train_time:370481ms step_avg:153.85ms step:2419/5100 train_loss:3.5574 train_time:370635ms step_avg:153.85ms step:2420/5100 train_loss:3.4331 train_time:370788ms step_avg:153.85ms step:2421/5100 train_loss:3.4730 train_time:370940ms step_avg:153.85ms step:2422/5100 train_loss:3.5704 train_time:371092ms step_avg:153.85ms step:2423/5100 train_loss:3.6164 train_time:371245ms step_avg:153.85ms step:2424/5100 train_loss:3.4342 train_time:371397ms step_avg:153.85ms step:2425/5100 train_loss:3.5249 train_time:371550ms step_avg:153.85ms step:2426/5100 train_loss:3.5321 train_time:371702ms step_avg:153.85ms step:2427/5100 train_loss:3.4562 train_time:371854ms step_avg:153.85ms step:2428/5100 train_loss:3.4039 train_time:372007ms step_avg:153.85ms step:2429/5100 train_loss:3.5314 train_time:372159ms step_avg:153.85ms step:2430/5100 train_loss:3.4239 train_time:372311ms step_avg:153.85ms step:2431/5100 train_loss:3.4793 train_time:372464ms step_avg:153.85ms step:2432/5100 train_loss:3.5407 train_time:372615ms step_avg:153.85ms step:2433/5100 train_loss:3.5056 train_time:372768ms step_avg:153.85ms step:2434/5100 train_loss:3.3765 train_time:372920ms step_avg:153.84ms step:2435/5100 train_loss:3.3395 train_time:373073ms step_avg:153.84ms step:2436/5100 train_loss:3.5087 train_time:373226ms step_avg:153.84ms step:2437/5100 train_loss:3.3628 train_time:373377ms step_avg:153.84ms step:2438/5100 train_loss:3.4393 train_time:373531ms step_avg:153.84ms step:2439/5100 train_loss:3.5276 train_time:373682ms step_avg:153.84ms step:2440/5100 train_loss:3.4499 train_time:373836ms step_avg:153.84ms step:2441/5100 train_loss:3.5368 train_time:373988ms step_avg:153.84ms step:2442/5100 train_loss:3.4278 train_time:374141ms step_avg:153.84ms step:2443/5100 train_loss:3.4849 train_time:374293ms step_avg:153.84ms step:2444/5100 train_loss:3.3688 train_time:374445ms step_avg:153.84ms step:2445/5100 train_loss:3.3783 train_time:374597ms step_avg:153.84ms step:2446/5100 train_loss:3.5410 train_time:374750ms step_avg:153.84ms step:2447/5100 train_loss:3.4003 train_time:374903ms step_avg:153.84ms step:2448/5100 train_loss:3.4718 train_time:375055ms step_avg:153.84ms step:2449/5100 train_loss:3.6359 train_time:375208ms step_avg:153.84ms step:2450/5100 train_loss:3.4628 train_time:375360ms step_avg:153.84ms step:2451/5100 train_loss:3.5373 train_time:375512ms step_avg:153.84ms step:2452/5100 train_loss:3.4395 train_time:375664ms step_avg:153.83ms step:2453/5100 train_loss:3.5421 train_time:375816ms step_avg:153.83ms step:2454/5100 train_loss:3.4315 train_time:375968ms step_avg:153.83ms step:2455/5100 train_loss:3.5741 train_time:376120ms step_avg:153.83ms step:2456/5100 train_loss:3.5002 train_time:376273ms step_avg:153.83ms step:2457/5100 train_loss:3.4194 train_time:376525ms step_avg:153.87ms step:2458/5100 train_loss:3.3486 train_time:376687ms step_avg:153.88ms step:2459/5100 train_loss:3.4766 train_time:376838ms step_avg:153.87ms step:2460/5100 train_loss:4.0714 train_time:376990ms step_avg:153.87ms step:2461/5100 train_loss:3.5425 train_time:377142ms step_avg:153.87ms step:2462/5100 train_loss:3.3551 train_time:377294ms step_avg:153.87ms step:2463/5100 train_loss:3.5571 train_time:377445ms step_avg:153.87ms step:2464/5100 train_loss:3.4698 train_time:377601ms step_avg:153.87ms step:2465/5100 train_loss:3.6670 train_time:377753ms step_avg:153.87ms step:2466/5100 train_loss:3.8454 train_time:377906ms step_avg:153.87ms step:2467/5100 train_loss:3.5789 train_time:378059ms step_avg:153.87ms step:2468/5100 train_loss:3.4533 train_time:378211ms step_avg:153.87ms step:2469/5100 train_loss:3.5687 train_time:378364ms step_avg:153.87ms step:2470/5100 train_loss:3.5867 train_time:378639ms step_avg:153.92ms step:2471/5100 train_loss:3.3879 train_time:378788ms step_avg:153.92ms step:2472/5100 train_loss:3.4786 train_time:378941ms step_avg:153.92ms step:2473/5100 train_loss:3.4699 train_time:379093ms step_avg:153.92ms step:2474/5100 train_loss:3.6179 train_time:379245ms step_avg:153.91ms step:2475/5100 train_loss:3.7471 train_time:379397ms step_avg:153.91ms step:2476/5100 train_loss:3.3328 train_time:379550ms step_avg:153.91ms step:2477/5100 train_loss:3.5488 train_time:379703ms step_avg:153.91ms step:2478/5100 train_loss:3.5110 train_time:379855ms step_avg:153.91ms step:2479/5100 train_loss:3.3522 train_time:380008ms step_avg:153.91ms step:2480/5100 train_loss:3.3426 train_time:380161ms step_avg:153.91ms step:2481/5100 train_loss:3.4880 train_time:380313ms step_avg:153.91ms step:2482/5100 train_loss:3.5089 train_time:380465ms step_avg:153.91ms step:2483/5100 train_loss:3.5145 train_time:380618ms step_avg:153.91ms step:2484/5100 train_loss:3.4798 train_time:380771ms step_avg:153.91ms step:2485/5100 train_loss:3.4902 train_time:380923ms step_avg:153.91ms step:2486/5100 train_loss:3.3698 train_time:381075ms step_avg:153.91ms step:2487/5100 train_loss:3.5675 train_time:381228ms step_avg:153.91ms step:2488/5100 train_loss:3.5260 train_time:381380ms step_avg:153.91ms step:2489/5100 train_loss:3.4312 train_time:381534ms step_avg:153.91ms step:2490/5100 train_loss:3.5357 train_time:381686ms step_avg:153.91ms step:2491/5100 train_loss:3.5863 train_time:381839ms step_avg:153.91ms step:2492/5100 train_loss:3.6692 train_time:381992ms step_avg:153.90ms step:2493/5100 train_loss:3.5200 train_time:382143ms step_avg:153.90ms step:2494/5100 train_loss:3.4433 train_time:382295ms step_avg:153.90ms step:2495/5100 train_loss:3.5744 train_time:382448ms step_avg:153.90ms step:2496/5100 train_loss:3.5172 train_time:382602ms step_avg:153.90ms step:2497/5100 train_loss:3.4311 train_time:382753ms step_avg:153.90ms step:2498/5100 train_loss:3.5297 train_time:382906ms step_avg:153.90ms step:2499/5100 train_loss:3.5787 train_time:383058ms step_avg:153.90ms step:2500/5100 train_loss:3.6020 train_time:383212ms step_avg:153.90ms step:2500/5100 val_loss:3.4755 train_time:383275ms step_avg:153.93ms step:2501/5100 train_loss:3.5442 train_time:383371ms step_avg:153.90ms step:2502/5100 train_loss:3.4943 train_time:383525ms step_avg:153.90ms step:2503/5100 train_loss:3.5174 train_time:383677ms step_avg:153.90ms step:2504/5100 train_loss:3.3864 train_time:383829ms step_avg:153.90ms step:2505/5100 train_loss:3.5804 train_time:383980ms step_avg:153.90ms step:2506/5100 train_loss:3.5305 train_time:384132ms step_avg:153.90ms step:2507/5100 train_loss:3.4714 train_time:384285ms step_avg:153.90ms step:2508/5100 train_loss:3.4823 train_time:384438ms step_avg:153.90ms step:2509/5100 train_loss:3.4407 train_time:384591ms step_avg:153.90ms step:2510/5100 train_loss:3.6057 train_time:384746ms step_avg:153.90ms step:2511/5100 train_loss:3.4384 train_time:384898ms step_avg:153.90ms step:2512/5100 train_loss:3.4271 train_time:385050ms step_avg:153.90ms step:2513/5100 train_loss:3.5005 train_time:385202ms step_avg:153.90ms step:2514/5100 train_loss:3.5364 train_time:385355ms step_avg:153.90ms step:2515/5100 train_loss:3.4326 train_time:385508ms step_avg:153.90ms step:2516/5100 train_loss:3.5277 train_time:385660ms step_avg:153.89ms step:2517/5100 train_loss:3.5169 train_time:385813ms step_avg:153.89ms step:2518/5100 train_loss:3.3986 train_time:385965ms step_avg:153.89ms step:2519/5100 train_loss:3.4206 train_time:386118ms step_avg:153.89ms step:2520/5100 train_loss:3.5477 train_time:386271ms step_avg:153.89ms step:2521/5100 train_loss:3.5401 train_time:386422ms step_avg:153.89ms step:2522/5100 train_loss:3.4167 train_time:386575ms step_avg:153.89ms step:2523/5100 train_loss:3.3997 train_time:386728ms step_avg:153.89ms step:2524/5100 train_loss:3.5000 train_time:386880ms step_avg:153.89ms step:2525/5100 train_loss:3.3427 train_time:387033ms step_avg:153.89ms step:2526/5100 train_loss:3.5659 train_time:387185ms step_avg:153.89ms step:2527/5100 train_loss:3.4738 train_time:387339ms step_avg:153.89ms step:2528/5100 train_loss:3.4790 train_time:387492ms step_avg:153.89ms step:2529/5100 train_loss:3.4571 train_time:387645ms step_avg:153.89ms step:2530/5100 train_loss:3.4863 train_time:387797ms step_avg:153.89ms step:2531/5100 train_loss:3.5274 train_time:387949ms step_avg:153.89ms step:2532/5100 train_loss:3.3384 train_time:388102ms step_avg:153.89ms step:2533/5100 train_loss:3.5034 train_time:388254ms step_avg:153.89ms step:2534/5100 train_loss:3.4001 train_time:388407ms step_avg:153.89ms step:2535/5100 train_loss:3.4316 train_time:388559ms step_avg:153.88ms step:2536/5100 train_loss:3.4931 train_time:388712ms step_avg:153.88ms step:2537/5100 train_loss:3.5005 train_time:388864ms step_avg:153.88ms step:2538/5100 train_loss:3.3276 train_time:389016ms step_avg:153.88ms step:2539/5100 train_loss:3.6324 train_time:389169ms step_avg:153.88ms step:2540/5100 train_loss:3.3140 train_time:389321ms step_avg:153.88ms step:2541/5100 train_loss:3.5004 train_time:389473ms step_avg:153.88ms step:2542/5100 train_loss:3.2708 train_time:389626ms step_avg:153.88ms step:2543/5100 train_loss:3.7093 train_time:389778ms step_avg:153.88ms step:2544/5100 train_loss:3.4657 train_time:389931ms step_avg:153.88ms step:2545/5100 train_loss:3.6321 train_time:390083ms step_avg:153.88ms step:2546/5100 train_loss:3.4587 train_time:390237ms step_avg:153.88ms step:2547/5100 train_loss:3.4452 train_time:390389ms step_avg:153.88ms step:2548/5100 train_loss:3.4466 train_time:390542ms step_avg:153.88ms step:2549/5100 train_loss:3.6192 train_time:390694ms step_avg:153.88ms step:2550/5100 train_loss:3.4726 train_time:390846ms step_avg:153.88ms step:2551/5100 train_loss:3.4645 train_time:390997ms step_avg:153.88ms step:2552/5100 train_loss:3.4952 train_time:391150ms step_avg:153.88ms step:2553/5100 train_loss:3.5185 train_time:391302ms step_avg:153.87ms step:2554/5100 train_loss:3.4265 train_time:391455ms step_avg:153.87ms step:2555/5100 train_loss:3.5335 train_time:391607ms step_avg:153.87ms step:2556/5100 train_loss:3.5858 train_time:391761ms step_avg:153.87ms step:2557/5100 train_loss:3.5776 train_time:391913ms step_avg:153.87ms step:2558/5100 train_loss:3.4179 train_time:392065ms step_avg:153.87ms step:2559/5100 train_loss:3.4160 train_time:392217ms step_avg:153.87ms step:2560/5100 train_loss:3.4324 train_time:392370ms step_avg:153.87ms step:2561/5100 train_loss:3.5549 train_time:392523ms step_avg:153.87ms step:2562/5100 train_loss:3.5852 train_time:392675ms step_avg:153.87ms step:2563/5100 train_loss:3.4679 train_time:392828ms step_avg:153.87ms step:2564/5100 train_loss:3.5021 train_time:392980ms step_avg:153.87ms step:2565/5100 train_loss:3.4127 train_time:393133ms step_avg:153.87ms step:2566/5100 train_loss:3.4283 train_time:393285ms step_avg:153.87ms step:2567/5100 train_loss:3.4210 train_time:393437ms step_avg:153.87ms step:2568/5100 train_loss:3.4714 train_time:393590ms step_avg:153.87ms step:2569/5100 train_loss:3.6198 train_time:393743ms step_avg:153.87ms step:2570/5100 train_loss:3.5103 train_time:393896ms step_avg:153.87ms step:2571/5100 train_loss:3.5976 train_time:394049ms step_avg:153.87ms step:2572/5100 train_loss:3.3509 train_time:394202ms step_avg:153.86ms step:2573/5100 train_loss:3.4604 train_time:394354ms step_avg:153.86ms step:2574/5100 train_loss:3.1360 train_time:394506ms step_avg:153.86ms step:2575/5100 train_loss:3.3719 train_time:394658ms step_avg:153.86ms step:2576/5100 train_loss:3.3080 train_time:394810ms step_avg:153.86ms step:2577/5100 train_loss:3.4278 train_time:394963ms step_avg:153.86ms step:2578/5100 train_loss:3.4824 train_time:395116ms step_avg:153.86ms step:2579/5100 train_loss:3.3866 train_time:395269ms step_avg:153.86ms step:2580/5100 train_loss:3.4419 train_time:395422ms step_avg:153.86ms step:2581/5100 train_loss:3.3892 train_time:395573ms step_avg:153.86ms step:2582/5100 train_loss:3.4928 train_time:395726ms step_avg:153.86ms step:2583/5100 train_loss:3.3725 train_time:395877ms step_avg:153.86ms step:2584/5100 train_loss:3.5646 train_time:396029ms step_avg:153.86ms step:2585/5100 train_loss:3.4869 train_time:396181ms step_avg:153.86ms step:2586/5100 train_loss:3.4897 train_time:396335ms step_avg:153.86ms step:2587/5100 train_loss:3.6175 train_time:396487ms step_avg:153.86ms step:2588/5100 train_loss:3.5027 train_time:396640ms step_avg:153.86ms step:2589/5100 train_loss:3.3623 train_time:396793ms step_avg:153.86ms step:2590/5100 train_loss:3.5252 train_time:396946ms step_avg:153.85ms step:2591/5100 train_loss:3.4393 train_time:397098ms step_avg:153.85ms step:2592/5100 train_loss:3.6369 train_time:397251ms step_avg:153.85ms step:2593/5100 train_loss:3.5159 train_time:397403ms step_avg:153.85ms step:2594/5100 train_loss:3.3329 train_time:397556ms step_avg:153.85ms step:2595/5100 train_loss:3.4073 train_time:397709ms step_avg:153.85ms step:2596/5100 train_loss:3.8654 train_time:397861ms step_avg:153.85ms step:2597/5100 train_loss:3.4923 train_time:398013ms step_avg:153.85ms step:2598/5100 train_loss:3.4920 train_time:398166ms step_avg:153.85ms step:2599/5100 train_loss:3.3468 train_time:398318ms step_avg:153.85ms step:2600/5100 train_loss:3.5875 train_time:398471ms step_avg:153.85ms step:2601/5100 train_loss:3.7520 train_time:398623ms step_avg:153.85ms step:2602/5100 train_loss:3.3274 train_time:398777ms step_avg:153.85ms step:2603/5100 train_loss:3.4703 train_time:398929ms step_avg:153.85ms step:2604/5100 train_loss:3.3008 train_time:399081ms step_avg:153.85ms step:2605/5100 train_loss:3.5957 train_time:399235ms step_avg:153.85ms step:2606/5100 train_loss:3.4683 train_time:399387ms step_avg:153.85ms step:2607/5100 train_loss:3.3639 train_time:399540ms step_avg:153.85ms step:2608/5100 train_loss:3.3183 train_time:399693ms step_avg:153.85ms step:2609/5100 train_loss:3.4357 train_time:399845ms step_avg:153.85ms step:2610/5100 train_loss:3.6166 train_time:399998ms step_avg:153.85ms step:2611/5100 train_loss:3.4908 train_time:400150ms step_avg:153.84ms step:2612/5100 train_loss:3.3169 train_time:400304ms step_avg:153.84ms step:2613/5100 train_loss:3.4231 train_time:400455ms step_avg:153.84ms step:2614/5100 train_loss:3.5304 train_time:400608ms step_avg:153.84ms step:2615/5100 train_loss:3.4620 train_time:400760ms step_avg:153.84ms step:2616/5100 train_loss:3.4549 train_time:400913ms step_avg:153.84ms step:2617/5100 train_loss:3.4936 train_time:401066ms step_avg:153.84ms step:2618/5100 train_loss:3.5333 train_time:401218ms step_avg:153.84ms step:2619/5100 train_loss:3.3771 train_time:401371ms step_avg:153.84ms step:2620/5100 train_loss:3.5588 train_time:401523ms step_avg:153.84ms step:2621/5100 train_loss:3.5186 train_time:401675ms step_avg:153.84ms step:2622/5100 train_loss:3.6455 train_time:401828ms step_avg:153.84ms step:2623/5100 train_loss:3.5601 train_time:401980ms step_avg:153.84ms step:2624/5100 train_loss:3.4789 train_time:402133ms step_avg:153.84ms step:2625/5100 train_loss:3.4385 train_time:402284ms step_avg:153.84ms step:2625/5100 val_loss:3.4655 train_time:402348ms step_avg:153.86ms step:2626/5100 train_loss:3.4545 train_time:402444ms step_avg:153.84ms step:2627/5100 train_loss:3.5198 train_time:402603ms step_avg:153.84ms step:2628/5100 train_loss:3.3511 train_time:402758ms step_avg:153.84ms step:2629/5100 train_loss:3.6020 train_time:402911ms step_avg:153.84ms step:2630/5100 train_loss:3.4912 train_time:403063ms step_avg:153.84ms step:2631/5100 train_loss:3.5398 train_time:403213ms step_avg:153.84ms step:2632/5100 train_loss:3.7688 train_time:403366ms step_avg:153.84ms step:2633/5100 train_loss:3.5132 train_time:403520ms step_avg:153.84ms step:2634/5100 train_loss:3.4324 train_time:403673ms step_avg:153.84ms step:2635/5100 train_loss:3.4007 train_time:403827ms step_avg:153.84ms step:2636/5100 train_loss:3.4458 train_time:403978ms step_avg:153.84ms step:2637/5100 train_loss:3.2304 train_time:404130ms step_avg:153.84ms step:2638/5100 train_loss:3.5420 train_time:404282ms step_avg:153.84ms step:2639/5100 train_loss:3.5188 train_time:404435ms step_avg:153.84ms step:2640/5100 train_loss:3.4097 train_time:404587ms step_avg:153.84ms step:2641/5100 train_loss:3.4915 train_time:404740ms step_avg:153.83ms step:2642/5100 train_loss:3.5269 train_time:404893ms step_avg:153.83ms step:2643/5100 train_loss:3.3170 train_time:405045ms step_avg:153.83ms step:2644/5100 train_loss:3.4407 train_time:405198ms step_avg:153.83ms step:2645/5100 train_loss:3.5044 train_time:405350ms step_avg:153.83ms step:2646/5100 train_loss:3.4796 train_time:405602ms step_avg:153.87ms step:2647/5100 train_loss:3.3590 train_time:405764ms step_avg:153.87ms step:2648/5100 train_loss:3.5946 train_time:405916ms step_avg:153.87ms step:2649/5100 train_loss:3.8489 train_time:406067ms step_avg:153.87ms step:2650/5100 train_loss:3.4863 train_time:406219ms step_avg:153.87ms step:2651/5100 train_loss:3.4451 train_time:406371ms step_avg:153.87ms step:2652/5100 train_loss:3.5805 train_time:406523ms step_avg:153.87ms step:2653/5100 train_loss:3.4175 train_time:406676ms step_avg:153.87ms step:2654/5100 train_loss:3.4015 train_time:406831ms step_avg:153.87ms step:2655/5100 train_loss:3.4714 train_time:406984ms step_avg:153.87ms step:2656/5100 train_loss:3.3936 train_time:407137ms step_avg:153.87ms step:2657/5100 train_loss:3.4356 train_time:407289ms step_avg:153.87ms step:2658/5100 train_loss:3.3999 train_time:407441ms step_avg:153.87ms step:2659/5100 train_loss:3.4792 train_time:407593ms step_avg:153.87ms step:2660/5100 train_loss:3.6262 train_time:407868ms step_avg:153.91ms step:2661/5100 train_loss:3.4254 train_time:408016ms step_avg:153.91ms step:2662/5100 train_loss:3.5755 train_time:408168ms step_avg:153.91ms step:2663/5100 train_loss:3.4316 train_time:408320ms step_avg:153.91ms step:2664/5100 train_loss:3.4311 train_time:408471ms step_avg:153.91ms step:2665/5100 train_loss:3.3607 train_time:408624ms step_avg:153.91ms step:2666/5100 train_loss:3.4137 train_time:408776ms step_avg:153.91ms step:2667/5100 train_loss:3.4493 train_time:408930ms step_avg:153.91ms step:2668/5100 train_loss:3.4916 train_time:409083ms step_avg:153.91ms step:2669/5100 train_loss:3.4069 train_time:409235ms step_avg:153.91ms step:2670/5100 train_loss:3.4710 train_time:409388ms step_avg:153.91ms step:2671/5100 train_loss:3.3508 train_time:409540ms step_avg:153.90ms step:2672/5100 train_loss:3.4196 train_time:409692ms step_avg:153.90ms step:2673/5100 train_loss:3.4106 train_time:409845ms step_avg:153.90ms step:2674/5100 train_loss:3.4672 train_time:409998ms step_avg:153.90ms step:2675/5100 train_loss:3.4922 train_time:410150ms step_avg:153.90ms step:2676/5100 train_loss:3.4580 train_time:410303ms step_avg:153.90ms step:2677/5100 train_loss:3.4543 train_time:410455ms step_avg:153.90ms step:2678/5100 train_loss:3.4853 train_time:410608ms step_avg:153.90ms step:2679/5100 train_loss:3.5294 train_time:410760ms step_avg:153.90ms step:2680/5100 train_loss:3.4398 train_time:410911ms step_avg:153.90ms step:2681/5100 train_loss:3.3633 train_time:411065ms step_avg:153.90ms step:2682/5100 train_loss:3.4101 train_time:411217ms step_avg:153.90ms step:2683/5100 train_loss:3.8808 train_time:411369ms step_avg:153.90ms step:2684/5100 train_loss:3.4637 train_time:411522ms step_avg:153.90ms step:2685/5100 train_loss:3.4954 train_time:411673ms step_avg:153.90ms step:2686/5100 train_loss:3.5468 train_time:411827ms step_avg:153.90ms step:2687/5100 train_loss:3.4604 train_time:411979ms step_avg:153.90ms step:2688/5100 train_loss:3.5447 train_time:412133ms step_avg:153.90ms step:2689/5100 train_loss:3.4747 train_time:412285ms step_avg:153.90ms step:2690/5100 train_loss:3.4619 train_time:412438ms step_avg:153.89ms step:2691/5100 train_loss:3.4885 train_time:412589ms step_avg:153.89ms step:2692/5100 train_loss:3.5585 train_time:412742ms step_avg:153.89ms step:2693/5100 train_loss:3.3639 train_time:412893ms step_avg:153.89ms step:2694/5100 train_loss:3.7344 train_time:413046ms step_avg:153.89ms step:2695/5100 train_loss:3.5372 train_time:413199ms step_avg:153.89ms step:2696/5100 train_loss:3.3571 train_time:413350ms step_avg:153.89ms step:2697/5100 train_loss:3.5239 train_time:413503ms step_avg:153.89ms step:2698/5100 train_loss:3.4817 train_time:413655ms step_avg:153.89ms step:2699/5100 train_loss:3.4370 train_time:413808ms step_avg:153.89ms step:2700/5100 train_loss:3.5384 train_time:413960ms step_avg:153.89ms step:2701/5100 train_loss:3.5025 train_time:414112ms step_avg:153.89ms step:2702/5100 train_loss:3.4089 train_time:414265ms step_avg:153.89ms step:2703/5100 train_loss:3.4393 train_time:414417ms step_avg:153.89ms step:2704/5100 train_loss:3.4498 train_time:414569ms step_avg:153.89ms step:2705/5100 train_loss:3.4185 train_time:414722ms step_avg:153.89ms step:2706/5100 train_loss:3.6011 train_time:414874ms step_avg:153.88ms step:2707/5100 train_loss:3.5523 train_time:415027ms step_avg:153.88ms step:2708/5100 train_loss:3.4655 train_time:415179ms step_avg:153.88ms step:2709/5100 train_loss:3.4522 train_time:415331ms step_avg:153.88ms step:2710/5100 train_loss:3.5598 train_time:415484ms step_avg:153.88ms step:2711/5100 train_loss:3.4334 train_time:415637ms step_avg:153.88ms step:2712/5100 train_loss:3.5493 train_time:415790ms step_avg:153.88ms step:2713/5100 train_loss:3.2897 train_time:415942ms step_avg:153.88ms step:2714/5100 train_loss:3.4849 train_time:416095ms step_avg:153.88ms step:2715/5100 train_loss:3.3791 train_time:416248ms step_avg:153.88ms step:2716/5100 train_loss:3.3855 train_time:416399ms step_avg:153.88ms step:2717/5100 train_loss:3.5733 train_time:416551ms step_avg:153.88ms step:2718/5100 train_loss:3.4719 train_time:416704ms step_avg:153.88ms step:2719/5100 train_loss:3.7049 train_time:416858ms step_avg:153.88ms step:2720/5100 train_loss:3.4481 train_time:417009ms step_avg:153.88ms step:2721/5100 train_loss:3.4466 train_time:417163ms step_avg:153.88ms step:2722/5100 train_loss:3.6685 train_time:417315ms step_avg:153.88ms step:2723/5100 train_loss:3.4433 train_time:417467ms step_avg:153.88ms step:2724/5100 train_loss:3.6152 train_time:417621ms step_avg:153.88ms step:2725/5100 train_loss:3.4957 train_time:417772ms step_avg:153.88ms step:2726/5100 train_loss:3.4540 train_time:417925ms step_avg:153.88ms step:2727/5100 train_loss:3.4512 train_time:418076ms step_avg:153.87ms step:2728/5100 train_loss:3.7966 train_time:418229ms step_avg:153.87ms step:2729/5100 train_loss:3.5281 train_time:418381ms step_avg:153.87ms step:2730/5100 train_loss:3.3995 train_time:418534ms step_avg:153.87ms step:2731/5100 train_loss:3.5015 train_time:418688ms step_avg:153.87ms step:2732/5100 train_loss:3.4131 train_time:418840ms step_avg:153.87ms step:2733/5100 train_loss:3.2985 train_time:418993ms step_avg:153.87ms step:2734/5100 train_loss:3.4086 train_time:419146ms step_avg:153.87ms step:2735/5100 train_loss:3.4808 train_time:419298ms step_avg:153.87ms step:2736/5100 train_loss:3.3758 train_time:419449ms step_avg:153.87ms step:2737/5100 train_loss:3.7823 train_time:419602ms step_avg:153.87ms step:2738/5100 train_loss:3.5215 train_time:419755ms step_avg:153.87ms step:2739/5100 train_loss:3.7175 train_time:419908ms step_avg:153.87ms step:2740/5100 train_loss:3.4701 train_time:420061ms step_avg:153.87ms step:2741/5100 train_loss:3.4689 train_time:420213ms step_avg:153.87ms step:2742/5100 train_loss:3.4019 train_time:420365ms step_avg:153.87ms step:2743/5100 train_loss:3.4793 train_time:420517ms step_avg:153.87ms step:2744/5100 train_loss:3.4857 train_time:420669ms step_avg:153.87ms step:2745/5100 train_loss:3.5915 train_time:420821ms step_avg:153.87ms step:2746/5100 train_loss:3.3581 train_time:420974ms step_avg:153.86ms step:2747/5100 train_loss:3.4477 train_time:421127ms step_avg:153.86ms step:2748/5100 train_loss:3.4856 train_time:421279ms step_avg:153.86ms step:2749/5100 train_loss:3.6004 train_time:421431ms step_avg:153.86ms step:2750/5100 train_loss:3.4400 train_time:421583ms step_avg:153.86ms step:2750/5100 val_loss:3.4576 train_time:421648ms step_avg:153.89ms step:2751/5100 train_loss:3.5051 train_time:421740ms step_avg:153.86ms step:2752/5100 train_loss:3.5676 train_time:421898ms step_avg:153.87ms step:2753/5100 train_loss:3.4761 train_time:422049ms step_avg:153.86ms step:2754/5100 train_loss:3.4045 train_time:422201ms step_avg:153.86ms step:2755/5100 train_loss:3.4089 train_time:422352ms step_avg:153.86ms step:2756/5100 train_loss:3.4898 train_time:422505ms step_avg:153.86ms step:2757/5100 train_loss:3.4324 train_time:422657ms step_avg:153.86ms step:2758/5100 train_loss:3.3067 train_time:422811ms step_avg:153.86ms step:2759/5100 train_loss:3.6977 train_time:422964ms step_avg:153.86ms step:2760/5100 train_loss:3.5080 train_time:423116ms step_avg:153.86ms step:2761/5100 train_loss:3.4685 train_time:423268ms step_avg:153.86ms step:2762/5100 train_loss:3.4408 train_time:423420ms step_avg:153.86ms step:2763/5100 train_loss:3.3628 train_time:423572ms step_avg:153.86ms step:2764/5100 train_loss:3.5269 train_time:423724ms step_avg:153.86ms step:2765/5100 train_loss:3.4437 train_time:423877ms step_avg:153.86ms step:2766/5100 train_loss:3.3461 train_time:424030ms step_avg:153.86ms step:2767/5100 train_loss:3.4302 train_time:424183ms step_avg:153.86ms step:2768/5100 train_loss:3.5210 train_time:424335ms step_avg:153.86ms step:2769/5100 train_loss:3.3952 train_time:424488ms step_avg:153.86ms step:2770/5100 train_loss:3.4750 train_time:424640ms step_avg:153.86ms step:2771/5100 train_loss:3.4477 train_time:424793ms step_avg:153.85ms step:2772/5100 train_loss:3.8841 train_time:424945ms step_avg:153.85ms step:2773/5100 train_loss:3.3534 train_time:425099ms step_avg:153.85ms step:2774/5100 train_loss:3.4854 train_time:425251ms step_avg:153.85ms step:2775/5100 train_loss:3.5459 train_time:425403ms step_avg:153.85ms step:2776/5100 train_loss:3.5159 train_time:425555ms step_avg:153.85ms step:2777/5100 train_loss:3.5940 train_time:425708ms step_avg:153.85ms step:2778/5100 train_loss:3.6023 train_time:425861ms step_avg:153.85ms step:2779/5100 train_loss:3.4597 train_time:426012ms step_avg:153.85ms step:2780/5100 train_loss:3.3336 train_time:426165ms step_avg:153.85ms step:2781/5100 train_loss:3.4824 train_time:426318ms step_avg:153.85ms step:2782/5100 train_loss:3.4994 train_time:426470ms step_avg:153.85ms step:2783/5100 train_loss:3.3647 train_time:426622ms step_avg:153.85ms step:2784/5100 train_loss:3.4814 train_time:426774ms step_avg:153.85ms step:2785/5100 train_loss:3.5300 train_time:426927ms step_avg:153.85ms step:2786/5100 train_loss:3.4005 train_time:427079ms step_avg:153.85ms step:2787/5100 train_loss:3.5391 train_time:427231ms step_avg:153.85ms step:2788/5100 train_loss:3.4895 train_time:427383ms step_avg:153.85ms step:2789/5100 train_loss:3.4289 train_time:427535ms step_avg:153.85ms step:2790/5100 train_loss:3.5080 train_time:427687ms step_avg:153.84ms step:2791/5100 train_loss:3.4375 train_time:427839ms step_avg:153.84ms step:2792/5100 train_loss:3.3375 train_time:427994ms step_avg:153.84ms step:2793/5100 train_loss:3.4364 train_time:428145ms step_avg:153.84ms step:2794/5100 train_loss:3.4754 train_time:428298ms step_avg:153.84ms step:2795/5100 train_loss:3.3915 train_time:428449ms step_avg:153.84ms step:2796/5100 train_loss:3.4319 train_time:428602ms step_avg:153.84ms step:2797/5100 train_loss:3.3594 train_time:428754ms step_avg:153.84ms step:2798/5100 train_loss:3.4597 train_time:428907ms step_avg:153.84ms step:2799/5100 train_loss:3.4105 train_time:429059ms step_avg:153.84ms step:2800/5100 train_loss:3.5858 train_time:429211ms step_avg:153.84ms step:2801/5100 train_loss:3.5338 train_time:429363ms step_avg:153.84ms step:2802/5100 train_loss:3.4983 train_time:429515ms step_avg:153.84ms step:2803/5100 train_loss:3.4427 train_time:429668ms step_avg:153.84ms step:2804/5100 train_loss:3.6282 train_time:429821ms step_avg:153.84ms step:2805/5100 train_loss:3.5943 train_time:429974ms step_avg:153.84ms step:2806/5100 train_loss:3.3178 train_time:430126ms step_avg:153.84ms step:2807/5100 train_loss:3.7207 train_time:430279ms step_avg:153.84ms step:2808/5100 train_loss:3.4686 train_time:430431ms step_avg:153.84ms step:2809/5100 train_loss:3.3944 train_time:430584ms step_avg:153.83ms step:2810/5100 train_loss:3.4306 train_time:430737ms step_avg:153.83ms step:2811/5100 train_loss:3.5772 train_time:430888ms step_avg:153.83ms step:2812/5100 train_loss:3.5638 train_time:431040ms step_avg:153.83ms step:2813/5100 train_loss:3.3124 train_time:431192ms step_avg:153.83ms step:2814/5100 train_loss:3.5361 train_time:431345ms step_avg:153.83ms step:2815/5100 train_loss:3.6120 train_time:431499ms step_avg:153.83ms step:2816/5100 train_loss:3.4175 train_time:431650ms step_avg:153.83ms step:2817/5100 train_loss:3.1691 train_time:431803ms step_avg:153.83ms step:2818/5100 train_loss:3.4419 train_time:431954ms step_avg:153.83ms step:2819/5100 train_loss:3.4142 train_time:432107ms step_avg:153.83ms step:2820/5100 train_loss:3.6073 train_time:432260ms step_avg:153.83ms step:2821/5100 train_loss:3.5615 train_time:432412ms step_avg:153.83ms step:2822/5100 train_loss:3.5324 train_time:432564ms step_avg:153.83ms step:2823/5100 train_loss:3.4702 train_time:432716ms step_avg:153.83ms step:2824/5100 train_loss:3.4283 train_time:432869ms step_avg:153.83ms step:2825/5100 train_loss:3.3228 train_time:433020ms step_avg:153.83ms step:2826/5100 train_loss:3.5925 train_time:433174ms step_avg:153.83ms step:2827/5100 train_loss:3.4900 train_time:433325ms step_avg:153.83ms step:2828/5100 train_loss:3.3613 train_time:433478ms step_avg:153.82ms step:2829/5100 train_loss:3.5054 train_time:433630ms step_avg:153.82ms step:2830/5100 train_loss:3.5017 train_time:433783ms step_avg:153.82ms step:2831/5100 train_loss:3.4436 train_time:433935ms step_avg:153.82ms step:2832/5100 train_loss:3.5823 train_time:434087ms step_avg:153.82ms step:2833/5100 train_loss:3.5008 train_time:434240ms step_avg:153.82ms step:2834/5100 train_loss:3.4835 train_time:434392ms step_avg:153.82ms step:2835/5100 train_loss:3.2959 train_time:434642ms step_avg:153.86ms step:2836/5100 train_loss:3.5186 train_time:434803ms step_avg:153.86ms step:2837/5100 train_loss:3.4467 train_time:434954ms step_avg:153.86ms step:2838/5100 train_loss:3.7660 train_time:435107ms step_avg:153.86ms step:2839/5100 train_loss:3.4043 train_time:435259ms step_avg:153.86ms step:2840/5100 train_loss:3.4132 train_time:435412ms step_avg:153.86ms step:2841/5100 train_loss:3.4716 train_time:435563ms step_avg:153.85ms step:2842/5100 train_loss:3.3958 train_time:435715ms step_avg:153.85ms step:2843/5100 train_loss:3.3958 train_time:435868ms step_avg:153.85ms step:2844/5100 train_loss:3.5767 train_time:436020ms step_avg:153.85ms step:2845/5100 train_loss:3.4604 train_time:436174ms step_avg:153.85ms step:2846/5100 train_loss:3.4873 train_time:436326ms step_avg:153.85ms step:2847/5100 train_loss:3.4389 train_time:436478ms step_avg:153.85ms step:2848/5100 train_loss:3.7160 train_time:436630ms step_avg:153.85ms step:2849/5100 train_loss:3.3850 train_time:436784ms step_avg:153.85ms step:2850/5100 train_loss:3.4093 train_time:437092ms step_avg:153.91ms step:2851/5100 train_loss:3.5128 train_time:437250ms step_avg:153.91ms step:2852/5100 train_loss:3.4771 train_time:437402ms step_avg:153.91ms step:2853/5100 train_loss:3.4534 train_time:437555ms step_avg:153.91ms step:2854/5100 train_loss:3.5235 train_time:437707ms step_avg:153.91ms step:2855/5100 train_loss:3.3432 train_time:437859ms step_avg:153.90ms step:2856/5100 train_loss:3.3657 train_time:438011ms step_avg:153.90ms step:2857/5100 train_loss:3.4539 train_time:438164ms step_avg:153.90ms step:2858/5100 train_loss:3.4582 train_time:438318ms step_avg:153.90ms step:2859/5100 train_loss:3.3383 train_time:438470ms step_avg:153.90ms step:2860/5100 train_loss:3.4473 train_time:438622ms step_avg:153.90ms step:2861/5100 train_loss:3.4066 train_time:438774ms step_avg:153.90ms step:2862/5100 train_loss:3.4514 train_time:438926ms step_avg:153.90ms step:2863/5100 train_loss:3.4953 train_time:439079ms step_avg:153.90ms step:2864/5100 train_loss:3.7611 train_time:439232ms step_avg:153.90ms step:2865/5100 train_loss:3.5820 train_time:439384ms step_avg:153.90ms step:2866/5100 train_loss:3.4540 train_time:439537ms step_avg:153.90ms step:2867/5100 train_loss:3.3440 train_time:439689ms step_avg:153.90ms step:2868/5100 train_loss:3.5435 train_time:439841ms step_avg:153.90ms step:2869/5100 train_loss:3.4947 train_time:439993ms step_avg:153.90ms step:2870/5100 train_loss:3.4528 train_time:440145ms step_avg:153.90ms step:2871/5100 train_loss:3.5907 train_time:440299ms step_avg:153.90ms step:2872/5100 train_loss:3.3789 train_time:440452ms step_avg:153.90ms step:2873/5100 train_loss:3.4264 train_time:440604ms step_avg:153.90ms step:2874/5100 train_loss:3.2954 train_time:440758ms step_avg:153.90ms step:2875/5100 train_loss:3.4485 train_time:440909ms step_avg:153.90ms step:2875/5100 val_loss:3.4518 train_time:440973ms step_avg:153.92ms step:2876/5100 train_loss:3.3660 train_time:441065ms step_avg:153.90ms step:2877/5100 train_loss:3.3559 train_time:441224ms step_avg:153.90ms step:2878/5100 train_loss:3.4395 train_time:441376ms step_avg:153.90ms step:2879/5100 train_loss:3.5620 train_time:441529ms step_avg:153.90ms step:2880/5100 train_loss:3.5107 train_time:441681ms step_avg:153.90ms step:2881/5100 train_loss:3.4483 train_time:441832ms step_avg:153.89ms step:2882/5100 train_loss:3.4352 train_time:441985ms step_avg:153.89ms step:2883/5100 train_loss:3.5653 train_time:442139ms step_avg:153.89ms step:2884/5100 train_loss:3.3442 train_time:442292ms step_avg:153.89ms step:2885/5100 train_loss:3.3733 train_time:442444ms step_avg:153.89ms step:2886/5100 train_loss:3.4107 train_time:442596ms step_avg:153.89ms step:2887/5100 train_loss:3.4106 train_time:442748ms step_avg:153.89ms step:2888/5100 train_loss:3.4221 train_time:442901ms step_avg:153.89ms step:2889/5100 train_loss:3.4424 train_time:443053ms step_avg:153.89ms step:2890/5100 train_loss:3.6260 train_time:443206ms step_avg:153.89ms step:2891/5100 train_loss:3.4699 train_time:443358ms step_avg:153.89ms step:2892/5100 train_loss:3.3135 train_time:443511ms step_avg:153.89ms step:2893/5100 train_loss:3.2417 train_time:443664ms step_avg:153.89ms step:2894/5100 train_loss:3.3852 train_time:443818ms step_avg:153.89ms step:2895/5100 train_loss:3.2619 train_time:443969ms step_avg:153.89ms step:2896/5100 train_loss:3.4442 train_time:444121ms step_avg:153.89ms step:2897/5100 train_loss:3.5724 train_time:444274ms step_avg:153.89ms step:2898/5100 train_loss:3.3979 train_time:444426ms step_avg:153.89ms step:2899/5100 train_loss:3.4827 train_time:444579ms step_avg:153.89ms step:2900/5100 train_loss:3.3737 train_time:444732ms step_avg:153.89ms step:2901/5100 train_loss:3.5585 train_time:444885ms step_avg:153.89ms step:2902/5100 train_loss:3.5499 train_time:445038ms step_avg:153.89ms step:2903/5100 train_loss:3.5956 train_time:445190ms step_avg:153.89ms step:2904/5100 train_loss:3.3013 train_time:445342ms step_avg:153.88ms step:2905/5100 train_loss:3.4432 train_time:445495ms step_avg:153.88ms step:2906/5100 train_loss:3.4234 train_time:445646ms step_avg:153.88ms step:2907/5100 train_loss:3.5042 train_time:445801ms step_avg:153.88ms step:2908/5100 train_loss:3.4389 train_time:445953ms step_avg:153.88ms step:2909/5100 train_loss:3.4057 train_time:446105ms step_avg:153.88ms step:2910/5100 train_loss:3.7406 train_time:446259ms step_avg:153.88ms step:2911/5100 train_loss:3.4532 train_time:446412ms step_avg:153.88ms step:2912/5100 train_loss:3.3565 train_time:446564ms step_avg:153.88ms step:2913/5100 train_loss:3.3460 train_time:446716ms step_avg:153.88ms step:2914/5100 train_loss:3.8333 train_time:446869ms step_avg:153.88ms step:2915/5100 train_loss:3.4198 train_time:447021ms step_avg:153.88ms step:2916/5100 train_loss:3.3688 train_time:447174ms step_avg:153.88ms step:2917/5100 train_loss:3.3524 train_time:447326ms step_avg:153.88ms step:2918/5100 train_loss:3.6461 train_time:447481ms step_avg:153.88ms step:2919/5100 train_loss:3.1445 train_time:447633ms step_avg:153.88ms step:2920/5100 train_loss:3.3471 train_time:447786ms step_avg:153.88ms step:2921/5100 train_loss:3.3643 train_time:447938ms step_avg:153.88ms step:2922/5100 train_loss:3.4598 train_time:448090ms step_avg:153.88ms step:2923/5100 train_loss:3.5053 train_time:448242ms step_avg:153.88ms step:2924/5100 train_loss:3.5364 train_time:448395ms step_avg:153.88ms step:2925/5100 train_loss:3.5404 train_time:448547ms step_avg:153.88ms step:2926/5100 train_loss:3.4211 train_time:448701ms step_avg:153.88ms step:2927/5100 train_loss:3.4342 train_time:448854ms step_avg:153.88ms step:2928/5100 train_loss:3.4228 train_time:449006ms step_avg:153.87ms step:2929/5100 train_loss:3.4196 train_time:449160ms step_avg:153.87ms step:2930/5100 train_loss:3.3886 train_time:449312ms step_avg:153.87ms step:2931/5100 train_loss:3.4153 train_time:449465ms step_avg:153.87ms step:2932/5100 train_loss:3.5524 train_time:449618ms step_avg:153.87ms step:2933/5100 train_loss:3.5927 train_time:449771ms step_avg:153.87ms step:2934/5100 train_loss:3.5625 train_time:449923ms step_avg:153.87ms step:2935/5100 train_loss:3.4023 train_time:450075ms step_avg:153.87ms step:2936/5100 train_loss:3.4575 train_time:450228ms step_avg:153.87ms step:2937/5100 train_loss:3.3878 train_time:450381ms step_avg:153.87ms step:2938/5100 train_loss:3.4238 train_time:450534ms step_avg:153.87ms step:2939/5100 train_loss:3.4471 train_time:450685ms step_avg:153.87ms step:2940/5100 train_loss:3.4914 train_time:450839ms step_avg:153.87ms step:2941/5100 train_loss:3.5337 train_time:450991ms step_avg:153.87ms step:2942/5100 train_loss:3.5177 train_time:451143ms step_avg:153.87ms step:2943/5100 train_loss:3.4486 train_time:451296ms step_avg:153.87ms step:2944/5100 train_loss:3.3292 train_time:451449ms step_avg:153.87ms step:2945/5100 train_loss:3.2607 train_time:451602ms step_avg:153.87ms step:2946/5100 train_loss:3.4732 train_time:451755ms step_avg:153.87ms step:2947/5100 train_loss:3.5381 train_time:451907ms step_avg:153.87ms step:2948/5100 train_loss:3.4719 train_time:452061ms step_avg:153.87ms step:2949/5100 train_loss:3.6515 train_time:452213ms step_avg:153.87ms step:2950/5100 train_loss:3.4764 train_time:452364ms step_avg:153.87ms step:2951/5100 train_loss:3.4744 train_time:452518ms step_avg:153.87ms step:2952/5100 train_loss:3.8868 train_time:452670ms step_avg:153.86ms step:2953/5100 train_loss:3.5550 train_time:452822ms step_avg:153.86ms step:2954/5100 train_loss:3.4977 train_time:452976ms step_avg:153.86ms step:2955/5100 train_loss:3.5047 train_time:453129ms step_avg:153.86ms step:2956/5100 train_loss:3.4371 train_time:453282ms step_avg:153.86ms step:2957/5100 train_loss:3.4717 train_time:453435ms step_avg:153.86ms step:2958/5100 train_loss:3.3338 train_time:453586ms step_avg:153.86ms step:2959/5100 train_loss:3.4207 train_time:453739ms step_avg:153.86ms step:2960/5100 train_loss:3.5557 train_time:453890ms step_avg:153.86ms step:2961/5100 train_loss:3.3680 train_time:454042ms step_avg:153.86ms step:2962/5100 train_loss:3.5004 train_time:454197ms step_avg:153.86ms step:2963/5100 train_loss:3.3560 train_time:454348ms step_avg:153.86ms step:2964/5100 train_loss:3.4124 train_time:454502ms step_avg:153.86ms step:2965/5100 train_loss:3.4041 train_time:454655ms step_avg:153.86ms step:2966/5100 train_loss:3.5187 train_time:454807ms step_avg:153.86ms step:2967/5100 train_loss:3.3862 train_time:454961ms step_avg:153.86ms step:2968/5100 train_loss:3.6367 train_time:455113ms step_avg:153.86ms step:2969/5100 train_loss:3.4857 train_time:455265ms step_avg:153.86ms step:2970/5100 train_loss:3.4986 train_time:455418ms step_avg:153.86ms step:2971/5100 train_loss:3.4873 train_time:455571ms step_avg:153.86ms step:2972/5100 train_loss:3.5634 train_time:455723ms step_avg:153.86ms step:2973/5100 train_loss:3.3860 train_time:455877ms step_avg:153.86ms step:2974/5100 train_loss:3.3922 train_time:456030ms step_avg:153.86ms step:2975/5100 train_loss:3.3068 train_time:456182ms step_avg:153.86ms step:2976/5100 train_loss:3.3896 train_time:456335ms step_avg:153.86ms step:2977/5100 train_loss:3.3807 train_time:456487ms step_avg:153.85ms step:2978/5100 train_loss:3.3909 train_time:456640ms step_avg:153.85ms step:2979/5100 train_loss:3.6774 train_time:456793ms step_avg:153.85ms step:2980/5100 train_loss:3.4850 train_time:456945ms step_avg:153.85ms step:2981/5100 train_loss:3.5250 train_time:457099ms step_avg:153.85ms step:2982/5100 train_loss:3.5399 train_time:457251ms step_avg:153.85ms step:2983/5100 train_loss:3.6260 train_time:457404ms step_avg:153.85ms step:2984/5100 train_loss:3.4223 train_time:457557ms step_avg:153.85ms step:2985/5100 train_loss:3.5215 train_time:457709ms step_avg:153.85ms step:2986/5100 train_loss:3.5194 train_time:457862ms step_avg:153.85ms step:2987/5100 train_loss:3.4724 train_time:458014ms step_avg:153.85ms step:2988/5100 train_loss:3.5928 train_time:458166ms step_avg:153.85ms step:2989/5100 train_loss:3.1964 train_time:458319ms step_avg:153.85ms step:2990/5100 train_loss:3.5336 train_time:458472ms step_avg:153.85ms step:2991/5100 train_loss:3.4888 train_time:458624ms step_avg:153.85ms step:2992/5100 train_loss:3.4651 train_time:458777ms step_avg:153.85ms step:2993/5100 train_loss:3.3796 train_time:458930ms step_avg:153.85ms step:2994/5100 train_loss:3.5216 train_time:459082ms step_avg:153.85ms step:2995/5100 train_loss:3.3399 train_time:459234ms step_avg:153.85ms step:2996/5100 train_loss:3.3694 train_time:459385ms step_avg:153.85ms step:2997/5100 train_loss:3.4412 train_time:459539ms step_avg:153.85ms step:2998/5100 train_loss:3.3836 train_time:459691ms step_avg:153.85ms step:2999/5100 train_loss:3.5056 train_time:459843ms step_avg:153.85ms step:3000/5100 train_loss:3.4084 train_time:459998ms step_avg:153.85ms step:3000/5100 val_loss:3.4433 train_time:460061ms step_avg:153.87ms step:3001/5100 train_loss:3.3914 train_time:460155ms step_avg:153.85ms step:3002/5100 train_loss:3.3408 train_time:460312ms step_avg:153.85ms step:3003/5100 train_loss:3.3837 train_time:460463ms step_avg:153.85ms step:3004/5100 train_loss:3.5180 train_time:460615ms step_avg:153.85ms step:3005/5100 train_loss:3.8662 train_time:460768ms step_avg:153.85ms step:3006/5100 train_loss:3.4271 train_time:460920ms step_avg:153.85ms step:3007/5100 train_loss:3.4939 train_time:461074ms step_avg:153.85ms step:3008/5100 train_loss:3.3090 train_time:461227ms step_avg:153.84ms step:3009/5100 train_loss:3.5221 train_time:461379ms step_avg:153.84ms step:3010/5100 train_loss:3.4137 train_time:461532ms step_avg:153.84ms step:3011/5100 train_loss:3.4809 train_time:461683ms step_avg:153.84ms step:3012/5100 train_loss:3.4740 train_time:461836ms step_avg:153.84ms step:3013/5100 train_loss:3.3746 train_time:461988ms step_avg:153.84ms step:3014/5100 train_loss:3.5658 train_time:462141ms step_avg:153.84ms step:3015/5100 train_loss:3.5285 train_time:462293ms step_avg:153.84ms step:3016/5100 train_loss:3.3968 train_time:462445ms step_avg:153.84ms step:3017/5100 train_loss:3.4288 train_time:462597ms step_avg:153.84ms step:3018/5100 train_loss:3.4710 train_time:462749ms step_avg:153.84ms step:3019/5100 train_loss:3.5114 train_time:462901ms step_avg:153.84ms step:3020/5100 train_loss:3.3023 train_time:463054ms step_avg:153.84ms step:3021/5100 train_loss:3.5952 train_time:463207ms step_avg:153.84ms step:3022/5100 train_loss:3.4286 train_time:463360ms step_avg:153.84ms step:3023/5100 train_loss:3.3509 train_time:463511ms step_avg:153.84ms step:3024/5100 train_loss:3.4427 train_time:463767ms step_avg:153.87ms step:3025/5100 train_loss:3.4310 train_time:463923ms step_avg:153.87ms step:3026/5100 train_loss:3.4758 train_time:464075ms step_avg:153.87ms step:3027/5100 train_loss:3.5041 train_time:464226ms step_avg:153.87ms step:3028/5100 train_loss:3.4076 train_time:464378ms step_avg:153.87ms step:3029/5100 train_loss:3.2125 train_time:464530ms step_avg:153.87ms step:3030/5100 train_loss:3.5614 train_time:464683ms step_avg:153.87ms step:3031/5100 train_loss:3.3143 train_time:464836ms step_avg:153.87ms step:3032/5100 train_loss:3.3068 train_time:464989ms step_avg:153.87ms step:3033/5100 train_loss:3.6464 train_time:465141ms step_avg:153.87ms step:3034/5100 train_loss:3.6407 train_time:465294ms step_avg:153.87ms step:3035/5100 train_loss:3.4115 train_time:465445ms step_avg:153.87ms step:3036/5100 train_loss:3.4900 train_time:465599ms step_avg:153.87ms step:3037/5100 train_loss:3.4359 train_time:465751ms step_avg:153.87ms step:3038/5100 train_loss:3.3365 train_time:465903ms step_avg:153.86ms step:3039/5100 train_loss:3.3960 train_time:466056ms step_avg:153.86ms step:3040/5100 train_loss:3.4867 train_time:466332ms step_avg:153.90ms step:3041/5100 train_loss:3.4799 train_time:466482ms step_avg:153.90ms step:3042/5100 train_loss:3.2921 train_time:466634ms step_avg:153.90ms step:3043/5100 train_loss:3.4346 train_time:466785ms step_avg:153.90ms step:3044/5100 train_loss:3.4623 train_time:466938ms step_avg:153.90ms step:3045/5100 train_loss:3.4710 train_time:467090ms step_avg:153.90ms step:3046/5100 train_loss:3.5527 train_time:467242ms step_avg:153.90ms step:3047/5100 train_loss:3.3623 train_time:467395ms step_avg:153.90ms step:3048/5100 train_loss:3.4904 train_time:467548ms step_avg:153.90ms step:3049/5100 train_loss:3.4423 train_time:467701ms step_avg:153.90ms step:3050/5100 train_loss:3.3672 train_time:467854ms step_avg:153.90ms step:3051/5100 train_loss:3.4977 train_time:468005ms step_avg:153.90ms step:3052/5100 train_loss:3.3427 train_time:468159ms step_avg:153.90ms step:3053/5100 train_loss:3.5793 train_time:468311ms step_avg:153.90ms step:3054/5100 train_loss:3.5227 train_time:468463ms step_avg:153.90ms step:3055/5100 train_loss:3.5087 train_time:468615ms step_avg:153.90ms step:3056/5100 train_loss:3.5003 train_time:468768ms step_avg:153.90ms step:3057/5100 train_loss:3.3912 train_time:468921ms step_avg:153.90ms step:3058/5100 train_loss:3.4222 train_time:469073ms step_avg:153.90ms step:3059/5100 train_loss:3.5011 train_time:469225ms step_avg:153.89ms step:3060/5100 train_loss:3.3955 train_time:469379ms step_avg:153.89ms step:3061/5100 train_loss:3.4552 train_time:469531ms step_avg:153.89ms step:3062/5100 train_loss:3.4541 train_time:469684ms step_avg:153.89ms step:3063/5100 train_loss:3.3904 train_time:469837ms step_avg:153.89ms step:3064/5100 train_loss:3.3672 train_time:469989ms step_avg:153.89ms step:3065/5100 train_loss:3.3780 train_time:470141ms step_avg:153.89ms step:3066/5100 train_loss:3.3636 train_time:470294ms step_avg:153.89ms step:3067/5100 train_loss:3.3553 train_time:470446ms step_avg:153.89ms step:3068/5100 train_loss:3.3076 train_time:470600ms step_avg:153.89ms step:3069/5100 train_loss:3.3548 train_time:470753ms step_avg:153.89ms step:3070/5100 train_loss:3.3503 train_time:470905ms step_avg:153.89ms step:3071/5100 train_loss:3.5339 train_time:471058ms step_avg:153.89ms step:3072/5100 train_loss:3.4649 train_time:471210ms step_avg:153.89ms step:3073/5100 train_loss:3.5078 train_time:471362ms step_avg:153.89ms step:3074/5100 train_loss:3.4936 train_time:471514ms step_avg:153.89ms step:3075/5100 train_loss:3.4440 train_time:471666ms step_avg:153.89ms step:3076/5100 train_loss:3.4934 train_time:471820ms step_avg:153.89ms step:3077/5100 train_loss:3.5453 train_time:471972ms step_avg:153.89ms step:3078/5100 train_loss:3.3482 train_time:472124ms step_avg:153.89ms step:3079/5100 train_loss:3.8948 train_time:472277ms step_avg:153.89ms step:3080/5100 train_loss:3.4450 train_time:472430ms step_avg:153.89ms step:3081/5100 train_loss:3.4054 train_time:472582ms step_avg:153.89ms step:3082/5100 train_loss:3.5466 train_time:472735ms step_avg:153.89ms step:3083/5100 train_loss:3.3595 train_time:472888ms step_avg:153.88ms step:3084/5100 train_loss:3.3896 train_time:473040ms step_avg:153.88ms step:3085/5100 train_loss:3.4366 train_time:473192ms step_avg:153.88ms step:3086/5100 train_loss:3.5288 train_time:473344ms step_avg:153.88ms step:3087/5100 train_loss:3.4464 train_time:473497ms step_avg:153.88ms step:3088/5100 train_loss:3.3548 train_time:473649ms step_avg:153.88ms step:3089/5100 train_loss:3.5068 train_time:473802ms step_avg:153.88ms step:3090/5100 train_loss:3.3707 train_time:473954ms step_avg:153.88ms step:3091/5100 train_loss:3.6269 train_time:474107ms step_avg:153.88ms step:3092/5100 train_loss:4.2019 train_time:474260ms step_avg:153.88ms step:3093/5100 train_loss:3.4698 train_time:474412ms step_avg:153.88ms step:3094/5100 train_loss:3.3609 train_time:474564ms step_avg:153.88ms step:3095/5100 train_loss:3.3190 train_time:474717ms step_avg:153.88ms step:3096/5100 train_loss:3.4848 train_time:474870ms step_avg:153.88ms step:3097/5100 train_loss:3.6097 train_time:475022ms step_avg:153.88ms step:3098/5100 train_loss:3.3865 train_time:475175ms step_avg:153.88ms step:3099/5100 train_loss:3.4260 train_time:475327ms step_avg:153.88ms step:3100/5100 train_loss:3.6069 train_time:475481ms step_avg:153.88ms step:3101/5100 train_loss:3.4976 train_time:475633ms step_avg:153.88ms step:3102/5100 train_loss:3.4979 train_time:475786ms step_avg:153.88ms step:3103/5100 train_loss:3.4024 train_time:475938ms step_avg:153.88ms step:3104/5100 train_loss:3.6563 train_time:476091ms step_avg:153.88ms step:3105/5100 train_loss:3.4779 train_time:476242ms step_avg:153.87ms step:3106/5100 train_loss:3.3296 train_time:476395ms step_avg:153.87ms step:3107/5100 train_loss:3.3628 train_time:476547ms step_avg:153.87ms step:3108/5100 train_loss:3.3206 train_time:476701ms step_avg:153.87ms step:3109/5100 train_loss:3.5425 train_time:476853ms step_avg:153.87ms step:3110/5100 train_loss:3.4301 train_time:477005ms step_avg:153.87ms step:3111/5100 train_loss:3.4684 train_time:477159ms step_avg:153.87ms step:3112/5100 train_loss:3.4481 train_time:477312ms step_avg:153.87ms step:3113/5100 train_loss:3.5017 train_time:477465ms step_avg:153.87ms step:3114/5100 train_loss:3.4418 train_time:477617ms step_avg:153.87ms step:3115/5100 train_loss:3.4634 train_time:477770ms step_avg:153.87ms step:3116/5100 train_loss:3.4900 train_time:477922ms step_avg:153.87ms step:3117/5100 train_loss:3.3513 train_time:478075ms step_avg:153.87ms step:3118/5100 train_loss:3.3609 train_time:478227ms step_avg:153.87ms step:3119/5100 train_loss:3.5531 train_time:478381ms step_avg:153.87ms step:3120/5100 train_loss:3.5253 train_time:478534ms step_avg:153.87ms step:3121/5100 train_loss:3.3179 train_time:478686ms step_avg:153.87ms step:3122/5100 train_loss:3.5104 train_time:478839ms step_avg:153.87ms step:3123/5100 train_loss:3.5605 train_time:478991ms step_avg:153.87ms step:3124/5100 train_loss:3.5336 train_time:479142ms step_avg:153.87ms step:3125/5100 train_loss:3.3223 train_time:479295ms step_avg:153.87ms step:3125/5100 val_loss:3.4360 train_time:479359ms step_avg:153.89ms step:3126/5100 train_loss:3.4084 train_time:479451ms step_avg:153.87ms step:3127/5100 train_loss:3.4504 train_time:479609ms step_avg:153.87ms step:3128/5100 train_loss:3.5288 train_time:479762ms step_avg:153.87ms step:3129/5100 train_loss:3.6042 train_time:479914ms step_avg:153.87ms step:3130/5100 train_loss:3.3098 train_time:480066ms step_avg:153.87ms step:3131/5100 train_loss:3.4748 train_time:480219ms step_avg:153.87ms step:3132/5100 train_loss:3.4721 train_time:480371ms step_avg:153.87ms step:3133/5100 train_loss:3.4938 train_time:480524ms step_avg:153.87ms step:3134/5100 train_loss:3.3916 train_time:480677ms step_avg:153.87ms step:3135/5100 train_loss:3.5135 train_time:480829ms step_avg:153.87ms step:3136/5100 train_loss:3.4257 train_time:480981ms step_avg:153.86ms step:3137/5100 train_loss:3.4900 train_time:481133ms step_avg:153.86ms step:3138/5100 train_loss:3.6809 train_time:481286ms step_avg:153.86ms step:3139/5100 train_loss:3.6483 train_time:481438ms step_avg:153.86ms step:3140/5100 train_loss:3.4157 train_time:481590ms step_avg:153.86ms step:3141/5100 train_loss:3.4309 train_time:481743ms step_avg:153.86ms step:3142/5100 train_loss:3.3567 train_time:481895ms step_avg:153.86ms step:3143/5100 train_loss:3.4497 train_time:482048ms step_avg:153.86ms step:3144/5100 train_loss:3.2468 train_time:482200ms step_avg:153.86ms step:3145/5100 train_loss:3.4847 train_time:482352ms step_avg:153.86ms step:3146/5100 train_loss:3.3949 train_time:482505ms step_avg:153.86ms step:3147/5100 train_loss:3.4180 train_time:482657ms step_avg:153.86ms step:3148/5100 train_loss:3.5935 train_time:482810ms step_avg:153.86ms step:3149/5100 train_loss:3.6910 train_time:482964ms step_avg:153.86ms step:3150/5100 train_loss:3.5446 train_time:483115ms step_avg:153.86ms step:3151/5100 train_loss:3.3641 train_time:483268ms step_avg:153.86ms step:3152/5100 train_loss:3.4041 train_time:483420ms step_avg:153.86ms step:3153/5100 train_loss:3.3857 train_time:483573ms step_avg:153.86ms step:3154/5100 train_loss:3.5018 train_time:483726ms step_avg:153.86ms step:3155/5100 train_loss:3.3235 train_time:483878ms step_avg:153.86ms step:3156/5100 train_loss:3.4610 train_time:484031ms step_avg:153.86ms step:3157/5100 train_loss:3.4073 train_time:484184ms step_avg:153.86ms step:3158/5100 train_loss:3.5238 train_time:484337ms step_avg:153.86ms step:3159/5100 train_loss:3.5919 train_time:484489ms step_avg:153.85ms step:3160/5100 train_loss:3.4296 train_time:484641ms step_avg:153.85ms step:3161/5100 train_loss:3.4975 train_time:484793ms step_avg:153.85ms step:3162/5100 train_loss:3.5715 train_time:484946ms step_avg:153.85ms step:3163/5100 train_loss:3.4750 train_time:485099ms step_avg:153.85ms step:3164/5100 train_loss:3.5318 train_time:485252ms step_avg:153.85ms step:3165/5100 train_loss:3.3486 train_time:485405ms step_avg:153.85ms step:3166/5100 train_loss:3.3388 train_time:485557ms step_avg:153.85ms step:3167/5100 train_loss:3.3746 train_time:485709ms step_avg:153.85ms step:3168/5100 train_loss:3.2022 train_time:485862ms step_avg:153.85ms step:3169/5100 train_loss:3.3705 train_time:486014ms step_avg:153.85ms step:3170/5100 train_loss:3.5132 train_time:486167ms step_avg:153.85ms step:3171/5100 train_loss:3.5403 train_time:486319ms step_avg:153.85ms step:3172/5100 train_loss:3.5041 train_time:486471ms step_avg:153.85ms step:3173/5100 train_loss:3.4738 train_time:486624ms step_avg:153.85ms step:3174/5100 train_loss:3.4426 train_time:486776ms step_avg:153.85ms step:3175/5100 train_loss:3.4465 train_time:486927ms step_avg:153.85ms step:3176/5100 train_loss:3.4475 train_time:487080ms step_avg:153.85ms step:3177/5100 train_loss:3.3743 train_time:487232ms step_avg:153.85ms step:3178/5100 train_loss:3.5037 train_time:487384ms step_avg:153.85ms step:3179/5100 train_loss:3.5757 train_time:487537ms step_avg:153.85ms step:3180/5100 train_loss:3.4219 train_time:487690ms step_avg:153.85ms step:3181/5100 train_loss:3.4039 train_time:487842ms step_avg:153.84ms step:3182/5100 train_loss:3.4529 train_time:487994ms step_avg:153.84ms step:3183/5100 train_loss:3.5623 train_time:488146ms step_avg:153.84ms step:3184/5100 train_loss:3.5709 train_time:488299ms step_avg:153.84ms step:3185/5100 train_loss:3.4684 train_time:488451ms step_avg:153.84ms step:3186/5100 train_loss:3.5425 train_time:488604ms step_avg:153.84ms step:3187/5100 train_loss:3.5211 train_time:488757ms step_avg:153.84ms step:3188/5100 train_loss:3.3175 train_time:488909ms step_avg:153.84ms step:3189/5100 train_loss:3.4905 train_time:489063ms step_avg:153.84ms step:3190/5100 train_loss:3.4270 train_time:489215ms step_avg:153.84ms step:3191/5100 train_loss:3.4522 train_time:489367ms step_avg:153.84ms step:3192/5100 train_loss:3.4098 train_time:489519ms step_avg:153.84ms step:3193/5100 train_loss:3.3444 train_time:489671ms step_avg:153.84ms step:3194/5100 train_loss:4.3712 train_time:489825ms step_avg:153.84ms step:3195/5100 train_loss:3.4600 train_time:489977ms step_avg:153.84ms step:3196/5100 train_loss:3.2491 train_time:490129ms step_avg:153.84ms step:3197/5100 train_loss:3.4154 train_time:490281ms step_avg:153.84ms step:3198/5100 train_loss:3.2951 train_time:490434ms step_avg:153.84ms step:3199/5100 train_loss:3.4039 train_time:490586ms step_avg:153.84ms step:3200/5100 train_loss:3.3318 train_time:490738ms step_avg:153.84ms step:3201/5100 train_loss:3.4126 train_time:490890ms step_avg:153.84ms step:3202/5100 train_loss:3.5093 train_time:491043ms step_avg:153.84ms step:3203/5100 train_loss:3.3609 train_time:491195ms step_avg:153.83ms step:3204/5100 train_loss:3.4032 train_time:491347ms step_avg:153.83ms step:3205/5100 train_loss:3.4952 train_time:491500ms step_avg:153.83ms step:3206/5100 train_loss:3.6473 train_time:491652ms step_avg:153.83ms step:3207/5100 train_loss:3.2452 train_time:491805ms step_avg:153.83ms step:3208/5100 train_loss:3.6020 train_time:491957ms step_avg:153.83ms step:3209/5100 train_loss:3.4504 train_time:492111ms step_avg:153.83ms step:3210/5100 train_loss:3.5192 train_time:492264ms step_avg:153.83ms step:3211/5100 train_loss:3.6088 train_time:492415ms step_avg:153.83ms step:3212/5100 train_loss:3.2888 train_time:492567ms step_avg:153.83ms step:3213/5100 train_loss:3.3461 train_time:492823ms step_avg:153.86ms step:3214/5100 train_loss:3.5531 train_time:492981ms step_avg:153.86ms step:3215/5100 train_loss:3.3452 train_time:493132ms step_avg:153.86ms step:3216/5100 train_loss:3.4154 train_time:493284ms step_avg:153.86ms step:3217/5100 train_loss:3.3147 train_time:493436ms step_avg:153.86ms step:3218/5100 train_loss:3.4456 train_time:493587ms step_avg:153.86ms step:3219/5100 train_loss:3.4902 train_time:493740ms step_avg:153.86ms step:3220/5100 train_loss:3.5395 train_time:493892ms step_avg:153.86ms step:3221/5100 train_loss:3.4776 train_time:494045ms step_avg:153.86ms step:3222/5100 train_loss:3.4813 train_time:494198ms step_avg:153.86ms step:3223/5100 train_loss:3.3489 train_time:494350ms step_avg:153.86ms step:3224/5100 train_loss:3.3707 train_time:494503ms step_avg:153.86ms step:3225/5100 train_loss:3.3651 train_time:494655ms step_avg:153.86ms step:3226/5100 train_loss:3.4070 train_time:494809ms step_avg:153.86ms step:3227/5100 train_loss:3.3512 train_time:494961ms step_avg:153.86ms step:3228/5100 train_loss:3.2616 train_time:495114ms step_avg:153.86ms step:3229/5100 train_loss:3.3862 train_time:495266ms step_avg:153.86ms step:3230/5100 train_loss:3.1537 train_time:495540ms step_avg:153.89ms step:3231/5100 train_loss:3.3366 train_time:495689ms step_avg:153.89ms step:3232/5100 train_loss:3.3287 train_time:495843ms step_avg:153.89ms step:3233/5100 train_loss:3.5722 train_time:495994ms step_avg:153.89ms step:3234/5100 train_loss:3.5549 train_time:496147ms step_avg:153.89ms step:3235/5100 train_loss:3.5200 train_time:496300ms step_avg:153.89ms step:3236/5100 train_loss:3.4144 train_time:496451ms step_avg:153.89ms step:3237/5100 train_loss:3.5667 train_time:496607ms step_avg:153.89ms step:3238/5100 train_loss:3.4310 train_time:496761ms step_avg:153.89ms step:3239/5100 train_loss:3.5543 train_time:496913ms step_avg:153.89ms step:3240/5100 train_loss:3.5247 train_time:497067ms step_avg:153.89ms step:3241/5100 train_loss:3.4244 train_time:497220ms step_avg:153.89ms step:3242/5100 train_loss:3.3717 train_time:497372ms step_avg:153.89ms step:3243/5100 train_loss:3.6031 train_time:497525ms step_avg:153.89ms step:3244/5100 train_loss:3.4750 train_time:497678ms step_avg:153.89ms step:3245/5100 train_loss:3.5168 train_time:497830ms step_avg:153.89ms step:3246/5100 train_loss:3.4064 train_time:497983ms step_avg:153.89ms step:3247/5100 train_loss:3.5396 train_time:498136ms step_avg:153.89ms step:3248/5100 train_loss:3.4641 train_time:498288ms step_avg:153.89ms step:3249/5100 train_loss:3.4089 train_time:498441ms step_avg:153.89ms step:3250/5100 train_loss:3.2842 train_time:498593ms step_avg:153.89ms step:3250/5100 val_loss:3.4276 train_time:498658ms step_avg:153.91ms step:3251/5100 train_loss:3.4828 train_time:498751ms step_avg:153.89ms step:3252/5100 train_loss:3.4835 train_time:498908ms step_avg:153.89ms step:3253/5100 train_loss:3.4427 train_time:499061ms step_avg:153.89ms step:3254/5100 train_loss:3.3650 train_time:499213ms step_avg:153.89ms step:3255/5100 train_loss:3.5128 train_time:499364ms step_avg:153.89ms step:3256/5100 train_loss:3.5390 train_time:499516ms step_avg:153.89ms step:3257/5100 train_loss:3.4829 train_time:499669ms step_avg:153.89ms step:3258/5100 train_loss:3.5109 train_time:499821ms step_avg:153.89ms step:3259/5100 train_loss:3.3560 train_time:499975ms step_avg:153.89ms step:3260/5100 train_loss:3.4345 train_time:500130ms step_avg:153.89ms step:3261/5100 train_loss:3.3067 train_time:500282ms step_avg:153.89ms step:3262/5100 train_loss:3.3418 train_time:500433ms step_avg:153.88ms step:3263/5100 train_loss:3.3681 train_time:500586ms step_avg:153.88ms step:3264/5100 train_loss:3.5315 train_time:500738ms step_avg:153.88ms step:3265/5100 train_loss:3.4093 train_time:500890ms step_avg:153.88ms step:3266/5100 train_loss:3.4612 train_time:501042ms step_avg:153.88ms step:3267/5100 train_loss:3.4772 train_time:501195ms step_avg:153.88ms step:3268/5100 train_loss:3.5650 train_time:501346ms step_avg:153.88ms step:3269/5100 train_loss:3.3777 train_time:501499ms step_avg:153.88ms step:3270/5100 train_loss:3.5039 train_time:501651ms step_avg:153.88ms step:3271/5100 train_loss:3.3699 train_time:501804ms step_avg:153.88ms step:3272/5100 train_loss:3.2710 train_time:501958ms step_avg:153.88ms step:3273/5100 train_loss:3.3816 train_time:502110ms step_avg:153.88ms step:3274/5100 train_loss:3.5254 train_time:502262ms step_avg:153.88ms step:3275/5100 train_loss:3.3168 train_time:502415ms step_avg:153.88ms step:3276/5100 train_loss:3.4685 train_time:502567ms step_avg:153.88ms step:3277/5100 train_loss:3.4622 train_time:502719ms step_avg:153.88ms step:3278/5100 train_loss:3.4550 train_time:502872ms step_avg:153.88ms step:3279/5100 train_loss:3.4241 train_time:503025ms step_avg:153.88ms step:3280/5100 train_loss:3.5805 train_time:503177ms step_avg:153.88ms step:3281/5100 train_loss:3.4286 train_time:503330ms step_avg:153.88ms step:3282/5100 train_loss:3.4710 train_time:503482ms step_avg:153.88ms step:3283/5100 train_loss:3.3355 train_time:503634ms step_avg:153.88ms step:3284/5100 train_loss:3.4649 train_time:503786ms step_avg:153.87ms step:3285/5100 train_loss:3.5088 train_time:503938ms step_avg:153.87ms step:3286/5100 train_loss:3.4994 train_time:504091ms step_avg:153.87ms step:3287/5100 train_loss:3.5339 train_time:504243ms step_avg:153.87ms step:3288/5100 train_loss:3.4065 train_time:504395ms step_avg:153.87ms step:3289/5100 train_loss:3.5200 train_time:504547ms step_avg:153.87ms step:3290/5100 train_loss:3.4418 train_time:504701ms step_avg:153.87ms step:3291/5100 train_loss:3.3195 train_time:504852ms step_avg:153.87ms step:3292/5100 train_loss:3.4482 train_time:505005ms step_avg:153.87ms step:3293/5100 train_loss:3.4830 train_time:505158ms step_avg:153.87ms step:3294/5100 train_loss:3.4688 train_time:505310ms step_avg:153.87ms step:3295/5100 train_loss:3.3498 train_time:505463ms step_avg:153.87ms step:3296/5100 train_loss:3.4100 train_time:505616ms step_avg:153.87ms step:3297/5100 train_loss:3.4591 train_time:505768ms step_avg:153.87ms step:3298/5100 train_loss:3.4496 train_time:505920ms step_avg:153.87ms step:3299/5100 train_loss:3.4328 train_time:506072ms step_avg:153.87ms step:3300/5100 train_loss:3.4957 train_time:506225ms step_avg:153.87ms step:3301/5100 train_loss:3.3930 train_time:506378ms step_avg:153.87ms step:3302/5100 train_loss:3.4614 train_time:506531ms step_avg:153.87ms step:3303/5100 train_loss:3.4154 train_time:506683ms step_avg:153.87ms step:3304/5100 train_loss:3.4190 train_time:506835ms step_avg:153.87ms step:3305/5100 train_loss:3.4223 train_time:506987ms step_avg:153.87ms step:3306/5100 train_loss:3.5182 train_time:507139ms step_avg:153.87ms step:3307/5100 train_loss:3.4429 train_time:507292ms step_avg:153.86ms step:3308/5100 train_loss:3.4119 train_time:507445ms step_avg:153.86ms step:3309/5100 train_loss:3.5324 train_time:507598ms step_avg:153.86ms step:3310/5100 train_loss:3.4066 train_time:507750ms step_avg:153.86ms step:3311/5100 train_loss:3.3503 train_time:507903ms step_avg:153.86ms step:3312/5100 train_loss:3.4625 train_time:508056ms step_avg:153.86ms step:3313/5100 train_loss:3.4267 train_time:508208ms step_avg:153.86ms step:3314/5100 train_loss:3.6345 train_time:508360ms step_avg:153.86ms step:3315/5100 train_loss:3.4569 train_time:508512ms step_avg:153.86ms step:3316/5100 train_loss:3.4246 train_time:508665ms step_avg:153.86ms step:3317/5100 train_loss:3.0595 train_time:508818ms step_avg:153.86ms step:3318/5100 train_loss:3.5639 train_time:508970ms step_avg:153.86ms step:3319/5100 train_loss:3.3998 train_time:509122ms step_avg:153.86ms step:3320/5100 train_loss:3.4790 train_time:509275ms step_avg:153.86ms step:3321/5100 train_loss:3.4103 train_time:509428ms step_avg:153.86ms step:3322/5100 train_loss:3.4770 train_time:509579ms step_avg:153.86ms step:3323/5100 train_loss:3.4105 train_time:509733ms step_avg:153.86ms step:3324/5100 train_loss:3.3357 train_time:509885ms step_avg:153.86ms step:3325/5100 train_loss:3.2765 train_time:510038ms step_avg:153.86ms step:3326/5100 train_loss:3.4376 train_time:510191ms step_avg:153.86ms step:3327/5100 train_loss:3.3994 train_time:510343ms step_avg:153.86ms step:3328/5100 train_loss:3.3293 train_time:510496ms step_avg:153.86ms step:3329/5100 train_loss:3.3596 train_time:510647ms step_avg:153.86ms step:3330/5100 train_loss:3.3155 train_time:510800ms step_avg:153.86ms step:3331/5100 train_loss:3.5689 train_time:510952ms step_avg:153.85ms step:3332/5100 train_loss:3.4703 train_time:511104ms step_avg:153.85ms step:3333/5100 train_loss:3.4474 train_time:511258ms step_avg:153.85ms step:3334/5100 train_loss:3.3073 train_time:511409ms step_avg:153.85ms step:3335/5100 train_loss:3.3771 train_time:511561ms step_avg:153.85ms step:3336/5100 train_loss:3.4794 train_time:511713ms step_avg:153.85ms step:3337/5100 train_loss:3.4507 train_time:511866ms step_avg:153.85ms step:3338/5100 train_loss:3.4853 train_time:512018ms step_avg:153.85ms step:3339/5100 train_loss:3.4166 train_time:512171ms step_avg:153.85ms step:3340/5100 train_loss:3.4349 train_time:512323ms step_avg:153.85ms step:3341/5100 train_loss:3.4492 train_time:512475ms step_avg:153.85ms step:3342/5100 train_loss:3.4593 train_time:512628ms step_avg:153.85ms step:3343/5100 train_loss:3.4537 train_time:512780ms step_avg:153.85ms step:3344/5100 train_loss:3.3822 train_time:512934ms step_avg:153.85ms step:3345/5100 train_loss:3.3000 train_time:513086ms step_avg:153.85ms step:3346/5100 train_loss:3.6274 train_time:513238ms step_avg:153.85ms step:3347/5100 train_loss:3.3817 train_time:513391ms step_avg:153.85ms step:3348/5100 train_loss:3.5458 train_time:513543ms step_avg:153.85ms step:3349/5100 train_loss:3.4165 train_time:513696ms step_avg:153.85ms step:3350/5100 train_loss:3.4931 train_time:513848ms step_avg:153.85ms step:3351/5100 train_loss:3.2322 train_time:514000ms step_avg:153.85ms step:3352/5100 train_loss:3.2593 train_time:514153ms step_avg:153.85ms step:3353/5100 train_loss:3.4461 train_time:514305ms step_avg:153.85ms step:3354/5100 train_loss:3.3111 train_time:514458ms step_avg:153.85ms step:3355/5100 train_loss:3.4686 train_time:514610ms step_avg:153.84ms step:3356/5100 train_loss:3.3282 train_time:514763ms step_avg:153.84ms step:3357/5100 train_loss:3.4987 train_time:514916ms step_avg:153.84ms step:3358/5100 train_loss:3.3559 train_time:515068ms step_avg:153.84ms step:3359/5100 train_loss:3.5249 train_time:515220ms step_avg:153.84ms step:3360/5100 train_loss:3.3364 train_time:515374ms step_avg:153.84ms step:3361/5100 train_loss:4.0477 train_time:515525ms step_avg:153.84ms step:3362/5100 train_loss:3.4920 train_time:515677ms step_avg:153.84ms step:3363/5100 train_loss:3.5261 train_time:515829ms step_avg:153.84ms step:3364/5100 train_loss:3.3988 train_time:515981ms step_avg:153.84ms step:3365/5100 train_loss:3.5218 train_time:516135ms step_avg:153.84ms step:3366/5100 train_loss:3.4205 train_time:516287ms step_avg:153.84ms step:3367/5100 train_loss:3.5952 train_time:516439ms step_avg:153.84ms step:3368/5100 train_loss:3.4002 train_time:516592ms step_avg:153.84ms step:3369/5100 train_loss:3.4164 train_time:516744ms step_avg:153.84ms step:3370/5100 train_loss:3.3924 train_time:516897ms step_avg:153.84ms step:3371/5100 train_loss:3.3458 train_time:517048ms step_avg:153.84ms step:3372/5100 train_loss:3.3495 train_time:517201ms step_avg:153.84ms step:3373/5100 train_loss:3.4122 train_time:517354ms step_avg:153.84ms step:3374/5100 train_loss:3.4443 train_time:517507ms step_avg:153.84ms step:3375/5100 train_loss:3.4104 train_time:517659ms step_avg:153.84ms step:3375/5100 val_loss:3.4232 train_time:517723ms step_avg:153.86ms step:3376/5100 train_loss:3.4559 train_time:517818ms step_avg:153.84ms step:3377/5100 train_loss:3.4561 train_time:517974ms step_avg:153.84ms step:3378/5100 train_loss:3.5357 train_time:518126ms step_avg:153.84ms step:3379/5100 train_loss:3.3919 train_time:518278ms step_avg:153.84ms step:3380/5100 train_loss:3.4009 train_time:518430ms step_avg:153.84ms step:3381/5100 train_loss:3.4132 train_time:518581ms step_avg:153.84ms step:3382/5100 train_loss:3.5139 train_time:518734ms step_avg:153.84ms step:3383/5100 train_loss:3.3507 train_time:518887ms step_avg:153.84ms step:3384/5100 train_loss:3.5244 train_time:519040ms step_avg:153.84ms step:3385/5100 train_loss:3.3842 train_time:519193ms step_avg:153.83ms step:3386/5100 train_loss:3.4009 train_time:519346ms step_avg:153.83ms step:3387/5100 train_loss:3.3461 train_time:519499ms step_avg:153.83ms step:3388/5100 train_loss:3.5221 train_time:519651ms step_avg:153.83ms step:3389/5100 train_loss:3.4849 train_time:519804ms step_avg:153.83ms step:3390/5100 train_loss:3.5027 train_time:519957ms step_avg:153.83ms step:3391/5100 train_loss:3.4810 train_time:520109ms step_avg:153.83ms step:3392/5100 train_loss:3.4185 train_time:520261ms step_avg:153.83ms step:3393/5100 train_loss:3.5454 train_time:520414ms step_avg:153.83ms step:3394/5100 train_loss:3.5015 train_time:520567ms step_avg:153.83ms step:3395/5100 train_loss:3.5934 train_time:520719ms step_avg:153.83ms step:3396/5100 train_loss:3.4446 train_time:520871ms step_avg:153.83ms step:3397/5100 train_loss:3.4343 train_time:521024ms step_avg:153.83ms step:3398/5100 train_loss:3.3880 train_time:521175ms step_avg:153.83ms step:3399/5100 train_loss:3.4480 train_time:521329ms step_avg:153.83ms step:3400/5100 train_loss:3.4428 train_time:521482ms step_avg:153.83ms step:3401/5100 train_loss:3.5309 train_time:521634ms step_avg:153.83ms step:3402/5100 train_loss:3.4013 train_time:521888ms step_avg:153.86ms step:3403/5100 train_loss:3.5862 train_time:522048ms step_avg:153.86ms step:3404/5100 train_loss:3.4113 train_time:522201ms step_avg:153.86ms step:3405/5100 train_loss:3.4224 train_time:522352ms step_avg:153.86ms step:3406/5100 train_loss:3.3727 train_time:522504ms step_avg:153.86ms step:3407/5100 train_loss:3.4317 train_time:522656ms step_avg:153.86ms step:3408/5100 train_loss:3.4350 train_time:522808ms step_avg:153.86ms step:3409/5100 train_loss:3.4218 train_time:522963ms step_avg:153.86ms step:3410/5100 train_loss:3.4359 train_time:523116ms step_avg:153.86ms step:3411/5100 train_loss:3.3987 train_time:523268ms step_avg:153.86ms step:3412/5100 train_loss:3.4319 train_time:523420ms step_avg:153.86ms step:3413/5100 train_loss:3.3580 train_time:523573ms step_avg:153.86ms step:3414/5100 train_loss:3.5735 train_time:523726ms step_avg:153.86ms step:3415/5100 train_loss:3.3242 train_time:523877ms step_avg:153.86ms step:3416/5100 train_loss:3.4822 train_time:524030ms step_avg:153.85ms step:3417/5100 train_loss:3.3656 train_time:524182ms step_avg:153.85ms step:3418/5100 train_loss:3.4661 train_time:524334ms step_avg:153.85ms step:3419/5100 train_loss:3.4667 train_time:524488ms step_avg:153.85ms step:3420/5100 train_loss:3.4996 train_time:524762ms step_avg:153.89ms step:3421/5100 train_loss:3.3724 train_time:524913ms step_avg:153.89ms step:3422/5100 train_loss:3.4083 train_time:525065ms step_avg:153.89ms step:3423/5100 train_loss:3.3427 train_time:525217ms step_avg:153.89ms step:3424/5100 train_loss:3.6627 train_time:525369ms step_avg:153.89ms step:3425/5100 train_loss:3.5504 train_time:525522ms step_avg:153.89ms step:3426/5100 train_loss:3.4249 train_time:525674ms step_avg:153.89ms step:3427/5100 train_loss:3.3709 train_time:525827ms step_avg:153.89ms step:3428/5100 train_loss:3.3545 train_time:525980ms step_avg:153.89ms step:3429/5100 train_loss:3.3515 train_time:526131ms step_avg:153.88ms step:3430/5100 train_loss:3.4172 train_time:526284ms step_avg:153.88ms step:3431/5100 train_loss:3.4492 train_time:526436ms step_avg:153.88ms step:3432/5100 train_loss:3.5433 train_time:526590ms step_avg:153.88ms step:3433/5100 train_loss:3.3575 train_time:526742ms step_avg:153.88ms step:3434/5100 train_loss:3.5752 train_time:526895ms step_avg:153.88ms step:3435/5100 train_loss:3.4919 train_time:527048ms step_avg:153.88ms step:3436/5100 train_loss:3.3431 train_time:527200ms step_avg:153.88ms step:3437/5100 train_loss:3.3843 train_time:527353ms step_avg:153.88ms step:3438/5100 train_loss:3.4300 train_time:527505ms step_avg:153.88ms step:3439/5100 train_loss:3.5261 train_time:527658ms step_avg:153.88ms step:3440/5100 train_loss:3.2986 train_time:527812ms step_avg:153.88ms step:3441/5100 train_loss:3.4805 train_time:527963ms step_avg:153.88ms step:3442/5100 train_loss:3.3814 train_time:528116ms step_avg:153.88ms step:3443/5100 train_loss:3.5636 train_time:528270ms step_avg:153.88ms step:3444/5100 train_loss:3.4304 train_time:528423ms step_avg:153.88ms step:3445/5100 train_loss:3.3125 train_time:528575ms step_avg:153.88ms step:3446/5100 train_loss:3.5269 train_time:528728ms step_avg:153.88ms step:3447/5100 train_loss:3.6010 train_time:528879ms step_avg:153.88ms step:3448/5100 train_loss:3.4128 train_time:529031ms step_avg:153.88ms step:3449/5100 train_loss:3.4308 train_time:529184ms step_avg:153.88ms step:3450/5100 train_loss:3.5210 train_time:529336ms step_avg:153.88ms step:3451/5100 train_loss:3.5168 train_time:529490ms step_avg:153.88ms step:3452/5100 train_loss:3.5218 train_time:529641ms step_avg:153.88ms step:3453/5100 train_loss:3.3298 train_time:529795ms step_avg:153.88ms step:3454/5100 train_loss:3.4414 train_time:529948ms step_avg:153.88ms step:3455/5100 train_loss:3.3431 train_time:530100ms step_avg:153.88ms step:3456/5100 train_loss:3.6165 train_time:530252ms step_avg:153.87ms step:3457/5100 train_loss:3.2935 train_time:530405ms step_avg:153.87ms step:3458/5100 train_loss:3.4355 train_time:530557ms step_avg:153.87ms step:3459/5100 train_loss:3.3814 train_time:530709ms step_avg:153.87ms step:3460/5100 train_loss:3.3861 train_time:530862ms step_avg:153.87ms step:3461/5100 train_loss:3.3788 train_time:531016ms step_avg:153.87ms step:3462/5100 train_loss:3.3896 train_time:531168ms step_avg:153.87ms step:3463/5100 train_loss:3.4916 train_time:531321ms step_avg:153.87ms step:3464/5100 train_loss:3.3629 train_time:531473ms step_avg:153.87ms step:3465/5100 train_loss:3.3807 train_time:531626ms step_avg:153.87ms step:3466/5100 train_loss:3.3543 train_time:531777ms step_avg:153.87ms step:3467/5100 train_loss:3.5163 train_time:531930ms step_avg:153.87ms step:3468/5100 train_loss:3.4016 train_time:532082ms step_avg:153.87ms step:3469/5100 train_loss:3.4177 train_time:532234ms step_avg:153.87ms step:3470/5100 train_loss:3.6040 train_time:532388ms step_avg:153.87ms step:3471/5100 train_loss:3.5005 train_time:532541ms step_avg:153.87ms step:3472/5100 train_loss:3.5439 train_time:532693ms step_avg:153.87ms step:3473/5100 train_loss:4.1817 train_time:532845ms step_avg:153.87ms step:3474/5100 train_loss:3.4189 train_time:532998ms step_avg:153.87ms step:3475/5100 train_loss:3.4298 train_time:533150ms step_avg:153.87ms step:3476/5100 train_loss:3.4103 train_time:533303ms step_avg:153.87ms step:3477/5100 train_loss:3.3530 train_time:533455ms step_avg:153.87ms step:3478/5100 train_loss:3.4316 train_time:533607ms step_avg:153.87ms step:3479/5100 train_loss:3.4218 train_time:533760ms step_avg:153.87ms step:3480/5100 train_loss:3.3229 train_time:533911ms step_avg:153.86ms step:3481/5100 train_loss:3.6273 train_time:534064ms step_avg:153.86ms step:3482/5100 train_loss:3.4908 train_time:534217ms step_avg:153.86ms step:3483/5100 train_loss:3.4418 train_time:534370ms step_avg:153.86ms step:3484/5100 train_loss:3.4633 train_time:534521ms step_avg:153.86ms step:3485/5100 train_loss:3.4283 train_time:534673ms step_avg:153.86ms step:3486/5100 train_loss:3.6150 train_time:534827ms step_avg:153.86ms step:3487/5100 train_loss:3.6391 train_time:534979ms step_avg:153.86ms step:3488/5100 train_loss:3.5002 train_time:535130ms step_avg:153.86ms step:3489/5100 train_loss:3.3501 train_time:535283ms step_avg:153.86ms step:3490/5100 train_loss:3.5269 train_time:535436ms step_avg:153.86ms step:3491/5100 train_loss:3.4317 train_time:535588ms step_avg:153.86ms step:3492/5100 train_loss:3.4660 train_time:535741ms step_avg:153.86ms step:3493/5100 train_loss:3.3047 train_time:535894ms step_avg:153.86ms step:3494/5100 train_loss:3.4483 train_time:536047ms step_avg:153.86ms step:3495/5100 train_loss:3.3997 train_time:536199ms step_avg:153.86ms step:3496/5100 train_loss:3.4193 train_time:536351ms step_avg:153.86ms step:3497/5100 train_loss:3.5886 train_time:536504ms step_avg:153.86ms step:3498/5100 train_loss:3.4263 train_time:536656ms step_avg:153.86ms step:3499/5100 train_loss:3.4498 train_time:536809ms step_avg:153.86ms step:3500/5100 train_loss:3.4442 train_time:536962ms step_avg:153.86ms step:3500/5100 val_loss:3.4145 train_time:537025ms step_avg:153.88ms step:3501/5100 train_loss:3.4877 train_time:537120ms step_avg:153.86ms step:3502/5100 train_loss:3.5587 train_time:537274ms step_avg:153.86ms step:3503/5100 train_loss:3.2711 train_time:537426ms step_avg:153.86ms step:3504/5100 train_loss:3.4301 train_time:537578ms step_avg:153.86ms step:3505/5100 train_loss:3.4513 train_time:537730ms step_avg:153.86ms step:3506/5100 train_loss:3.4780 train_time:537883ms step_avg:153.86ms step:3507/5100 train_loss:3.3538 train_time:538034ms step_avg:153.86ms step:3508/5100 train_loss:3.5053 train_time:538188ms step_avg:153.86ms step:3509/5100 train_loss:3.3924 train_time:538340ms step_avg:153.86ms step:3510/5100 train_loss:3.6091 train_time:538493ms step_avg:153.86ms step:3511/5100 train_loss:3.4117 train_time:538645ms step_avg:153.85ms step:3512/5100 train_loss:3.3705 train_time:538798ms step_avg:153.85ms step:3513/5100 train_loss:3.4289 train_time:538952ms step_avg:153.85ms step:3514/5100 train_loss:3.3929 train_time:539103ms step_avg:153.85ms step:3515/5100 train_loss:3.4489 train_time:539255ms step_avg:153.85ms step:3516/5100 train_loss:3.4413 train_time:539407ms step_avg:153.85ms step:3517/5100 train_loss:3.4080 train_time:539560ms step_avg:153.85ms step:3518/5100 train_loss:3.4391 train_time:539712ms step_avg:153.85ms step:3519/5100 train_loss:3.4300 train_time:539866ms step_avg:153.85ms step:3520/5100 train_loss:3.4466 train_time:540019ms step_avg:153.85ms step:3521/5100 train_loss:3.5191 train_time:540172ms step_avg:153.85ms step:3522/5100 train_loss:3.4317 train_time:540324ms step_avg:153.85ms step:3523/5100 train_loss:3.3586 train_time:540476ms step_avg:153.85ms step:3524/5100 train_loss:3.3997 train_time:540629ms step_avg:153.85ms step:3525/5100 train_loss:3.3940 train_time:540781ms step_avg:153.85ms step:3526/5100 train_loss:3.3967 train_time:540932ms step_avg:153.85ms step:3527/5100 train_loss:3.5266 train_time:541086ms step_avg:153.85ms step:3528/5100 train_loss:3.3594 train_time:541238ms step_avg:153.85ms step:3529/5100 train_loss:3.2440 train_time:541392ms step_avg:153.85ms step:3530/5100 train_loss:3.5316 train_time:541543ms step_avg:153.85ms step:3531/5100 train_loss:3.3367 train_time:541696ms step_avg:153.85ms step:3532/5100 train_loss:3.3760 train_time:541849ms step_avg:153.85ms step:3533/5100 train_loss:3.2927 train_time:542000ms step_avg:153.85ms step:3534/5100 train_loss:3.3553 train_time:542152ms step_avg:153.85ms step:3535/5100 train_loss:3.3236 train_time:542305ms step_avg:153.85ms step:3536/5100 train_loss:3.5048 train_time:542457ms step_avg:153.84ms step:3537/5100 train_loss:3.4533 train_time:542610ms step_avg:153.84ms step:3538/5100 train_loss:3.4833 train_time:542762ms step_avg:153.84ms step:3539/5100 train_loss:3.3875 train_time:542913ms step_avg:153.84ms step:3540/5100 train_loss:3.3573 train_time:543066ms step_avg:153.84ms step:3541/5100 train_loss:3.4923 train_time:543219ms step_avg:153.84ms step:3542/5100 train_loss:3.3403 train_time:543372ms step_avg:153.84ms step:3543/5100 train_loss:3.5665 train_time:543526ms step_avg:153.84ms step:3544/5100 train_loss:3.6772 train_time:543677ms step_avg:153.84ms step:3545/5100 train_loss:3.4929 train_time:543831ms step_avg:153.84ms step:3546/5100 train_loss:3.5250 train_time:543984ms step_avg:153.84ms step:3547/5100 train_loss:3.2905 train_time:544136ms step_avg:153.84ms step:3548/5100 train_loss:3.3721 train_time:544290ms step_avg:153.84ms step:3549/5100 train_loss:3.4024 train_time:544442ms step_avg:153.84ms step:3550/5100 train_loss:3.4866 train_time:544594ms step_avg:153.84ms step:3551/5100 train_loss:3.4526 train_time:544747ms step_avg:153.84ms step:3552/5100 train_loss:3.3809 train_time:544899ms step_avg:153.84ms step:3553/5100 train_loss:3.4932 train_time:545052ms step_avg:153.84ms step:3554/5100 train_loss:3.4244 train_time:545204ms step_avg:153.84ms step:3555/5100 train_loss:3.3834 train_time:545357ms step_avg:153.84ms step:3556/5100 train_loss:3.3249 train_time:545509ms step_avg:153.84ms step:3557/5100 train_loss:3.2850 train_time:545661ms step_avg:153.84ms step:3558/5100 train_loss:3.3970 train_time:545813ms step_avg:153.84ms step:3559/5100 train_loss:3.4180 train_time:545965ms step_avg:153.84ms step:3560/5100 train_loss:3.6252 train_time:546118ms step_avg:153.84ms step:3561/5100 train_loss:3.4914 train_time:546270ms step_avg:153.84ms step:3562/5100 train_loss:3.4006 train_time:546423ms step_avg:153.84ms step:3563/5100 train_loss:3.2782 train_time:546575ms step_avg:153.83ms step:3564/5100 train_loss:3.7888 train_time:546730ms step_avg:153.84ms step:3565/5100 train_loss:3.3603 train_time:546881ms step_avg:153.83ms step:3566/5100 train_loss:3.3015 train_time:547033ms step_avg:153.83ms step:3567/5100 train_loss:3.3275 train_time:547185ms step_avg:153.83ms step:3568/5100 train_loss:3.4690 train_time:547337ms step_avg:153.83ms step:3569/5100 train_loss:3.3948 train_time:547491ms step_avg:153.83ms step:3570/5100 train_loss:3.5162 train_time:547644ms step_avg:153.83ms step:3571/5100 train_loss:3.4364 train_time:547796ms step_avg:153.83ms step:3572/5100 train_loss:3.7163 train_time:547950ms step_avg:153.83ms step:3573/5100 train_loss:3.3442 train_time:548102ms step_avg:153.83ms step:3574/5100 train_loss:3.4237 train_time:548254ms step_avg:153.83ms step:3575/5100 train_loss:3.5862 train_time:548406ms step_avg:153.83ms step:3576/5100 train_loss:3.4584 train_time:548560ms step_avg:153.83ms step:3577/5100 train_loss:3.4035 train_time:548712ms step_avg:153.83ms step:3578/5100 train_loss:3.3666 train_time:548865ms step_avg:153.83ms step:3579/5100 train_loss:3.4433 train_time:549017ms step_avg:153.83ms step:3580/5100 train_loss:3.3919 train_time:549170ms step_avg:153.83ms step:3581/5100 train_loss:3.2894 train_time:549322ms step_avg:153.83ms step:3582/5100 train_loss:3.3653 train_time:549474ms step_avg:153.83ms step:3583/5100 train_loss:3.3199 train_time:549627ms step_avg:153.83ms step:3584/5100 train_loss:3.4261 train_time:549778ms step_avg:153.83ms step:3585/5100 train_loss:3.5131 train_time:549931ms step_avg:153.83ms step:3586/5100 train_loss:3.3670 train_time:550084ms step_avg:153.83ms step:3587/5100 train_loss:3.4161 train_time:550236ms step_avg:153.83ms step:3588/5100 train_loss:3.4185 train_time:550390ms step_avg:153.83ms step:3589/5100 train_loss:3.3958 train_time:550542ms step_avg:153.83ms step:3590/5100 train_loss:3.3958 train_time:550695ms step_avg:153.83ms step:3591/5100 train_loss:3.5184 train_time:550950ms step_avg:153.85ms step:3592/5100 train_loss:3.3902 train_time:551108ms step_avg:153.85ms step:3593/5100 train_loss:3.4816 train_time:551261ms step_avg:153.85ms step:3594/5100 train_loss:3.4617 train_time:551412ms step_avg:153.85ms step:3595/5100 train_loss:3.4210 train_time:551564ms step_avg:153.85ms step:3596/5100 train_loss:3.3425 train_time:551716ms step_avg:153.85ms step:3597/5100 train_loss:3.3528 train_time:551868ms step_avg:153.85ms step:3598/5100 train_loss:3.6169 train_time:552022ms step_avg:153.85ms step:3599/5100 train_loss:3.4035 train_time:552175ms step_avg:153.85ms step:3600/5100 train_loss:3.4175 train_time:552328ms step_avg:153.85ms step:3601/5100 train_loss:3.2814 train_time:552479ms step_avg:153.85ms step:3602/5100 train_loss:3.4537 train_time:552632ms step_avg:153.85ms step:3603/5100 train_loss:3.4057 train_time:552785ms step_avg:153.85ms step:3604/5100 train_loss:3.5634 train_time:552936ms step_avg:153.85ms step:3605/5100 train_loss:3.6019 train_time:553089ms step_avg:153.85ms step:3606/5100 train_loss:3.3961 train_time:553241ms step_avg:153.85ms step:3607/5100 train_loss:3.4371 train_time:553393ms step_avg:153.85ms step:3608/5100 train_loss:3.6914 train_time:553545ms step_avg:153.85ms step:3609/5100 train_loss:3.3915 train_time:553698ms step_avg:153.85ms step:3610/5100 train_loss:3.5379 train_time:553972ms step_avg:153.88ms step:3611/5100 train_loss:3.2840 train_time:554123ms step_avg:153.88ms step:3612/5100 train_loss:3.3972 train_time:554276ms step_avg:153.88ms step:3613/5100 train_loss:3.4611 train_time:554427ms step_avg:153.88ms step:3614/5100 train_loss:3.6664 train_time:554580ms step_avg:153.88ms step:3615/5100 train_loss:3.6786 train_time:554732ms step_avg:153.88ms step:3616/5100 train_loss:3.3316 train_time:554884ms step_avg:153.88ms step:3617/5100 train_loss:3.4186 train_time:555038ms step_avg:153.88ms step:3618/5100 train_loss:3.3935 train_time:555192ms step_avg:153.88ms step:3619/5100 train_loss:3.5160 train_time:555345ms step_avg:153.88ms step:3620/5100 train_loss:3.4553 train_time:555499ms step_avg:153.88ms step:3621/5100 train_loss:3.2864 train_time:555650ms step_avg:153.88ms step:3622/5100 train_loss:3.4261 train_time:555803ms step_avg:153.88ms step:3623/5100 train_loss:3.4210 train_time:555956ms step_avg:153.88ms step:3624/5100 train_loss:3.3637 train_time:556108ms step_avg:153.88ms step:3625/5100 train_loss:3.4861 train_time:556261ms step_avg:153.88ms step:3625/5100 val_loss:3.4122 train_time:556325ms step_avg:153.89ms step:3626/5100 train_loss:3.5516 train_time:556421ms step_avg:153.88ms step:3627/5100 train_loss:3.5570 train_time:556577ms step_avg:153.88ms step:3628/5100 train_loss:3.4418 train_time:556728ms step_avg:153.88ms step:3629/5100 train_loss:3.5940 train_time:556880ms step_avg:153.88ms step:3630/5100 train_loss:3.4246 train_time:557032ms step_avg:153.88ms step:3631/5100 train_loss:3.4237 train_time:557184ms step_avg:153.88ms step:3632/5100 train_loss:3.4970 train_time:557335ms step_avg:153.88ms step:3633/5100 train_loss:3.4863 train_time:557488ms step_avg:153.87ms step:3634/5100 train_loss:3.4085 train_time:557641ms step_avg:153.87ms step:3635/5100 train_loss:3.4128 train_time:557794ms step_avg:153.87ms step:3636/5100 train_loss:3.4564 train_time:557947ms step_avg:153.87ms step:3637/5100 train_loss:3.6414 train_time:558099ms step_avg:153.87ms step:3638/5100 train_loss:3.4385 train_time:558251ms step_avg:153.87ms step:3639/5100 train_loss:3.3958 train_time:558403ms step_avg:153.87ms step:3640/5100 train_loss:3.3946 train_time:558555ms step_avg:153.87ms step:3641/5100 train_loss:3.6861 train_time:558707ms step_avg:153.87ms step:3642/5100 train_loss:3.4184 train_time:558860ms step_avg:153.87ms step:3643/5100 train_loss:3.4630 train_time:559013ms step_avg:153.87ms step:3644/5100 train_loss:3.4473 train_time:559166ms step_avg:153.87ms step:3645/5100 train_loss:3.3649 train_time:559318ms step_avg:153.87ms step:3646/5100 train_loss:3.5610 train_time:559470ms step_avg:153.87ms step:3647/5100 train_loss:3.3390 train_time:559622ms step_avg:153.87ms step:3648/5100 train_loss:3.4159 train_time:559775ms step_avg:153.87ms step:3649/5100 train_loss:3.4869 train_time:559927ms step_avg:153.87ms step:3650/5100 train_loss:3.4347 train_time:560079ms step_avg:153.87ms step:3651/5100 train_loss:3.4722 train_time:560230ms step_avg:153.87ms step:3652/5100 train_loss:3.5314 train_time:560384ms step_avg:153.87ms step:3653/5100 train_loss:3.3541 train_time:560535ms step_avg:153.87ms step:3654/5100 train_loss:3.4579 train_time:560688ms step_avg:153.87ms step:3655/5100 train_loss:3.4818 train_time:560840ms step_avg:153.87ms step:3656/5100 train_loss:4.1682 train_time:560993ms step_avg:153.87ms step:3657/5100 train_loss:3.5375 train_time:561146ms step_avg:153.87ms step:3658/5100 train_loss:3.4512 train_time:561297ms step_avg:153.86ms step:3659/5100 train_loss:3.4409 train_time:561450ms step_avg:153.86ms step:3660/5100 train_loss:3.3227 train_time:561603ms step_avg:153.86ms step:3661/5100 train_loss:3.4613 train_time:561755ms step_avg:153.86ms step:3662/5100 train_loss:3.3302 train_time:561907ms step_avg:153.86ms step:3663/5100 train_loss:3.4756 train_time:562059ms step_avg:153.86ms step:3664/5100 train_loss:3.4862 train_time:562211ms step_avg:153.86ms step:3665/5100 train_loss:3.3216 train_time:562364ms step_avg:153.86ms step:3666/5100 train_loss:3.2587 train_time:562517ms step_avg:153.86ms step:3667/5100 train_loss:3.6830 train_time:562669ms step_avg:153.86ms step:3668/5100 train_loss:3.4721 train_time:562821ms step_avg:153.86ms step:3669/5100 train_loss:3.4937 train_time:562973ms step_avg:153.86ms step:3670/5100 train_loss:3.4155 train_time:563126ms step_avg:153.86ms step:3671/5100 train_loss:3.4821 train_time:563279ms step_avg:153.86ms step:3672/5100 train_loss:3.3793 train_time:563430ms step_avg:153.86ms step:3673/5100 train_loss:3.3854 train_time:563583ms step_avg:153.86ms step:3674/5100 train_loss:3.2779 train_time:563735ms step_avg:153.86ms step:3675/5100 train_loss:3.3576 train_time:563887ms step_avg:153.86ms step:3676/5100 train_loss:3.5158 train_time:564040ms step_avg:153.86ms step:3677/5100 train_loss:3.3167 train_time:564194ms step_avg:153.86ms step:3678/5100 train_loss:3.4820 train_time:564346ms step_avg:153.86ms step:3679/5100 train_loss:3.4687 train_time:564498ms step_avg:153.86ms step:3680/5100 train_loss:3.3697 train_time:564651ms step_avg:153.86ms step:3681/5100 train_loss:3.4370 train_time:564803ms step_avg:153.86ms step:3682/5100 train_loss:3.5049 train_time:564955ms step_avg:153.85ms step:3683/5100 train_loss:3.5993 train_time:565107ms step_avg:153.85ms step:3684/5100 train_loss:3.3448 train_time:565260ms step_avg:153.85ms step:3685/5100 train_loss:3.4224 train_time:565413ms step_avg:153.85ms step:3686/5100 train_loss:3.5684 train_time:565566ms step_avg:153.85ms step:3687/5100 train_loss:3.3419 train_time:565718ms step_avg:153.85ms step:3688/5100 train_loss:3.5503 train_time:565871ms step_avg:153.85ms step:3689/5100 train_loss:3.2821 train_time:566024ms step_avg:153.85ms step:3690/5100 train_loss:3.3749 train_time:566176ms step_avg:153.85ms step:3691/5100 train_loss:3.5046 train_time:566328ms step_avg:153.85ms step:3692/5100 train_loss:3.2963 train_time:566481ms step_avg:153.85ms step:3693/5100 train_loss:3.4455 train_time:566633ms step_avg:153.85ms step:3694/5100 train_loss:3.4282 train_time:566786ms step_avg:153.85ms step:3695/5100 train_loss:3.4262 train_time:566938ms step_avg:153.85ms step:3696/5100 train_loss:3.4648 train_time:567091ms step_avg:153.85ms step:3697/5100 train_loss:3.3030 train_time:567243ms step_avg:153.85ms step:3698/5100 train_loss:3.4416 train_time:567396ms step_avg:153.85ms step:3699/5100 train_loss:3.4530 train_time:567548ms step_avg:153.85ms step:3700/5100 train_loss:3.4317 train_time:567701ms step_avg:153.85ms step:3701/5100 train_loss:3.4920 train_time:567852ms step_avg:153.85ms step:3702/5100 train_loss:3.4533 train_time:568005ms step_avg:153.85ms step:3703/5100 train_loss:3.3726 train_time:568158ms step_avg:153.85ms step:3704/5100 train_loss:3.3471 train_time:568310ms step_avg:153.85ms step:3705/5100 train_loss:3.4864 train_time:568463ms step_avg:153.85ms step:3706/5100 train_loss:3.4918 train_time:568615ms step_avg:153.85ms step:3707/5100 train_loss:3.4950 train_time:568768ms step_avg:153.85ms step:3708/5100 train_loss:3.4527 train_time:568921ms step_avg:153.85ms step:3709/5100 train_loss:3.3132 train_time:569073ms step_avg:153.85ms step:3710/5100 train_loss:3.6253 train_time:569226ms step_avg:153.84ms step:3711/5100 train_loss:3.2048 train_time:569378ms step_avg:153.84ms step:3712/5100 train_loss:3.4782 train_time:569530ms step_avg:153.84ms step:3713/5100 train_loss:3.3635 train_time:569683ms step_avg:153.84ms step:3714/5100 train_loss:3.3944 train_time:569835ms step_avg:153.84ms step:3715/5100 train_loss:3.7774 train_time:569988ms step_avg:153.84ms step:3716/5100 train_loss:3.6261 train_time:570141ms step_avg:153.84ms step:3717/5100 train_loss:3.8896 train_time:570294ms step_avg:153.84ms step:3718/5100 train_loss:3.3959 train_time:570447ms step_avg:153.84ms step:3719/5100 train_loss:3.3133 train_time:570599ms step_avg:153.84ms step:3720/5100 train_loss:3.5689 train_time:570751ms step_avg:153.84ms step:3721/5100 train_loss:3.3343 train_time:570903ms step_avg:153.84ms step:3722/5100 train_loss:3.4220 train_time:571055ms step_avg:153.84ms step:3723/5100 train_loss:3.2919 train_time:571207ms step_avg:153.84ms step:3724/5100 train_loss:3.2816 train_time:571359ms step_avg:153.84ms step:3725/5100 train_loss:3.4112 train_time:571513ms step_avg:153.84ms step:3726/5100 train_loss:3.3644 train_time:571666ms step_avg:153.84ms step:3727/5100 train_loss:3.6318 train_time:571819ms step_avg:153.84ms step:3728/5100 train_loss:3.3598 train_time:571971ms step_avg:153.84ms step:3729/5100 train_loss:3.3516 train_time:572124ms step_avg:153.84ms step:3730/5100 train_loss:3.7059 train_time:572276ms step_avg:153.84ms step:3731/5100 train_loss:3.4674 train_time:572428ms step_avg:153.84ms step:3732/5100 train_loss:3.3676 train_time:572581ms step_avg:153.84ms step:3733/5100 train_loss:3.1717 train_time:572733ms step_avg:153.84ms step:3734/5100 train_loss:3.4722 train_time:572887ms step_avg:153.84ms step:3735/5100 train_loss:3.3396 train_time:573038ms step_avg:153.84ms step:3736/5100 train_loss:3.4391 train_time:573191ms step_avg:153.84ms step:3737/5100 train_loss:3.3515 train_time:573343ms step_avg:153.84ms step:3738/5100 train_loss:3.4437 train_time:573496ms step_avg:153.83ms step:3739/5100 train_loss:3.3563 train_time:573648ms step_avg:153.83ms step:3740/5100 train_loss:3.3985 train_time:573800ms step_avg:153.83ms step:3741/5100 train_loss:3.7006 train_time:573953ms step_avg:153.83ms step:3742/5100 train_loss:3.3504 train_time:574105ms step_avg:153.83ms step:3743/5100 train_loss:3.4043 train_time:574257ms step_avg:153.83ms step:3744/5100 train_loss:3.6108 train_time:574409ms step_avg:153.83ms step:3745/5100 train_loss:3.3364 train_time:574561ms step_avg:153.83ms step:3746/5100 train_loss:3.2771 train_time:574713ms step_avg:153.83ms step:3747/5100 train_loss:3.4506 train_time:574866ms step_avg:153.83ms step:3748/5100 train_loss:3.3068 train_time:575018ms step_avg:153.83ms step:3749/5100 train_loss:3.3444 train_time:575170ms step_avg:153.83ms step:3750/5100 train_loss:3.5451 train_time:575324ms step_avg:153.83ms step:3750/5100 val_loss:3.4007 train_time:575386ms step_avg:153.85ms step:3751/5100 train_loss:3.4232 train_time:575481ms step_avg:153.83ms step:3752/5100 train_loss:3.6578 train_time:575636ms step_avg:153.83ms step:3753/5100 train_loss:3.3804 train_time:575788ms step_avg:153.83ms step:3754/5100 train_loss:3.3800 train_time:575941ms step_avg:153.83ms step:3755/5100 train_loss:3.3467 train_time:576093ms step_avg:153.83ms step:3756/5100 train_loss:3.4445 train_time:576245ms step_avg:153.83ms step:3757/5100 train_loss:3.3937 train_time:576397ms step_avg:153.83ms step:3758/5100 train_loss:3.4039 train_time:576550ms step_avg:153.83ms step:3759/5100 train_loss:3.5843 train_time:576702ms step_avg:153.83ms step:3760/5100 train_loss:3.4771 train_time:576855ms step_avg:153.83ms step:3761/5100 train_loss:3.5856 train_time:577008ms step_avg:153.83ms step:3762/5100 train_loss:3.3530 train_time:577160ms step_avg:153.83ms step:3763/5100 train_loss:3.3698 train_time:577313ms step_avg:153.83ms step:3764/5100 train_loss:3.5359 train_time:577465ms step_avg:153.83ms step:3765/5100 train_loss:3.2877 train_time:577619ms step_avg:153.83ms step:3766/5100 train_loss:3.3810 train_time:577771ms step_avg:153.83ms step:3767/5100 train_loss:3.4743 train_time:577923ms step_avg:153.83ms step:3768/5100 train_loss:3.2730 train_time:578076ms step_avg:153.83ms step:3769/5100 train_loss:3.5514 train_time:578229ms step_avg:153.83ms step:3770/5100 train_loss:3.3640 train_time:578381ms step_avg:153.82ms step:3771/5100 train_loss:3.2384 train_time:578533ms step_avg:153.82ms step:3772/5100 train_loss:3.4849 train_time:578687ms step_avg:153.82ms step:3773/5100 train_loss:3.4124 train_time:578841ms step_avg:153.82ms step:3774/5100 train_loss:3.4051 train_time:578994ms step_avg:153.82ms step:3775/5100 train_loss:3.4020 train_time:579147ms step_avg:153.82ms step:3776/5100 train_loss:3.4450 train_time:579299ms step_avg:153.82ms step:3777/5100 train_loss:3.2901 train_time:579451ms step_avg:153.82ms step:3778/5100 train_loss:3.3975 train_time:579604ms step_avg:153.82ms step:3779/5100 train_loss:3.5121 train_time:579757ms step_avg:153.82ms step:3780/5100 train_loss:3.4855 train_time:580014ms step_avg:153.85ms step:3781/5100 train_loss:3.4846 train_time:580171ms step_avg:153.85ms step:3782/5100 train_loss:3.4374 train_time:580323ms step_avg:153.85ms step:3783/5100 train_loss:3.4354 train_time:580475ms step_avg:153.85ms step:3784/5100 train_loss:3.4046 train_time:580626ms step_avg:153.85ms step:3785/5100 train_loss:3.2831 train_time:580779ms step_avg:153.85ms step:3786/5100 train_loss:3.3697 train_time:580931ms step_avg:153.85ms step:3787/5100 train_loss:3.4170 train_time:581084ms step_avg:153.85ms step:3788/5100 train_loss:3.4004 train_time:581237ms step_avg:153.85ms step:3789/5100 train_loss:3.3497 train_time:581390ms step_avg:153.85ms step:3790/5100 train_loss:3.3780 train_time:581542ms step_avg:153.85ms step:3791/5100 train_loss:3.2487 train_time:581694ms step_avg:153.85ms step:3792/5100 train_loss:3.4678 train_time:581846ms step_avg:153.85ms step:3793/5100 train_loss:3.4523 train_time:581999ms step_avg:153.85ms step:3794/5100 train_loss:3.3819 train_time:582150ms step_avg:153.85ms step:3795/5100 train_loss:3.3395 train_time:582302ms step_avg:153.84ms step:3796/5100 train_loss:3.1880 train_time:582455ms step_avg:153.84ms step:3797/5100 train_loss:3.3792 train_time:582609ms step_avg:153.84ms step:3798/5100 train_loss:3.3826 train_time:582761ms step_avg:153.84ms step:3799/5100 train_loss:3.4449 train_time:582913ms step_avg:153.84ms step:3800/5100 train_loss:3.3516 train_time:583188ms step_avg:153.88ms step:3801/5100 train_loss:3.3317 train_time:583341ms step_avg:153.88ms step:3802/5100 train_loss:3.2868 train_time:583492ms step_avg:153.87ms step:3803/5100 train_loss:3.5929 train_time:583644ms step_avg:153.87ms step:3804/5100 train_loss:3.4480 train_time:583795ms step_avg:153.87ms step:3805/5100 train_loss:3.2940 train_time:583948ms step_avg:153.87ms step:3806/5100 train_loss:3.5351 train_time:584100ms step_avg:153.87ms step:3807/5100 train_loss:3.5047 train_time:584254ms step_avg:153.87ms step:3808/5100 train_loss:3.3760 train_time:584407ms step_avg:153.87ms step:3809/5100 train_loss:3.4357 train_time:584559ms step_avg:153.87ms step:3810/5100 train_loss:3.3462 train_time:584711ms step_avg:153.87ms step:3811/5100 train_loss:3.4175 train_time:584863ms step_avg:153.87ms step:3812/5100 train_loss:3.3919 train_time:585016ms step_avg:153.87ms step:3813/5100 train_loss:3.4252 train_time:585168ms step_avg:153.87ms step:3814/5100 train_loss:3.4221 train_time:585321ms step_avg:153.87ms step:3815/5100 train_loss:3.3162 train_time:585475ms step_avg:153.87ms step:3816/5100 train_loss:3.6445 train_time:585627ms step_avg:153.87ms step:3817/5100 train_loss:3.2468 train_time:585780ms step_avg:153.87ms step:3818/5100 train_loss:3.4267 train_time:585932ms step_avg:153.87ms step:3819/5100 train_loss:3.4004 train_time:586085ms step_avg:153.87ms step:3820/5100 train_loss:3.3843 train_time:586238ms step_avg:153.87ms step:3821/5100 train_loss:3.3167 train_time:586390ms step_avg:153.87ms step:3822/5100 train_loss:3.4908 train_time:586543ms step_avg:153.87ms step:3823/5100 train_loss:3.2246 train_time:586696ms step_avg:153.87ms step:3824/5100 train_loss:3.3370 train_time:586848ms step_avg:153.87ms step:3825/5100 train_loss:3.3872 train_time:587001ms step_avg:153.87ms step:3826/5100 train_loss:3.5331 train_time:587153ms step_avg:153.87ms step:3827/5100 train_loss:3.4693 train_time:587305ms step_avg:153.87ms step:3828/5100 train_loss:3.8456 train_time:587458ms step_avg:153.87ms step:3829/5100 train_loss:3.4637 train_time:587610ms step_avg:153.87ms step:3830/5100 train_loss:3.2895 train_time:587762ms step_avg:153.86ms step:3831/5100 train_loss:3.3334 train_time:587915ms step_avg:153.86ms step:3832/5100 train_loss:3.5444 train_time:588067ms step_avg:153.86ms step:3833/5100 train_loss:3.3589 train_time:588221ms step_avg:153.86ms step:3834/5100 train_loss:3.4760 train_time:588373ms step_avg:153.86ms step:3835/5100 train_loss:3.4112 train_time:588525ms step_avg:153.86ms step:3836/5100 train_loss:3.2212 train_time:588678ms step_avg:153.86ms step:3837/5100 train_loss:3.5009 train_time:588830ms step_avg:153.86ms step:3838/5100 train_loss:3.4759 train_time:588983ms step_avg:153.86ms step:3839/5100 train_loss:3.4318 train_time:589135ms step_avg:153.86ms step:3840/5100 train_loss:3.4964 train_time:589288ms step_avg:153.86ms step:3841/5100 train_loss:3.6197 train_time:589441ms step_avg:153.86ms step:3842/5100 train_loss:3.3662 train_time:589593ms step_avg:153.86ms step:3843/5100 train_loss:3.4188 train_time:589745ms step_avg:153.86ms step:3844/5100 train_loss:3.5620 train_time:589898ms step_avg:153.86ms step:3845/5100 train_loss:3.3643 train_time:590050ms step_avg:153.86ms step:3846/5100 train_loss:3.2302 train_time:590202ms step_avg:153.86ms step:3847/5100 train_loss:3.4682 train_time:590355ms step_avg:153.86ms step:3848/5100 train_loss:3.3847 train_time:590507ms step_avg:153.86ms step:3849/5100 train_loss:3.4375 train_time:590660ms step_avg:153.86ms step:3850/5100 train_loss:3.3248 train_time:590813ms step_avg:153.86ms step:3851/5100 train_loss:3.3148 train_time:590965ms step_avg:153.86ms step:3852/5100 train_loss:3.4751 train_time:591118ms step_avg:153.86ms step:3853/5100 train_loss:3.3234 train_time:591270ms step_avg:153.86ms step:3854/5100 train_loss:3.2936 train_time:591421ms step_avg:153.86ms step:3855/5100 train_loss:3.3752 train_time:591575ms step_avg:153.86ms step:3856/5100 train_loss:3.4135 train_time:591726ms step_avg:153.86ms step:3857/5100 train_loss:3.3811 train_time:591880ms step_avg:153.85ms step:3858/5100 train_loss:3.4098 train_time:592031ms step_avg:153.85ms step:3859/5100 train_loss:3.3926 train_time:592184ms step_avg:153.85ms step:3860/5100 train_loss:3.4037 train_time:592336ms step_avg:153.85ms step:3861/5100 train_loss:3.5579 train_time:592489ms step_avg:153.85ms step:3862/5100 train_loss:3.3620 train_time:592642ms step_avg:153.85ms step:3863/5100 train_loss:3.4872 train_time:592793ms step_avg:153.85ms step:3864/5100 train_loss:3.4407 train_time:592946ms step_avg:153.85ms step:3865/5100 train_loss:3.4864 train_time:593098ms step_avg:153.85ms step:3866/5100 train_loss:3.4536 train_time:593251ms step_avg:153.85ms step:3867/5100 train_loss:3.3943 train_time:593403ms step_avg:153.85ms step:3868/5100 train_loss:3.4864 train_time:593556ms step_avg:153.85ms step:3869/5100 train_loss:3.6466 train_time:593708ms step_avg:153.85ms step:3870/5100 train_loss:3.4928 train_time:593861ms step_avg:153.85ms step:3871/5100 train_loss:3.3867 train_time:594014ms step_avg:153.85ms step:3872/5100 train_loss:3.5291 train_time:594166ms step_avg:153.85ms step:3873/5100 train_loss:3.4261 train_time:594320ms step_avg:153.85ms step:3874/5100 train_loss:3.3685 train_time:594471ms step_avg:153.85ms step:3875/5100 train_loss:3.4720 train_time:594624ms step_avg:153.85ms step:3875/5100 val_loss:3.3876 train_time:594688ms step_avg:153.86ms step:3876/5100 train_loss:3.9934 train_time:594781ms step_avg:153.85ms step:3877/5100 train_loss:3.4069 train_time:594938ms step_avg:153.85ms step:3878/5100 train_loss:3.4067 train_time:595091ms step_avg:153.85ms step:3879/5100 train_loss:3.3888 train_time:595242ms step_avg:153.85ms step:3880/5100 train_loss:3.5869 train_time:595394ms step_avg:153.85ms step:3881/5100 train_loss:3.4018 train_time:595547ms step_avg:153.85ms step:3882/5100 train_loss:3.4706 train_time:595698ms step_avg:153.85ms step:3883/5100 train_loss:3.5157 train_time:595850ms step_avg:153.85ms step:3884/5100 train_loss:3.3303 train_time:596003ms step_avg:153.85ms step:3885/5100 train_loss:3.3165 train_time:596156ms step_avg:153.85ms step:3886/5100 train_loss:3.3585 train_time:596308ms step_avg:153.85ms step:3887/5100 train_loss:3.3962 train_time:596460ms step_avg:153.85ms step:3888/5100 train_loss:3.5698 train_time:596614ms step_avg:153.85ms step:3889/5100 train_loss:3.4225 train_time:596767ms step_avg:153.85ms step:3890/5100 train_loss:3.3581 train_time:596919ms step_avg:153.85ms step:3891/5100 train_loss:3.5023 train_time:597073ms step_avg:153.85ms step:3892/5100 train_loss:3.3624 train_time:597225ms step_avg:153.84ms step:3893/5100 train_loss:3.6079 train_time:597378ms step_avg:153.84ms step:3894/5100 train_loss:3.3438 train_time:597531ms step_avg:153.84ms step:3895/5100 train_loss:3.3571 train_time:597683ms step_avg:153.84ms step:3896/5100 train_loss:3.4353 train_time:597835ms step_avg:153.84ms step:3897/5100 train_loss:3.6794 train_time:597987ms step_avg:153.84ms step:3898/5100 train_loss:3.2343 train_time:598140ms step_avg:153.84ms step:3899/5100 train_loss:3.3576 train_time:598292ms step_avg:153.84ms step:3900/5100 train_loss:3.5031 train_time:598444ms step_avg:153.84ms step:3901/5100 train_loss:3.4275 train_time:598596ms step_avg:153.84ms step:3902/5100 train_loss:3.4642 train_time:598749ms step_avg:153.84ms step:3903/5100 train_loss:3.7514 train_time:598901ms step_avg:153.84ms step:3904/5100 train_loss:3.3425 train_time:599053ms step_avg:153.84ms step:3905/5100 train_loss:3.3634 train_time:599206ms step_avg:153.84ms step:3906/5100 train_loss:3.3157 train_time:599358ms step_avg:153.84ms step:3907/5100 train_loss:3.4773 train_time:599510ms step_avg:153.84ms step:3908/5100 train_loss:3.4941 train_time:599662ms step_avg:153.84ms step:3909/5100 train_loss:3.4843 train_time:599814ms step_avg:153.84ms step:3910/5100 train_loss:3.4272 train_time:599966ms step_avg:153.84ms step:3911/5100 train_loss:3.3630 train_time:600119ms step_avg:153.84ms step:3912/5100 train_loss:3.3856 train_time:600272ms step_avg:153.84ms step:3913/5100 train_loss:3.3759 train_time:600424ms step_avg:153.84ms step:3914/5100 train_loss:3.4932 train_time:600577ms step_avg:153.84ms step:3915/5100 train_loss:3.3363 train_time:600729ms step_avg:153.84ms step:3916/5100 train_loss:3.3097 train_time:600882ms step_avg:153.84ms step:3917/5100 train_loss:3.3058 train_time:601035ms step_avg:153.84ms step:3918/5100 train_loss:3.4256 train_time:601187ms step_avg:153.83ms step:3919/5100 train_loss:3.5457 train_time:601338ms step_avg:153.83ms step:3920/5100 train_loss:3.3256 train_time:601491ms step_avg:153.83ms step:3921/5100 train_loss:3.3008 train_time:601644ms step_avg:153.83ms step:3922/5100 train_loss:3.3767 train_time:601797ms step_avg:153.83ms step:3923/5100 train_loss:3.3772 train_time:601951ms step_avg:153.83ms step:3924/5100 train_loss:3.3936 train_time:602106ms step_avg:153.83ms step:3925/5100 train_loss:3.4722 train_time:602258ms step_avg:153.83ms step:3926/5100 train_loss:3.4311 train_time:602410ms step_avg:153.83ms step:3927/5100 train_loss:3.5267 train_time:602562ms step_avg:153.83ms step:3928/5100 train_loss:3.4118 train_time:602713ms step_avg:153.83ms step:3929/5100 train_loss:3.2683 train_time:602867ms step_avg:153.83ms step:3930/5100 train_loss:3.5978 train_time:603019ms step_avg:153.83ms step:3931/5100 train_loss:3.3806 train_time:603173ms step_avg:153.83ms step:3932/5100 train_loss:3.4292 train_time:603326ms step_avg:153.83ms step:3933/5100 train_loss:3.4658 train_time:603478ms step_avg:153.83ms step:3934/5100 train_loss:3.3427 train_time:603631ms step_avg:153.83ms step:3935/5100 train_loss:3.4705 train_time:603783ms step_avg:153.83ms step:3936/5100 train_loss:3.4793 train_time:603935ms step_avg:153.83ms step:3937/5100 train_loss:3.4033 train_time:604087ms step_avg:153.83ms step:3938/5100 train_loss:3.4626 train_time:604241ms step_avg:153.83ms step:3939/5100 train_loss:3.3855 train_time:604393ms step_avg:153.83ms step:3940/5100 train_loss:3.1532 train_time:604546ms step_avg:153.83ms step:3941/5100 train_loss:3.3521 train_time:604698ms step_avg:153.83ms step:3942/5100 train_loss:3.4570 train_time:604851ms step_avg:153.83ms step:3943/5100 train_loss:3.5569 train_time:605004ms step_avg:153.83ms step:3944/5100 train_loss:3.5918 train_time:605156ms step_avg:153.83ms step:3945/5100 train_loss:3.4330 train_time:605309ms step_avg:153.83ms step:3946/5100 train_loss:3.3470 train_time:605463ms step_avg:153.83ms step:3947/5100 train_loss:3.3664 train_time:605615ms step_avg:153.83ms step:3948/5100 train_loss:3.4358 train_time:605768ms step_avg:153.83ms step:3949/5100 train_loss:3.2467 train_time:605920ms step_avg:153.83ms step:3950/5100 train_loss:3.4489 train_time:606074ms step_avg:153.83ms step:3951/5100 train_loss:3.3797 train_time:606226ms step_avg:153.83ms step:3952/5100 train_loss:3.1886 train_time:606379ms step_avg:153.83ms step:3953/5100 train_loss:3.2589 train_time:606532ms step_avg:153.83ms step:3954/5100 train_loss:3.5254 train_time:606684ms step_avg:153.82ms step:3955/5100 train_loss:3.4412 train_time:606836ms step_avg:153.82ms step:3956/5100 train_loss:3.3822 train_time:606990ms step_avg:153.82ms step:3957/5100 train_loss:3.4350 train_time:607141ms step_avg:153.82ms step:3958/5100 train_loss:3.1548 train_time:607294ms step_avg:153.82ms step:3959/5100 train_loss:3.4461 train_time:607448ms step_avg:153.82ms step:3960/5100 train_loss:3.3916 train_time:607602ms step_avg:153.82ms step:3961/5100 train_loss:3.3603 train_time:607754ms step_avg:153.82ms step:3962/5100 train_loss:3.3773 train_time:607906ms step_avg:153.82ms step:3963/5100 train_loss:3.4017 train_time:608058ms step_avg:153.82ms step:3964/5100 train_loss:3.4322 train_time:608212ms step_avg:153.82ms step:3965/5100 train_loss:3.2759 train_time:608363ms step_avg:153.82ms step:3966/5100 train_loss:3.4043 train_time:608516ms step_avg:153.82ms step:3967/5100 train_loss:3.4748 train_time:608670ms step_avg:153.82ms step:3968/5100 train_loss:3.3863 train_time:608822ms step_avg:153.82ms step:3969/5100 train_loss:3.4730 train_time:609071ms step_avg:153.84ms step:3970/5100 train_loss:3.3594 train_time:609232ms step_avg:153.85ms step:3971/5100 train_loss:3.5550 train_time:609383ms step_avg:153.85ms step:3972/5100 train_loss:3.4736 train_time:609535ms step_avg:153.85ms step:3973/5100 train_loss:3.4269 train_time:609687ms step_avg:153.84ms step:3974/5100 train_loss:3.3214 train_time:609838ms step_avg:153.84ms step:3975/5100 train_loss:3.3731 train_time:609990ms step_avg:153.84ms step:3976/5100 train_loss:3.4409 train_time:610144ms step_avg:153.84ms step:3977/5100 train_loss:3.3535 train_time:610296ms step_avg:153.84ms step:3978/5100 train_loss:3.4121 train_time:610450ms step_avg:153.84ms step:3979/5100 train_loss:3.4907 train_time:610602ms step_avg:153.84ms step:3980/5100 train_loss:3.4262 train_time:610754ms step_avg:153.84ms step:3981/5100 train_loss:3.4366 train_time:610906ms step_avg:153.84ms step:3982/5100 train_loss:3.6327 train_time:611058ms step_avg:153.84ms step:3983/5100 train_loss:3.3677 train_time:611211ms step_avg:153.84ms step:3984/5100 train_loss:3.4381 train_time:611363ms step_avg:153.84ms step:3985/5100 train_loss:3.3696 train_time:611515ms step_avg:153.84ms step:3986/5100 train_loss:3.2953 train_time:611668ms step_avg:153.84ms step:3987/5100 train_loss:3.3381 train_time:611821ms step_avg:153.84ms step:3988/5100 train_loss:3.3640 train_time:611974ms step_avg:153.84ms step:3989/5100 train_loss:3.0928 train_time:612126ms step_avg:153.84ms step:3990/5100 train_loss:3.4126 train_time:612402ms step_avg:153.87ms step:3991/5100 train_loss:3.3880 train_time:612553ms step_avg:153.87ms step:3992/5100 train_loss:3.2365 train_time:612704ms step_avg:153.87ms step:3993/5100 train_loss:3.3414 train_time:612856ms step_avg:153.87ms step:3994/5100 train_loss:3.5318 train_time:613008ms step_avg:153.87ms step:3995/5100 train_loss:3.3567 train_time:613160ms step_avg:153.87ms step:3996/5100 train_loss:3.2739 train_time:613312ms step_avg:153.87ms step:3997/5100 train_loss:3.4208 train_time:613466ms step_avg:153.87ms step:3998/5100 train_loss:3.3339 train_time:613619ms step_avg:153.87ms step:3999/5100 train_loss:3.2972 train_time:613772ms step_avg:153.87ms step:4000/5100 train_loss:3.3673 train_time:613924ms step_avg:153.87ms step:4000/5100 val_loss:3.3718 train_time:613988ms step_avg:153.88ms step:4001/5100 train_loss:3.4921 train_time:614082ms step_avg:153.87ms step:4002/5100 train_loss:3.5540 train_time:614242ms step_avg:153.87ms step:4003/5100 train_loss:3.2321 train_time:614394ms step_avg:153.87ms step:4004/5100 train_loss:3.4329 train_time:614546ms step_avg:153.87ms step:4005/5100 train_loss:3.3303 train_time:614698ms step_avg:153.87ms step:4006/5100 train_loss:3.3850 train_time:614850ms step_avg:153.87ms step:4007/5100 train_loss:3.3695 train_time:615002ms step_avg:153.87ms step:4008/5100 train_loss:3.5668 train_time:615156ms step_avg:153.87ms step:4009/5100 train_loss:3.1528 train_time:615310ms step_avg:153.87ms step:4010/5100 train_loss:3.3565 train_time:615462ms step_avg:153.87ms step:4011/5100 train_loss:3.3383 train_time:615615ms step_avg:153.87ms step:4012/5100 train_loss:3.3060 train_time:615766ms step_avg:153.86ms step:4013/5100 train_loss:3.4689 train_time:615918ms step_avg:153.86ms step:4014/5100 train_loss:3.3391 train_time:616071ms step_avg:153.86ms step:4015/5100 train_loss:3.4407 train_time:616223ms step_avg:153.86ms step:4016/5100 train_loss:3.5296 train_time:616376ms step_avg:153.86ms step:4017/5100 train_loss:3.5151 train_time:616529ms step_avg:153.86ms step:4018/5100 train_loss:3.2747 train_time:616681ms step_avg:153.86ms step:4019/5100 train_loss:3.4020 train_time:616834ms step_avg:153.86ms step:4020/5100 train_loss:3.3208 train_time:616985ms step_avg:153.86ms step:4021/5100 train_loss:3.5902 train_time:617138ms step_avg:153.86ms step:4022/5100 train_loss:3.4693 train_time:617291ms step_avg:153.86ms step:4023/5100 train_loss:3.4511 train_time:617443ms step_avg:153.86ms step:4024/5100 train_loss:3.4225 train_time:617595ms step_avg:153.86ms step:4025/5100 train_loss:3.4551 train_time:617748ms step_avg:153.86ms step:4026/5100 train_loss:3.2117 train_time:617900ms step_avg:153.86ms step:4027/5100 train_loss:3.4296 train_time:618053ms step_avg:153.86ms step:4028/5100 train_loss:3.3807 train_time:618205ms step_avg:153.86ms step:4029/5100 train_loss:3.2591 train_time:618357ms step_avg:153.86ms step:4030/5100 train_loss:3.2913 train_time:618510ms step_avg:153.86ms step:4031/5100 train_loss:3.3428 train_time:618662ms step_avg:153.86ms step:4032/5100 train_loss:3.4366 train_time:618814ms step_avg:153.86ms step:4033/5100 train_loss:3.4029 train_time:618966ms step_avg:153.86ms step:4034/5100 train_loss:3.3804 train_time:619118ms step_avg:153.86ms step:4035/5100 train_loss:3.3772 train_time:619271ms step_avg:153.86ms step:4036/5100 train_loss:3.3184 train_time:619424ms step_avg:153.86ms step:4037/5100 train_loss:3.4793 train_time:619576ms step_avg:153.86ms step:4038/5100 train_loss:3.4123 train_time:619728ms step_avg:153.86ms step:4039/5100 train_loss:3.3939 train_time:619881ms step_avg:153.85ms step:4040/5100 train_loss:3.3889 train_time:620034ms step_avg:153.85ms step:4041/5100 train_loss:3.4301 train_time:620186ms step_avg:153.85ms step:4042/5100 train_loss:3.6196 train_time:620339ms step_avg:153.85ms step:4043/5100 train_loss:3.5189 train_time:620491ms step_avg:153.85ms step:4044/5100 train_loss:3.3030 train_time:620643ms step_avg:153.85ms step:4045/5100 train_loss:3.4755 train_time:620795ms step_avg:153.85ms step:4046/5100 train_loss:3.1776 train_time:620948ms step_avg:153.85ms step:4047/5100 train_loss:3.4397 train_time:621100ms step_avg:153.85ms step:4048/5100 train_loss:3.5238 train_time:621253ms step_avg:153.85ms step:4049/5100 train_loss:3.3980 train_time:621405ms step_avg:153.85ms step:4050/5100 train_loss:3.3279 train_time:621558ms step_avg:153.85ms step:4051/5100 train_loss:3.3658 train_time:621710ms step_avg:153.85ms step:4052/5100 train_loss:3.3019 train_time:621862ms step_avg:153.85ms step:4053/5100 train_loss:3.5082 train_time:622015ms step_avg:153.85ms step:4054/5100 train_loss:3.3642 train_time:622168ms step_avg:153.85ms step:4055/5100 train_loss:3.4495 train_time:622319ms step_avg:153.85ms step:4056/5100 train_loss:3.4278 train_time:622473ms step_avg:153.85ms step:4057/5100 train_loss:3.4009 train_time:622626ms step_avg:153.85ms step:4058/5100 train_loss:3.2710 train_time:622778ms step_avg:153.85ms step:4059/5100 train_loss:3.4230 train_time:622929ms step_avg:153.85ms step:4060/5100 train_loss:3.2855 train_time:623082ms step_avg:153.85ms step:4061/5100 train_loss:3.3686 train_time:623236ms step_avg:153.85ms step:4062/5100 train_loss:3.4838 train_time:623388ms step_avg:153.85ms step:4063/5100 train_loss:3.6628 train_time:623540ms step_avg:153.85ms step:4064/5100 train_loss:3.0452 train_time:623693ms step_avg:153.85ms step:4065/5100 train_loss:3.4062 train_time:623844ms step_avg:153.85ms step:4066/5100 train_loss:3.2897 train_time:623997ms step_avg:153.85ms step:4067/5100 train_loss:3.4462 train_time:624150ms step_avg:153.85ms step:4068/5100 train_loss:3.4514 train_time:624302ms step_avg:153.84ms step:4069/5100 train_loss:3.2608 train_time:624454ms step_avg:153.84ms step:4070/5100 train_loss:3.4259 train_time:624606ms step_avg:153.84ms step:4071/5100 train_loss:3.2388 train_time:624758ms step_avg:153.84ms step:4072/5100 train_loss:3.4243 train_time:624911ms step_avg:153.84ms step:4073/5100 train_loss:3.5280 train_time:625064ms step_avg:153.84ms step:4074/5100 train_loss:3.4548 train_time:625216ms step_avg:153.84ms step:4075/5100 train_loss:3.3703 train_time:625369ms step_avg:153.84ms step:4076/5100 train_loss:3.3632 train_time:625521ms step_avg:153.84ms step:4077/5100 train_loss:3.2370 train_time:625676ms step_avg:153.84ms step:4078/5100 train_loss:3.4021 train_time:625828ms step_avg:153.84ms step:4079/5100 train_loss:3.4172 train_time:625980ms step_avg:153.84ms step:4080/5100 train_loss:3.2124 train_time:626133ms step_avg:153.84ms step:4081/5100 train_loss:3.3847 train_time:626285ms step_avg:153.84ms step:4082/5100 train_loss:3.3363 train_time:626438ms step_avg:153.84ms step:4083/5100 train_loss:3.3931 train_time:626590ms step_avg:153.84ms step:4084/5100 train_loss:3.3915 train_time:626743ms step_avg:153.84ms step:4085/5100 train_loss:3.4264 train_time:626895ms step_avg:153.84ms step:4086/5100 train_loss:3.3827 train_time:627048ms step_avg:153.84ms step:4087/5100 train_loss:3.3540 train_time:627199ms step_avg:153.84ms step:4088/5100 train_loss:3.4735 train_time:627353ms step_avg:153.84ms step:4089/5100 train_loss:3.3098 train_time:627505ms step_avg:153.84ms step:4090/5100 train_loss:3.3363 train_time:627658ms step_avg:153.84ms step:4091/5100 train_loss:3.3475 train_time:627811ms step_avg:153.84ms step:4092/5100 train_loss:3.3015 train_time:627962ms step_avg:153.84ms step:4093/5100 train_loss:3.2994 train_time:628115ms step_avg:153.84ms step:4094/5100 train_loss:3.4815 train_time:628266ms step_avg:153.84ms step:4095/5100 train_loss:3.4660 train_time:628419ms step_avg:153.84ms step:4096/5100 train_loss:3.3826 train_time:628572ms step_avg:153.84ms step:4097/5100 train_loss:3.4409 train_time:628724ms step_avg:153.84ms step:4098/5100 train_loss:3.2156 train_time:628877ms step_avg:153.83ms step:4099/5100 train_loss:3.3500 train_time:629029ms step_avg:153.83ms step:4100/5100 train_loss:3.3263 train_time:629182ms step_avg:153.83ms step:4101/5100 train_loss:3.1265 train_time:629334ms step_avg:153.83ms step:4102/5100 train_loss:3.4067 train_time:629487ms step_avg:153.83ms step:4103/5100 train_loss:3.3644 train_time:629639ms step_avg:153.83ms step:4104/5100 train_loss:3.1989 train_time:629792ms step_avg:153.83ms step:4105/5100 train_loss:3.2904 train_time:629945ms step_avg:153.83ms step:4106/5100 train_loss:3.4429 train_time:630098ms step_avg:153.83ms step:4107/5100 train_loss:3.4921 train_time:630250ms step_avg:153.83ms step:4108/5100 train_loss:3.3913 train_time:630403ms step_avg:153.83ms step:4109/5100 train_loss:3.4776 train_time:630555ms step_avg:153.83ms step:4110/5100 train_loss:3.4727 train_time:630707ms step_avg:153.83ms step:4111/5100 train_loss:3.6320 train_time:630859ms step_avg:153.83ms step:4112/5100 train_loss:3.2922 train_time:631013ms step_avg:153.83ms step:4113/5100 train_loss:3.4244 train_time:631165ms step_avg:153.83ms step:4114/5100 train_loss:3.3228 train_time:631317ms step_avg:153.83ms step:4115/5100 train_loss:3.4307 train_time:631470ms step_avg:153.83ms step:4116/5100 train_loss:3.4303 train_time:631622ms step_avg:153.83ms step:4117/5100 train_loss:3.6686 train_time:631775ms step_avg:153.83ms step:4118/5100 train_loss:3.2427 train_time:631927ms step_avg:153.83ms step:4119/5100 train_loss:3.4106 train_time:632080ms step_avg:153.83ms step:4120/5100 train_loss:3.3239 train_time:632233ms step_avg:153.83ms step:4121/5100 train_loss:3.4280 train_time:632385ms step_avg:153.83ms step:4122/5100 train_loss:3.4256 train_time:632538ms step_avg:153.83ms step:4123/5100 train_loss:3.4125 train_time:632690ms step_avg:153.83ms step:4124/5100 train_loss:3.2514 train_time:632842ms step_avg:153.83ms step:4125/5100 train_loss:3.2593 train_time:632995ms step_avg:153.83ms step:4125/5100 val_loss:3.3587 train_time:633058ms step_avg:153.84ms step:4126/5100 train_loss:3.3839 train_time:633152ms step_avg:153.83ms step:4127/5100 train_loss:3.3151 train_time:633310ms step_avg:153.83ms step:4128/5100 train_loss:3.3876 train_time:633464ms step_avg:153.83ms step:4129/5100 train_loss:3.3812 train_time:633616ms step_avg:153.83ms step:4130/5100 train_loss:3.1673 train_time:633769ms step_avg:153.83ms step:4131/5100 train_loss:3.4577 train_time:633920ms step_avg:153.83ms step:4132/5100 train_loss:3.4118 train_time:634072ms step_avg:153.83ms step:4133/5100 train_loss:3.3295 train_time:634225ms step_avg:153.83ms step:4134/5100 train_loss:3.5371 train_time:634379ms step_avg:153.83ms step:4135/5100 train_loss:3.3648 train_time:634531ms step_avg:153.83ms step:4136/5100 train_loss:3.3455 train_time:634683ms step_avg:153.83ms step:4137/5100 train_loss:3.4955 train_time:634835ms step_avg:153.82ms step:4138/5100 train_loss:3.3345 train_time:634987ms step_avg:153.82ms step:4139/5100 train_loss:3.3978 train_time:635140ms step_avg:153.82ms step:4140/5100 train_loss:3.4886 train_time:635293ms step_avg:153.82ms step:4141/5100 train_loss:3.5116 train_time:635446ms step_avg:153.82ms step:4142/5100 train_loss:3.4762 train_time:635598ms step_avg:153.82ms step:4143/5100 train_loss:3.4541 train_time:635751ms step_avg:153.82ms step:4144/5100 train_loss:3.3559 train_time:635903ms step_avg:153.82ms step:4145/5100 train_loss:3.3232 train_time:636055ms step_avg:153.82ms step:4146/5100 train_loss:3.4337 train_time:636208ms step_avg:153.82ms step:4147/5100 train_loss:2.9926 train_time:636360ms step_avg:153.82ms step:4148/5100 train_loss:3.3582 train_time:636512ms step_avg:153.82ms step:4149/5100 train_loss:3.3903 train_time:636664ms step_avg:153.82ms step:4150/5100 train_loss:3.2048 train_time:636816ms step_avg:153.82ms step:4151/5100 train_loss:3.2366 train_time:636970ms step_avg:153.82ms step:4152/5100 train_loss:3.2748 train_time:637122ms step_avg:153.82ms step:4153/5100 train_loss:3.3274 train_time:637275ms step_avg:153.82ms step:4154/5100 train_loss:3.3866 train_time:637428ms step_avg:153.82ms step:4155/5100 train_loss:3.4934 train_time:637579ms step_avg:153.82ms step:4156/5100 train_loss:3.2995 train_time:637732ms step_avg:153.82ms step:4157/5100 train_loss:3.2515 train_time:637884ms step_avg:153.82ms step:4158/5100 train_loss:3.3729 train_time:638142ms step_avg:153.84ms step:4159/5100 train_loss:3.3728 train_time:638296ms step_avg:153.84ms step:4160/5100 train_loss:3.2959 train_time:638449ms step_avg:153.84ms step:4161/5100 train_loss:3.3803 train_time:638600ms step_avg:153.84ms step:4162/5100 train_loss:3.3095 train_time:638752ms step_avg:153.84ms step:4163/5100 train_loss:3.5383 train_time:638904ms step_avg:153.84ms step:4164/5100 train_loss:3.2199 train_time:639055ms step_avg:153.84ms step:4165/5100 train_loss:3.3243 train_time:639208ms step_avg:153.84ms step:4166/5100 train_loss:3.3118 train_time:639361ms step_avg:153.84ms step:4167/5100 train_loss:3.3537 train_time:639513ms step_avg:153.84ms step:4168/5100 train_loss:3.3441 train_time:639666ms step_avg:153.84ms step:4169/5100 train_loss:3.3700 train_time:639818ms step_avg:153.84ms step:4170/5100 train_loss:3.2117 train_time:639972ms step_avg:153.84ms step:4171/5100 train_loss:3.3138 train_time:640124ms step_avg:153.84ms step:4172/5100 train_loss:3.4399 train_time:640276ms step_avg:153.84ms step:4173/5100 train_loss:3.5096 train_time:640429ms step_avg:153.84ms step:4174/5100 train_loss:3.8832 train_time:640581ms step_avg:153.84ms step:4175/5100 train_loss:3.3173 train_time:640734ms step_avg:153.84ms step:4176/5100 train_loss:3.4758 train_time:640886ms step_avg:153.84ms step:4177/5100 train_loss:3.2755 train_time:641039ms step_avg:153.84ms step:4178/5100 train_loss:3.3000 train_time:641191ms step_avg:153.84ms step:4179/5100 train_loss:3.4601 train_time:641344ms step_avg:153.84ms step:4180/5100 train_loss:3.4068 train_time:641617ms step_avg:153.87ms step:4181/5100 train_loss:3.3931 train_time:641769ms step_avg:153.86ms step:4182/5100 train_loss:3.3940 train_time:641920ms step_avg:153.86ms step:4183/5100 train_loss:3.4268 train_time:642074ms step_avg:153.86ms step:4184/5100 train_loss:3.8621 train_time:642226ms step_avg:153.86ms step:4185/5100 train_loss:3.3767 train_time:642378ms step_avg:153.86ms step:4186/5100 train_loss:3.4246 train_time:642530ms step_avg:153.86ms step:4187/5100 train_loss:3.4657 train_time:642684ms step_avg:153.86ms step:4188/5100 train_loss:3.4596 train_time:642837ms step_avg:153.86ms step:4189/5100 train_loss:3.0964 train_time:642990ms step_avg:153.86ms step:4190/5100 train_loss:3.4522 train_time:643142ms step_avg:153.86ms step:4191/5100 train_loss:3.4645 train_time:643295ms step_avg:153.86ms step:4192/5100 train_loss:3.4375 train_time:643448ms step_avg:153.86ms step:4193/5100 train_loss:3.3689 train_time:643601ms step_avg:153.86ms step:4194/5100 train_loss:3.3891 train_time:643754ms step_avg:153.86ms step:4195/5100 train_loss:3.3620 train_time:643907ms step_avg:153.86ms step:4196/5100 train_loss:3.3149 train_time:644059ms step_avg:153.86ms step:4197/5100 train_loss:3.6602 train_time:644211ms step_avg:153.86ms step:4198/5100 train_loss:3.1069 train_time:644364ms step_avg:153.86ms step:4199/5100 train_loss:3.5105 train_time:644516ms step_avg:153.86ms step:4200/5100 train_loss:3.3826 train_time:644670ms step_avg:153.86ms step:4201/5100 train_loss:3.2694 train_time:644823ms step_avg:153.86ms step:4202/5100 train_loss:3.4155 train_time:644977ms step_avg:153.86ms step:4203/5100 train_loss:3.2743 train_time:645130ms step_avg:153.86ms step:4204/5100 train_loss:3.2880 train_time:645282ms step_avg:153.86ms step:4205/5100 train_loss:3.2958 train_time:645435ms step_avg:153.86ms step:4206/5100 train_loss:3.2984 train_time:645588ms step_avg:153.86ms step:4207/5100 train_loss:3.7497 train_time:645739ms step_avg:153.86ms step:4208/5100 train_loss:3.3277 train_time:645891ms step_avg:153.86ms step:4209/5100 train_loss:3.4576 train_time:646044ms step_avg:153.86ms step:4210/5100 train_loss:3.3494 train_time:646197ms step_avg:153.86ms step:4211/5100 train_loss:3.7435 train_time:646350ms step_avg:153.86ms step:4212/5100 train_loss:3.4013 train_time:646503ms step_avg:153.86ms step:4213/5100 train_loss:3.4001 train_time:646654ms step_avg:153.86ms step:4214/5100 train_loss:3.2746 train_time:646807ms step_avg:153.86ms step:4215/5100 train_loss:3.3307 train_time:646960ms step_avg:153.85ms step:4216/5100 train_loss:3.4120 train_time:647112ms step_avg:153.85ms step:4217/5100 train_loss:3.2674 train_time:647267ms step_avg:153.85ms step:4218/5100 train_loss:3.3350 train_time:647420ms step_avg:153.85ms step:4219/5100 train_loss:3.3831 train_time:647572ms step_avg:153.85ms step:4220/5100 train_loss:3.1955 train_time:647725ms step_avg:153.85ms step:4221/5100 train_loss:3.3590 train_time:647877ms step_avg:153.85ms step:4222/5100 train_loss:3.3904 train_time:648031ms step_avg:153.85ms step:4223/5100 train_loss:3.3563 train_time:648184ms step_avg:153.85ms step:4224/5100 train_loss:3.5627 train_time:648336ms step_avg:153.85ms step:4225/5100 train_loss:3.4411 train_time:648490ms step_avg:153.85ms step:4226/5100 train_loss:3.4882 train_time:648641ms step_avg:153.85ms step:4227/5100 train_loss:3.2680 train_time:648793ms step_avg:153.85ms step:4228/5100 train_loss:3.3410 train_time:648946ms step_avg:153.85ms step:4229/5100 train_loss:3.3747 train_time:649099ms step_avg:153.85ms step:4230/5100 train_loss:3.2865 train_time:649251ms step_avg:153.85ms step:4231/5100 train_loss:3.4770 train_time:649405ms step_avg:153.85ms step:4232/5100 train_loss:3.4818 train_time:649557ms step_avg:153.85ms step:4233/5100 train_loss:3.4712 train_time:649710ms step_avg:153.85ms step:4234/5100 train_loss:3.5774 train_time:649863ms step_avg:153.85ms step:4235/5100 train_loss:3.4303 train_time:650015ms step_avg:153.85ms step:4236/5100 train_loss:3.3785 train_time:650170ms step_avg:153.85ms step:4237/5100 train_loss:3.2298 train_time:650322ms step_avg:153.85ms step:4238/5100 train_loss:3.4397 train_time:650475ms step_avg:153.85ms step:4239/5100 train_loss:3.3489 train_time:650628ms step_avg:153.85ms step:4240/5100 train_loss:3.2662 train_time:650780ms step_avg:153.85ms step:4241/5100 train_loss:3.3188 train_time:650933ms step_avg:153.85ms step:4242/5100 train_loss:3.2622 train_time:651086ms step_avg:153.85ms step:4243/5100 train_loss:3.3387 train_time:651238ms step_avg:153.85ms step:4244/5100 train_loss:3.2624 train_time:651392ms step_avg:153.85ms step:4245/5100 train_loss:3.1679 train_time:651545ms step_avg:153.85ms step:4246/5100 train_loss:3.4894 train_time:651698ms step_avg:153.85ms step:4247/5100 train_loss:3.2762 train_time:651850ms step_avg:153.85ms step:4248/5100 train_loss:3.2099 train_time:652004ms step_avg:153.85ms step:4249/5100 train_loss:3.4238 train_time:652155ms step_avg:153.85ms step:4250/5100 train_loss:3.7105 train_time:652308ms step_avg:153.85ms step:4250/5100 val_loss:3.3466 train_time:652372ms step_avg:153.86ms step:4251/5100 train_loss:3.3329 train_time:652466ms step_avg:153.85ms step:4252/5100 train_loss:3.5780 train_time:652620ms step_avg:153.85ms step:4253/5100 train_loss:3.4175 train_time:652772ms step_avg:153.85ms step:4254/5100 train_loss:3.2314 train_time:652925ms step_avg:153.85ms step:4255/5100 train_loss:3.3104 train_time:653076ms step_avg:153.85ms step:4256/5100 train_loss:3.2364 train_time:653229ms step_avg:153.85ms step:4257/5100 train_loss:3.4743 train_time:653381ms step_avg:153.85ms step:4258/5100 train_loss:3.3578 train_time:653536ms step_avg:153.85ms step:4259/5100 train_loss:3.4094 train_time:653690ms step_avg:153.85ms step:4260/5100 train_loss:3.2289 train_time:653844ms step_avg:153.85ms step:4261/5100 train_loss:3.5340 train_time:653996ms step_avg:153.85ms step:4262/5100 train_loss:3.3579 train_time:654149ms step_avg:153.84ms step:4263/5100 train_loss:3.3655 train_time:654302ms step_avg:153.84ms step:4264/5100 train_loss:3.4167 train_time:654454ms step_avg:153.84ms step:4265/5100 train_loss:3.3597 train_time:654606ms step_avg:153.84ms step:4266/5100 train_loss:3.3693 train_time:654758ms step_avg:153.84ms step:4267/5100 train_loss:3.4921 train_time:654911ms step_avg:153.84ms step:4268/5100 train_loss:3.3153 train_time:655064ms step_avg:153.84ms step:4269/5100 train_loss:3.8534 train_time:655216ms step_avg:153.84ms step:4270/5100 train_loss:3.2975 train_time:655369ms step_avg:153.84ms step:4271/5100 train_loss:3.3943 train_time:655521ms step_avg:153.84ms step:4272/5100 train_loss:3.3254 train_time:655674ms step_avg:153.84ms step:4273/5100 train_loss:3.5297 train_time:655826ms step_avg:153.84ms step:4274/5100 train_loss:3.4505 train_time:655979ms step_avg:153.84ms step:4275/5100 train_loss:3.3066 train_time:656132ms step_avg:153.84ms step:4276/5100 train_loss:3.3625 train_time:656284ms step_avg:153.84ms step:4277/5100 train_loss:3.2854 train_time:656437ms step_avg:153.84ms step:4278/5100 train_loss:3.3208 train_time:656589ms step_avg:153.84ms step:4279/5100 train_loss:3.3288 train_time:656743ms step_avg:153.84ms step:4280/5100 train_loss:3.3885 train_time:656895ms step_avg:153.84ms step:4281/5100 train_loss:3.3803 train_time:657047ms step_avg:153.84ms step:4282/5100 train_loss:3.3961 train_time:657200ms step_avg:153.84ms step:4283/5100 train_loss:3.3210 train_time:657351ms step_avg:153.84ms step:4284/5100 train_loss:3.3663 train_time:657503ms step_avg:153.84ms step:4285/5100 train_loss:3.4407 train_time:657656ms step_avg:153.84ms step:4286/5100 train_loss:3.3820 train_time:657809ms step_avg:153.84ms step:4287/5100 train_loss:3.2712 train_time:657962ms step_avg:153.84ms step:4288/5100 train_loss:3.3103 train_time:658114ms step_avg:153.84ms step:4289/5100 train_loss:3.4022 train_time:658266ms step_avg:153.84ms step:4290/5100 train_loss:3.3658 train_time:658418ms step_avg:153.84ms step:4291/5100 train_loss:3.2640 train_time:658571ms step_avg:153.84ms step:4292/5100 train_loss:3.2969 train_time:658723ms step_avg:153.84ms step:4293/5100 train_loss:3.3628 train_time:658875ms step_avg:153.83ms step:4294/5100 train_loss:3.1426 train_time:659028ms step_avg:153.83ms step:4295/5100 train_loss:3.4983 train_time:659181ms step_avg:153.83ms step:4296/5100 train_loss:3.3913 train_time:659334ms step_avg:153.83ms step:4297/5100 train_loss:3.3431 train_time:659487ms step_avg:153.83ms step:4298/5100 train_loss:3.5035 train_time:659639ms step_avg:153.83ms step:4299/5100 train_loss:3.4304 train_time:659792ms step_avg:153.83ms step:4300/5100 train_loss:3.2591 train_time:659945ms step_avg:153.83ms step:4301/5100 train_loss:3.2526 train_time:660098ms step_avg:153.83ms step:4302/5100 train_loss:3.4002 train_time:660249ms step_avg:153.83ms step:4303/5100 train_loss:3.2399 train_time:660403ms step_avg:153.83ms step:4304/5100 train_loss:3.3903 train_time:660555ms step_avg:153.83ms step:4305/5100 train_loss:3.4681 train_time:660707ms step_avg:153.83ms step:4306/5100 train_loss:3.2253 train_time:660862ms step_avg:153.83ms step:4307/5100 train_loss:3.7463 train_time:661013ms step_avg:153.83ms step:4308/5100 train_loss:3.3559 train_time:661166ms step_avg:153.83ms step:4309/5100 train_loss:3.2790 train_time:661318ms step_avg:153.83ms step:4310/5100 train_loss:3.2964 train_time:661470ms step_avg:153.83ms step:4311/5100 train_loss:3.5860 train_time:661623ms step_avg:153.83ms step:4312/5100 train_loss:3.4242 train_time:661775ms step_avg:153.83ms step:4313/5100 train_loss:3.2796 train_time:661928ms step_avg:153.83ms step:4314/5100 train_loss:3.4786 train_time:662080ms step_avg:153.83ms step:4315/5100 train_loss:3.4157 train_time:662233ms step_avg:153.83ms step:4316/5100 train_loss:3.3239 train_time:662385ms step_avg:153.83ms step:4317/5100 train_loss:3.3691 train_time:662539ms step_avg:153.83ms step:4318/5100 train_loss:3.3231 train_time:662692ms step_avg:153.83ms step:4319/5100 train_loss:3.4402 train_time:662844ms step_avg:153.83ms step:4320/5100 train_loss:3.4863 train_time:662996ms step_avg:153.83ms step:4321/5100 train_loss:3.3074 train_time:663148ms step_avg:153.83ms step:4322/5100 train_loss:3.4834 train_time:663301ms step_avg:153.83ms step:4323/5100 train_loss:3.3412 train_time:663453ms step_avg:153.83ms step:4324/5100 train_loss:3.2635 train_time:663605ms step_avg:153.83ms step:4325/5100 train_loss:3.2078 train_time:663758ms step_avg:153.83ms step:4326/5100 train_loss:3.3037 train_time:663910ms step_avg:153.83ms step:4327/5100 train_loss:3.1894 train_time:664063ms step_avg:153.83ms step:4328/5100 train_loss:3.3103 train_time:664215ms step_avg:153.82ms step:4329/5100 train_loss:3.3413 train_time:664368ms step_avg:153.82ms step:4330/5100 train_loss:3.2839 train_time:664521ms step_avg:153.82ms step:4331/5100 train_loss:3.5303 train_time:664674ms step_avg:153.82ms step:4332/5100 train_loss:3.3345 train_time:664827ms step_avg:153.82ms step:4333/5100 train_loss:3.4501 train_time:664981ms step_avg:153.82ms step:4334/5100 train_loss:3.8154 train_time:665133ms step_avg:153.82ms step:4335/5100 train_loss:3.3455 train_time:665285ms step_avg:153.82ms step:4336/5100 train_loss:3.4533 train_time:665437ms step_avg:153.82ms step:4337/5100 train_loss:3.3445 train_time:665589ms step_avg:153.82ms step:4338/5100 train_loss:3.2353 train_time:665741ms step_avg:153.82ms step:4339/5100 train_loss:3.3755 train_time:665895ms step_avg:153.82ms step:4340/5100 train_loss:3.2668 train_time:666046ms step_avg:153.82ms step:4341/5100 train_loss:3.3663 train_time:666199ms step_avg:153.82ms step:4342/5100 train_loss:3.3870 train_time:666350ms step_avg:153.82ms step:4343/5100 train_loss:3.3828 train_time:666503ms step_avg:153.82ms step:4344/5100 train_loss:3.3798 train_time:666656ms step_avg:153.82ms step:4345/5100 train_loss:4.0105 train_time:666808ms step_avg:153.82ms step:4346/5100 train_loss:3.4475 train_time:666962ms step_avg:153.82ms step:4347/5100 train_loss:3.2436 train_time:667217ms step_avg:153.84ms step:4348/5100 train_loss:3.3875 train_time:667380ms step_avg:153.85ms step:4349/5100 train_loss:3.3277 train_time:667531ms step_avg:153.84ms step:4350/5100 train_loss:3.2474 train_time:667683ms step_avg:153.84ms step:4351/5100 train_loss:3.4112 train_time:667834ms step_avg:153.84ms step:4352/5100 train_loss:3.3600 train_time:667986ms step_avg:153.84ms step:4353/5100 train_loss:3.4342 train_time:668138ms step_avg:153.84ms step:4354/5100 train_loss:3.4736 train_time:668291ms step_avg:153.84ms step:4355/5100 train_loss:3.2911 train_time:668445ms step_avg:153.84ms step:4356/5100 train_loss:3.2435 train_time:668598ms step_avg:153.84ms step:4357/5100 train_loss:3.3808 train_time:668750ms step_avg:153.84ms step:4358/5100 train_loss:3.3207 train_time:668903ms step_avg:153.84ms step:4359/5100 train_loss:3.5082 train_time:669054ms step_avg:153.84ms step:4360/5100 train_loss:3.3745 train_time:669206ms step_avg:153.84ms step:4361/5100 train_loss:3.4551 train_time:669359ms step_avg:153.84ms step:4362/5100 train_loss:3.6123 train_time:669511ms step_avg:153.84ms step:4363/5100 train_loss:3.4052 train_time:669664ms step_avg:153.84ms step:4364/5100 train_loss:3.3703 train_time:669815ms step_avg:153.84ms step:4365/5100 train_loss:3.5770 train_time:669968ms step_avg:153.84ms step:4366/5100 train_loss:3.4677 train_time:670122ms step_avg:153.84ms step:4367/5100 train_loss:3.2897 train_time:670273ms step_avg:153.84ms step:4368/5100 train_loss:3.3084 train_time:670426ms step_avg:153.84ms step:4369/5100 train_loss:3.4034 train_time:670578ms step_avg:153.84ms step:4370/5100 train_loss:3.4018 train_time:670854ms step_avg:153.87ms step:4371/5100 train_loss:3.5490 train_time:671004ms step_avg:153.86ms step:4372/5100 train_loss:3.2450 train_time:671157ms step_avg:153.86ms step:4373/5100 train_loss:3.2384 train_time:671311ms step_avg:153.86ms step:4374/5100 train_loss:3.3965 train_time:671462ms step_avg:153.86ms step:4375/5100 train_loss:3.4143 train_time:671613ms step_avg:153.86ms step:4375/5100 val_loss:3.3323 train_time:671677ms step_avg:153.88ms step:4376/5100 train_loss:3.4816 train_time:671771ms step_avg:153.86ms step:4377/5100 train_loss:3.2729 train_time:671929ms step_avg:153.87ms step:4378/5100 train_loss:3.3193 train_time:672081ms step_avg:153.86ms step:4379/5100 train_loss:3.3346 train_time:672233ms step_avg:153.86ms step:4380/5100 train_loss:3.3798 train_time:672386ms step_avg:153.86ms step:4381/5100 train_loss:3.2757 train_time:672537ms step_avg:153.86ms step:4382/5100 train_loss:3.5151 train_time:672689ms step_avg:153.86ms step:4383/5100 train_loss:3.4339 train_time:672844ms step_avg:153.86ms step:4384/5100 train_loss:3.4239 train_time:672998ms step_avg:153.86ms step:4385/5100 train_loss:3.3071 train_time:673150ms step_avg:153.86ms step:4386/5100 train_loss:3.4326 train_time:673302ms step_avg:153.86ms step:4387/5100 train_loss:3.3249 train_time:673454ms step_avg:153.86ms step:4388/5100 train_loss:3.4514 train_time:673606ms step_avg:153.86ms step:4389/5100 train_loss:3.2790 train_time:673758ms step_avg:153.86ms step:4390/5100 train_loss:3.4005 train_time:673910ms step_avg:153.86ms step:4391/5100 train_loss:3.4201 train_time:674063ms step_avg:153.86ms step:4392/5100 train_loss:3.2393 train_time:674216ms step_avg:153.86ms step:4393/5100 train_loss:3.9248 train_time:674369ms step_avg:153.86ms step:4394/5100 train_loss:3.3124 train_time:674521ms step_avg:153.86ms step:4395/5100 train_loss:3.4974 train_time:674674ms step_avg:153.86ms step:4396/5100 train_loss:3.2773 train_time:674827ms step_avg:153.86ms step:4397/5100 train_loss:3.4065 train_time:674980ms step_avg:153.86ms step:4398/5100 train_loss:3.2036 train_time:675132ms step_avg:153.86ms step:4399/5100 train_loss:3.4230 train_time:675284ms step_avg:153.86ms step:4400/5100 train_loss:3.2479 train_time:675437ms step_avg:153.86ms step:4401/5100 train_loss:3.3190 train_time:675589ms step_avg:153.86ms step:4402/5100 train_loss:3.3964 train_time:675742ms step_avg:153.86ms step:4403/5100 train_loss:3.2254 train_time:675896ms step_avg:153.86ms step:4404/5100 train_loss:3.2821 train_time:676049ms step_avg:153.86ms step:4405/5100 train_loss:3.4746 train_time:676202ms step_avg:153.86ms step:4406/5100 train_loss:3.3375 train_time:676354ms step_avg:153.86ms step:4407/5100 train_loss:3.3532 train_time:676506ms step_avg:153.86ms step:4408/5100 train_loss:3.3140 train_time:676659ms step_avg:153.86ms step:4409/5100 train_loss:3.3972 train_time:676811ms step_avg:153.86ms step:4410/5100 train_loss:3.3840 train_time:676964ms step_avg:153.86ms step:4411/5100 train_loss:3.5029 train_time:677116ms step_avg:153.86ms step:4412/5100 train_loss:3.3452 train_time:677269ms step_avg:153.85ms step:4413/5100 train_loss:3.3741 train_time:677423ms step_avg:153.85ms step:4414/5100 train_loss:3.3534 train_time:677575ms step_avg:153.85ms step:4415/5100 train_loss:3.4101 train_time:677727ms step_avg:153.85ms step:4416/5100 train_loss:3.3511 train_time:677879ms step_avg:153.85ms step:4417/5100 train_loss:3.4220 train_time:678032ms step_avg:153.85ms step:4418/5100 train_loss:3.3482 train_time:678184ms step_avg:153.85ms step:4419/5100 train_loss:3.2551 train_time:678337ms step_avg:153.85ms step:4420/5100 train_loss:3.3103 train_time:678489ms step_avg:153.85ms step:4421/5100 train_loss:3.5424 train_time:678641ms step_avg:153.85ms step:4422/5100 train_loss:3.3392 train_time:678794ms step_avg:153.85ms step:4423/5100 train_loss:3.2586 train_time:678946ms step_avg:153.85ms step:4424/5100 train_loss:3.2844 train_time:679099ms step_avg:153.85ms step:4425/5100 train_loss:3.4629 train_time:679250ms step_avg:153.85ms step:4426/5100 train_loss:3.4141 train_time:679404ms step_avg:153.85ms step:4427/5100 train_loss:3.3246 train_time:679557ms step_avg:153.85ms step:4428/5100 train_loss:3.5335 train_time:679709ms step_avg:153.85ms step:4429/5100 train_loss:3.4422 train_time:679863ms step_avg:153.85ms step:4430/5100 train_loss:3.2458 train_time:680015ms step_avg:153.85ms step:4431/5100 train_loss:3.2381 train_time:680168ms step_avg:153.85ms step:4432/5100 train_loss:3.3765 train_time:680320ms step_avg:153.85ms step:4433/5100 train_loss:3.2556 train_time:680473ms step_avg:153.85ms step:4434/5100 train_loss:3.3763 train_time:680625ms step_avg:153.85ms step:4435/5100 train_loss:3.4330 train_time:680778ms step_avg:153.85ms step:4436/5100 train_loss:3.3258 train_time:680930ms step_avg:153.85ms step:4437/5100 train_loss:3.2425 train_time:681083ms step_avg:153.85ms step:4438/5100 train_loss:3.4892 train_time:681235ms step_avg:153.85ms step:4439/5100 train_loss:3.4132 train_time:681389ms step_avg:153.85ms step:4440/5100 train_loss:3.3108 train_time:681542ms step_avg:153.85ms step:4441/5100 train_loss:3.4184 train_time:681694ms step_avg:153.85ms step:4442/5100 train_loss:3.4438 train_time:681846ms step_avg:153.85ms step:4443/5100 train_loss:3.4958 train_time:682000ms step_avg:153.85ms step:4444/5100 train_loss:3.3874 train_time:682151ms step_avg:153.85ms step:4445/5100 train_loss:3.2022 train_time:682304ms step_avg:153.85ms step:4446/5100 train_loss:3.4792 train_time:682458ms step_avg:153.85ms step:4447/5100 train_loss:3.3593 train_time:682610ms step_avg:153.84ms step:4448/5100 train_loss:3.2514 train_time:682763ms step_avg:153.84ms step:4449/5100 train_loss:3.3819 train_time:682915ms step_avg:153.84ms step:4450/5100 train_loss:3.3668 train_time:683067ms step_avg:153.84ms step:4451/5100 train_loss:3.3900 train_time:683220ms step_avg:153.84ms step:4452/5100 train_loss:3.4157 train_time:683374ms step_avg:153.84ms step:4453/5100 train_loss:3.2837 train_time:683527ms step_avg:153.84ms step:4454/5100 train_loss:3.3144 train_time:683679ms step_avg:153.84ms step:4455/5100 train_loss:3.3118 train_time:683831ms step_avg:153.84ms step:4456/5100 train_loss:3.2066 train_time:683984ms step_avg:153.84ms step:4457/5100 train_loss:3.4243 train_time:684138ms step_avg:153.84ms step:4458/5100 train_loss:3.2862 train_time:684290ms step_avg:153.84ms step:4459/5100 train_loss:3.2509 train_time:684442ms step_avg:153.84ms step:4460/5100 train_loss:3.3778 train_time:684595ms step_avg:153.84ms step:4461/5100 train_loss:3.8720 train_time:684748ms step_avg:153.84ms step:4462/5100 train_loss:3.3639 train_time:684900ms step_avg:153.84ms step:4463/5100 train_loss:3.4791 train_time:685052ms step_avg:153.84ms step:4464/5100 train_loss:3.3880 train_time:685204ms step_avg:153.84ms step:4465/5100 train_loss:3.3644 train_time:685357ms step_avg:153.84ms step:4466/5100 train_loss:3.4343 train_time:685509ms step_avg:153.84ms step:4467/5100 train_loss:3.2415 train_time:685663ms step_avg:153.84ms step:4468/5100 train_loss:3.2878 train_time:685814ms step_avg:153.84ms step:4469/5100 train_loss:3.4151 train_time:685967ms step_avg:153.84ms step:4470/5100 train_loss:3.3977 train_time:686120ms step_avg:153.84ms step:4471/5100 train_loss:3.3170 train_time:686273ms step_avg:153.84ms step:4472/5100 train_loss:3.2846 train_time:686425ms step_avg:153.84ms step:4473/5100 train_loss:3.3571 train_time:686577ms step_avg:153.84ms step:4474/5100 train_loss:3.2104 train_time:686730ms step_avg:153.84ms step:4475/5100 train_loss:3.2864 train_time:686883ms step_avg:153.84ms step:4476/5100 train_loss:3.3091 train_time:687035ms step_avg:153.84ms step:4477/5100 train_loss:3.4771 train_time:687189ms step_avg:153.84ms step:4478/5100 train_loss:3.2112 train_time:687341ms step_avg:153.84ms step:4479/5100 train_loss:3.3360 train_time:687494ms step_avg:153.84ms step:4480/5100 train_loss:3.3725 train_time:687648ms step_avg:153.84ms step:4481/5100 train_loss:3.3454 train_time:687799ms step_avg:153.84ms step:4482/5100 train_loss:3.3385 train_time:687951ms step_avg:153.84ms step:4483/5100 train_loss:3.1506 train_time:688104ms step_avg:153.83ms step:4484/5100 train_loss:3.3079 train_time:688257ms step_avg:153.83ms step:4485/5100 train_loss:3.2515 train_time:688408ms step_avg:153.83ms step:4486/5100 train_loss:3.3810 train_time:688561ms step_avg:153.83ms step:4487/5100 train_loss:3.2618 train_time:688714ms step_avg:153.83ms step:4488/5100 train_loss:3.3330 train_time:688866ms step_avg:153.83ms step:4489/5100 train_loss:3.4711 train_time:689019ms step_avg:153.83ms step:4490/5100 train_loss:3.4394 train_time:689172ms step_avg:153.83ms step:4491/5100 train_loss:3.3125 train_time:689325ms step_avg:153.83ms step:4492/5100 train_loss:3.2731 train_time:689477ms step_avg:153.83ms step:4493/5100 train_loss:3.3243 train_time:689629ms step_avg:153.83ms step:4494/5100 train_loss:3.3532 train_time:689782ms step_avg:153.83ms step:4495/5100 train_loss:3.3516 train_time:689934ms step_avg:153.83ms step:4496/5100 train_loss:3.2927 train_time:690085ms step_avg:153.83ms step:4497/5100 train_loss:3.4533 train_time:690237ms step_avg:153.83ms step:4498/5100 train_loss:3.3264 train_time:690390ms step_avg:153.83ms step:4499/5100 train_loss:3.1726 train_time:690543ms step_avg:153.83ms step:4500/5100 train_loss:3.4753 train_time:690696ms step_avg:153.83ms step:4500/5100 val_loss:3.3205 train_time:690759ms step_avg:153.84ms step:4501/5100 train_loss:3.2713 train_time:690855ms step_avg:153.83ms step:4502/5100 train_loss:3.2354 train_time:691008ms step_avg:153.83ms step:4503/5100 train_loss:3.4261 train_time:691161ms step_avg:153.83ms step:4504/5100 train_loss:3.3059 train_time:691313ms step_avg:153.83ms step:4505/5100 train_loss:3.3997 train_time:691465ms step_avg:153.83ms step:4506/5100 train_loss:3.3175 train_time:691618ms step_avg:153.83ms step:4507/5100 train_loss:3.3944 train_time:691769ms step_avg:153.83ms step:4508/5100 train_loss:3.1209 train_time:691922ms step_avg:153.83ms step:4509/5100 train_loss:3.3995 train_time:692075ms step_avg:153.83ms step:4510/5100 train_loss:3.2333 train_time:692227ms step_avg:153.83ms step:4511/5100 train_loss:3.3070 train_time:692380ms step_avg:153.83ms step:4512/5100 train_loss:3.2551 train_time:692532ms step_avg:153.83ms step:4513/5100 train_loss:3.2523 train_time:692685ms step_avg:153.83ms step:4514/5100 train_loss:3.2080 train_time:692838ms step_avg:153.83ms step:4515/5100 train_loss:3.3527 train_time:692991ms step_avg:153.83ms step:4516/5100 train_loss:3.2053 train_time:693143ms step_avg:153.83ms step:4517/5100 train_loss:3.3084 train_time:693296ms step_avg:153.83ms step:4518/5100 train_loss:3.3207 train_time:693448ms step_avg:153.83ms step:4519/5100 train_loss:3.3273 train_time:693600ms step_avg:153.83ms step:4520/5100 train_loss:3.2447 train_time:693752ms step_avg:153.83ms step:4521/5100 train_loss:3.4314 train_time:693905ms step_avg:153.83ms step:4522/5100 train_loss:3.4987 train_time:694058ms step_avg:153.82ms step:4523/5100 train_loss:3.8446 train_time:694210ms step_avg:153.82ms step:4524/5100 train_loss:3.5794 train_time:694363ms step_avg:153.82ms step:4525/5100 train_loss:3.3381 train_time:694516ms step_avg:153.82ms step:4526/5100 train_loss:3.2907 train_time:694669ms step_avg:153.82ms step:4527/5100 train_loss:3.3592 train_time:694821ms step_avg:153.82ms step:4528/5100 train_loss:3.3239 train_time:694974ms step_avg:153.82ms step:4529/5100 train_loss:3.2238 train_time:695126ms step_avg:153.82ms step:4530/5100 train_loss:3.9377 train_time:695279ms step_avg:153.82ms step:4531/5100 train_loss:3.4018 train_time:695432ms step_avg:153.82ms step:4532/5100 train_loss:3.1298 train_time:695584ms step_avg:153.82ms step:4533/5100 train_loss:3.2405 train_time:695736ms step_avg:153.82ms step:4534/5100 train_loss:3.3555 train_time:695889ms step_avg:153.82ms step:4535/5100 train_loss:3.5703 train_time:696041ms step_avg:153.82ms step:4536/5100 train_loss:3.5680 train_time:696295ms step_avg:153.84ms step:4537/5100 train_loss:3.3067 train_time:696455ms step_avg:153.84ms step:4538/5100 train_loss:3.2986 train_time:696606ms step_avg:153.84ms step:4539/5100 train_loss:3.3357 train_time:696760ms step_avg:153.84ms step:4540/5100 train_loss:3.9065 train_time:696912ms step_avg:153.84ms step:4541/5100 train_loss:3.3863 train_time:697064ms step_avg:153.84ms step:4542/5100 train_loss:3.2981 train_time:697216ms step_avg:153.84ms step:4543/5100 train_loss:3.4602 train_time:697370ms step_avg:153.84ms step:4544/5100 train_loss:3.2428 train_time:697523ms step_avg:153.84ms step:4545/5100 train_loss:3.3550 train_time:697675ms step_avg:153.84ms step:4546/5100 train_loss:3.5320 train_time:697827ms step_avg:153.84ms step:4547/5100 train_loss:3.3886 train_time:697980ms step_avg:153.84ms step:4548/5100 train_loss:3.3391 train_time:698132ms step_avg:153.84ms step:4549/5100 train_loss:3.3320 train_time:698284ms step_avg:153.84ms step:4550/5100 train_loss:3.2765 train_time:698438ms step_avg:153.84ms step:4551/5100 train_loss:3.2555 train_time:698590ms step_avg:153.84ms step:4552/5100 train_loss:3.2302 train_time:698742ms step_avg:153.84ms step:4553/5100 train_loss:3.3472 train_time:698896ms step_avg:153.84ms step:4554/5100 train_loss:3.5353 train_time:699048ms step_avg:153.84ms step:4555/5100 train_loss:3.4181 train_time:699200ms step_avg:153.84ms step:4556/5100 train_loss:3.1726 train_time:699353ms step_avg:153.84ms step:4557/5100 train_loss:3.3759 train_time:699505ms step_avg:153.84ms step:4558/5100 train_loss:3.3848 train_time:699659ms step_avg:153.84ms step:4559/5100 train_loss:3.3735 train_time:699811ms step_avg:153.84ms step:4560/5100 train_loss:3.4758 train_time:700090ms step_avg:153.87ms step:4561/5100 train_loss:3.2969 train_time:700239ms step_avg:153.86ms step:4562/5100 train_loss:3.3110 train_time:700390ms step_avg:153.86ms step:4563/5100 train_loss:3.3348 train_time:700541ms step_avg:153.86ms step:4564/5100 train_loss:3.3721 train_time:700693ms step_avg:153.86ms step:4565/5100 train_loss:3.4601 train_time:700844ms step_avg:153.86ms step:4566/5100 train_loss:3.5175 train_time:700997ms step_avg:153.86ms step:4567/5100 train_loss:3.3694 train_time:701149ms step_avg:153.86ms step:4568/5100 train_loss:3.2420 train_time:701303ms step_avg:153.86ms step:4569/5100 train_loss:3.3414 train_time:701455ms step_avg:153.86ms step:4570/5100 train_loss:3.2311 train_time:701607ms step_avg:153.86ms step:4571/5100 train_loss:3.2529 train_time:701760ms step_avg:153.86ms step:4572/5100 train_loss:3.4610 train_time:701912ms step_avg:153.86ms step:4573/5100 train_loss:3.1681 train_time:702065ms step_avg:153.86ms step:4574/5100 train_loss:3.2515 train_time:702217ms step_avg:153.86ms step:4575/5100 train_loss:3.3778 train_time:702369ms step_avg:153.86ms step:4576/5100 train_loss:3.4058 train_time:702522ms step_avg:153.86ms step:4577/5100 train_loss:3.3649 train_time:702674ms step_avg:153.86ms step:4578/5100 train_loss:3.3242 train_time:702827ms step_avg:153.86ms step:4579/5100 train_loss:3.3509 train_time:702979ms step_avg:153.86ms step:4580/5100 train_loss:3.4486 train_time:703132ms step_avg:153.86ms step:4581/5100 train_loss:3.2770 train_time:703286ms step_avg:153.86ms step:4582/5100 train_loss:3.3051 train_time:703439ms step_avg:153.86ms step:4583/5100 train_loss:3.4080 train_time:703591ms step_avg:153.86ms step:4584/5100 train_loss:3.2664 train_time:703743ms step_avg:153.86ms step:4585/5100 train_loss:3.3782 train_time:703896ms step_avg:153.86ms step:4586/5100 train_loss:3.3492 train_time:704047ms step_avg:153.86ms step:4587/5100 train_loss:3.3367 train_time:704200ms step_avg:153.86ms step:4588/5100 train_loss:3.1874 train_time:704352ms step_avg:153.86ms step:4589/5100 train_loss:3.3142 train_time:704504ms step_avg:153.86ms step:4590/5100 train_loss:3.5068 train_time:704657ms step_avg:153.86ms step:4591/5100 train_loss:3.3371 train_time:704810ms step_avg:153.85ms step:4592/5100 train_loss:3.3297 train_time:704962ms step_avg:153.85ms step:4593/5100 train_loss:3.2929 train_time:705114ms step_avg:153.85ms step:4594/5100 train_loss:3.4520 train_time:705267ms step_avg:153.85ms step:4595/5100 train_loss:3.3343 train_time:705419ms step_avg:153.85ms step:4596/5100 train_loss:3.2362 train_time:705572ms step_avg:153.85ms step:4597/5100 train_loss:3.2209 train_time:705726ms step_avg:153.85ms step:4598/5100 train_loss:3.4133 train_time:705878ms step_avg:153.85ms step:4599/5100 train_loss:3.3337 train_time:706031ms step_avg:153.85ms step:4600/5100 train_loss:3.4643 train_time:706183ms step_avg:153.85ms step:4601/5100 train_loss:3.3561 train_time:706336ms step_avg:153.85ms step:4602/5100 train_loss:3.1695 train_time:706489ms step_avg:153.85ms step:4603/5100 train_loss:3.2807 train_time:706641ms step_avg:153.85ms step:4604/5100 train_loss:3.3646 train_time:706794ms step_avg:153.85ms step:4605/5100 train_loss:3.3768 train_time:706947ms step_avg:153.85ms step:4606/5100 train_loss:3.2960 train_time:707099ms step_avg:153.85ms step:4607/5100 train_loss:3.4110 train_time:707252ms step_avg:153.85ms step:4608/5100 train_loss:3.2841 train_time:707403ms step_avg:153.85ms step:4609/5100 train_loss:3.3915 train_time:707556ms step_avg:153.85ms step:4610/5100 train_loss:3.3340 train_time:707709ms step_avg:153.85ms step:4611/5100 train_loss:3.3838 train_time:707862ms step_avg:153.85ms step:4612/5100 train_loss:3.5196 train_time:708015ms step_avg:153.85ms step:4613/5100 train_loss:3.2191 train_time:708168ms step_avg:153.85ms step:4614/5100 train_loss:3.0649 train_time:708320ms step_avg:153.85ms step:4615/5100 train_loss:3.2765 train_time:708473ms step_avg:153.85ms step:4616/5100 train_loss:3.1965 train_time:708625ms step_avg:153.85ms step:4617/5100 train_loss:3.3049 train_time:708778ms step_avg:153.85ms step:4618/5100 train_loss:3.1761 train_time:708930ms step_avg:153.85ms step:4619/5100 train_loss:3.3871 train_time:709083ms step_avg:153.85ms step:4620/5100 train_loss:3.4348 train_time:709235ms step_avg:153.85ms step:4621/5100 train_loss:3.4753 train_time:709389ms step_avg:153.85ms step:4622/5100 train_loss:3.2620 train_time:709541ms step_avg:153.85ms step:4623/5100 train_loss:3.2655 train_time:709694ms step_avg:153.85ms step:4624/5100 train_loss:3.3012 train_time:709846ms step_avg:153.85ms step:4625/5100 train_loss:3.2049 train_time:709999ms step_avg:153.85ms step:4625/5100 val_loss:3.3086 train_time:710062ms step_avg:153.86ms step:4626/5100 train_loss:3.3855 train_time:710157ms step_avg:153.85ms step:4627/5100 train_loss:3.2700 train_time:710316ms step_avg:153.85ms step:4628/5100 train_loss:3.3278 train_time:710469ms step_avg:153.85ms step:4629/5100 train_loss:3.5264 train_time:710621ms step_avg:153.85ms step:4630/5100 train_loss:3.3652 train_time:710773ms step_avg:153.85ms step:4631/5100 train_loss:3.4565 train_time:710925ms step_avg:153.85ms step:4632/5100 train_loss:3.2616 train_time:711077ms step_avg:153.85ms step:4633/5100 train_loss:3.4514 train_time:711231ms step_avg:153.85ms step:4634/5100 train_loss:3.3111 train_time:711385ms step_avg:153.85ms step:4635/5100 train_loss:3.3606 train_time:711538ms step_avg:153.85ms step:4636/5100 train_loss:3.3757 train_time:711690ms step_avg:153.85ms step:4637/5100 train_loss:3.2077 train_time:711842ms step_avg:153.85ms step:4638/5100 train_loss:3.3730 train_time:711993ms step_avg:153.84ms step:4639/5100 train_loss:3.3241 train_time:712146ms step_avg:153.84ms step:4640/5100 train_loss:3.3260 train_time:712300ms step_avg:153.84ms step:4641/5100 train_loss:3.2689 train_time:712452ms step_avg:153.84ms step:4642/5100 train_loss:3.2967 train_time:712605ms step_avg:153.84ms step:4643/5100 train_loss:3.3169 train_time:712756ms step_avg:153.84ms step:4644/5100 train_loss:3.5417 train_time:712908ms step_avg:153.84ms step:4645/5100 train_loss:3.4034 train_time:713061ms step_avg:153.84ms step:4646/5100 train_loss:3.4331 train_time:713213ms step_avg:153.84ms step:4647/5100 train_loss:3.2654 train_time:713366ms step_avg:153.84ms step:4648/5100 train_loss:3.3722 train_time:713517ms step_avg:153.84ms step:4649/5100 train_loss:3.3237 train_time:713670ms step_avg:153.84ms step:4650/5100 train_loss:3.3723 train_time:713823ms step_avg:153.84ms step:4651/5100 train_loss:3.4919 train_time:713975ms step_avg:153.84ms step:4652/5100 train_loss:3.3233 train_time:714127ms step_avg:153.84ms step:4653/5100 train_loss:3.4264 train_time:714279ms step_avg:153.84ms step:4654/5100 train_loss:3.2722 train_time:714432ms step_avg:153.84ms step:4655/5100 train_loss:3.3150 train_time:714585ms step_avg:153.84ms step:4656/5100 train_loss:3.3427 train_time:714738ms step_avg:153.84ms step:4657/5100 train_loss:3.2828 train_time:714891ms step_avg:153.84ms step:4658/5100 train_loss:3.2149 train_time:715043ms step_avg:153.84ms step:4659/5100 train_loss:3.2539 train_time:715196ms step_avg:153.84ms step:4660/5100 train_loss:3.1866 train_time:715349ms step_avg:153.84ms step:4661/5100 train_loss:3.3623 train_time:715501ms step_avg:153.84ms step:4662/5100 train_loss:3.3361 train_time:715653ms step_avg:153.84ms step:4663/5100 train_loss:3.2973 train_time:715805ms step_avg:153.84ms step:4664/5100 train_loss:3.2313 train_time:715958ms step_avg:153.84ms step:4665/5100 train_loss:3.2385 train_time:716110ms step_avg:153.84ms step:4666/5100 train_loss:3.2797 train_time:716264ms step_avg:153.84ms step:4667/5100 train_loss:3.3731 train_time:716417ms step_avg:153.84ms step:4668/5100 train_loss:3.2997 train_time:716569ms step_avg:153.84ms step:4669/5100 train_loss:3.2706 train_time:716723ms step_avg:153.84ms step:4670/5100 train_loss:3.3325 train_time:716874ms step_avg:153.84ms step:4671/5100 train_loss:3.4154 train_time:717027ms step_avg:153.84ms step:4672/5100 train_loss:3.3111 train_time:717179ms step_avg:153.84ms step:4673/5100 train_loss:3.3599 train_time:717331ms step_avg:153.83ms step:4674/5100 train_loss:3.2977 train_time:717484ms step_avg:153.83ms step:4675/5100 train_loss:3.3303 train_time:717636ms step_avg:153.83ms step:4676/5100 train_loss:3.4157 train_time:717788ms step_avg:153.83ms step:4677/5100 train_loss:3.1425 train_time:717940ms step_avg:153.83ms step:4678/5100 train_loss:3.1973 train_time:718093ms step_avg:153.83ms step:4679/5100 train_loss:3.3273 train_time:718245ms step_avg:153.83ms step:4680/5100 train_loss:3.2938 train_time:718397ms step_avg:153.83ms step:4681/5100 train_loss:3.3158 train_time:718550ms step_avg:153.83ms step:4682/5100 train_loss:3.3053 train_time:718702ms step_avg:153.83ms step:4683/5100 train_loss:3.2352 train_time:718855ms step_avg:153.83ms step:4684/5100 train_loss:3.2082 train_time:719006ms step_avg:153.83ms step:4685/5100 train_loss:3.4659 train_time:719159ms step_avg:153.83ms step:4686/5100 train_loss:3.5411 train_time:719312ms step_avg:153.83ms step:4687/5100 train_loss:3.2335 train_time:719465ms step_avg:153.83ms step:4688/5100 train_loss:3.2383 train_time:719617ms step_avg:153.83ms step:4689/5100 train_loss:3.4304 train_time:719769ms step_avg:153.83ms step:4690/5100 train_loss:3.2558 train_time:719923ms step_avg:153.83ms step:4691/5100 train_loss:3.1150 train_time:720074ms step_avg:153.83ms step:4692/5100 train_loss:3.2204 train_time:720226ms step_avg:153.83ms step:4693/5100 train_loss:3.2253 train_time:720379ms step_avg:153.83ms step:4694/5100 train_loss:3.2516 train_time:720531ms step_avg:153.83ms step:4695/5100 train_loss:3.2594 train_time:720684ms step_avg:153.83ms step:4696/5100 train_loss:3.2826 train_time:720837ms step_avg:153.83ms step:4697/5100 train_loss:3.3448 train_time:720989ms step_avg:153.83ms step:4698/5100 train_loss:3.2564 train_time:721142ms step_avg:153.83ms step:4699/5100 train_loss:3.2847 train_time:721295ms step_avg:153.83ms step:4700/5100 train_loss:3.3769 train_time:721447ms step_avg:153.83ms step:4701/5100 train_loss:3.3063 train_time:721599ms step_avg:153.83ms step:4702/5100 train_loss:3.2914 train_time:721752ms step_avg:153.83ms step:4703/5100 train_loss:3.2436 train_time:721905ms step_avg:153.83ms step:4704/5100 train_loss:3.3320 train_time:722057ms step_avg:153.83ms step:4705/5100 train_loss:3.2853 train_time:722210ms step_avg:153.83ms step:4706/5100 train_loss:3.2122 train_time:722363ms step_avg:153.83ms step:4707/5100 train_loss:3.3416 train_time:722515ms step_avg:153.82ms step:4708/5100 train_loss:3.4308 train_time:722668ms step_avg:153.82ms step:4709/5100 train_loss:3.2390 train_time:722821ms step_avg:153.82ms step:4710/5100 train_loss:3.2256 train_time:722974ms step_avg:153.82ms step:4711/5100 train_loss:3.2400 train_time:723126ms step_avg:153.82ms step:4712/5100 train_loss:3.2622 train_time:723278ms step_avg:153.82ms step:4713/5100 train_loss:3.3825 train_time:723431ms step_avg:153.82ms step:4714/5100 train_loss:3.2258 train_time:723583ms step_avg:153.82ms step:4715/5100 train_loss:3.2951 train_time:723736ms step_avg:153.82ms step:4716/5100 train_loss:3.2263 train_time:723889ms step_avg:153.82ms step:4717/5100 train_loss:3.2909 train_time:724042ms step_avg:153.82ms step:4718/5100 train_loss:3.2258 train_time:724195ms step_avg:153.82ms step:4719/5100 train_loss:3.1786 train_time:724347ms step_avg:153.82ms step:4720/5100 train_loss:3.3538 train_time:724499ms step_avg:153.82ms step:4721/5100 train_loss:3.3392 train_time:724651ms step_avg:153.82ms step:4722/5100 train_loss:3.3432 train_time:724804ms step_avg:153.82ms step:4723/5100 train_loss:3.1895 train_time:724957ms step_avg:153.82ms step:4724/5100 train_loss:3.3731 train_time:725108ms step_avg:153.82ms step:4725/5100 train_loss:3.2557 train_time:725363ms step_avg:153.84ms step:4726/5100 train_loss:3.5286 train_time:725524ms step_avg:153.84ms step:4727/5100 train_loss:3.3876 train_time:725675ms step_avg:153.84ms step:4728/5100 train_loss:3.2669 train_time:725827ms step_avg:153.84ms step:4729/5100 train_loss:3.1896 train_time:725979ms step_avg:153.84ms step:4730/5100 train_loss:3.1419 train_time:726131ms step_avg:153.84ms step:4731/5100 train_loss:3.2424 train_time:726283ms step_avg:153.84ms step:4732/5100 train_loss:3.3145 train_time:726439ms step_avg:153.84ms step:4733/5100 train_loss:3.2131 train_time:726592ms step_avg:153.84ms step:4734/5100 train_loss:3.0901 train_time:726744ms step_avg:153.84ms step:4735/5100 train_loss:3.3887 train_time:726896ms step_avg:153.84ms step:4736/5100 train_loss:3.2664 train_time:727048ms step_avg:153.84ms step:4737/5100 train_loss:3.4365 train_time:727201ms step_avg:153.84ms step:4738/5100 train_loss:3.3557 train_time:727353ms step_avg:153.84ms step:4739/5100 train_loss:3.3102 train_time:727506ms step_avg:153.84ms step:4740/5100 train_loss:3.2773 train_time:727660ms step_avg:153.84ms step:4741/5100 train_loss:3.2880 train_time:727811ms step_avg:153.84ms step:4742/5100 train_loss:3.2864 train_time:727964ms step_avg:153.84ms step:4743/5100 train_loss:3.1653 train_time:728116ms step_avg:153.84ms step:4744/5100 train_loss:3.3048 train_time:728269ms step_avg:153.84ms step:4745/5100 train_loss:3.2662 train_time:728423ms step_avg:153.84ms step:4746/5100 train_loss:3.2679 train_time:728574ms step_avg:153.84ms step:4747/5100 train_loss:3.2428 train_time:728726ms step_avg:153.84ms step:4748/5100 train_loss:3.4301 train_time:728878ms step_avg:153.84ms step:4749/5100 train_loss:3.2703 train_time:729031ms step_avg:153.84ms step:4750/5100 train_loss:3.3729 train_time:729307ms step_avg:153.86ms step:4750/5100 val_loss:3.2968 train_time:729369ms step_avg:153.88ms step:4751/5100 train_loss:3.1914 train_time:729465ms step_avg:153.86ms step:4752/5100 train_loss:3.1133 train_time:729623ms step_avg:153.86ms step:4753/5100 train_loss:3.1916 train_time:729775ms step_avg:153.86ms step:4754/5100 train_loss:3.3964 train_time:729927ms step_avg:153.86ms step:4755/5100 train_loss:3.2774 train_time:730078ms step_avg:153.86ms step:4756/5100 train_loss:3.5205 train_time:730233ms step_avg:153.86ms step:4757/5100 train_loss:3.3816 train_time:730384ms step_avg:153.86ms step:4758/5100 train_loss:3.2939 train_time:730537ms step_avg:153.86ms step:4759/5100 train_loss:3.3431 train_time:730691ms step_avg:153.86ms step:4760/5100 train_loss:3.3299 train_time:730843ms step_avg:153.86ms step:4761/5100 train_loss:3.2667 train_time:730995ms step_avg:153.86ms step:4762/5100 train_loss:3.3100 train_time:731148ms step_avg:153.86ms step:4763/5100 train_loss:3.2819 train_time:731300ms step_avg:153.86ms step:4764/5100 train_loss:3.1332 train_time:731451ms step_avg:153.86ms step:4765/5100 train_loss:3.1522 train_time:731605ms step_avg:153.86ms step:4766/5100 train_loss:3.1503 train_time:731758ms step_avg:153.86ms step:4767/5100 train_loss:3.3844 train_time:731910ms step_avg:153.86ms step:4768/5100 train_loss:3.6557 train_time:732063ms step_avg:153.86ms step:4769/5100 train_loss:3.3687 train_time:732215ms step_avg:153.86ms step:4770/5100 train_loss:3.2691 train_time:732368ms step_avg:153.86ms step:4771/5100 train_loss:3.3333 train_time:732520ms step_avg:153.86ms step:4772/5100 train_loss:3.2932 train_time:732673ms step_avg:153.86ms step:4773/5100 train_loss:3.2652 train_time:732826ms step_avg:153.86ms step:4774/5100 train_loss:3.4528 train_time:732978ms step_avg:153.86ms step:4775/5100 train_loss:3.2686 train_time:733132ms step_avg:153.86ms step:4776/5100 train_loss:3.4059 train_time:733284ms step_avg:153.86ms step:4777/5100 train_loss:3.3198 train_time:733437ms step_avg:153.86ms step:4778/5100 train_loss:3.1572 train_time:733591ms step_avg:153.86ms step:4779/5100 train_loss:3.3335 train_time:733744ms step_avg:153.86ms step:4780/5100 train_loss:3.2665 train_time:733896ms step_avg:153.86ms step:4781/5100 train_loss:3.3386 train_time:734049ms step_avg:153.86ms step:4782/5100 train_loss:3.2574 train_time:734201ms step_avg:153.86ms step:4783/5100 train_loss:3.2081 train_time:734353ms step_avg:153.86ms step:4784/5100 train_loss:3.2649 train_time:734506ms step_avg:153.86ms step:4785/5100 train_loss:3.1913 train_time:734660ms step_avg:153.86ms step:4786/5100 train_loss:3.5244 train_time:734811ms step_avg:153.85ms step:4787/5100 train_loss:3.4104 train_time:734964ms step_avg:153.85ms step:4788/5100 train_loss:3.3284 train_time:735116ms step_avg:153.85ms step:4789/5100 train_loss:3.3144 train_time:735269ms step_avg:153.85ms step:4790/5100 train_loss:3.2377 train_time:735422ms step_avg:153.85ms step:4791/5100 train_loss:3.3409 train_time:735574ms step_avg:153.85ms step:4792/5100 train_loss:3.3510 train_time:735727ms step_avg:153.85ms step:4793/5100 train_loss:3.2636 train_time:735880ms step_avg:153.85ms step:4794/5100 train_loss:3.3389 train_time:736032ms step_avg:153.85ms step:4795/5100 train_loss:3.1837 train_time:736185ms step_avg:153.85ms step:4796/5100 train_loss:3.3272 train_time:736338ms step_avg:153.85ms step:4797/5100 train_loss:3.4047 train_time:736489ms step_avg:153.85ms step:4798/5100 train_loss:3.0621 train_time:736641ms step_avg:153.85ms step:4799/5100 train_loss:3.2478 train_time:736794ms step_avg:153.85ms step:4800/5100 train_loss:3.2271 train_time:736947ms step_avg:153.85ms step:4801/5100 train_loss:3.3275 train_time:737100ms step_avg:153.85ms step:4802/5100 train_loss:3.1461 train_time:737252ms step_avg:153.85ms step:4803/5100 train_loss:3.1844 train_time:737406ms step_avg:153.85ms step:4804/5100 train_loss:3.3809 train_time:737557ms step_avg:153.85ms step:4805/5100 train_loss:3.3319 train_time:737710ms step_avg:153.85ms step:4806/5100 train_loss:3.3996 train_time:737864ms step_avg:153.85ms step:4807/5100 train_loss:3.4086 train_time:738015ms step_avg:153.85ms step:4808/5100 train_loss:3.1781 train_time:738169ms step_avg:153.85ms step:4809/5100 train_loss:3.2924 train_time:738321ms step_avg:153.85ms step:4810/5100 train_loss:3.2435 train_time:738473ms step_avg:153.85ms step:4811/5100 train_loss:3.4760 train_time:738626ms step_avg:153.85ms step:4812/5100 train_loss:3.2765 train_time:738778ms step_avg:153.85ms step:4813/5100 train_loss:3.3119 train_time:738930ms step_avg:153.85ms step:4814/5100 train_loss:3.2104 train_time:739082ms step_avg:153.85ms step:4815/5100 train_loss:3.2559 train_time:739235ms step_avg:153.85ms step:4816/5100 train_loss:3.6908 train_time:739387ms step_avg:153.85ms step:4817/5100 train_loss:3.3642 train_time:739541ms step_avg:153.85ms step:4818/5100 train_loss:3.3058 train_time:739693ms step_avg:153.85ms step:4819/5100 train_loss:3.1567 train_time:739847ms step_avg:153.85ms step:4820/5100 train_loss:3.2821 train_time:740000ms step_avg:153.85ms step:4821/5100 train_loss:3.2946 train_time:740152ms step_avg:153.85ms step:4822/5100 train_loss:3.3482 train_time:740304ms step_avg:153.85ms step:4823/5100 train_loss:3.3964 train_time:740456ms step_avg:153.85ms step:4824/5100 train_loss:3.2779 train_time:740608ms step_avg:153.84ms step:4825/5100 train_loss:3.2597 train_time:740762ms step_avg:153.84ms step:4826/5100 train_loss:3.1878 train_time:740914ms step_avg:153.84ms step:4827/5100 train_loss:3.1500 train_time:741067ms step_avg:153.84ms step:4828/5100 train_loss:3.3366 train_time:741220ms step_avg:153.84ms step:4829/5100 train_loss:3.2203 train_time:741372ms step_avg:153.84ms step:4830/5100 train_loss:3.3387 train_time:741526ms step_avg:153.84ms step:4831/5100 train_loss:3.5012 train_time:741678ms step_avg:153.84ms step:4832/5100 train_loss:3.2347 train_time:741830ms step_avg:153.84ms step:4833/5100 train_loss:3.3183 train_time:741982ms step_avg:153.84ms step:4834/5100 train_loss:3.2762 train_time:742135ms step_avg:153.84ms step:4835/5100 train_loss:3.4615 train_time:742288ms step_avg:153.84ms step:4836/5100 train_loss:3.2786 train_time:742441ms step_avg:153.84ms step:4837/5100 train_loss:3.5367 train_time:742593ms step_avg:153.84ms step:4838/5100 train_loss:3.4881 train_time:742747ms step_avg:153.84ms step:4839/5100 train_loss:3.3133 train_time:742900ms step_avg:153.84ms step:4840/5100 train_loss:3.3108 train_time:743052ms step_avg:153.84ms step:4841/5100 train_loss:3.2964 train_time:743206ms step_avg:153.84ms step:4842/5100 train_loss:3.3421 train_time:743358ms step_avg:153.84ms step:4843/5100 train_loss:3.3302 train_time:743511ms step_avg:153.84ms step:4844/5100 train_loss:3.1931 train_time:743665ms step_avg:153.84ms step:4845/5100 train_loss:3.2204 train_time:743817ms step_avg:153.84ms step:4846/5100 train_loss:3.1936 train_time:743969ms step_avg:153.84ms step:4847/5100 train_loss:3.3561 train_time:744123ms step_avg:153.84ms step:4848/5100 train_loss:3.2290 train_time:744276ms step_avg:153.84ms step:4849/5100 train_loss:3.2608 train_time:744429ms step_avg:153.84ms step:4850/5100 train_loss:3.3935 train_time:744581ms step_avg:153.84ms step:4851/5100 train_loss:3.2850 train_time:744734ms step_avg:153.84ms step:4852/5100 train_loss:3.0863 train_time:744886ms step_avg:153.84ms step:4853/5100 train_loss:3.1804 train_time:745038ms step_avg:153.84ms step:4854/5100 train_loss:3.3138 train_time:745189ms step_avg:153.84ms step:4855/5100 train_loss:3.2715 train_time:745343ms step_avg:153.84ms step:4856/5100 train_loss:3.4082 train_time:745496ms step_avg:153.84ms step:4857/5100 train_loss:3.2464 train_time:745648ms step_avg:153.84ms step:4858/5100 train_loss:3.2849 train_time:745801ms step_avg:153.84ms step:4859/5100 train_loss:3.2303 train_time:745953ms step_avg:153.84ms step:4860/5100 train_loss:3.3660 train_time:746106ms step_avg:153.84ms step:4861/5100 train_loss:3.2293 train_time:746258ms step_avg:153.84ms step:4862/5100 train_loss:3.2862 train_time:746411ms step_avg:153.84ms step:4863/5100 train_loss:3.2957 train_time:746563ms step_avg:153.84ms step:4864/5100 train_loss:3.2514 train_time:746716ms step_avg:153.84ms step:4865/5100 train_loss:3.3240 train_time:746868ms step_avg:153.83ms step:4866/5100 train_loss:2.9770 train_time:747020ms step_avg:153.83ms step:4867/5100 train_loss:3.2024 train_time:747173ms step_avg:153.83ms step:4868/5100 train_loss:3.2538 train_time:747327ms step_avg:153.83ms step:4869/5100 train_loss:3.2735 train_time:747479ms step_avg:153.83ms step:4870/5100 train_loss:3.2823 train_time:747632ms step_avg:153.83ms step:4871/5100 train_loss:3.2809 train_time:747784ms step_avg:153.83ms step:4872/5100 train_loss:3.3987 train_time:747937ms step_avg:153.83ms step:4873/5100 train_loss:3.3848 train_time:748090ms step_avg:153.83ms step:4874/5100 train_loss:3.4123 train_time:748244ms step_avg:153.83ms step:4875/5100 train_loss:3.4906 train_time:748397ms step_avg:153.83ms step:4875/5100 val_loss:3.2866 train_time:748460ms step_avg:153.85ms step:4876/5100 train_loss:3.3094 train_time:748553ms step_avg:153.83ms step:4877/5100 train_loss:3.2175 train_time:748710ms step_avg:153.83ms step:4878/5100 train_loss:3.1793 train_time:748866ms step_avg:153.83ms step:4879/5100 train_loss:3.2349 train_time:749018ms step_avg:153.83ms step:4880/5100 train_loss:3.3676 train_time:749170ms step_avg:153.83ms step:4881/5100 train_loss:3.2108 train_time:749322ms step_avg:153.83ms step:4882/5100 train_loss:3.3394 train_time:749475ms step_avg:153.83ms step:4883/5100 train_loss:3.3491 train_time:749627ms step_avg:153.83ms step:4884/5100 train_loss:3.2534 train_time:749780ms step_avg:153.83ms step:4885/5100 train_loss:3.2546 train_time:749932ms step_avg:153.83ms step:4886/5100 train_loss:3.3659 train_time:750085ms step_avg:153.83ms step:4887/5100 train_loss:3.3837 train_time:750238ms step_avg:153.83ms step:4888/5100 train_loss:3.2603 train_time:750390ms step_avg:153.83ms step:4889/5100 train_loss:3.2302 train_time:750543ms step_avg:153.83ms step:4890/5100 train_loss:3.3142 train_time:750696ms step_avg:153.83ms step:4891/5100 train_loss:3.2255 train_time:750848ms step_avg:153.83ms step:4892/5100 train_loss:3.3179 train_time:751001ms step_avg:153.83ms step:4893/5100 train_loss:3.3155 train_time:751153ms step_avg:153.83ms step:4894/5100 train_loss:3.3446 train_time:751306ms step_avg:153.83ms step:4895/5100 train_loss:3.4074 train_time:751458ms step_avg:153.83ms step:4896/5100 train_loss:3.3090 train_time:751610ms step_avg:153.83ms step:4897/5100 train_loss:3.2565 train_time:751763ms step_avg:153.83ms step:4898/5100 train_loss:3.4473 train_time:751915ms step_avg:153.83ms step:4899/5100 train_loss:3.2184 train_time:752068ms step_avg:153.83ms step:4900/5100 train_loss:3.2721 train_time:752222ms step_avg:153.83ms step:4901/5100 train_loss:3.1774 train_time:752373ms step_avg:153.83ms step:4902/5100 train_loss:3.1478 train_time:752526ms step_avg:153.83ms step:4903/5100 train_loss:3.2845 train_time:752678ms step_avg:153.83ms step:4904/5100 train_loss:3.2310 train_time:752831ms step_avg:153.83ms step:4905/5100 train_loss:3.3028 train_time:752984ms step_avg:153.83ms step:4906/5100 train_loss:3.3444 train_time:753136ms step_avg:153.83ms step:4907/5100 train_loss:3.2210 train_time:753288ms step_avg:153.83ms step:4908/5100 train_loss:3.3101 train_time:753441ms step_avg:153.83ms step:4909/5100 train_loss:3.2100 train_time:753594ms step_avg:153.83ms step:4910/5100 train_loss:3.3287 train_time:753746ms step_avg:153.83ms step:4911/5100 train_loss:3.3646 train_time:753899ms step_avg:153.83ms step:4912/5100 train_loss:3.2738 train_time:754051ms step_avg:153.83ms step:4913/5100 train_loss:3.2345 train_time:754203ms step_avg:153.82ms step:4914/5100 train_loss:3.2343 train_time:754457ms step_avg:153.85ms step:4915/5100 train_loss:3.1512 train_time:754621ms step_avg:153.85ms step:4916/5100 train_loss:3.3570 train_time:754773ms step_avg:153.85ms step:4917/5100 train_loss:3.3355 train_time:754926ms step_avg:153.85ms step:4918/5100 train_loss:3.2603 train_time:755077ms step_avg:153.85ms step:4919/5100 train_loss:3.2732 train_time:755229ms step_avg:153.85ms step:4920/5100 train_loss:3.2771 train_time:755381ms step_avg:153.85ms step:4921/5100 train_loss:3.3546 train_time:755534ms step_avg:153.85ms step:4922/5100 train_loss:3.5145 train_time:755688ms step_avg:153.85ms step:4923/5100 train_loss:3.3739 train_time:755841ms step_avg:153.85ms step:4924/5100 train_loss:3.2398 train_time:755993ms step_avg:153.84ms step:4925/5100 train_loss:3.5437 train_time:756146ms step_avg:153.84ms step:4926/5100 train_loss:3.2801 train_time:756299ms step_avg:153.84ms step:4927/5100 train_loss:3.2662 train_time:756450ms step_avg:153.84ms step:4928/5100 train_loss:3.1837 train_time:756603ms step_avg:153.84ms step:4929/5100 train_loss:3.1911 train_time:756756ms step_avg:153.84ms step:4930/5100 train_loss:3.3416 train_time:756909ms step_avg:153.84ms step:4931/5100 train_loss:3.5844 train_time:757062ms step_avg:153.84ms step:4932/5100 train_loss:3.1908 train_time:757214ms step_avg:153.84ms step:4933/5100 train_loss:3.2928 train_time:757367ms step_avg:153.84ms step:4934/5100 train_loss:3.3545 train_time:757520ms step_avg:153.84ms step:4935/5100 train_loss:3.1481 train_time:757673ms step_avg:153.84ms step:4936/5100 train_loss:3.2972 train_time:757826ms step_avg:153.84ms step:4937/5100 train_loss:3.3517 train_time:757979ms step_avg:153.84ms step:4938/5100 train_loss:3.3323 train_time:758132ms step_avg:153.84ms step:4939/5100 train_loss:3.3268 train_time:758284ms step_avg:153.84ms step:4940/5100 train_loss:3.4490 train_time:758561ms step_avg:153.87ms step:4941/5100 train_loss:3.2864 train_time:758712ms step_avg:153.87ms step:4942/5100 train_loss:3.2977 train_time:758865ms step_avg:153.87ms step:4943/5100 train_loss:3.0443 train_time:759016ms step_avg:153.87ms step:4944/5100 train_loss:3.5399 train_time:759168ms step_avg:153.86ms step:4945/5100 train_loss:3.4964 train_time:759321ms step_avg:153.86ms step:4946/5100 train_loss:3.1218 train_time:759473ms step_avg:153.86ms step:4947/5100 train_loss:3.3709 train_time:759626ms step_avg:153.86ms step:4948/5100 train_loss:3.3939 train_time:759780ms step_avg:153.86ms step:4949/5100 train_loss:3.2285 train_time:759934ms step_avg:153.86ms step:4950/5100 train_loss:3.3531 train_time:760086ms step_avg:153.86ms step:4951/5100 train_loss:3.2177 train_time:760238ms step_avg:153.86ms step:4952/5100 train_loss:3.3431 train_time:760390ms step_avg:153.86ms step:4953/5100 train_loss:3.3030 train_time:760543ms step_avg:153.86ms step:4954/5100 train_loss:3.1914 train_time:760696ms step_avg:153.86ms step:4955/5100 train_loss:3.3270 train_time:760848ms step_avg:153.86ms step:4956/5100 train_loss:3.1569 train_time:761003ms step_avg:153.86ms step:4957/5100 train_loss:3.2597 train_time:761154ms step_avg:153.86ms step:4958/5100 train_loss:3.2356 train_time:761306ms step_avg:153.86ms step:4959/5100 train_loss:3.2472 train_time:761459ms step_avg:153.86ms step:4960/5100 train_loss:3.2880 train_time:761611ms step_avg:153.86ms step:4961/5100 train_loss:3.4390 train_time:761765ms step_avg:153.86ms step:4962/5100 train_loss:3.1984 train_time:761917ms step_avg:153.86ms step:4963/5100 train_loss:3.3382 train_time:762070ms step_avg:153.86ms step:4964/5100 train_loss:3.1773 train_time:762224ms step_avg:153.86ms step:4965/5100 train_loss:3.9044 train_time:762376ms step_avg:153.86ms step:4966/5100 train_loss:3.1688 train_time:762528ms step_avg:153.86ms step:4967/5100 train_loss:3.3081 train_time:762681ms step_avg:153.86ms step:4968/5100 train_loss:3.1311 train_time:762835ms step_avg:153.86ms step:4969/5100 train_loss:3.8477 train_time:762988ms step_avg:153.86ms step:4970/5100 train_loss:3.3780 train_time:763141ms step_avg:153.86ms step:4971/5100 train_loss:3.2942 train_time:763294ms step_avg:153.86ms step:4972/5100 train_loss:3.2373 train_time:763446ms step_avg:153.86ms step:4973/5100 train_loss:3.3235 train_time:763599ms step_avg:153.86ms step:4974/5100 train_loss:3.1956 train_time:763751ms step_avg:153.86ms step:4975/5100 train_loss:3.1943 train_time:763904ms step_avg:153.86ms step:4976/5100 train_loss:3.3438 train_time:764058ms step_avg:153.86ms step:4977/5100 train_loss:3.2711 train_time:764210ms step_avg:153.86ms step:4978/5100 train_loss:3.2263 train_time:764364ms step_avg:153.86ms step:4979/5100 train_loss:3.2846 train_time:764515ms step_avg:153.86ms step:4980/5100 train_loss:3.2208 train_time:764667ms step_avg:153.86ms step:4981/5100 train_loss:3.3631 train_time:764821ms step_avg:153.86ms step:4982/5100 train_loss:3.3309 train_time:764973ms step_avg:153.86ms step:4983/5100 train_loss:3.1347 train_time:765126ms step_avg:153.86ms step:4984/5100 train_loss:3.1627 train_time:765279ms step_avg:153.86ms step:4985/5100 train_loss:3.4501 train_time:765432ms step_avg:153.86ms step:4986/5100 train_loss:3.3343 train_time:765585ms step_avg:153.86ms step:4987/5100 train_loss:3.2528 train_time:765738ms step_avg:153.86ms step:4988/5100 train_loss:3.2816 train_time:765890ms step_avg:153.85ms step:4989/5100 train_loss:3.2687 train_time:766043ms step_avg:153.85ms step:4990/5100 train_loss:3.2423 train_time:766197ms step_avg:153.85ms step:4991/5100 train_loss:3.3009 train_time:766349ms step_avg:153.85ms step:4992/5100 train_loss:3.3404 train_time:766503ms step_avg:153.85ms step:4993/5100 train_loss:3.1669 train_time:766654ms step_avg:153.85ms step:4994/5100 train_loss:3.2941 train_time:766807ms step_avg:153.85ms step:4995/5100 train_loss:3.2202 train_time:766960ms step_avg:153.85ms step:4996/5100 train_loss:3.3652 train_time:767111ms step_avg:153.85ms step:4997/5100 train_loss:3.2406 train_time:767265ms step_avg:153.85ms step:4998/5100 train_loss:3.4018 train_time:767417ms step_avg:153.85ms step:4999/5100 train_loss:3.2628 train_time:767570ms step_avg:153.85ms step:5000/5100 train_loss:3.3906 train_time:767725ms step_avg:153.85ms step:5000/5100 val_loss:3.2783 train_time:767787ms step_avg:153.87ms step:5001/5100 train_loss:3.3181 train_time:767882ms step_avg:153.85ms step:5002/5100 train_loss:3.3135 train_time:768037ms step_avg:153.85ms step:5003/5100 train_loss:3.2025 train_time:768189ms step_avg:153.85ms step:5004/5100 train_loss:3.2776 train_time:768342ms step_avg:153.85ms step:5005/5100 train_loss:3.2929 train_time:768494ms step_avg:153.85ms step:5006/5100 train_loss:3.1656 train_time:768645ms step_avg:153.85ms step:5007/5100 train_loss:3.3932 train_time:768797ms step_avg:153.85ms step:5008/5100 train_loss:3.2418 train_time:768950ms step_avg:153.85ms step:5009/5100 train_loss:3.2705 train_time:769102ms step_avg:153.85ms step:5010/5100 train_loss:3.2407 train_time:769256ms step_avg:153.85ms step:5011/5100 train_loss:3.4437 train_time:769408ms step_avg:153.85ms step:5012/5100 train_loss:3.2436 train_time:769562ms step_avg:153.85ms step:5013/5100 train_loss:3.2283 train_time:769714ms step_avg:153.85ms step:5014/5100 train_loss:3.1807 train_time:769867ms step_avg:153.85ms step:5015/5100 train_loss:3.2893 train_time:770020ms step_avg:153.85ms step:5016/5100 train_loss:3.2742 train_time:770172ms step_avg:153.85ms step:5017/5100 train_loss:3.3072 train_time:770326ms step_avg:153.85ms step:5018/5100 train_loss:3.3262 train_time:770478ms step_avg:153.85ms step:5019/5100 train_loss:3.2832 train_time:770631ms step_avg:153.85ms step:5020/5100 train_loss:3.7949 train_time:770784ms step_avg:153.85ms step:5021/5100 train_loss:3.2345 train_time:770936ms step_avg:153.85ms step:5022/5100 train_loss:3.3405 train_time:771089ms step_avg:153.85ms step:5023/5100 train_loss:3.2580 train_time:771240ms step_avg:153.85ms step:5024/5100 train_loss:3.4060 train_time:771393ms step_avg:153.85ms step:5025/5100 train_loss:3.1980 train_time:771545ms step_avg:153.85ms step:5026/5100 train_loss:3.3604 train_time:771698ms step_avg:153.85ms step:5027/5100 train_loss:3.2093 train_time:771850ms step_avg:153.85ms step:5028/5100 train_loss:3.4230 train_time:772002ms step_avg:153.85ms step:5029/5100 train_loss:3.3293 train_time:772155ms step_avg:153.85ms step:5030/5100 train_loss:3.3609 train_time:772308ms step_avg:153.85ms step:5031/5100 train_loss:3.2150 train_time:772461ms step_avg:153.85ms step:5032/5100 train_loss:3.2542 train_time:772614ms step_avg:153.85ms step:5033/5100 train_loss:3.1951 train_time:772766ms step_avg:153.85ms step:5034/5100 train_loss:3.3898 train_time:772919ms step_avg:153.85ms step:5035/5100 train_loss:3.3836 train_time:773072ms step_avg:153.85ms step:5036/5100 train_loss:3.2357 train_time:773224ms step_avg:153.84ms step:5037/5100 train_loss:3.1634 train_time:773377ms step_avg:153.84ms step:5038/5100 train_loss:3.2018 train_time:773529ms step_avg:153.84ms step:5039/5100 train_loss:3.3329 train_time:773682ms step_avg:153.84ms step:5040/5100 train_loss:3.2609 train_time:773835ms step_avg:153.84ms step:5041/5100 train_loss:3.4298 train_time:773988ms step_avg:153.84ms step:5042/5100 train_loss:3.2262 train_time:774141ms step_avg:153.84ms step:5043/5100 train_loss:3.4130 train_time:774293ms step_avg:153.84ms step:5044/5100 train_loss:3.3244 train_time:774445ms step_avg:153.84ms step:5045/5100 train_loss:3.3864 train_time:774598ms step_avg:153.84ms step:5046/5100 train_loss:3.2129 train_time:774751ms step_avg:153.84ms step:5047/5100 train_loss:3.3475 train_time:774904ms step_avg:153.84ms step:5048/5100 train_loss:3.0939 train_time:775057ms step_avg:153.84ms step:5049/5100 train_loss:3.2619 train_time:775209ms step_avg:153.84ms step:5050/5100 train_loss:3.2667 train_time:775362ms step_avg:153.84ms step:5051/5100 train_loss:3.2041 train_time:775514ms step_avg:153.84ms step:5052/5100 train_loss:3.2464 train_time:775666ms step_avg:153.84ms step:5053/5100 train_loss:3.2953 train_time:775818ms step_avg:153.84ms step:5054/5100 train_loss:3.3353 train_time:775970ms step_avg:153.84ms step:5055/5100 train_loss:3.4079 train_time:776123ms step_avg:153.84ms step:5056/5100 train_loss:3.3563 train_time:776275ms step_avg:153.84ms step:5057/5100 train_loss:3.2391 train_time:776428ms step_avg:153.84ms step:5058/5100 train_loss:3.1417 train_time:776581ms step_avg:153.84ms step:5059/5100 train_loss:3.0567 train_time:776733ms step_avg:153.84ms step:5060/5100 train_loss:3.2677 train_time:776886ms step_avg:153.84ms step:5061/5100 train_loss:3.3660 train_time:777039ms step_avg:153.84ms step:5062/5100 train_loss:3.2891 train_time:777192ms step_avg:153.84ms step:5063/5100 train_loss:3.4073 train_time:777344ms step_avg:153.84ms step:5064/5100 train_loss:3.4107 train_time:777498ms step_avg:153.84ms step:5065/5100 train_loss:3.2713 train_time:777650ms step_avg:153.84ms step:5066/5100 train_loss:3.3754 train_time:777803ms step_avg:153.84ms step:5067/5100 train_loss:3.5801 train_time:777957ms step_avg:153.84ms step:5068/5100 train_loss:3.2260 train_time:778108ms step_avg:153.84ms step:5069/5100 train_loss:3.5747 train_time:778261ms step_avg:153.84ms step:5070/5100 train_loss:3.2792 train_time:778412ms step_avg:153.84ms step:5071/5100 train_loss:3.7046 train_time:778565ms step_avg:153.84ms step:5072/5100 train_loss:3.2276 train_time:778719ms step_avg:153.84ms step:5073/5100 train_loss:3.2860 train_time:778872ms step_avg:153.84ms step:5074/5100 train_loss:3.4260 train_time:779025ms step_avg:153.84ms step:5075/5100 train_loss:3.2563 train_time:779177ms step_avg:153.84ms step:5076/5100 train_loss:3.2526 train_time:779329ms step_avg:153.84ms step:5077/5100 train_loss:3.2248 train_time:779482ms step_avg:153.83ms step:5078/5100 train_loss:3.3046 train_time:779634ms step_avg:153.83ms step:5079/5100 train_loss:3.4334 train_time:779787ms step_avg:153.83ms step:5080/5100 train_loss:3.4146 train_time:779941ms step_avg:153.83ms step:5081/5100 train_loss:3.2396 train_time:780093ms step_avg:153.83ms step:5082/5100 train_loss:3.3598 train_time:780245ms step_avg:153.83ms step:5083/5100 train_loss:3.2184 train_time:780398ms step_avg:153.83ms step:5084/5100 train_loss:3.3073 train_time:780550ms step_avg:153.83ms step:5085/5100 train_loss:3.1952 train_time:780703ms step_avg:153.83ms step:5086/5100 train_loss:4.0473 train_time:780856ms step_avg:153.83ms step:5087/5100 train_loss:3.3330 train_time:781007ms step_avg:153.83ms step:5088/5100 train_loss:3.2453 train_time:781161ms step_avg:153.83ms step:5089/5100 train_loss:3.2572 train_time:781313ms step_avg:153.83ms step:5090/5100 train_loss:3.3922 train_time:781466ms step_avg:153.83ms step:5091/5100 train_loss:3.3117 train_time:781621ms step_avg:153.83ms step:5092/5100 train_loss:3.2134 train_time:781772ms step_avg:153.83ms step:5093/5100 train_loss:3.2305 train_time:781925ms step_avg:153.83ms step:5094/5100 train_loss:3.2275 train_time:782078ms step_avg:153.83ms step:5095/5100 train_loss:3.1546 train_time:782231ms step_avg:153.83ms step:5096/5100 train_loss:3.2793 train_time:782384ms step_avg:153.83ms step:5097/5100 train_loss:3.0651 train_time:782537ms step_avg:153.83ms step:5098/5100 train_loss:3.3613 train_time:782689ms step_avg:153.83ms step:5099/5100 train_loss:3.2215 train_time:782842ms step_avg:153.83ms step:5100/5100 train_loss:3.2718 train_time:782995ms step_avg:153.83ms step:5100/5100 val_loss:3.2747 train_time:783059ms step_avg:153.84ms