==================================================================================================== 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): defaults = dict(lr=lr, momentum=momentum, nesterov=nesterov, backend=backend, backend_steps=backend_steps) super().__init__(params, defaults) def step(self): for group in self.param_groups: lr = group['lr'] momentum = group['momentum'] zeropower_backend = zeropower_backends[group['backend']] for p in group['params']: 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) if g.size(0) == 3 * g.size(1): # split grouped QKV parameters g = torch.cat([zeropower_backend(g1, steps=group['backend_steps']) for g1 in g.split(g.size(1))]) scale = g.size(1)**0.5 else: g = zeropower_backend(g, steps=group['backend_steps']) scale = max(g.size(0), g.size(1))**0.5 # scale to have update.square().mean() == 1 p.data.add_(g, alpha=-lr * scale) # ----------------------------------------------------------------------------- # 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 = apply_rotary_emb(q, cos, sin), apply_rotary_emb(k, cos, sin) q, k = F.rms_norm(q, (q.size(-1),)), F.rms_norm(k, (k.size(-1),)) # QK norm suggested by @Grad62304977 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) 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: Mon Oct 14 05:01:30 2024 +---------------------------------------------------------------------------------------+ | NVIDIA-SMI 535.129.03 Driver Version: 535.129.03 CUDA Version: 12.2 | |-----------------------------------------+----------------------+----------------------+ | 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 On | 00000000:61:00.0 Off | 0 | | N/A 26C P0 113W / 700W | 5787MiB / 81559MiB | 1% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 1 NVIDIA H100 80GB HBM3 On | 00000000:62:00.0 Off | 0 | | N/A 32C P0 121W / 700W | 5835MiB / 81559MiB | 4% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 2 NVIDIA H100 80GB HBM3 On | 00000000:63:00.0 Off | 0 | | N/A 35C P0 115W / 700W | 5835MiB / 81559MiB | 7% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 3 NVIDIA H100 80GB HBM3 On | 00000000:64:00.0 Off | 0 | | N/A 31C P0 116W / 700W | 5835MiB / 81559MiB | 6% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 4 NVIDIA H100 80GB HBM3 On | 00000000:6A:00.0 Off | 0 | | N/A 28C P0 118W / 700W | 5835MiB / 81559MiB | 4% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 5 NVIDIA H100 80GB HBM3 On | 00000000:6B:00.0 Off | 0 | | N/A 33C P0 115W / 700W | 5835MiB / 81559MiB | 4% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 6 NVIDIA H100 80GB HBM3 On | 00000000:6C:00.0 Off | 0 | | N/A 31C P0 115W / 700W | 5835MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 7 NVIDIA H100 80GB HBM3 On | 00000000:6D:00.0 Off | 0 | | N/A 27C P0 117W / 700W | 5595MiB / 81559MiB | 2% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ +---------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=======================================================================================| | 0 N/A N/A 67786 C /usr/bin/python3 5774MiB | | 1 N/A N/A 67787 C /usr/bin/python3 5822MiB | | 2 N/A N/A 67788 C /usr/bin/python3 5822MiB | | 3 N/A N/A 67789 C /usr/bin/python3 5822MiB | | 4 N/A N/A 67790 C /usr/bin/python3 5822MiB | | 5 N/A N/A 67791 C /usr/bin/python3 5822MiB | | 6 N/A N/A 67792 C /usr/bin/python3 5822MiB | | 7 N/A N/A 67793 C /usr/bin/python3 5582MiB | +---------------------------------------------------------------------------------------+ ==================================================================================================== step:0/5100 val_loss:16.0058 train_time:278ms step_avg:nanms step:1/5100 train_loss:16.0026 train_time:42703ms step_avg:nanms step:2/5100 train_loss:9.4947 train_time:53466ms step_avg:nanms step:3/5100 train_loss:8.7243 train_time:53645ms step_avg:nanms step:4/5100 train_loss:8.0034 train_time:53821ms step_avg:nanms step:5/5100 train_loss:7.5714 train_time:53997ms step_avg:nanms step:6/5100 train_loss:7.5361 train_time:54176ms step_avg:nanms step:7/5100 train_loss:7.4515 train_time:54354ms step_avg:nanms step:8/5100 train_loss:7.4935 train_time:54529ms step_avg:nanms step:9/5100 train_loss:7.2292 train_time:54704ms step_avg:nanms step:10/5100 train_loss:7.0049 train_time:54881ms step_avg:nanms step:11/5100 train_loss:6.9134 train_time:176ms step_avg:nanms step:12/5100 train_loss:6.8756 train_time:353ms step_avg:nanms step:13/5100 train_loss:6.7598 train_time:530ms step_avg:176.69ms step:14/5100 train_loss:6.6413 train_time:706ms step_avg:176.50ms step:15/5100 train_loss:6.6229 train_time:883ms step_avg:176.64ms step:16/5100 train_loss:6.5247 train_time:1059ms step_avg:176.49ms step:17/5100 train_loss:6.5430 train_time:1236ms step_avg:176.50ms step:18/5100 train_loss:6.5634 train_time:1411ms step_avg:176.33ms step:19/5100 train_loss:6.3690 train_time:1588ms step_avg:176.40ms step:20/5100 train_loss:6.4064 train_time:1763ms step_avg:176.34ms step:21/5100 train_loss:6.0726 train_time:1940ms step_avg:176.34ms step:22/5100 train_loss:6.4316 train_time:2116ms step_avg:176.31ms step:23/5100 train_loss:6.6346 train_time:2292ms step_avg:176.29ms step:24/5100 train_loss:6.3251 train_time:2469ms step_avg:176.38ms step:25/5100 train_loss:6.4596 train_time:2645ms step_avg:176.34ms step:26/5100 train_loss:6.1703 train_time:2821ms step_avg:176.33ms step:27/5100 train_loss:6.0790 train_time:2997ms step_avg:176.30ms step:28/5100 train_loss:6.2160 train_time:3173ms step_avg:176.28ms step:29/5100 train_loss:5.9012 train_time:3350ms step_avg:176.31ms step:30/5100 train_loss:6.1927 train_time:3526ms step_avg:176.30ms step:31/5100 train_loss:6.0178 train_time:3703ms step_avg:176.34ms step:32/5100 train_loss:5.9916 train_time:3880ms step_avg:176.36ms step:33/5100 train_loss:5.8184 train_time:4056ms step_avg:176.37ms step:34/5100 train_loss:6.0974 train_time:4233ms step_avg:176.37ms step:35/5100 train_loss:6.0359 train_time:4409ms step_avg:176.37ms step:36/5100 train_loss:6.1705 train_time:4585ms step_avg:176.35ms step:37/5100 train_loss:6.1132 train_time:4762ms step_avg:176.38ms step:38/5100 train_loss:6.0114 train_time:4941ms step_avg:176.48ms step:39/5100 train_loss:5.8895 train_time:5117ms step_avg:176.44ms step:40/5100 train_loss:5.9137 train_time:5293ms step_avg:176.44ms step:41/5100 train_loss:5.8300 train_time:5469ms step_avg:176.42ms step:42/5100 train_loss:5.8464 train_time:5645ms step_avg:176.41ms step:43/5100 train_loss:5.7293 train_time:5821ms step_avg:176.40ms step:44/5100 train_loss:5.8348 train_time:5998ms step_avg:176.42ms step:45/5100 train_loss:5.7872 train_time:6175ms step_avg:176.42ms step:46/5100 train_loss:5.9451 train_time:6351ms step_avg:176.43ms step:47/5100 train_loss:5.7408 train_time:6528ms step_avg:176.42ms step:48/5100 train_loss:5.6171 train_time:6704ms step_avg:176.42ms step:49/5100 train_loss:5.8214 train_time:6881ms step_avg:176.43ms step:50/5100 train_loss:5.7003 train_time:7057ms step_avg:176.43ms step:51/5100 train_loss:5.8409 train_time:7234ms step_avg:176.43ms step:52/5100 train_loss:5.7024 train_time:7410ms step_avg:176.43ms step:53/5100 train_loss:5.5714 train_time:7587ms step_avg:176.44ms step:54/5100 train_loss:5.7099 train_time:7764ms step_avg:176.46ms step:55/5100 train_loss:5.5848 train_time:7940ms step_avg:176.44ms step:56/5100 train_loss:5.9282 train_time:8116ms step_avg:176.43ms step:57/5100 train_loss:5.5857 train_time:8292ms step_avg:176.42ms step:58/5100 train_loss:5.4541 train_time:8468ms step_avg:176.42ms step:59/5100 train_loss:5.5927 train_time:8645ms step_avg:176.43ms step:60/5100 train_loss:5.5636 train_time:8821ms step_avg:176.42ms step:61/5100 train_loss:5.6773 train_time:8998ms step_avg:176.43ms step:62/5100 train_loss:5.4317 train_time:9175ms step_avg:176.44ms step:63/5100 train_loss:5.5337 train_time:9352ms step_avg:176.45ms step:64/5100 train_loss:5.5094 train_time:9529ms step_avg:176.46ms step:65/5100 train_loss:5.1816 train_time:9706ms step_avg:176.47ms step:66/5100 train_loss:5.3273 train_time:9882ms step_avg:176.47ms step:67/5100 train_loss:5.4873 train_time:10059ms step_avg:176.48ms step:68/5100 train_loss:5.3574 train_time:10236ms step_avg:176.48ms step:69/5100 train_loss:5.6205 train_time:10413ms step_avg:176.49ms step:70/5100 train_loss:5.2609 train_time:10590ms step_avg:176.50ms step:71/5100 train_loss:5.2774 train_time:10765ms step_avg:176.48ms step:72/5100 train_loss:5.4959 train_time:10942ms step_avg:176.49ms step:73/5100 train_loss:5.4203 train_time:11123ms step_avg:176.55ms step:74/5100 train_loss:5.3024 train_time:11296ms step_avg:176.51ms step:75/5100 train_loss:5.4277 train_time:11473ms step_avg:176.51ms step:76/5100 train_loss:5.3970 train_time:11649ms step_avg:176.51ms step:77/5100 train_loss:5.3551 train_time:11826ms step_avg:176.50ms step:78/5100 train_loss:5.4435 train_time:12002ms step_avg:176.51ms step:79/5100 train_loss:5.5162 train_time:12179ms step_avg:176.51ms step:80/5100 train_loss:5.2904 train_time:12356ms step_avg:176.51ms step:81/5100 train_loss:5.4086 train_time:12533ms step_avg:176.53ms step:82/5100 train_loss:5.1663 train_time:12709ms step_avg:176.52ms step:83/5100 train_loss:5.3478 train_time:12885ms step_avg:176.51ms step:84/5100 train_loss:5.2974 train_time:13062ms step_avg:176.52ms step:85/5100 train_loss:5.2750 train_time:13240ms step_avg:176.53ms step:86/5100 train_loss:5.1372 train_time:13416ms step_avg:176.52ms step:87/5100 train_loss:5.3540 train_time:13592ms step_avg:176.52ms step:88/5100 train_loss:5.2620 train_time:13768ms step_avg:176.51ms step:89/5100 train_loss:5.3097 train_time:13944ms step_avg:176.51ms step:90/5100 train_loss:5.2766 train_time:14120ms step_avg:176.51ms step:91/5100 train_loss:5.2022 train_time:14297ms step_avg:176.50ms step:92/5100 train_loss:5.1902 train_time:14475ms step_avg:176.52ms step:93/5100 train_loss:5.3196 train_time:14649ms step_avg:176.50ms step:94/5100 train_loss:5.1340 train_time:14825ms step_avg:176.49ms step:95/5100 train_loss:5.1428 train_time:15001ms step_avg:176.49ms step:96/5100 train_loss:5.1916 train_time:15177ms step_avg:176.48ms step:97/5100 train_loss:5.1030 train_time:15353ms step_avg:176.48ms step:98/5100 train_loss:5.1782 train_time:15530ms step_avg:176.47ms step:99/5100 train_loss:5.1015 train_time:15706ms step_avg:176.47ms step:100/5100 train_loss:5.2243 train_time:15882ms step_avg:176.47ms step:101/5100 train_loss:5.1945 train_time:16059ms step_avg:176.47ms step:102/5100 train_loss:5.1008 train_time:16235ms step_avg:176.47ms step:103/5100 train_loss:5.1931 train_time:16411ms step_avg:176.47ms step:104/5100 train_loss:5.1403 train_time:16587ms step_avg:176.46ms step:105/5100 train_loss:4.9992 train_time:16763ms step_avg:176.46ms step:106/5100 train_loss:5.1001 train_time:16939ms step_avg:176.45ms step:107/5100 train_loss:5.3160 train_time:17115ms step_avg:176.44ms step:108/5100 train_loss:5.0735 train_time:17293ms step_avg:176.45ms step:109/5100 train_loss:4.8698 train_time:17469ms step_avg:176.45ms step:110/5100 train_loss:5.0438 train_time:17645ms step_avg:176.45ms step:111/5100 train_loss:5.0285 train_time:17821ms step_avg:176.45ms step:112/5100 train_loss:4.9934 train_time:17997ms step_avg:176.44ms step:113/5100 train_loss:5.1061 train_time:18173ms step_avg:176.44ms step:114/5100 train_loss:5.0397 train_time:18349ms step_avg:176.43ms step:115/5100 train_loss:4.8932 train_time:18525ms step_avg:176.43ms step:116/5100 train_loss:5.0457 train_time:18701ms step_avg:176.43ms step:117/5100 train_loss:4.9561 train_time:18878ms step_avg:176.43ms step:118/5100 train_loss:4.9087 train_time:19055ms step_avg:176.43ms step:119/5100 train_loss:5.0531 train_time:19231ms step_avg:176.44ms step:120/5100 train_loss:5.0061 train_time:19408ms step_avg:176.43ms step:121/5100 train_loss:4.9401 train_time:19584ms step_avg:176.43ms step:122/5100 train_loss:4.8356 train_time:19760ms step_avg:176.43ms step:123/5100 train_loss:4.9557 train_time:19936ms step_avg:176.43ms step:124/5100 train_loss:4.8063 train_time:20113ms step_avg:176.43ms step:125/5100 train_loss:5.1188 train_time:20289ms step_avg:176.42ms step:125/5100 val_loss:4.9493 train_time:20289ms step_avg:176.43ms step:126/5100 train_loss:4.9946 train_time:20480ms step_avg:176.55ms step:127/5100 train_loss:4.9437 train_time:20656ms step_avg:176.55ms step:128/5100 train_loss:4.9967 train_time:20833ms step_avg:176.55ms step:129/5100 train_loss:4.8718 train_time:21009ms step_avg:176.55ms step:130/5100 train_loss:5.1766 train_time:21185ms step_avg:176.54ms step:131/5100 train_loss:4.9408 train_time:21362ms step_avg:176.54ms step:132/5100 train_loss:4.9384 train_time:21539ms step_avg:176.55ms step:133/5100 train_loss:4.8975 train_time:21715ms step_avg:176.54ms step:134/5100 train_loss:4.9352 train_time:21891ms step_avg:176.54ms step:135/5100 train_loss:4.8185 train_time:22068ms step_avg:176.54ms step:136/5100 train_loss:4.9446 train_time:22244ms step_avg:176.54ms step:137/5100 train_loss:4.7276 train_time:22420ms step_avg:176.53ms step:138/5100 train_loss:4.8822 train_time:22596ms step_avg:176.53ms step:139/5100 train_loss:4.8319 train_time:22772ms step_avg:176.53ms step:140/5100 train_loss:4.8668 train_time:22948ms step_avg:176.52ms step:141/5100 train_loss:4.9399 train_time:23124ms step_avg:176.52ms step:142/5100 train_loss:4.8072 train_time:23300ms step_avg:176.51ms step:143/5100 train_loss:4.8598 train_time:23477ms step_avg:176.52ms step:144/5100 train_loss:4.7252 train_time:23652ms step_avg:176.51ms step:145/5100 train_loss:4.8631 train_time:23828ms step_avg:176.50ms step:146/5100 train_loss:4.8110 train_time:24004ms step_avg:176.50ms step:147/5100 train_loss:4.6902 train_time:24181ms step_avg:176.51ms step:148/5100 train_loss:4.8383 train_time:24358ms step_avg:176.51ms step:149/5100 train_loss:4.8407 train_time:24535ms step_avg:176.51ms step:150/5100 train_loss:4.8537 train_time:24711ms step_avg:176.51ms step:151/5100 train_loss:4.8991 train_time:24890ms step_avg:176.52ms step:152/5100 train_loss:4.7800 train_time:25065ms step_avg:176.51ms step:153/5100 train_loss:4.7851 train_time:25241ms step_avg:176.51ms step:154/5100 train_loss:4.8742 train_time:25418ms step_avg:176.51ms step:155/5100 train_loss:4.8277 train_time:25594ms step_avg:176.51ms step:156/5100 train_loss:4.7787 train_time:25770ms step_avg:176.51ms step:157/5100 train_loss:4.8128 train_time:25946ms step_avg:176.50ms step:158/5100 train_loss:4.9299 train_time:26122ms step_avg:176.50ms step:159/5100 train_loss:4.7138 train_time:26298ms step_avg:176.50ms step:160/5100 train_loss:4.7836 train_time:26474ms step_avg:176.49ms step:161/5100 train_loss:4.6138 train_time:26649ms step_avg:176.48ms step:162/5100 train_loss:4.8028 train_time:26826ms step_avg:176.48ms step:163/5100 train_loss:4.8306 train_time:27010ms step_avg:176.53ms step:164/5100 train_loss:4.8191 train_time:27186ms step_avg:176.53ms step:165/5100 train_loss:4.6339 train_time:27361ms step_avg:176.52ms step:166/5100 train_loss:4.7510 train_time:27537ms step_avg:176.52ms step:167/5100 train_loss:4.8860 train_time:27713ms step_avg:176.51ms step:168/5100 train_loss:4.6828 train_time:27889ms step_avg:176.51ms step:169/5100 train_loss:4.7675 train_time:28065ms step_avg:176.51ms step:170/5100 train_loss:4.6191 train_time:28242ms step_avg:176.52ms step:171/5100 train_loss:4.5174 train_time:28418ms step_avg:176.51ms step:172/5100 train_loss:4.6884 train_time:28594ms step_avg:176.51ms step:173/5100 train_loss:4.6659 train_time:28771ms step_avg:176.51ms step:174/5100 train_loss:4.7133 train_time:28947ms step_avg:176.50ms step:175/5100 train_loss:4.8760 train_time:29122ms step_avg:176.50ms step:176/5100 train_loss:4.7220 train_time:29298ms step_avg:176.49ms step:177/5100 train_loss:4.5810 train_time:29474ms step_avg:176.49ms step:178/5100 train_loss:4.5401 train_time:29650ms step_avg:176.49ms step:179/5100 train_loss:4.6193 train_time:29827ms step_avg:176.49ms step:180/5100 train_loss:4.6132 train_time:30002ms step_avg:176.48ms step:181/5100 train_loss:4.6202 train_time:30178ms step_avg:176.48ms step:182/5100 train_loss:4.7540 train_time:30355ms step_avg:176.48ms step:183/5100 train_loss:4.6130 train_time:30532ms step_avg:176.49ms step:184/5100 train_loss:4.5672 train_time:30708ms step_avg:176.48ms step:185/5100 train_loss:4.5755 train_time:30884ms step_avg:176.48ms step:186/5100 train_loss:4.7012 train_time:31060ms step_avg:176.48ms step:187/5100 train_loss:4.6133 train_time:31236ms step_avg:176.48ms step:188/5100 train_loss:4.7916 train_time:31412ms step_avg:176.47ms step:189/5100 train_loss:4.6349 train_time:31819ms step_avg:177.76ms step:190/5100 train_loss:4.5500 train_time:32239ms step_avg:179.10ms step:191/5100 train_loss:4.6854 train_time:32414ms step_avg:179.09ms step:192/5100 train_loss:4.5359 train_time:32592ms step_avg:179.07ms step:193/5100 train_loss:4.4593 train_time:32769ms step_avg:179.06ms step:194/5100 train_loss:4.6817 train_time:32945ms step_avg:179.05ms step:195/5100 train_loss:4.6099 train_time:33121ms step_avg:179.03ms step:196/5100 train_loss:4.8048 train_time:33297ms step_avg:179.02ms step:197/5100 train_loss:4.6597 train_time:33473ms step_avg:179.00ms step:198/5100 train_loss:4.5105 train_time:33649ms step_avg:178.98ms step:199/5100 train_loss:4.5878 train_time:33825ms step_avg:178.97ms step:200/5100 train_loss:4.4466 train_time:34001ms step_avg:178.95ms step:201/5100 train_loss:4.5474 train_time:34177ms step_avg:178.94ms step:202/5100 train_loss:4.4441 train_time:34353ms step_avg:178.92ms step:203/5100 train_loss:4.6859 train_time:34529ms step_avg:178.91ms step:204/5100 train_loss:4.5510 train_time:34705ms step_avg:178.89ms step:205/5100 train_loss:4.5904 train_time:34882ms step_avg:178.88ms step:206/5100 train_loss:4.7008 train_time:35058ms step_avg:178.87ms step:207/5100 train_loss:4.3588 train_time:35234ms step_avg:178.85ms step:208/5100 train_loss:4.5183 train_time:35410ms step_avg:178.84ms step:209/5100 train_loss:4.4942 train_time:35586ms step_avg:178.82ms step:210/5100 train_loss:4.6552 train_time:35762ms step_avg:178.81ms step:211/5100 train_loss:4.5780 train_time:35937ms step_avg:178.79ms step:212/5100 train_loss:4.4570 train_time:36114ms step_avg:178.78ms step:213/5100 train_loss:4.5755 train_time:36290ms step_avg:178.77ms step:214/5100 train_loss:4.4309 train_time:36466ms step_avg:178.75ms step:215/5100 train_loss:4.5025 train_time:36642ms step_avg:178.74ms step:216/5100 train_loss:4.3593 train_time:36817ms step_avg:178.72ms step:217/5100 train_loss:4.4584 train_time:36993ms step_avg:178.71ms step:218/5100 train_loss:4.4218 train_time:37168ms step_avg:178.69ms step:219/5100 train_loss:4.4525 train_time:37344ms step_avg:178.68ms step:220/5100 train_loss:4.4454 train_time:37521ms step_avg:178.67ms step:221/5100 train_loss:4.4802 train_time:37696ms step_avg:178.65ms step:222/5100 train_loss:4.4963 train_time:37873ms step_avg:178.64ms step:223/5100 train_loss:4.4229 train_time:38048ms step_avg:178.63ms step:224/5100 train_loss:4.4148 train_time:38224ms step_avg:178.62ms step:225/5100 train_loss:4.6404 train_time:38399ms step_avg:178.60ms step:226/5100 train_loss:4.2775 train_time:38575ms step_avg:178.59ms step:227/5100 train_loss:4.3405 train_time:38751ms step_avg:178.58ms step:228/5100 train_loss:4.3456 train_time:38928ms step_avg:178.57ms step:229/5100 train_loss:4.5040 train_time:39104ms step_avg:178.56ms step:230/5100 train_loss:4.2919 train_time:39280ms step_avg:178.55ms step:231/5100 train_loss:4.4256 train_time:39458ms step_avg:178.54ms step:232/5100 train_loss:4.2853 train_time:39634ms step_avg:178.53ms step:233/5100 train_loss:4.3117 train_time:39809ms step_avg:178.52ms step:234/5100 train_loss:4.4765 train_time:39987ms step_avg:178.51ms step:235/5100 train_loss:4.3544 train_time:40162ms step_avg:178.50ms step:236/5100 train_loss:4.2542 train_time:40338ms step_avg:178.49ms step:237/5100 train_loss:4.4520 train_time:40514ms step_avg:178.48ms step:238/5100 train_loss:4.4211 train_time:40691ms step_avg:178.47ms step:239/5100 train_loss:4.2868 train_time:40866ms step_avg:178.45ms step:240/5100 train_loss:4.4460 train_time:41042ms step_avg:178.44ms step:241/5100 train_loss:4.4481 train_time:41218ms step_avg:178.44ms step:242/5100 train_loss:4.3234 train_time:41394ms step_avg:178.42ms step:243/5100 train_loss:4.5057 train_time:41570ms step_avg:178.41ms step:244/5100 train_loss:4.3363 train_time:41745ms step_avg:178.40ms step:245/5100 train_loss:4.3795 train_time:41921ms step_avg:178.39ms step:246/5100 train_loss:4.4645 train_time:42097ms step_avg:178.38ms step:247/5100 train_loss:4.3934 train_time:42273ms step_avg:178.37ms step:248/5100 train_loss:4.3306 train_time:42448ms step_avg:178.35ms step:249/5100 train_loss:4.4495 train_time:42624ms step_avg:178.34ms step:250/5100 train_loss:4.2343 train_time:42800ms step_avg:178.33ms step:250/5100 val_loss:4.3294 train_time:42801ms step_avg:178.34ms step:251/5100 train_loss:4.2866 train_time:42990ms step_avg:178.38ms step:252/5100 train_loss:4.3910 train_time:43167ms step_avg:178.38ms step:253/5100 train_loss:4.4375 train_time:43343ms step_avg:178.37ms step:254/5100 train_loss:4.2618 train_time:43519ms step_avg:178.36ms step:255/5100 train_loss:4.2056 train_time:43695ms step_avg:178.35ms step:256/5100 train_loss:4.3795 train_time:43870ms step_avg:178.33ms step:257/5100 train_loss:4.2962 train_time:44045ms step_avg:178.32ms step:258/5100 train_loss:4.3110 train_time:44221ms step_avg:178.31ms step:259/5100 train_loss:4.2766 train_time:44397ms step_avg:178.30ms step:260/5100 train_loss:4.3112 train_time:44573ms step_avg:178.29ms step:261/5100 train_loss:4.3578 train_time:44749ms step_avg:178.28ms step:262/5100 train_loss:4.3110 train_time:44924ms step_avg:178.27ms step:263/5100 train_loss:4.2845 train_time:45100ms step_avg:178.26ms step:264/5100 train_loss:4.1933 train_time:45276ms step_avg:178.25ms step:265/5100 train_loss:4.2789 train_time:45452ms step_avg:178.24ms step:266/5100 train_loss:4.1411 train_time:45628ms step_avg:178.23ms step:267/5100 train_loss:4.2057 train_time:45804ms step_avg:178.23ms step:268/5100 train_loss:4.2190 train_time:45980ms step_avg:178.22ms step:269/5100 train_loss:4.2343 train_time:46156ms step_avg:178.21ms step:270/5100 train_loss:4.1531 train_time:46332ms step_avg:178.20ms step:271/5100 train_loss:4.3739 train_time:46508ms step_avg:178.19ms step:272/5100 train_loss:4.2763 train_time:46685ms step_avg:178.19ms step:273/5100 train_loss:4.1906 train_time:46861ms step_avg:178.18ms step:274/5100 train_loss:4.2347 train_time:47037ms step_avg:178.17ms step:275/5100 train_loss:4.3184 train_time:47212ms step_avg:178.16ms step:276/5100 train_loss:4.3326 train_time:47389ms step_avg:178.15ms step:277/5100 train_loss:4.5033 train_time:47565ms step_avg:178.15ms step:278/5100 train_loss:4.3076 train_time:47741ms step_avg:178.14ms step:279/5100 train_loss:4.3757 train_time:47917ms step_avg:178.13ms step:280/5100 train_loss:4.2652 train_time:48093ms step_avg:178.12ms step:281/5100 train_loss:4.3849 train_time:48269ms step_avg:178.11ms step:282/5100 train_loss:4.2231 train_time:48445ms step_avg:178.11ms step:283/5100 train_loss:4.2406 train_time:48621ms step_avg:178.10ms step:284/5100 train_loss:4.1802 train_time:48798ms step_avg:178.09ms step:285/5100 train_loss:4.3255 train_time:48974ms step_avg:178.09ms step:286/5100 train_loss:4.3297 train_time:49149ms step_avg:178.08ms step:287/5100 train_loss:4.3605 train_time:49325ms step_avg:178.07ms step:288/5100 train_loss:4.1870 train_time:49502ms step_avg:178.06ms step:289/5100 train_loss:4.2773 train_time:49677ms step_avg:178.05ms step:290/5100 train_loss:4.1463 train_time:49854ms step_avg:178.05ms step:291/5100 train_loss:4.1352 train_time:50030ms step_avg:178.04ms step:292/5100 train_loss:4.2215 train_time:50206ms step_avg:178.04ms step:293/5100 train_loss:4.1353 train_time:50382ms step_avg:178.03ms step:294/5100 train_loss:4.1765 train_time:50558ms step_avg:178.02ms step:295/5100 train_loss:4.2147 train_time:50736ms step_avg:178.02ms step:296/5100 train_loss:4.1041 train_time:50911ms step_avg:178.01ms step:297/5100 train_loss:4.1091 train_time:51088ms step_avg:178.01ms step:298/5100 train_loss:4.1188 train_time:51264ms step_avg:178.00ms step:299/5100 train_loss:4.2266 train_time:51439ms step_avg:177.99ms step:300/5100 train_loss:4.0911 train_time:51617ms step_avg:177.99ms step:301/5100 train_loss:4.2290 train_time:51792ms step_avg:177.98ms step:302/5100 train_loss:4.2452 train_time:51968ms step_avg:177.97ms step:303/5100 train_loss:4.1818 train_time:52144ms step_avg:177.97ms step:304/5100 train_loss:4.2385 train_time:52320ms step_avg:177.96ms step:305/5100 train_loss:4.2183 train_time:52496ms step_avg:177.95ms step:306/5100 train_loss:4.6906 train_time:52671ms step_avg:177.94ms step:307/5100 train_loss:4.1859 train_time:52847ms step_avg:177.94ms step:308/5100 train_loss:4.0954 train_time:53023ms step_avg:177.93ms step:309/5100 train_loss:4.2516 train_time:53200ms step_avg:177.93ms step:310/5100 train_loss:4.1107 train_time:53376ms step_avg:177.92ms step:311/5100 train_loss:4.3280 train_time:53552ms step_avg:177.91ms step:312/5100 train_loss:4.1847 train_time:53728ms step_avg:177.91ms step:313/5100 train_loss:4.1178 train_time:53904ms step_avg:177.90ms step:314/5100 train_loss:4.2313 train_time:54080ms step_avg:177.89ms step:315/5100 train_loss:4.3307 train_time:54256ms step_avg:177.89ms step:316/5100 train_loss:4.2040 train_time:54433ms step_avg:177.88ms step:317/5100 train_loss:4.0369 train_time:54608ms step_avg:177.88ms step:318/5100 train_loss:4.1194 train_time:54784ms step_avg:177.87ms step:319/5100 train_loss:4.1512 train_time:54960ms step_avg:177.86ms step:320/5100 train_loss:4.1312 train_time:55136ms step_avg:177.86ms step:321/5100 train_loss:4.2382 train_time:55312ms step_avg:177.85ms step:322/5100 train_loss:4.1943 train_time:55487ms step_avg:177.84ms step:323/5100 train_loss:4.1569 train_time:55663ms step_avg:177.84ms step:324/5100 train_loss:4.2454 train_time:55840ms step_avg:177.83ms step:325/5100 train_loss:4.2094 train_time:56015ms step_avg:177.83ms step:326/5100 train_loss:4.2689 train_time:56191ms step_avg:177.82ms step:327/5100 train_loss:4.1268 train_time:56368ms step_avg:177.82ms step:328/5100 train_loss:4.6237 train_time:56545ms step_avg:177.81ms step:329/5100 train_loss:4.3142 train_time:56722ms step_avg:177.81ms step:330/5100 train_loss:4.0510 train_time:56898ms step_avg:177.81ms step:331/5100 train_loss:3.9927 train_time:57075ms step_avg:177.80ms step:332/5100 train_loss:4.2155 train_time:57252ms step_avg:177.80ms step:333/5100 train_loss:4.1338 train_time:57427ms step_avg:177.79ms step:334/5100 train_loss:4.1184 train_time:57603ms step_avg:177.79ms step:335/5100 train_loss:4.0777 train_time:57780ms step_avg:177.78ms step:336/5100 train_loss:4.2559 train_time:57956ms step_avg:177.78ms step:337/5100 train_loss:4.1938 train_time:58132ms step_avg:177.77ms step:338/5100 train_loss:4.6657 train_time:58308ms step_avg:177.77ms step:339/5100 train_loss:4.1724 train_time:58484ms step_avg:177.76ms step:340/5100 train_loss:4.1241 train_time:58661ms step_avg:177.76ms step:341/5100 train_loss:4.1550 train_time:58837ms step_avg:177.76ms step:342/5100 train_loss:4.0779 train_time:59014ms step_avg:177.75ms step:343/5100 train_loss:4.0418 train_time:59190ms step_avg:177.75ms step:344/5100 train_loss:4.0895 train_time:59366ms step_avg:177.74ms step:345/5100 train_loss:4.2267 train_time:59542ms step_avg:177.74ms step:346/5100 train_loss:4.0679 train_time:59719ms step_avg:177.73ms step:347/5100 train_loss:4.0054 train_time:59895ms step_avg:177.73ms step:348/5100 train_loss:4.0440 train_time:60071ms step_avg:177.73ms step:349/5100 train_loss:4.0900 train_time:60248ms step_avg:177.72ms step:350/5100 train_loss:4.0462 train_time:60423ms step_avg:177.72ms step:351/5100 train_loss:3.7708 train_time:60599ms step_avg:177.71ms step:352/5100 train_loss:4.0472 train_time:60776ms step_avg:177.71ms step:353/5100 train_loss:4.3921 train_time:60952ms step_avg:177.70ms step:354/5100 train_loss:3.8966 train_time:61129ms step_avg:177.70ms step:355/5100 train_loss:4.1555 train_time:61305ms step_avg:177.70ms step:356/5100 train_loss:4.0261 train_time:61481ms step_avg:177.69ms step:357/5100 train_loss:4.1168 train_time:61657ms step_avg:177.69ms step:358/5100 train_loss:4.0748 train_time:61833ms step_avg:177.68ms step:359/5100 train_loss:4.0650 train_time:62009ms step_avg:177.68ms step:360/5100 train_loss:4.1344 train_time:62185ms step_avg:177.67ms step:361/5100 train_loss:3.6909 train_time:62362ms step_avg:177.67ms step:362/5100 train_loss:4.2503 train_time:62537ms step_avg:177.66ms step:363/5100 train_loss:4.1436 train_time:62712ms step_avg:177.66ms step:364/5100 train_loss:4.0631 train_time:62888ms step_avg:177.65ms step:365/5100 train_loss:3.9781 train_time:63065ms step_avg:177.65ms step:366/5100 train_loss:4.1396 train_time:63241ms step_avg:177.64ms step:367/5100 train_loss:4.0982 train_time:63416ms step_avg:177.64ms step:368/5100 train_loss:4.0841 train_time:63593ms step_avg:177.63ms step:369/5100 train_loss:4.0658 train_time:63768ms step_avg:177.63ms step:370/5100 train_loss:3.9644 train_time:63944ms step_avg:177.62ms step:371/5100 train_loss:4.1065 train_time:64120ms step_avg:177.62ms step:372/5100 train_loss:3.9905 train_time:64296ms step_avg:177.61ms step:373/5100 train_loss:3.9210 train_time:64472ms step_avg:177.61ms step:374/5100 train_loss:4.1341 train_time:64648ms step_avg:177.60ms step:375/5100 train_loss:4.0597 train_time:64824ms step_avg:177.60ms step:375/5100 val_loss:4.0562 train_time:64824ms step_avg:177.60ms step:376/5100 train_loss:4.0301 train_time:65010ms step_avg:177.62ms step:377/5100 train_loss:4.0905 train_time:65187ms step_avg:177.62ms step:378/5100 train_loss:4.0025 train_time:65595ms step_avg:178.25ms step:379/5100 train_loss:4.0635 train_time:65772ms step_avg:178.24ms step:380/5100 train_loss:4.1011 train_time:66192ms step_avg:178.90ms step:381/5100 train_loss:4.1625 train_time:66369ms step_avg:178.89ms step:382/5100 train_loss:4.0703 train_time:66544ms step_avg:178.88ms step:383/5100 train_loss:4.0429 train_time:66720ms step_avg:178.87ms step:384/5100 train_loss:4.0037 train_time:66896ms step_avg:178.87ms step:385/5100 train_loss:4.0878 train_time:67072ms step_avg:178.86ms step:386/5100 train_loss:3.9992 train_time:67248ms step_avg:178.85ms step:387/5100 train_loss:4.1152 train_time:67425ms step_avg:178.85ms step:388/5100 train_loss:4.3059 train_time:67601ms step_avg:178.84ms step:389/5100 train_loss:4.0138 train_time:67776ms step_avg:178.83ms step:390/5100 train_loss:4.0023 train_time:67953ms step_avg:178.82ms step:391/5100 train_loss:4.1095 train_time:68131ms step_avg:178.82ms step:392/5100 train_loss:4.0304 train_time:68307ms step_avg:178.81ms step:393/5100 train_loss:4.1374 train_time:68484ms step_avg:178.81ms step:394/5100 train_loss:3.9696 train_time:68659ms step_avg:178.80ms step:395/5100 train_loss:4.1109 train_time:68835ms step_avg:178.79ms step:396/5100 train_loss:3.8514 train_time:69010ms step_avg:178.78ms step:397/5100 train_loss:4.0552 train_time:69188ms step_avg:178.78ms step:398/5100 train_loss:4.1052 train_time:69363ms step_avg:178.77ms step:399/5100 train_loss:4.1100 train_time:69539ms step_avg:178.76ms step:400/5100 train_loss:4.0020 train_time:69715ms step_avg:178.76ms step:401/5100 train_loss:4.0727 train_time:69891ms step_avg:178.75ms step:402/5100 train_loss:4.1263 train_time:70066ms step_avg:178.74ms step:403/5100 train_loss:4.0611 train_time:70243ms step_avg:178.73ms step:404/5100 train_loss:4.1727 train_time:70418ms step_avg:178.73ms step:405/5100 train_loss:3.9276 train_time:70594ms step_avg:178.72ms step:406/5100 train_loss:4.0151 train_time:70771ms step_avg:178.71ms step:407/5100 train_loss:4.2963 train_time:70946ms step_avg:178.71ms step:408/5100 train_loss:4.0113 train_time:71122ms step_avg:178.70ms step:409/5100 train_loss:4.0379 train_time:71298ms step_avg:178.69ms step:410/5100 train_loss:4.0852 train_time:71474ms step_avg:178.68ms step:411/5100 train_loss:3.9602 train_time:71650ms step_avg:178.68ms step:412/5100 train_loss:3.9804 train_time:71826ms step_avg:178.67ms step:413/5100 train_loss:4.4052 train_time:72002ms step_avg:178.67ms step:414/5100 train_loss:3.8528 train_time:72178ms step_avg:178.66ms step:415/5100 train_loss:4.2238 train_time:72354ms step_avg:178.65ms step:416/5100 train_loss:3.9783 train_time:72530ms step_avg:178.65ms step:417/5100 train_loss:3.9719 train_time:72706ms step_avg:178.64ms step:418/5100 train_loss:4.1763 train_time:72882ms step_avg:178.63ms step:419/5100 train_loss:3.9039 train_time:73060ms step_avg:178.63ms step:420/5100 train_loss:4.0124 train_time:73237ms step_avg:178.63ms step:421/5100 train_loss:3.9444 train_time:73414ms step_avg:178.62ms step:422/5100 train_loss:3.8631 train_time:73589ms step_avg:178.61ms step:423/5100 train_loss:3.9975 train_time:73765ms step_avg:178.61ms step:424/5100 train_loss:4.0903 train_time:73941ms step_avg:178.60ms step:425/5100 train_loss:3.8450 train_time:74117ms step_avg:178.59ms step:426/5100 train_loss:4.0260 train_time:74293ms step_avg:178.59ms step:427/5100 train_loss:3.9031 train_time:74468ms step_avg:178.58ms step:428/5100 train_loss:4.1161 train_time:74644ms step_avg:178.58ms step:429/5100 train_loss:4.0311 train_time:74820ms step_avg:178.57ms step:430/5100 train_loss:3.9688 train_time:74996ms step_avg:178.56ms step:431/5100 train_loss:3.9387 train_time:75171ms step_avg:178.55ms step:432/5100 train_loss:3.8476 train_time:75347ms step_avg:178.55ms step:433/5100 train_loss:3.9787 train_time:75524ms step_avg:178.54ms step:434/5100 train_loss:4.0403 train_time:75699ms step_avg:178.54ms step:435/5100 train_loss:3.9796 train_time:75876ms step_avg:178.53ms step:436/5100 train_loss:4.0276 train_time:76051ms step_avg:178.52ms step:437/5100 train_loss:4.0453 train_time:76227ms step_avg:178.52ms step:438/5100 train_loss:3.9220 train_time:76403ms step_avg:178.51ms step:439/5100 train_loss:3.9371 train_time:76579ms step_avg:178.51ms step:440/5100 train_loss:3.9143 train_time:76755ms step_avg:178.50ms step:441/5100 train_loss:4.0932 train_time:76931ms step_avg:178.49ms step:442/5100 train_loss:3.9810 train_time:77109ms step_avg:178.49ms step:443/5100 train_loss:3.9660 train_time:77283ms step_avg:178.48ms step:444/5100 train_loss:3.8542 train_time:77459ms step_avg:178.48ms step:445/5100 train_loss:4.1184 train_time:77635ms step_avg:178.47ms step:446/5100 train_loss:4.0536 train_time:77811ms step_avg:178.46ms step:447/5100 train_loss:4.0509 train_time:77987ms step_avg:178.46ms step:448/5100 train_loss:3.9700 train_time:78162ms step_avg:178.45ms step:449/5100 train_loss:4.0701 train_time:78339ms step_avg:178.45ms step:450/5100 train_loss:3.8881 train_time:78514ms step_avg:178.44ms step:451/5100 train_loss:3.9325 train_time:78691ms step_avg:178.44ms step:452/5100 train_loss:3.7946 train_time:78867ms step_avg:178.43ms step:453/5100 train_loss:3.9125 train_time:79042ms step_avg:178.42ms step:454/5100 train_loss:3.8894 train_time:79218ms step_avg:178.42ms step:455/5100 train_loss:3.8431 train_time:79394ms step_avg:178.41ms step:456/5100 train_loss:4.0640 train_time:79570ms step_avg:178.41ms step:457/5100 train_loss:3.9311 train_time:79747ms step_avg:178.40ms step:458/5100 train_loss:4.0070 train_time:79922ms step_avg:178.40ms step:459/5100 train_loss:4.0460 train_time:80098ms step_avg:178.39ms step:460/5100 train_loss:3.8477 train_time:80273ms step_avg:178.39ms step:461/5100 train_loss:4.0130 train_time:80449ms step_avg:178.38ms step:462/5100 train_loss:3.9171 train_time:80625ms step_avg:178.37ms step:463/5100 train_loss:3.9309 train_time:80801ms step_avg:178.37ms step:464/5100 train_loss:3.9856 train_time:80977ms step_avg:178.36ms step:465/5100 train_loss:3.9250 train_time:81153ms step_avg:178.36ms step:466/5100 train_loss:3.9311 train_time:81330ms step_avg:178.35ms step:467/5100 train_loss:4.0260 train_time:81507ms step_avg:178.35ms step:468/5100 train_loss:4.0369 train_time:81682ms step_avg:178.35ms step:469/5100 train_loss:4.0104 train_time:81858ms step_avg:178.34ms step:470/5100 train_loss:3.9015 train_time:82035ms step_avg:178.34ms step:471/5100 train_loss:3.9822 train_time:82212ms step_avg:178.33ms step:472/5100 train_loss:4.0404 train_time:82389ms step_avg:178.33ms step:473/5100 train_loss:3.9839 train_time:82566ms step_avg:178.33ms step:474/5100 train_loss:3.9352 train_time:82741ms step_avg:178.32ms step:475/5100 train_loss:3.7898 train_time:82917ms step_avg:178.32ms step:476/5100 train_loss:4.2312 train_time:83093ms step_avg:178.31ms step:477/5100 train_loss:3.9809 train_time:83270ms step_avg:178.31ms step:478/5100 train_loss:3.7910 train_time:83446ms step_avg:178.30ms step:479/5100 train_loss:4.0216 train_time:83622ms step_avg:178.30ms step:480/5100 train_loss:3.9827 train_time:83798ms step_avg:178.29ms step:481/5100 train_loss:4.1206 train_time:83974ms step_avg:178.29ms step:482/5100 train_loss:3.9301 train_time:84150ms step_avg:178.28ms step:483/5100 train_loss:3.7330 train_time:84326ms step_avg:178.28ms step:484/5100 train_loss:4.0216 train_time:84502ms step_avg:178.27ms step:485/5100 train_loss:3.8728 train_time:84678ms step_avg:178.27ms step:486/5100 train_loss:3.8822 train_time:84854ms step_avg:178.26ms step:487/5100 train_loss:3.8154 train_time:85030ms step_avg:178.26ms step:488/5100 train_loss:3.8778 train_time:85206ms step_avg:178.26ms step:489/5100 train_loss:4.0819 train_time:85382ms step_avg:178.25ms step:490/5100 train_loss:3.9244 train_time:85558ms step_avg:178.25ms step:491/5100 train_loss:3.8150 train_time:85736ms step_avg:178.25ms step:492/5100 train_loss:3.8308 train_time:85911ms step_avg:178.24ms step:493/5100 train_loss:3.9448 train_time:86087ms step_avg:178.23ms step:494/5100 train_loss:3.7912 train_time:86263ms step_avg:178.23ms step:495/5100 train_loss:3.9244 train_time:86438ms step_avg:178.22ms step:496/5100 train_loss:3.8618 train_time:86614ms step_avg:178.22ms step:497/5100 train_loss:3.7519 train_time:86791ms step_avg:178.21ms step:498/5100 train_loss:3.9380 train_time:86967ms step_avg:178.21ms step:499/5100 train_loss:4.0106 train_time:87142ms step_avg:178.20ms step:500/5100 train_loss:4.0431 train_time:87317ms step_avg:178.20ms step:500/5100 val_loss:3.9188 train_time:87317ms step_avg:178.20ms step:501/5100 train_loss:3.9502 train_time:87501ms step_avg:178.21ms step:502/5100 train_loss:4.0077 train_time:87677ms step_avg:178.21ms step:503/5100 train_loss:3.9455 train_time:87854ms step_avg:178.20ms step:504/5100 train_loss:3.9953 train_time:88030ms step_avg:178.20ms step:505/5100 train_loss:3.9433 train_time:88206ms step_avg:178.19ms step:506/5100 train_loss:4.0279 train_time:88383ms step_avg:178.19ms step:507/5100 train_loss:3.8603 train_time:88559ms step_avg:178.19ms step:508/5100 train_loss:3.9701 train_time:88736ms step_avg:178.18ms step:509/5100 train_loss:4.0460 train_time:88912ms step_avg:178.18ms step:510/5100 train_loss:3.9859 train_time:89088ms step_avg:178.18ms step:511/5100 train_loss:3.7874 train_time:89264ms step_avg:178.17ms step:512/5100 train_loss:3.9935 train_time:89441ms step_avg:178.17ms step:513/5100 train_loss:3.9401 train_time:89616ms step_avg:178.16ms step:514/5100 train_loss:3.8890 train_time:89793ms step_avg:178.16ms step:515/5100 train_loss:3.9833 train_time:89968ms step_avg:178.16ms step:516/5100 train_loss:3.9539 train_time:90144ms step_avg:178.15ms step:517/5100 train_loss:4.2907 train_time:90321ms step_avg:178.15ms step:518/5100 train_loss:3.8917 train_time:90497ms step_avg:178.14ms step:519/5100 train_loss:3.9915 train_time:90673ms step_avg:178.14ms step:520/5100 train_loss:3.8904 train_time:90848ms step_avg:178.13ms step:521/5100 train_loss:3.8882 train_time:91025ms step_avg:178.13ms step:522/5100 train_loss:3.8509 train_time:91202ms step_avg:178.13ms step:523/5100 train_loss:3.8554 train_time:91379ms step_avg:178.13ms step:524/5100 train_loss:4.4753 train_time:91555ms step_avg:178.12ms step:525/5100 train_loss:3.9501 train_time:91731ms step_avg:178.12ms step:526/5100 train_loss:3.8923 train_time:91907ms step_avg:178.11ms step:527/5100 train_loss:3.9072 train_time:92083ms step_avg:178.11ms step:528/5100 train_loss:3.8605 train_time:92259ms step_avg:178.11ms step:529/5100 train_loss:3.8397 train_time:92436ms step_avg:178.10ms step:530/5100 train_loss:4.0584 train_time:92612ms step_avg:178.10ms step:531/5100 train_loss:3.8426 train_time:92788ms step_avg:178.10ms step:532/5100 train_loss:4.1246 train_time:92964ms step_avg:178.09ms step:533/5100 train_loss:3.9433 train_time:93140ms step_avg:178.09ms step:534/5100 train_loss:3.8672 train_time:93316ms step_avg:178.08ms step:535/5100 train_loss:3.8882 train_time:93492ms step_avg:178.08ms step:536/5100 train_loss:3.8200 train_time:93667ms step_avg:178.07ms step:537/5100 train_loss:3.9513 train_time:93843ms step_avg:178.07ms step:538/5100 train_loss:3.9414 train_time:94020ms step_avg:178.07ms step:539/5100 train_loss:3.8408 train_time:94196ms step_avg:178.06ms step:540/5100 train_loss:4.3288 train_time:94372ms step_avg:178.06ms step:541/5100 train_loss:3.8795 train_time:94548ms step_avg:178.06ms step:542/5100 train_loss:3.9896 train_time:94723ms step_avg:178.05ms step:543/5100 train_loss:3.8220 train_time:94901ms step_avg:178.05ms step:544/5100 train_loss:3.7910 train_time:95076ms step_avg:178.05ms step:545/5100 train_loss:3.8777 train_time:95253ms step_avg:178.04ms step:546/5100 train_loss:3.8054 train_time:95429ms step_avg:178.04ms step:547/5100 train_loss:3.8444 train_time:95605ms step_avg:178.04ms step:548/5100 train_loss:3.8558 train_time:95781ms step_avg:178.03ms step:549/5100 train_loss:3.8342 train_time:95958ms step_avg:178.03ms step:550/5100 train_loss:3.9301 train_time:96134ms step_avg:178.03ms step:551/5100 train_loss:3.8126 train_time:96313ms step_avg:178.03ms step:552/5100 train_loss:3.8332 train_time:96487ms step_avg:178.02ms step:553/5100 train_loss:4.1573 train_time:96663ms step_avg:178.02ms step:554/5100 train_loss:3.9562 train_time:96839ms step_avg:178.01ms step:555/5100 train_loss:3.9223 train_time:97016ms step_avg:178.01ms step:556/5100 train_loss:3.8623 train_time:97192ms step_avg:178.01ms step:557/5100 train_loss:3.8957 train_time:97368ms step_avg:178.00ms step:558/5100 train_loss:3.5677 train_time:97544ms step_avg:178.00ms step:559/5100 train_loss:3.8166 train_time:97720ms step_avg:178.00ms step:560/5100 train_loss:3.8610 train_time:97897ms step_avg:177.99ms step:561/5100 train_loss:3.9043 train_time:98072ms step_avg:177.99ms step:562/5100 train_loss:3.8105 train_time:98249ms step_avg:177.99ms step:563/5100 train_loss:3.7575 train_time:98425ms step_avg:177.98ms step:564/5100 train_loss:3.9658 train_time:98602ms step_avg:177.98ms step:565/5100 train_loss:3.7789 train_time:98778ms step_avg:177.98ms step:566/5100 train_loss:3.8929 train_time:98954ms step_avg:177.98ms step:567/5100 train_loss:3.8315 train_time:99361ms step_avg:178.39ms step:568/5100 train_loss:3.7933 train_time:99540ms step_avg:178.39ms step:569/5100 train_loss:3.8881 train_time:99714ms step_avg:178.38ms step:570/5100 train_loss:3.8603 train_time:100136ms step_avg:178.81ms step:571/5100 train_loss:3.8846 train_time:100313ms step_avg:178.81ms step:572/5100 train_loss:3.9689 train_time:100489ms step_avg:178.81ms step:573/5100 train_loss:3.9245 train_time:100664ms step_avg:178.80ms step:574/5100 train_loss:3.9269 train_time:100840ms step_avg:178.79ms step:575/5100 train_loss:3.9786 train_time:101016ms step_avg:178.79ms step:576/5100 train_loss:3.9338 train_time:101191ms step_avg:178.78ms step:577/5100 train_loss:3.9538 train_time:101367ms step_avg:178.78ms step:578/5100 train_loss:3.8873 train_time:101543ms step_avg:178.77ms step:579/5100 train_loss:3.8800 train_time:101719ms step_avg:178.77ms step:580/5100 train_loss:3.8628 train_time:101895ms step_avg:178.76ms step:581/5100 train_loss:3.8061 train_time:102070ms step_avg:178.76ms step:582/5100 train_loss:3.8292 train_time:102246ms step_avg:178.75ms step:583/5100 train_loss:4.0653 train_time:102423ms step_avg:178.75ms step:584/5100 train_loss:3.8324 train_time:102601ms step_avg:178.75ms step:585/5100 train_loss:3.7893 train_time:102777ms step_avg:178.74ms step:586/5100 train_loss:3.9812 train_time:102953ms step_avg:178.74ms step:587/5100 train_loss:3.7345 train_time:103128ms step_avg:178.73ms step:588/5100 train_loss:3.8702 train_time:103304ms step_avg:178.73ms step:589/5100 train_loss:3.8562 train_time:103480ms step_avg:178.72ms step:590/5100 train_loss:4.2108 train_time:103657ms step_avg:178.72ms step:591/5100 train_loss:3.9837 train_time:103833ms step_avg:178.71ms step:592/5100 train_loss:3.7251 train_time:104009ms step_avg:178.71ms step:593/5100 train_loss:3.7382 train_time:104184ms step_avg:178.70ms step:594/5100 train_loss:3.7307 train_time:104361ms step_avg:178.70ms step:595/5100 train_loss:3.7701 train_time:104537ms step_avg:178.70ms step:596/5100 train_loss:4.1275 train_time:104712ms step_avg:178.69ms step:597/5100 train_loss:3.8481 train_time:104887ms step_avg:178.68ms step:598/5100 train_loss:3.7841 train_time:105063ms step_avg:178.68ms step:599/5100 train_loss:3.8644 train_time:105240ms step_avg:178.68ms step:600/5100 train_loss:3.6791 train_time:105417ms step_avg:178.67ms step:601/5100 train_loss:3.8077 train_time:105593ms step_avg:178.67ms step:602/5100 train_loss:3.8333 train_time:105768ms step_avg:178.66ms step:603/5100 train_loss:3.8581 train_time:105944ms step_avg:178.66ms step:604/5100 train_loss:3.9846 train_time:106120ms step_avg:178.65ms step:605/5100 train_loss:3.8375 train_time:106296ms step_avg:178.65ms step:606/5100 train_loss:3.8197 train_time:106472ms step_avg:178.64ms step:607/5100 train_loss:3.7651 train_time:106648ms step_avg:178.64ms step:608/5100 train_loss:4.0189 train_time:106825ms step_avg:178.64ms step:609/5100 train_loss:3.8548 train_time:107001ms step_avg:178.63ms step:610/5100 train_loss:3.8221 train_time:107177ms step_avg:178.63ms step:611/5100 train_loss:3.9189 train_time:107353ms step_avg:178.62ms step:612/5100 train_loss:3.8258 train_time:107529ms step_avg:178.62ms step:613/5100 train_loss:3.8080 train_time:107706ms step_avg:178.62ms step:614/5100 train_loss:3.9684 train_time:107883ms step_avg:178.61ms step:615/5100 train_loss:3.9376 train_time:108059ms step_avg:178.61ms step:616/5100 train_loss:3.8930 train_time:108235ms step_avg:178.61ms step:617/5100 train_loss:3.8240 train_time:108411ms step_avg:178.60ms step:618/5100 train_loss:3.7706 train_time:108586ms step_avg:178.60ms step:619/5100 train_loss:3.8839 train_time:108763ms step_avg:178.59ms step:620/5100 train_loss:3.7825 train_time:108939ms step_avg:178.59ms step:621/5100 train_loss:3.7916 train_time:109115ms step_avg:178.58ms step:622/5100 train_loss:4.1079 train_time:109291ms step_avg:178.58ms step:623/5100 train_loss:3.7936 train_time:109466ms step_avg:178.57ms step:624/5100 train_loss:3.8199 train_time:109643ms step_avg:178.57ms step:625/5100 train_loss:3.9023 train_time:109819ms step_avg:178.57ms step:625/5100 val_loss:3.8316 train_time:109819ms step_avg:178.57ms step:626/5100 train_loss:3.9217 train_time:110005ms step_avg:178.58ms step:627/5100 train_loss:3.9544 train_time:110181ms step_avg:178.57ms step:628/5100 train_loss:3.9287 train_time:110357ms step_avg:178.57ms step:629/5100 train_loss:3.9728 train_time:110533ms step_avg:178.57ms step:630/5100 train_loss:3.7960 train_time:110710ms step_avg:178.56ms step:631/5100 train_loss:3.9224 train_time:110886ms step_avg:178.56ms step:632/5100 train_loss:3.9588 train_time:111062ms step_avg:178.56ms step:633/5100 train_loss:3.8611 train_time:111238ms step_avg:178.55ms step:634/5100 train_loss:3.7921 train_time:111415ms step_avg:178.55ms step:635/5100 train_loss:3.8929 train_time:111590ms step_avg:178.54ms step:636/5100 train_loss:4.1425 train_time:111769ms step_avg:178.54ms step:637/5100 train_loss:3.7385 train_time:111943ms step_avg:178.54ms step:638/5100 train_loss:3.5656 train_time:112121ms step_avg:178.54ms step:639/5100 train_loss:3.7848 train_time:112297ms step_avg:178.53ms step:640/5100 train_loss:3.8227 train_time:112473ms step_avg:178.53ms step:641/5100 train_loss:3.7758 train_time:112650ms step_avg:178.53ms step:642/5100 train_loss:3.7817 train_time:112826ms step_avg:178.52ms step:643/5100 train_loss:3.8214 train_time:113001ms step_avg:178.52ms step:644/5100 train_loss:3.8391 train_time:113178ms step_avg:178.51ms step:645/5100 train_loss:3.7587 train_time:113364ms step_avg:178.53ms step:646/5100 train_loss:3.9784 train_time:113538ms step_avg:178.52ms step:647/5100 train_loss:3.8779 train_time:113714ms step_avg:178.52ms step:648/5100 train_loss:3.8735 train_time:113892ms step_avg:178.51ms step:649/5100 train_loss:3.9024 train_time:114068ms step_avg:178.51ms step:650/5100 train_loss:3.9675 train_time:114243ms step_avg:178.51ms step:651/5100 train_loss:3.8205 train_time:114421ms step_avg:178.50ms step:652/5100 train_loss:3.9638 train_time:114597ms step_avg:178.50ms step:653/5100 train_loss:3.7917 train_time:114773ms step_avg:178.50ms step:654/5100 train_loss:3.8630 train_time:114949ms step_avg:178.49ms step:655/5100 train_loss:3.6263 train_time:115124ms step_avg:178.49ms step:656/5100 train_loss:3.7785 train_time:115300ms step_avg:178.48ms step:657/5100 train_loss:3.7825 train_time:115477ms step_avg:178.48ms step:658/5100 train_loss:3.7134 train_time:115653ms step_avg:178.48ms step:659/5100 train_loss:3.8927 train_time:115831ms step_avg:178.48ms step:660/5100 train_loss:3.7916 train_time:116008ms step_avg:178.47ms step:661/5100 train_loss:3.8882 train_time:116183ms step_avg:178.47ms step:662/5100 train_loss:3.9538 train_time:116361ms step_avg:178.47ms step:663/5100 train_loss:3.8716 train_time:116537ms step_avg:178.46ms step:664/5100 train_loss:3.7479 train_time:116713ms step_avg:178.46ms step:665/5100 train_loss:3.8379 train_time:116889ms step_avg:178.46ms step:666/5100 train_loss:3.7019 train_time:117066ms step_avg:178.45ms step:667/5100 train_loss:3.9954 train_time:117241ms step_avg:178.45ms step:668/5100 train_loss:3.8234 train_time:117418ms step_avg:178.45ms step:669/5100 train_loss:3.8341 train_time:117594ms step_avg:178.44ms step:670/5100 train_loss:3.6867 train_time:117771ms step_avg:178.44ms step:671/5100 train_loss:3.8028 train_time:117947ms step_avg:178.44ms step:672/5100 train_loss:3.7575 train_time:118123ms step_avg:178.43ms step:673/5100 train_loss:3.7780 train_time:118300ms step_avg:178.43ms step:674/5100 train_loss:4.0540 train_time:118476ms step_avg:178.43ms step:675/5100 train_loss:3.8469 train_time:118653ms step_avg:178.43ms step:676/5100 train_loss:3.9212 train_time:118829ms step_avg:178.42ms step:677/5100 train_loss:3.6947 train_time:119004ms step_avg:178.42ms step:678/5100 train_loss:3.7985 train_time:119180ms step_avg:178.41ms step:679/5100 train_loss:3.7390 train_time:119358ms step_avg:178.41ms step:680/5100 train_loss:3.8888 train_time:119534ms step_avg:178.41ms step:681/5100 train_loss:3.7878 train_time:119711ms step_avg:178.41ms step:682/5100 train_loss:3.8174 train_time:119888ms step_avg:178.40ms step:683/5100 train_loss:3.8879 train_time:120063ms step_avg:178.40ms step:684/5100 train_loss:3.9420 train_time:120239ms step_avg:178.40ms step:685/5100 train_loss:3.8344 train_time:120416ms step_avg:178.39ms step:686/5100 train_loss:3.9065 train_time:120592ms step_avg:178.39ms step:687/5100 train_loss:3.8296 train_time:120768ms step_avg:178.39ms step:688/5100 train_loss:3.8819 train_time:120944ms step_avg:178.38ms step:689/5100 train_loss:3.5134 train_time:121120ms step_avg:178.38ms step:690/5100 train_loss:3.6195 train_time:121296ms step_avg:178.38ms step:691/5100 train_loss:3.7576 train_time:121473ms step_avg:178.37ms step:692/5100 train_loss:3.6331 train_time:121649ms step_avg:178.37ms step:693/5100 train_loss:3.8461 train_time:121825ms step_avg:178.37ms step:694/5100 train_loss:3.8628 train_time:122000ms step_avg:178.36ms step:695/5100 train_loss:3.7531 train_time:122177ms step_avg:178.36ms step:696/5100 train_loss:3.7422 train_time:122353ms step_avg:178.36ms step:697/5100 train_loss:4.0591 train_time:122529ms step_avg:178.35ms step:698/5100 train_loss:3.7978 train_time:122705ms step_avg:178.35ms step:699/5100 train_loss:3.8505 train_time:122881ms step_avg:178.35ms step:700/5100 train_loss:4.0068 train_time:123057ms step_avg:178.34ms step:701/5100 train_loss:3.7724 train_time:123234ms step_avg:178.34ms step:702/5100 train_loss:3.7374 train_time:123412ms step_avg:178.34ms step:703/5100 train_loss:3.7262 train_time:123588ms step_avg:178.34ms step:704/5100 train_loss:3.6785 train_time:123764ms step_avg:178.33ms step:705/5100 train_loss:3.7686 train_time:123940ms step_avg:178.33ms step:706/5100 train_loss:3.7649 train_time:124116ms step_avg:178.33ms step:707/5100 train_loss:3.7761 train_time:124291ms step_avg:178.32ms step:708/5100 train_loss:3.8495 train_time:124468ms step_avg:178.32ms step:709/5100 train_loss:3.7945 train_time:124644ms step_avg:178.32ms step:710/5100 train_loss:3.7774 train_time:124820ms step_avg:178.31ms step:711/5100 train_loss:3.7487 train_time:124998ms step_avg:178.31ms step:712/5100 train_loss:3.7922 train_time:125173ms step_avg:178.31ms step:713/5100 train_loss:3.8492 train_time:125350ms step_avg:178.31ms step:714/5100 train_loss:3.8569 train_time:125527ms step_avg:178.31ms step:715/5100 train_loss:3.7710 train_time:125703ms step_avg:178.30ms step:716/5100 train_loss:3.7724 train_time:125880ms step_avg:178.30ms step:717/5100 train_loss:3.7843 train_time:126056ms step_avg:178.30ms step:718/5100 train_loss:3.9345 train_time:126232ms step_avg:178.29ms step:719/5100 train_loss:3.7964 train_time:126409ms step_avg:178.29ms step:720/5100 train_loss:3.8656 train_time:126585ms step_avg:178.29ms step:721/5100 train_loss:4.0261 train_time:126760ms step_avg:178.28ms step:722/5100 train_loss:3.6642 train_time:126937ms step_avg:178.28ms step:723/5100 train_loss:3.9247 train_time:127113ms step_avg:178.28ms step:724/5100 train_loss:3.9793 train_time:127289ms step_avg:178.28ms step:725/5100 train_loss:3.7649 train_time:127465ms step_avg:178.27ms step:726/5100 train_loss:3.8481 train_time:127641ms step_avg:178.27ms step:727/5100 train_loss:3.7442 train_time:127817ms step_avg:178.27ms step:728/5100 train_loss:3.7619 train_time:127993ms step_avg:178.26ms step:729/5100 train_loss:3.9435 train_time:128170ms step_avg:178.26ms step:730/5100 train_loss:3.8820 train_time:128347ms step_avg:178.26ms step:731/5100 train_loss:3.8790 train_time:128522ms step_avg:178.26ms step:732/5100 train_loss:3.7641 train_time:128699ms step_avg:178.25ms step:733/5100 train_loss:3.7867 train_time:128876ms step_avg:178.25ms step:734/5100 train_loss:4.0305 train_time:129052ms step_avg:178.25ms step:735/5100 train_loss:3.7614 train_time:129227ms step_avg:178.24ms step:736/5100 train_loss:3.8175 train_time:129404ms step_avg:178.24ms step:737/5100 train_loss:3.9439 train_time:129580ms step_avg:178.24ms step:738/5100 train_loss:3.8647 train_time:129756ms step_avg:178.24ms step:739/5100 train_loss:3.8005 train_time:129932ms step_avg:178.23ms step:740/5100 train_loss:3.7041 train_time:130108ms step_avg:178.23ms step:741/5100 train_loss:4.3426 train_time:130284ms step_avg:178.23ms step:742/5100 train_loss:3.6990 train_time:130460ms step_avg:178.22ms step:743/5100 train_loss:3.7837 train_time:130636ms step_avg:178.22ms step:744/5100 train_loss:3.7852 train_time:130813ms step_avg:178.22ms step:745/5100 train_loss:3.8413 train_time:130989ms step_avg:178.22ms step:746/5100 train_loss:3.8140 train_time:131165ms step_avg:178.21ms step:747/5100 train_loss:3.7991 train_time:131342ms step_avg:178.21ms step:748/5100 train_loss:3.8339 train_time:131518ms step_avg:178.21ms step:749/5100 train_loss:3.7628 train_time:131694ms step_avg:178.21ms step:750/5100 train_loss:3.7665 train_time:131870ms step_avg:178.20ms step:750/5100 val_loss:3.7724 train_time:131870ms step_avg:178.20ms step:751/5100 train_loss:3.8001 train_time:132054ms step_avg:178.21ms step:752/5100 train_loss:3.7614 train_time:132230ms step_avg:178.21ms step:753/5100 train_loss:3.7970 train_time:132407ms step_avg:178.21ms step:754/5100 train_loss:3.8242 train_time:132583ms step_avg:178.20ms step:755/5100 train_loss:3.7891 train_time:132759ms step_avg:178.20ms step:756/5100 train_loss:3.8573 train_time:133165ms step_avg:178.51ms step:757/5100 train_loss:3.6875 train_time:133342ms step_avg:178.50ms step:758/5100 train_loss:3.9319 train_time:133518ms step_avg:178.50ms step:759/5100 train_loss:3.8500 train_time:133695ms step_avg:178.50ms step:760/5100 train_loss:3.7792 train_time:134113ms step_avg:178.82ms step:761/5100 train_loss:3.8854 train_time:134290ms step_avg:178.82ms step:762/5100 train_loss:3.5989 train_time:134467ms step_avg:178.81ms step:763/5100 train_loss:3.7536 train_time:134644ms step_avg:178.81ms step:764/5100 train_loss:3.8662 train_time:134819ms step_avg:178.81ms step:765/5100 train_loss:3.5227 train_time:134996ms step_avg:178.80ms step:766/5100 train_loss:3.9456 train_time:135172ms step_avg:178.80ms step:767/5100 train_loss:3.7913 train_time:135348ms step_avg:178.79ms step:768/5100 train_loss:3.7574 train_time:135524ms step_avg:178.79ms step:769/5100 train_loss:3.7791 train_time:135700ms step_avg:178.79ms step:770/5100 train_loss:3.7953 train_time:135876ms step_avg:178.78ms step:771/5100 train_loss:3.8515 train_time:136051ms step_avg:178.78ms step:772/5100 train_loss:4.0771 train_time:136228ms step_avg:178.78ms step:773/5100 train_loss:3.6565 train_time:136404ms step_avg:178.77ms step:774/5100 train_loss:3.8489 train_time:136579ms step_avg:178.77ms step:775/5100 train_loss:3.8360 train_time:136756ms step_avg:178.77ms step:776/5100 train_loss:3.8125 train_time:136932ms step_avg:178.76ms step:777/5100 train_loss:3.6126 train_time:137108ms step_avg:178.76ms step:778/5100 train_loss:3.6064 train_time:137285ms step_avg:178.76ms step:779/5100 train_loss:3.6798 train_time:137461ms step_avg:178.75ms step:780/5100 train_loss:3.7733 train_time:137636ms step_avg:178.75ms step:781/5100 train_loss:3.7998 train_time:137812ms step_avg:178.74ms step:782/5100 train_loss:3.8581 train_time:137988ms step_avg:178.74ms step:783/5100 train_loss:3.7742 train_time:138164ms step_avg:178.74ms step:784/5100 train_loss:3.7736 train_time:138340ms step_avg:178.73ms step:785/5100 train_loss:3.7850 train_time:138515ms step_avg:178.73ms step:786/5100 train_loss:3.7540 train_time:138691ms step_avg:178.73ms step:787/5100 train_loss:3.6498 train_time:138868ms step_avg:178.72ms step:788/5100 train_loss:3.9576 train_time:139044ms step_avg:178.72ms step:789/5100 train_loss:3.6972 train_time:139220ms step_avg:178.72ms step:790/5100 train_loss:3.7679 train_time:139396ms step_avg:178.71ms step:791/5100 train_loss:3.8186 train_time:139571ms step_avg:178.71ms step:792/5100 train_loss:3.9593 train_time:139748ms step_avg:178.71ms step:793/5100 train_loss:3.9631 train_time:139925ms step_avg:178.70ms step:794/5100 train_loss:3.6705 train_time:140101ms step_avg:178.70ms step:795/5100 train_loss:3.8011 train_time:140276ms step_avg:178.70ms step:796/5100 train_loss:3.8576 train_time:140452ms step_avg:178.69ms step:797/5100 train_loss:3.9622 train_time:140628ms step_avg:178.69ms step:798/5100 train_loss:3.7095 train_time:140805ms step_avg:178.69ms step:799/5100 train_loss:3.8598 train_time:140981ms step_avg:178.68ms step:800/5100 train_loss:3.7532 train_time:141157ms step_avg:178.68ms step:801/5100 train_loss:3.7409 train_time:141333ms step_avg:178.68ms step:802/5100 train_loss:3.8416 train_time:141509ms step_avg:178.67ms step:803/5100 train_loss:3.6935 train_time:141685ms step_avg:178.67ms step:804/5100 train_loss:3.7270 train_time:141860ms step_avg:178.67ms step:805/5100 train_loss:3.8300 train_time:142039ms step_avg:178.66ms step:806/5100 train_loss:3.7338 train_time:142214ms step_avg:178.66ms step:807/5100 train_loss:3.7431 train_time:142389ms step_avg:178.66ms step:808/5100 train_loss:3.8428 train_time:142565ms step_avg:178.65ms step:809/5100 train_loss:3.7615 train_time:142743ms step_avg:178.65ms step:810/5100 train_loss:3.6810 train_time:142918ms step_avg:178.65ms step:811/5100 train_loss:3.7670 train_time:143094ms step_avg:178.64ms step:812/5100 train_loss:3.7987 train_time:143270ms step_avg:178.64ms step:813/5100 train_loss:3.7931 train_time:143446ms step_avg:178.64ms step:814/5100 train_loss:3.8273 train_time:143622ms step_avg:178.63ms step:815/5100 train_loss:3.7766 train_time:143799ms step_avg:178.63ms step:816/5100 train_loss:3.7595 train_time:143975ms step_avg:178.63ms step:817/5100 train_loss:3.8616 train_time:144151ms step_avg:178.63ms step:818/5100 train_loss:3.9603 train_time:144327ms step_avg:178.62ms step:819/5100 train_loss:3.7211 train_time:144502ms step_avg:178.62ms step:820/5100 train_loss:3.9185 train_time:144678ms step_avg:178.62ms step:821/5100 train_loss:3.7057 train_time:144855ms step_avg:178.61ms step:822/5100 train_loss:3.7474 train_time:145031ms step_avg:178.61ms step:823/5100 train_loss:3.8715 train_time:145207ms step_avg:178.61ms step:824/5100 train_loss:3.7876 train_time:145383ms step_avg:178.60ms step:825/5100 train_loss:3.7095 train_time:145559ms step_avg:178.60ms step:826/5100 train_loss:3.8119 train_time:145735ms step_avg:178.60ms step:827/5100 train_loss:3.7025 train_time:145910ms step_avg:178.59ms step:828/5100 train_loss:3.9294 train_time:146087ms step_avg:178.59ms step:829/5100 train_loss:3.8195 train_time:146263ms step_avg:178.59ms step:830/5100 train_loss:3.8809 train_time:146440ms step_avg:178.58ms step:831/5100 train_loss:3.7290 train_time:146615ms step_avg:178.58ms step:832/5100 train_loss:3.7821 train_time:146791ms step_avg:178.58ms step:833/5100 train_loss:3.7153 train_time:146968ms step_avg:178.58ms step:834/5100 train_loss:3.8433 train_time:147144ms step_avg:178.57ms step:835/5100 train_loss:3.6812 train_time:147320ms step_avg:178.57ms step:836/5100 train_loss:3.6619 train_time:147497ms step_avg:178.57ms step:837/5100 train_loss:3.9268 train_time:147673ms step_avg:178.56ms step:838/5100 train_loss:3.6182 train_time:147849ms step_avg:178.56ms step:839/5100 train_loss:3.7861 train_time:148024ms step_avg:178.56ms step:840/5100 train_loss:3.6276 train_time:148201ms step_avg:178.56ms step:841/5100 train_loss:3.6720 train_time:148377ms step_avg:178.55ms step:842/5100 train_loss:3.7597 train_time:148552ms step_avg:178.55ms step:843/5100 train_loss:3.7748 train_time:148728ms step_avg:178.54ms step:844/5100 train_loss:3.7752 train_time:148904ms step_avg:178.54ms step:845/5100 train_loss:3.6248 train_time:149081ms step_avg:178.54ms step:846/5100 train_loss:3.8661 train_time:149255ms step_avg:178.54ms step:847/5100 train_loss:3.7251 train_time:149431ms step_avg:178.53ms step:848/5100 train_loss:3.6945 train_time:149607ms step_avg:178.53ms step:849/5100 train_loss:3.8303 train_time:149783ms step_avg:178.53ms step:850/5100 train_loss:3.6933 train_time:149958ms step_avg:178.52ms step:851/5100 train_loss:3.6424 train_time:150134ms step_avg:178.52ms step:852/5100 train_loss:3.9382 train_time:150310ms step_avg:178.52ms step:853/5100 train_loss:3.6484 train_time:150487ms step_avg:178.51ms step:854/5100 train_loss:3.7619 train_time:150663ms step_avg:178.51ms step:855/5100 train_loss:3.8423 train_time:150839ms step_avg:178.51ms step:856/5100 train_loss:3.7308 train_time:151014ms step_avg:178.50ms step:857/5100 train_loss:3.7455 train_time:151191ms step_avg:178.50ms step:858/5100 train_loss:3.7952 train_time:151367ms step_avg:178.50ms step:859/5100 train_loss:3.6844 train_time:151544ms step_avg:178.50ms step:860/5100 train_loss:3.7610 train_time:151719ms step_avg:178.49ms step:861/5100 train_loss:3.7959 train_time:151896ms step_avg:178.49ms step:862/5100 train_loss:3.8425 train_time:152071ms step_avg:178.49ms step:863/5100 train_loss:3.7868 train_time:152247ms step_avg:178.48ms step:864/5100 train_loss:3.7734 train_time:152424ms step_avg:178.48ms step:865/5100 train_loss:3.5996 train_time:152600ms step_avg:178.48ms step:866/5100 train_loss:3.7863 train_time:152776ms step_avg:178.48ms step:867/5100 train_loss:4.0655 train_time:152952ms step_avg:178.47ms step:868/5100 train_loss:3.6421 train_time:153128ms step_avg:178.47ms step:869/5100 train_loss:3.8358 train_time:153305ms step_avg:178.47ms step:870/5100 train_loss:3.8119 train_time:153482ms step_avg:178.47ms step:871/5100 train_loss:3.6456 train_time:153657ms step_avg:178.46ms step:872/5100 train_loss:3.6297 train_time:153843ms step_avg:178.47ms step:873/5100 train_loss:3.8635 train_time:154018ms step_avg:178.47ms step:874/5100 train_loss:3.6479 train_time:154194ms step_avg:178.46ms step:875/5100 train_loss:3.3732 train_time:154370ms step_avg:178.46ms step:875/5100 val_loss:3.7241 train_time:154370ms step_avg:178.46ms step:876/5100 train_loss:3.8410 train_time:154557ms step_avg:178.47ms step:877/5100 train_loss:3.6484 train_time:154733ms step_avg:178.47ms step:878/5100 train_loss:3.8252 train_time:154910ms step_avg:178.47ms step:879/5100 train_loss:3.6782 train_time:155086ms step_avg:178.46ms step:880/5100 train_loss:3.8633 train_time:155262ms step_avg:178.46ms step:881/5100 train_loss:3.5205 train_time:155438ms step_avg:178.46ms step:882/5100 train_loss:3.7010 train_time:155615ms step_avg:178.46ms step:883/5100 train_loss:3.8862 train_time:155791ms step_avg:178.45ms step:884/5100 train_loss:4.0451 train_time:155967ms step_avg:178.45ms step:885/5100 train_loss:3.7659 train_time:156143ms step_avg:178.45ms step:886/5100 train_loss:3.6809 train_time:156319ms step_avg:178.45ms step:887/5100 train_loss:3.7797 train_time:156494ms step_avg:178.44ms step:888/5100 train_loss:4.2812 train_time:156670ms step_avg:178.44ms step:889/5100 train_loss:4.0381 train_time:156846ms step_avg:178.44ms step:890/5100 train_loss:3.7178 train_time:157021ms step_avg:178.43ms step:891/5100 train_loss:3.7297 train_time:157198ms step_avg:178.43ms step:892/5100 train_loss:3.5595 train_time:157375ms step_avg:178.43ms step:893/5100 train_loss:3.9017 train_time:157550ms step_avg:178.43ms step:894/5100 train_loss:3.6344 train_time:157726ms step_avg:178.42ms step:895/5100 train_loss:3.8792 train_time:157902ms step_avg:178.42ms step:896/5100 train_loss:3.8940 train_time:158078ms step_avg:178.42ms step:897/5100 train_loss:3.6900 train_time:158255ms step_avg:178.42ms step:898/5100 train_loss:3.7329 train_time:158431ms step_avg:178.41ms step:899/5100 train_loss:3.7900 train_time:158607ms step_avg:178.41ms step:900/5100 train_loss:3.6762 train_time:158783ms step_avg:178.41ms step:901/5100 train_loss:3.6154 train_time:158959ms step_avg:178.40ms step:902/5100 train_loss:3.8266 train_time:159134ms step_avg:178.40ms step:903/5100 train_loss:3.8291 train_time:159313ms step_avg:178.40ms step:904/5100 train_loss:3.7323 train_time:159487ms step_avg:178.40ms step:905/5100 train_loss:3.7011 train_time:159664ms step_avg:178.40ms step:906/5100 train_loss:3.6940 train_time:159839ms step_avg:178.39ms step:907/5100 train_loss:3.9204 train_time:160015ms step_avg:178.39ms step:908/5100 train_loss:3.7093 train_time:160192ms step_avg:178.39ms step:909/5100 train_loss:3.7485 train_time:160368ms step_avg:178.38ms step:910/5100 train_loss:3.6563 train_time:160544ms step_avg:178.38ms step:911/5100 train_loss:3.7473 train_time:160720ms step_avg:178.38ms step:912/5100 train_loss:3.8180 train_time:160896ms step_avg:178.38ms step:913/5100 train_loss:3.8221 train_time:161072ms step_avg:178.37ms step:914/5100 train_loss:3.6847 train_time:161248ms step_avg:178.37ms step:915/5100 train_loss:3.9343 train_time:161424ms step_avg:178.37ms step:916/5100 train_loss:3.7254 train_time:161600ms step_avg:178.37ms step:917/5100 train_loss:3.8266 train_time:161777ms step_avg:178.36ms step:918/5100 train_loss:3.7991 train_time:161953ms step_avg:178.36ms step:919/5100 train_loss:5.0580 train_time:162128ms step_avg:178.36ms step:920/5100 train_loss:3.7205 train_time:162305ms step_avg:178.36ms step:921/5100 train_loss:3.7682 train_time:162481ms step_avg:178.35ms step:922/5100 train_loss:3.7315 train_time:162657ms step_avg:178.35ms step:923/5100 train_loss:3.7852 train_time:162833ms step_avg:178.35ms step:924/5100 train_loss:3.7944 train_time:163009ms step_avg:178.35ms step:925/5100 train_loss:3.8824 train_time:163185ms step_avg:178.34ms step:926/5100 train_loss:3.8587 train_time:163360ms step_avg:178.34ms step:927/5100 train_loss:3.7522 train_time:163537ms step_avg:178.34ms step:928/5100 train_loss:3.7438 train_time:163713ms step_avg:178.34ms step:929/5100 train_loss:3.9685 train_time:163890ms step_avg:178.34ms step:930/5100 train_loss:3.8116 train_time:164066ms step_avg:178.33ms step:931/5100 train_loss:3.5959 train_time:164242ms step_avg:178.33ms step:932/5100 train_loss:3.6870 train_time:164419ms step_avg:178.33ms step:933/5100 train_loss:3.8724 train_time:164595ms step_avg:178.33ms step:934/5100 train_loss:3.5886 train_time:164772ms step_avg:178.32ms step:935/5100 train_loss:3.7687 train_time:164947ms step_avg:178.32ms step:936/5100 train_loss:3.6460 train_time:165124ms step_avg:178.32ms step:937/5100 train_loss:3.7098 train_time:165300ms step_avg:178.32ms step:938/5100 train_loss:3.8074 train_time:165476ms step_avg:178.31ms step:939/5100 train_loss:3.7385 train_time:165653ms step_avg:178.31ms step:940/5100 train_loss:3.8965 train_time:165828ms step_avg:178.31ms step:941/5100 train_loss:3.6826 train_time:166007ms step_avg:178.31ms step:942/5100 train_loss:3.7473 train_time:166183ms step_avg:178.31ms step:943/5100 train_loss:3.5443 train_time:166359ms step_avg:178.31ms step:944/5100 train_loss:3.8951 train_time:166535ms step_avg:178.30ms step:945/5100 train_loss:3.6099 train_time:166944ms step_avg:178.55ms step:946/5100 train_loss:3.6235 train_time:167120ms step_avg:178.55ms step:947/5100 train_loss:5.2465 train_time:167295ms step_avg:178.54ms step:948/5100 train_loss:3.7941 train_time:167472ms step_avg:178.54ms step:949/5100 train_loss:3.6902 train_time:167648ms step_avg:178.54ms step:950/5100 train_loss:3.5960 train_time:168070ms step_avg:178.80ms step:951/5100 train_loss:3.6499 train_time:168248ms step_avg:178.80ms step:952/5100 train_loss:3.6064 train_time:168425ms step_avg:178.79ms step:953/5100 train_loss:3.6774 train_time:168601ms step_avg:178.79ms step:954/5100 train_loss:3.7513 train_time:168778ms step_avg:178.79ms step:955/5100 train_loss:3.6357 train_time:168955ms step_avg:178.79ms step:956/5100 train_loss:3.6710 train_time:169131ms step_avg:178.79ms step:957/5100 train_loss:3.6333 train_time:169307ms step_avg:178.78ms step:958/5100 train_loss:3.7005 train_time:169482ms step_avg:178.78ms step:959/5100 train_loss:3.6894 train_time:169658ms step_avg:178.78ms step:960/5100 train_loss:3.7117 train_time:169835ms step_avg:178.77ms step:961/5100 train_loss:3.5905 train_time:170010ms step_avg:178.77ms step:962/5100 train_loss:3.8492 train_time:170186ms step_avg:178.77ms step:963/5100 train_loss:3.8070 train_time:170362ms step_avg:178.76ms step:964/5100 train_loss:3.7099 train_time:170538ms step_avg:178.76ms step:965/5100 train_loss:3.6424 train_time:170715ms step_avg:178.76ms step:966/5100 train_loss:3.6797 train_time:170894ms step_avg:178.76ms step:967/5100 train_loss:3.9060 train_time:171068ms step_avg:178.75ms step:968/5100 train_loss:3.7298 train_time:171244ms step_avg:178.75ms step:969/5100 train_loss:3.7156 train_time:171420ms step_avg:178.75ms step:970/5100 train_loss:3.7727 train_time:171596ms step_avg:178.75ms step:971/5100 train_loss:3.5865 train_time:171772ms step_avg:178.74ms step:972/5100 train_loss:3.7431 train_time:171948ms step_avg:178.74ms step:973/5100 train_loss:3.6983 train_time:172124ms step_avg:178.74ms step:974/5100 train_loss:3.7400 train_time:172300ms step_avg:178.73ms step:975/5100 train_loss:3.8128 train_time:172477ms step_avg:178.73ms step:976/5100 train_loss:3.6847 train_time:172654ms step_avg:178.73ms step:977/5100 train_loss:3.8817 train_time:172830ms step_avg:178.73ms step:978/5100 train_loss:3.7644 train_time:173005ms step_avg:178.72ms step:979/5100 train_loss:3.6019 train_time:173181ms step_avg:178.72ms step:980/5100 train_loss:3.8791 train_time:173358ms step_avg:178.72ms step:981/5100 train_loss:3.6165 train_time:173534ms step_avg:178.72ms step:982/5100 train_loss:3.7841 train_time:173710ms step_avg:178.71ms step:983/5100 train_loss:3.7644 train_time:173886ms step_avg:178.71ms step:984/5100 train_loss:3.7663 train_time:174062ms step_avg:178.71ms step:985/5100 train_loss:3.7061 train_time:174238ms step_avg:178.71ms step:986/5100 train_loss:3.7931 train_time:174415ms step_avg:178.70ms step:987/5100 train_loss:3.6164 train_time:174592ms step_avg:178.70ms step:988/5100 train_loss:3.6945 train_time:174768ms step_avg:178.70ms step:989/5100 train_loss:3.7059 train_time:174944ms step_avg:178.70ms step:990/5100 train_loss:3.6294 train_time:175120ms step_avg:178.69ms step:991/5100 train_loss:3.8454 train_time:175297ms step_avg:178.69ms step:992/5100 train_loss:3.6656 train_time:175473ms step_avg:178.69ms step:993/5100 train_loss:3.6404 train_time:175650ms step_avg:178.69ms step:994/5100 train_loss:3.7154 train_time:175825ms step_avg:178.68ms step:995/5100 train_loss:3.8008 train_time:176001ms step_avg:178.68ms step:996/5100 train_loss:3.7442 train_time:176177ms step_avg:178.68ms step:997/5100 train_loss:3.6508 train_time:176354ms step_avg:178.68ms step:998/5100 train_loss:4.0033 train_time:176530ms step_avg:178.67ms step:999/5100 train_loss:3.6628 train_time:176707ms step_avg:178.67ms step:1000/5100 train_loss:3.7824 train_time:176882ms step_avg:178.67ms step:1000/5100 val_loss:3.6815 train_time:176882ms step_avg:178.67ms step:1001/5100 train_loss:3.6527 train_time:177066ms step_avg:178.67ms step:1002/5100 train_loss:3.7124 train_time:177242ms step_avg:178.67ms step:1003/5100 train_loss:3.5912 train_time:177419ms step_avg:178.67ms step:1004/5100 train_loss:3.7802 train_time:177594ms step_avg:178.67ms step:1005/5100 train_loss:3.8287 train_time:177772ms step_avg:178.67ms step:1006/5100 train_loss:3.6028 train_time:177949ms step_avg:178.66ms step:1007/5100 train_loss:3.6786 train_time:178125ms step_avg:178.66ms step:1008/5100 train_loss:3.6465 train_time:178301ms step_avg:178.66ms step:1009/5100 train_loss:3.7702 train_time:178477ms step_avg:178.66ms step:1010/5100 train_loss:3.8737 train_time:178653ms step_avg:178.65ms step:1011/5100 train_loss:3.7718 train_time:178830ms step_avg:178.65ms step:1012/5100 train_loss:3.7266 train_time:179005ms step_avg:178.65ms step:1013/5100 train_loss:3.5906 train_time:179181ms step_avg:178.65ms step:1014/5100 train_loss:3.7346 train_time:179357ms step_avg:178.64ms step:1015/5100 train_loss:3.8483 train_time:179532ms step_avg:178.64ms step:1016/5100 train_loss:3.5551 train_time:179708ms step_avg:178.64ms step:1017/5100 train_loss:3.6434 train_time:179885ms step_avg:178.63ms step:1018/5100 train_loss:3.6538 train_time:180061ms step_avg:178.63ms step:1019/5100 train_loss:3.5932 train_time:180237ms step_avg:178.63ms step:1020/5100 train_loss:3.7328 train_time:180413ms step_avg:178.63ms step:1021/5100 train_loss:3.6433 train_time:180590ms step_avg:178.62ms step:1022/5100 train_loss:3.5755 train_time:180766ms step_avg:178.62ms step:1023/5100 train_loss:3.6889 train_time:180942ms step_avg:178.62ms step:1024/5100 train_loss:3.7118 train_time:181118ms step_avg:178.62ms step:1025/5100 train_loss:3.6900 train_time:181293ms step_avg:178.61ms step:1026/5100 train_loss:3.7018 train_time:181470ms step_avg:178.61ms step:1027/5100 train_loss:3.8665 train_time:181646ms step_avg:178.61ms step:1028/5100 train_loss:3.5376 train_time:181822ms step_avg:178.61ms step:1029/5100 train_loss:3.6029 train_time:181997ms step_avg:178.60ms step:1030/5100 train_loss:3.5592 train_time:182174ms step_avg:178.60ms step:1031/5100 train_loss:3.7260 train_time:182350ms step_avg:178.60ms step:1032/5100 train_loss:3.7216 train_time:182526ms step_avg:178.60ms step:1033/5100 train_loss:3.8958 train_time:182702ms step_avg:178.59ms step:1034/5100 train_loss:3.7050 train_time:182877ms step_avg:178.59ms step:1035/5100 train_loss:3.6314 train_time:183053ms step_avg:178.59ms step:1036/5100 train_loss:3.6454 train_time:183229ms step_avg:178.59ms step:1037/5100 train_loss:3.7084 train_time:183406ms step_avg:178.58ms step:1038/5100 train_loss:4.0124 train_time:183582ms step_avg:178.58ms step:1039/5100 train_loss:3.8378 train_time:183758ms step_avg:178.58ms step:1040/5100 train_loss:3.7293 train_time:183934ms step_avg:178.58ms step:1041/5100 train_loss:3.6256 train_time:184110ms step_avg:178.57ms step:1042/5100 train_loss:3.6966 train_time:184286ms step_avg:178.57ms step:1043/5100 train_loss:3.7349 train_time:184463ms step_avg:178.57ms step:1044/5100 train_loss:3.6610 train_time:184639ms step_avg:178.57ms step:1045/5100 train_loss:3.6692 train_time:184814ms step_avg:178.56ms step:1046/5100 train_loss:3.7536 train_time:184990ms step_avg:178.56ms step:1047/5100 train_loss:3.6583 train_time:185166ms step_avg:178.56ms step:1048/5100 train_loss:3.8655 train_time:185341ms step_avg:178.56ms step:1049/5100 train_loss:3.7171 train_time:185517ms step_avg:178.55ms step:1050/5100 train_loss:3.6392 train_time:185693ms step_avg:178.55ms step:1051/5100 train_loss:3.6020 train_time:185870ms step_avg:178.55ms step:1052/5100 train_loss:3.7278 train_time:186047ms step_avg:178.55ms step:1053/5100 train_loss:3.6008 train_time:186223ms step_avg:178.55ms step:1054/5100 train_loss:3.9263 train_time:186399ms step_avg:178.54ms step:1055/5100 train_loss:3.7591 train_time:186574ms step_avg:178.54ms step:1056/5100 train_loss:3.6162 train_time:186751ms step_avg:178.54ms step:1057/5100 train_loss:3.7182 train_time:186927ms step_avg:178.54ms step:1058/5100 train_loss:3.7921 train_time:187103ms step_avg:178.53ms step:1059/5100 train_loss:3.5170 train_time:187279ms step_avg:178.53ms step:1060/5100 train_loss:3.6458 train_time:187454ms step_avg:178.53ms step:1061/5100 train_loss:3.6602 train_time:187631ms step_avg:178.53ms step:1062/5100 train_loss:3.6338 train_time:187807ms step_avg:178.52ms step:1063/5100 train_loss:3.6075 train_time:187983ms step_avg:178.52ms step:1064/5100 train_loss:3.7147 train_time:188158ms step_avg:178.52ms step:1065/5100 train_loss:3.6063 train_time:188335ms step_avg:178.52ms step:1066/5100 train_loss:3.5943 train_time:188510ms step_avg:178.51ms step:1067/5100 train_loss:3.6140 train_time:188687ms step_avg:178.51ms step:1068/5100 train_loss:3.5344 train_time:188864ms step_avg:178.51ms step:1069/5100 train_loss:3.6454 train_time:189039ms step_avg:178.51ms step:1070/5100 train_loss:3.5184 train_time:189215ms step_avg:178.50ms step:1071/5100 train_loss:3.7750 train_time:189393ms step_avg:178.50ms step:1072/5100 train_loss:3.7235 train_time:189569ms step_avg:178.50ms step:1073/5100 train_loss:3.6716 train_time:189746ms step_avg:178.50ms step:1074/5100 train_loss:3.7369 train_time:189922ms step_avg:178.50ms step:1075/5100 train_loss:3.6861 train_time:190098ms step_avg:178.50ms step:1076/5100 train_loss:3.6207 train_time:190274ms step_avg:178.49ms step:1077/5100 train_loss:4.0182 train_time:190450ms step_avg:178.49ms step:1078/5100 train_loss:3.6858 train_time:190626ms step_avg:178.49ms step:1079/5100 train_loss:3.3655 train_time:190801ms step_avg:178.49ms step:1080/5100 train_loss:3.7476 train_time:190978ms step_avg:178.48ms step:1081/5100 train_loss:3.6747 train_time:191154ms step_avg:178.48ms step:1082/5100 train_loss:3.7391 train_time:191330ms step_avg:178.48ms step:1083/5100 train_loss:3.8352 train_time:191507ms step_avg:178.48ms step:1084/5100 train_loss:3.7277 train_time:191683ms step_avg:178.48ms step:1085/5100 train_loss:3.7045 train_time:191859ms step_avg:178.47ms step:1086/5100 train_loss:3.6631 train_time:192034ms step_avg:178.47ms step:1087/5100 train_loss:3.8592 train_time:192210ms step_avg:178.47ms step:1088/5100 train_loss:3.7587 train_time:192387ms step_avg:178.47ms step:1089/5100 train_loss:3.5844 train_time:192563ms step_avg:178.46ms step:1090/5100 train_loss:3.6071 train_time:192740ms step_avg:178.46ms step:1091/5100 train_loss:3.7280 train_time:192916ms step_avg:178.46ms step:1092/5100 train_loss:3.5193 train_time:193092ms step_avg:178.46ms step:1093/5100 train_loss:3.7200 train_time:193269ms step_avg:178.46ms step:1094/5100 train_loss:3.8524 train_time:193445ms step_avg:178.45ms step:1095/5100 train_loss:3.6885 train_time:193621ms step_avg:178.45ms step:1096/5100 train_loss:3.6419 train_time:193797ms step_avg:178.45ms step:1097/5100 train_loss:3.6663 train_time:193973ms step_avg:178.45ms step:1098/5100 train_loss:3.7162 train_time:194149ms step_avg:178.45ms step:1099/5100 train_loss:3.7847 train_time:194326ms step_avg:178.44ms step:1100/5100 train_loss:3.7454 train_time:194501ms step_avg:178.44ms step:1101/5100 train_loss:3.6740 train_time:194677ms step_avg:178.44ms step:1102/5100 train_loss:3.5300 train_time:194853ms step_avg:178.44ms step:1103/5100 train_loss:3.5890 train_time:195030ms step_avg:178.44ms step:1104/5100 train_loss:3.6855 train_time:195206ms step_avg:178.43ms step:1105/5100 train_loss:3.5518 train_time:195382ms step_avg:178.43ms step:1106/5100 train_loss:4.3024 train_time:195558ms step_avg:178.43ms step:1107/5100 train_loss:3.4621 train_time:195734ms step_avg:178.43ms step:1108/5100 train_loss:3.8020 train_time:195910ms step_avg:178.42ms step:1109/5100 train_loss:3.5801 train_time:196087ms step_avg:178.42ms step:1110/5100 train_loss:3.7354 train_time:196263ms step_avg:178.42ms step:1111/5100 train_loss:3.6592 train_time:196439ms step_avg:178.42ms step:1112/5100 train_loss:3.7043 train_time:196615ms step_avg:178.42ms step:1113/5100 train_loss:3.7951 train_time:196791ms step_avg:178.41ms step:1114/5100 train_loss:3.6634 train_time:196968ms step_avg:178.41ms step:1115/5100 train_loss:3.5943 train_time:197144ms step_avg:178.41ms step:1116/5100 train_loss:3.5097 train_time:197321ms step_avg:178.41ms step:1117/5100 train_loss:3.6697 train_time:197497ms step_avg:178.41ms step:1118/5100 train_loss:3.8179 train_time:197672ms step_avg:178.40ms step:1119/5100 train_loss:3.8600 train_time:197849ms step_avg:178.40ms step:1120/5100 train_loss:3.6978 train_time:198025ms step_avg:178.40ms step:1121/5100 train_loss:3.7258 train_time:198201ms step_avg:178.40ms step:1122/5100 train_loss:3.6280 train_time:198377ms step_avg:178.40ms step:1123/5100 train_loss:3.6831 train_time:198553ms step_avg:178.39ms step:1124/5100 train_loss:3.8224 train_time:198730ms step_avg:178.39ms step:1125/5100 train_loss:3.5868 train_time:198906ms step_avg:178.39ms step:1125/5100 val_loss:3.6530 train_time:198906ms step_avg:178.39ms step:1126/5100 train_loss:3.4825 train_time:199091ms step_avg:178.40ms step:1127/5100 train_loss:3.7177 train_time:199267ms step_avg:178.39ms step:1128/5100 train_loss:3.9228 train_time:199444ms step_avg:178.39ms step:1129/5100 train_loss:3.4700 train_time:199619ms step_avg:178.39ms step:1130/5100 train_loss:3.7866 train_time:199797ms step_avg:178.39ms step:1131/5100 train_loss:3.6194 train_time:199972ms step_avg:178.39ms step:1132/5100 train_loss:3.6529 train_time:200148ms step_avg:178.38ms step:1133/5100 train_loss:3.6057 train_time:200324ms step_avg:178.38ms step:1134/5100 train_loss:3.7626 train_time:200732ms step_avg:178.59ms step:1135/5100 train_loss:3.7005 train_time:200908ms step_avg:178.58ms step:1136/5100 train_loss:3.7515 train_time:201084ms step_avg:178.58ms step:1137/5100 train_loss:3.7848 train_time:201260ms step_avg:178.58ms step:1138/5100 train_loss:3.6950 train_time:201436ms step_avg:178.58ms step:1139/5100 train_loss:3.5939 train_time:201612ms step_avg:178.58ms step:1140/5100 train_loss:3.9014 train_time:202038ms step_avg:178.79ms step:1141/5100 train_loss:3.7069 train_time:202214ms step_avg:178.79ms step:1142/5100 train_loss:3.8012 train_time:202391ms step_avg:178.79ms step:1143/5100 train_loss:3.6941 train_time:202567ms step_avg:178.79ms step:1144/5100 train_loss:3.6029 train_time:202744ms step_avg:178.79ms step:1145/5100 train_loss:3.7105 train_time:202920ms step_avg:178.78ms step:1146/5100 train_loss:3.8341 train_time:203097ms step_avg:178.78ms step:1147/5100 train_loss:3.7953 train_time:203273ms step_avg:178.78ms step:1148/5100 train_loss:3.7172 train_time:203450ms step_avg:178.78ms step:1149/5100 train_loss:3.7393 train_time:203627ms step_avg:178.78ms step:1150/5100 train_loss:3.5863 train_time:203805ms step_avg:178.78ms step:1151/5100 train_loss:3.6106 train_time:203980ms step_avg:178.77ms step:1152/5100 train_loss:3.5728 train_time:204156ms step_avg:178.77ms step:1153/5100 train_loss:3.7261 train_time:204332ms step_avg:178.77ms step:1154/5100 train_loss:3.6941 train_time:204508ms step_avg:178.77ms step:1155/5100 train_loss:3.7466 train_time:204686ms step_avg:178.76ms step:1156/5100 train_loss:3.5999 train_time:204864ms step_avg:178.76ms step:1157/5100 train_loss:3.7741 train_time:205040ms step_avg:178.76ms step:1158/5100 train_loss:3.7312 train_time:205216ms step_avg:178.76ms step:1159/5100 train_loss:3.5439 train_time:205392ms step_avg:178.76ms step:1160/5100 train_loss:3.5769 train_time:205568ms step_avg:178.75ms step:1161/5100 train_loss:3.5659 train_time:205744ms step_avg:178.75ms step:1162/5100 train_loss:3.3886 train_time:205921ms step_avg:178.75ms step:1163/5100 train_loss:3.6876 train_time:206097ms step_avg:178.75ms step:1164/5100 train_loss:3.6561 train_time:206275ms step_avg:178.75ms step:1165/5100 train_loss:3.5168 train_time:206451ms step_avg:178.75ms step:1166/5100 train_loss:3.5172 train_time:206627ms step_avg:178.74ms step:1167/5100 train_loss:3.6239 train_time:206803ms step_avg:178.74ms step:1168/5100 train_loss:3.6279 train_time:206979ms step_avg:178.74ms step:1169/5100 train_loss:3.9497 train_time:207155ms step_avg:178.74ms step:1170/5100 train_loss:3.6379 train_time:207331ms step_avg:178.73ms step:1171/5100 train_loss:3.6477 train_time:207507ms step_avg:178.73ms step:1172/5100 train_loss:3.5580 train_time:207683ms step_avg:178.73ms step:1173/5100 train_loss:3.6526 train_time:207860ms step_avg:178.73ms step:1174/5100 train_loss:3.7932 train_time:208036ms step_avg:178.73ms step:1175/5100 train_loss:3.6251 train_time:208212ms step_avg:178.72ms step:1176/5100 train_loss:3.6460 train_time:208388ms step_avg:178.72ms step:1177/5100 train_loss:3.6974 train_time:208564ms step_avg:178.72ms step:1178/5100 train_loss:3.6747 train_time:208741ms step_avg:178.72ms step:1179/5100 train_loss:3.7325 train_time:208917ms step_avg:178.71ms step:1180/5100 train_loss:3.6513 train_time:209093ms step_avg:178.71ms step:1181/5100 train_loss:3.6458 train_time:209268ms step_avg:178.71ms step:1182/5100 train_loss:3.5978 train_time:209445ms step_avg:178.71ms step:1183/5100 train_loss:3.6486 train_time:209621ms step_avg:178.70ms step:1184/5100 train_loss:3.5745 train_time:209797ms step_avg:178.70ms step:1185/5100 train_loss:3.7471 train_time:209974ms step_avg:178.70ms step:1186/5100 train_loss:3.8095 train_time:210150ms step_avg:178.70ms step:1187/5100 train_loss:3.6037 train_time:210327ms step_avg:178.70ms step:1188/5100 train_loss:3.6559 train_time:210503ms step_avg:178.69ms step:1189/5100 train_loss:3.6824 train_time:210678ms step_avg:178.69ms step:1190/5100 train_loss:3.5178 train_time:210855ms step_avg:178.69ms step:1191/5100 train_loss:3.6950 train_time:211031ms step_avg:178.69ms step:1192/5100 train_loss:3.8454 train_time:211208ms step_avg:178.69ms step:1193/5100 train_loss:3.6413 train_time:211384ms step_avg:178.68ms step:1194/5100 train_loss:3.5221 train_time:211561ms step_avg:178.68ms step:1195/5100 train_loss:3.8162 train_time:211739ms step_avg:178.68ms step:1196/5100 train_loss:3.6273 train_time:211915ms step_avg:178.68ms step:1197/5100 train_loss:3.6355 train_time:212091ms step_avg:178.68ms step:1198/5100 train_loss:3.5249 train_time:212267ms step_avg:178.68ms step:1199/5100 train_loss:3.5413 train_time:212443ms step_avg:178.67ms step:1200/5100 train_loss:3.5935 train_time:212619ms step_avg:178.67ms step:1201/5100 train_loss:3.6779 train_time:212796ms step_avg:178.67ms step:1202/5100 train_loss:3.7500 train_time:212973ms step_avg:178.67ms step:1203/5100 train_loss:3.8305 train_time:213150ms step_avg:178.67ms step:1204/5100 train_loss:3.6596 train_time:213327ms step_avg:178.67ms step:1205/5100 train_loss:3.5848 train_time:213503ms step_avg:178.66ms step:1206/5100 train_loss:3.6713 train_time:213680ms step_avg:178.66ms step:1207/5100 train_loss:3.7221 train_time:213856ms step_avg:178.66ms step:1208/5100 train_loss:3.7704 train_time:214031ms step_avg:178.66ms step:1209/5100 train_loss:3.6465 train_time:214208ms step_avg:178.66ms step:1210/5100 train_loss:3.5068 train_time:214383ms step_avg:178.65ms step:1211/5100 train_loss:3.5606 train_time:214560ms step_avg:178.65ms step:1212/5100 train_loss:3.6501 train_time:214735ms step_avg:178.65ms step:1213/5100 train_loss:3.6691 train_time:214911ms step_avg:178.65ms step:1214/5100 train_loss:3.6990 train_time:215087ms step_avg:178.64ms step:1215/5100 train_loss:3.5872 train_time:215263ms step_avg:178.64ms step:1216/5100 train_loss:3.6496 train_time:215439ms step_avg:178.64ms step:1217/5100 train_loss:3.5940 train_time:215615ms step_avg:178.64ms step:1218/5100 train_loss:3.5798 train_time:215793ms step_avg:178.64ms step:1219/5100 train_loss:3.6728 train_time:215968ms step_avg:178.63ms step:1220/5100 train_loss:3.5300 train_time:216144ms step_avg:178.63ms step:1221/5100 train_loss:3.7499 train_time:216320ms step_avg:178.63ms step:1222/5100 train_loss:3.7733 train_time:216497ms step_avg:178.63ms step:1223/5100 train_loss:3.7034 train_time:216674ms step_avg:178.63ms step:1224/5100 train_loss:3.5453 train_time:216849ms step_avg:178.62ms step:1225/5100 train_loss:3.5365 train_time:217026ms step_avg:178.62ms step:1226/5100 train_loss:3.6209 train_time:217202ms step_avg:178.62ms step:1227/5100 train_loss:3.5967 train_time:217378ms step_avg:178.62ms step:1228/5100 train_loss:3.5364 train_time:217554ms step_avg:178.62ms step:1229/5100 train_loss:3.7115 train_time:217731ms step_avg:178.61ms step:1230/5100 train_loss:3.6351 train_time:217907ms step_avg:178.61ms step:1231/5100 train_loss:3.6942 train_time:218083ms step_avg:178.61ms step:1232/5100 train_loss:3.8499 train_time:218260ms step_avg:178.61ms step:1233/5100 train_loss:3.7450 train_time:218436ms step_avg:178.61ms step:1234/5100 train_loss:3.6809 train_time:218612ms step_avg:178.60ms step:1235/5100 train_loss:3.8385 train_time:218788ms step_avg:178.60ms step:1236/5100 train_loss:3.5953 train_time:218965ms step_avg:178.60ms step:1237/5100 train_loss:3.5623 train_time:219141ms step_avg:178.60ms step:1238/5100 train_loss:3.5100 train_time:219318ms step_avg:178.60ms step:1239/5100 train_loss:3.5931 train_time:219494ms step_avg:178.60ms step:1240/5100 train_loss:3.6001 train_time:219670ms step_avg:178.59ms step:1241/5100 train_loss:3.6401 train_time:219846ms step_avg:178.59ms step:1242/5100 train_loss:3.6913 train_time:220022ms step_avg:178.59ms step:1243/5100 train_loss:3.5631 train_time:220198ms step_avg:178.59ms step:1244/5100 train_loss:3.6548 train_time:220375ms step_avg:178.59ms step:1245/5100 train_loss:3.6765 train_time:220550ms step_avg:178.58ms step:1246/5100 train_loss:3.6730 train_time:220727ms step_avg:178.58ms step:1247/5100 train_loss:3.5009 train_time:220903ms step_avg:178.58ms step:1248/5100 train_loss:3.6447 train_time:221079ms step_avg:178.58ms step:1249/5100 train_loss:3.7100 train_time:221256ms step_avg:178.58ms step:1250/5100 train_loss:3.6671 train_time:221431ms step_avg:178.57ms step:1250/5100 val_loss:3.6230 train_time:221431ms step_avg:178.57ms step:1251/5100 train_loss:3.5626 train_time:221616ms step_avg:178.58ms step:1252/5100 train_loss:3.7756 train_time:221792ms step_avg:178.58ms step:1253/5100 train_loss:3.6461 train_time:221968ms step_avg:178.57ms step:1254/5100 train_loss:3.5706 train_time:222144ms step_avg:178.57ms step:1255/5100 train_loss:3.7047 train_time:222320ms step_avg:178.57ms step:1256/5100 train_loss:3.7690 train_time:222496ms step_avg:178.57ms step:1257/5100 train_loss:3.5737 train_time:222673ms step_avg:178.57ms step:1258/5100 train_loss:3.6061 train_time:222848ms step_avg:178.56ms step:1259/5100 train_loss:3.6364 train_time:223024ms step_avg:178.56ms step:1260/5100 train_loss:3.6096 train_time:223200ms step_avg:178.56ms step:1261/5100 train_loss:3.4602 train_time:223376ms step_avg:178.56ms step:1262/5100 train_loss:3.5711 train_time:223552ms step_avg:178.56ms step:1263/5100 train_loss:3.6325 train_time:223728ms step_avg:178.55ms step:1264/5100 train_loss:3.4833 train_time:223904ms step_avg:178.55ms step:1265/5100 train_loss:3.7129 train_time:224080ms step_avg:178.55ms step:1266/5100 train_loss:3.6865 train_time:224256ms step_avg:178.55ms step:1267/5100 train_loss:3.6944 train_time:224432ms step_avg:178.55ms step:1268/5100 train_loss:3.6329 train_time:224608ms step_avg:178.54ms step:1269/5100 train_loss:3.6694 train_time:224784ms step_avg:178.54ms step:1270/5100 train_loss:3.5246 train_time:224960ms step_avg:178.54ms step:1271/5100 train_loss:3.3769 train_time:225136ms step_avg:178.54ms step:1272/5100 train_loss:3.6590 train_time:225312ms step_avg:178.54ms step:1273/5100 train_loss:3.6180 train_time:225488ms step_avg:178.53ms step:1274/5100 train_loss:3.6786 train_time:225664ms step_avg:178.53ms step:1275/5100 train_loss:3.6165 train_time:225840ms step_avg:178.53ms step:1276/5100 train_loss:3.7129 train_time:226017ms step_avg:178.53ms step:1277/5100 train_loss:3.7311 train_time:226193ms step_avg:178.53ms step:1278/5100 train_loss:3.6876 train_time:226370ms step_avg:178.53ms step:1279/5100 train_loss:3.6830 train_time:226546ms step_avg:178.52ms step:1280/5100 train_loss:3.5146 train_time:226722ms step_avg:178.52ms step:1281/5100 train_loss:3.6393 train_time:226898ms step_avg:178.52ms step:1282/5100 train_loss:3.6991 train_time:227075ms step_avg:178.52ms step:1283/5100 train_loss:3.7362 train_time:227251ms step_avg:178.52ms step:1284/5100 train_loss:3.6198 train_time:227428ms step_avg:178.51ms step:1285/5100 train_loss:3.6490 train_time:227603ms step_avg:178.51ms step:1286/5100 train_loss:3.6303 train_time:227779ms step_avg:178.51ms step:1287/5100 train_loss:3.6071 train_time:227956ms step_avg:178.51ms step:1288/5100 train_loss:3.7378 train_time:228132ms step_avg:178.51ms step:1289/5100 train_loss:3.5776 train_time:228308ms step_avg:178.51ms step:1290/5100 train_loss:3.6588 train_time:228485ms step_avg:178.50ms step:1291/5100 train_loss:3.7279 train_time:228660ms step_avg:178.50ms step:1292/5100 train_loss:3.6568 train_time:228837ms step_avg:178.50ms step:1293/5100 train_loss:3.7534 train_time:229013ms step_avg:178.50ms step:1294/5100 train_loss:3.7769 train_time:229190ms step_avg:178.50ms step:1295/5100 train_loss:3.7383 train_time:229365ms step_avg:178.49ms step:1296/5100 train_loss:3.5505 train_time:229541ms step_avg:178.49ms step:1297/5100 train_loss:3.6292 train_time:229718ms step_avg:178.49ms step:1298/5100 train_loss:3.5332 train_time:229894ms step_avg:178.49ms step:1299/5100 train_loss:3.5968 train_time:230070ms step_avg:178.49ms step:1300/5100 train_loss:3.6681 train_time:230246ms step_avg:178.49ms step:1301/5100 train_loss:3.6758 train_time:230422ms step_avg:178.48ms step:1302/5100 train_loss:3.6816 train_time:230599ms step_avg:178.48ms step:1303/5100 train_loss:3.8336 train_time:230775ms step_avg:178.48ms step:1304/5100 train_loss:3.6061 train_time:230951ms step_avg:178.48ms step:1305/5100 train_loss:3.8180 train_time:231127ms step_avg:178.48ms step:1306/5100 train_loss:3.5473 train_time:231304ms step_avg:178.48ms step:1307/5100 train_loss:3.7279 train_time:231479ms step_avg:178.47ms step:1308/5100 train_loss:3.7247 train_time:231656ms step_avg:178.47ms step:1309/5100 train_loss:3.6006 train_time:231833ms step_avg:178.47ms step:1310/5100 train_loss:3.5641 train_time:232010ms step_avg:178.47ms step:1311/5100 train_loss:3.5901 train_time:232186ms step_avg:178.47ms step:1312/5100 train_loss:3.5581 train_time:232362ms step_avg:178.47ms step:1313/5100 train_loss:3.6798 train_time:232538ms step_avg:178.46ms step:1314/5100 train_loss:3.6193 train_time:232715ms step_avg:178.46ms step:1315/5100 train_loss:3.3393 train_time:232891ms step_avg:178.46ms step:1316/5100 train_loss:3.5732 train_time:233067ms step_avg:178.46ms step:1317/5100 train_loss:3.6448 train_time:233244ms step_avg:178.46ms step:1318/5100 train_loss:3.6807 train_time:233420ms step_avg:178.46ms step:1319/5100 train_loss:3.5537 train_time:233597ms step_avg:178.45ms step:1320/5100 train_loss:3.6915 train_time:233773ms step_avg:178.45ms step:1321/5100 train_loss:3.7437 train_time:233949ms step_avg:178.45ms step:1322/5100 train_loss:3.6324 train_time:234125ms step_avg:178.45ms step:1323/5100 train_loss:3.5756 train_time:234531ms step_avg:178.62ms step:1324/5100 train_loss:3.6098 train_time:234706ms step_avg:178.62ms step:1325/5100 train_loss:3.7061 train_time:234882ms step_avg:178.62ms step:1326/5100 train_loss:3.7627 train_time:235058ms step_avg:178.62ms step:1327/5100 train_loss:3.5272 train_time:235236ms step_avg:178.61ms step:1328/5100 train_loss:3.4428 train_time:235412ms step_avg:178.61ms step:1329/5100 train_loss:3.7473 train_time:235588ms step_avg:178.61ms step:1330/5100 train_loss:3.6005 train_time:236006ms step_avg:178.79ms step:1331/5100 train_loss:3.7223 train_time:236183ms step_avg:178.79ms step:1332/5100 train_loss:3.6263 train_time:236360ms step_avg:178.79ms step:1333/5100 train_loss:4.0278 train_time:236536ms step_avg:178.79ms step:1334/5100 train_loss:3.7205 train_time:236713ms step_avg:178.79ms step:1335/5100 train_loss:3.6322 train_time:236889ms step_avg:178.78ms step:1336/5100 train_loss:3.5834 train_time:237066ms step_avg:178.78ms step:1337/5100 train_loss:3.5779 train_time:237242ms step_avg:178.78ms step:1338/5100 train_loss:3.8330 train_time:237418ms step_avg:178.78ms step:1339/5100 train_loss:3.7688 train_time:237594ms step_avg:178.78ms step:1340/5100 train_loss:3.6158 train_time:237771ms step_avg:178.77ms step:1341/5100 train_loss:3.5679 train_time:237947ms step_avg:178.77ms step:1342/5100 train_loss:3.8781 train_time:238122ms step_avg:178.77ms step:1343/5100 train_loss:3.6442 train_time:238298ms step_avg:178.77ms step:1344/5100 train_loss:3.6450 train_time:238475ms step_avg:178.77ms step:1345/5100 train_loss:3.7053 train_time:238651ms step_avg:178.76ms step:1346/5100 train_loss:3.6632 train_time:238828ms step_avg:178.76ms step:1347/5100 train_loss:3.5663 train_time:239003ms step_avg:178.76ms step:1348/5100 train_loss:3.5162 train_time:239180ms step_avg:178.76ms step:1349/5100 train_loss:3.6177 train_time:239357ms step_avg:178.76ms step:1350/5100 train_loss:3.5415 train_time:239534ms step_avg:178.76ms step:1351/5100 train_loss:3.6830 train_time:239710ms step_avg:178.75ms step:1352/5100 train_loss:3.5257 train_time:239886ms step_avg:178.75ms step:1353/5100 train_loss:3.5830 train_time:240062ms step_avg:178.75ms step:1354/5100 train_loss:3.6989 train_time:240238ms step_avg:178.75ms step:1355/5100 train_loss:3.5310 train_time:240415ms step_avg:178.75ms step:1356/5100 train_loss:3.4614 train_time:240591ms step_avg:178.75ms step:1357/5100 train_loss:3.8080 train_time:240768ms step_avg:178.74ms step:1358/5100 train_loss:3.7339 train_time:240944ms step_avg:178.74ms step:1359/5100 train_loss:3.4461 train_time:241120ms step_avg:178.74ms step:1360/5100 train_loss:3.7318 train_time:241297ms step_avg:178.74ms step:1361/5100 train_loss:3.6246 train_time:241473ms step_avg:178.74ms step:1362/5100 train_loss:3.4832 train_time:241648ms step_avg:178.73ms step:1363/5100 train_loss:3.6590 train_time:241824ms step_avg:178.73ms step:1364/5100 train_loss:3.5498 train_time:242000ms step_avg:178.73ms step:1365/5100 train_loss:3.5809 train_time:242177ms step_avg:178.73ms step:1366/5100 train_loss:3.5949 train_time:242353ms step_avg:178.73ms step:1367/5100 train_loss:3.7008 train_time:242529ms step_avg:178.72ms step:1368/5100 train_loss:3.6796 train_time:242705ms step_avg:178.72ms step:1369/5100 train_loss:3.6430 train_time:242881ms step_avg:178.72ms step:1370/5100 train_loss:3.5439 train_time:243057ms step_avg:178.72ms step:1371/5100 train_loss:3.8653 train_time:243234ms step_avg:178.72ms step:1372/5100 train_loss:3.6125 train_time:243410ms step_avg:178.72ms step:1373/5100 train_loss:3.6442 train_time:243586ms step_avg:178.71ms step:1374/5100 train_loss:3.6471 train_time:243762ms step_avg:178.71ms step:1375/5100 train_loss:3.4386 train_time:243938ms step_avg:178.71ms step:1375/5100 val_loss:3.6031 train_time:243938ms step_avg:178.71ms step:1376/5100 train_loss:3.8436 train_time:244124ms step_avg:178.71ms step:1377/5100 train_loss:3.6237 train_time:244300ms step_avg:178.71ms step:1378/5100 train_loss:3.7687 train_time:244476ms step_avg:178.71ms step:1379/5100 train_loss:3.8252 train_time:244653ms step_avg:178.71ms step:1380/5100 train_loss:3.4814 train_time:244829ms step_avg:178.71ms step:1381/5100 train_loss:3.6114 train_time:245006ms step_avg:178.71ms step:1382/5100 train_loss:4.0504 train_time:245182ms step_avg:178.70ms step:1383/5100 train_loss:3.5224 train_time:245358ms step_avg:178.70ms step:1384/5100 train_loss:3.6819 train_time:245536ms step_avg:178.70ms step:1385/5100 train_loss:3.7565 train_time:245712ms step_avg:178.70ms step:1386/5100 train_loss:3.6636 train_time:245889ms step_avg:178.70ms step:1387/5100 train_loss:3.6683 train_time:246066ms step_avg:178.70ms step:1388/5100 train_loss:3.4877 train_time:246243ms step_avg:178.70ms step:1389/5100 train_loss:3.6300 train_time:246419ms step_avg:178.69ms step:1390/5100 train_loss:3.6005 train_time:246595ms step_avg:178.69ms step:1391/5100 train_loss:3.8660 train_time:246771ms step_avg:178.69ms step:1392/5100 train_loss:3.5758 train_time:246949ms step_avg:178.69ms step:1393/5100 train_loss:3.5712 train_time:247124ms step_avg:178.69ms step:1394/5100 train_loss:3.5392 train_time:247300ms step_avg:178.69ms step:1395/5100 train_loss:3.8263 train_time:247477ms step_avg:178.68ms step:1396/5100 train_loss:3.7186 train_time:247653ms step_avg:178.68ms step:1397/5100 train_loss:3.7221 train_time:247829ms step_avg:178.68ms step:1398/5100 train_loss:3.5887 train_time:248006ms step_avg:178.68ms step:1399/5100 train_loss:3.5617 train_time:248183ms step_avg:178.68ms step:1400/5100 train_loss:3.6206 train_time:248360ms step_avg:178.68ms step:1401/5100 train_loss:3.5920 train_time:248537ms step_avg:178.67ms step:1402/5100 train_loss:3.6144 train_time:248713ms step_avg:178.67ms step:1403/5100 train_loss:3.5830 train_time:248889ms step_avg:178.67ms step:1404/5100 train_loss:3.8169 train_time:249066ms step_avg:178.67ms step:1405/5100 train_loss:3.5573 train_time:249242ms step_avg:178.67ms step:1406/5100 train_loss:3.6146 train_time:249418ms step_avg:178.67ms step:1407/5100 train_loss:3.6011 train_time:249594ms step_avg:178.66ms step:1408/5100 train_loss:3.4682 train_time:249771ms step_avg:178.66ms step:1409/5100 train_loss:3.5847 train_time:249948ms step_avg:178.66ms step:1410/5100 train_loss:3.5664 train_time:250123ms step_avg:178.66ms step:1411/5100 train_loss:3.5635 train_time:250300ms step_avg:178.66ms step:1412/5100 train_loss:3.6631 train_time:250476ms step_avg:178.66ms step:1413/5100 train_loss:3.5923 train_time:250653ms step_avg:178.65ms step:1414/5100 train_loss:3.6438 train_time:250829ms step_avg:178.65ms step:1415/5100 train_loss:3.6381 train_time:251006ms step_avg:178.65ms step:1416/5100 train_loss:3.7173 train_time:251183ms step_avg:178.65ms step:1417/5100 train_loss:3.5107 train_time:251360ms step_avg:178.65ms step:1418/5100 train_loss:3.5751 train_time:251536ms step_avg:178.65ms step:1419/5100 train_loss:3.6680 train_time:251712ms step_avg:178.65ms step:1420/5100 train_loss:3.6937 train_time:251888ms step_avg:178.64ms step:1421/5100 train_loss:3.6678 train_time:252065ms step_avg:178.64ms step:1422/5100 train_loss:3.6670 train_time:252242ms step_avg:178.64ms step:1423/5100 train_loss:3.6423 train_time:252418ms step_avg:178.64ms step:1424/5100 train_loss:3.6308 train_time:252594ms step_avg:178.64ms step:1425/5100 train_loss:3.6245 train_time:252770ms step_avg:178.64ms step:1426/5100 train_loss:3.4946 train_time:252947ms step_avg:178.63ms step:1427/5100 train_loss:3.6105 train_time:253123ms step_avg:178.63ms step:1428/5100 train_loss:3.5510 train_time:253299ms step_avg:178.63ms step:1429/5100 train_loss:3.6661 train_time:253476ms step_avg:178.63ms step:1430/5100 train_loss:3.6282 train_time:253652ms step_avg:178.63ms step:1431/5100 train_loss:3.5629 train_time:253828ms step_avg:178.63ms step:1432/5100 train_loss:3.6037 train_time:254005ms step_avg:178.63ms step:1433/5100 train_loss:3.6438 train_time:254181ms step_avg:178.62ms step:1434/5100 train_loss:3.5263 train_time:254358ms step_avg:178.62ms step:1435/5100 train_loss:3.6125 train_time:254534ms step_avg:178.62ms step:1436/5100 train_loss:3.4404 train_time:254710ms step_avg:178.62ms step:1437/5100 train_loss:3.5090 train_time:254887ms step_avg:178.62ms step:1438/5100 train_loss:3.7030 train_time:255063ms step_avg:178.62ms step:1439/5100 train_loss:3.6565 train_time:255240ms step_avg:178.61ms step:1440/5100 train_loss:3.6092 train_time:255416ms step_avg:178.61ms step:1441/5100 train_loss:3.4656 train_time:255593ms step_avg:178.61ms step:1442/5100 train_loss:3.6330 train_time:255769ms step_avg:178.61ms step:1443/5100 train_loss:3.6880 train_time:255946ms step_avg:178.61ms step:1444/5100 train_loss:3.7688 train_time:256122ms step_avg:178.61ms step:1445/5100 train_loss:3.7396 train_time:256299ms step_avg:178.61ms step:1446/5100 train_loss:3.6196 train_time:256475ms step_avg:178.60ms step:1447/5100 train_loss:3.4967 train_time:256651ms step_avg:178.60ms step:1448/5100 train_loss:3.5691 train_time:256827ms step_avg:178.60ms step:1449/5100 train_loss:3.5924 train_time:257004ms step_avg:178.60ms step:1450/5100 train_loss:3.7072 train_time:257181ms step_avg:178.60ms step:1451/5100 train_loss:3.6952 train_time:257357ms step_avg:178.60ms step:1452/5100 train_loss:3.5139 train_time:257533ms step_avg:178.59ms step:1453/5100 train_loss:3.6318 train_time:257710ms step_avg:178.59ms step:1454/5100 train_loss:3.5382 train_time:257886ms step_avg:178.59ms step:1455/5100 train_loss:3.5699 train_time:258062ms step_avg:178.59ms step:1456/5100 train_loss:3.6197 train_time:258239ms step_avg:178.59ms step:1457/5100 train_loss:3.5529 train_time:258415ms step_avg:178.59ms step:1458/5100 train_loss:3.4538 train_time:258591ms step_avg:178.58ms step:1459/5100 train_loss:3.6879 train_time:258768ms step_avg:178.58ms step:1460/5100 train_loss:3.5614 train_time:258944ms step_avg:178.58ms step:1461/5100 train_loss:3.6141 train_time:259121ms step_avg:178.58ms step:1462/5100 train_loss:3.7396 train_time:259296ms step_avg:178.58ms step:1463/5100 train_loss:3.5596 train_time:259472ms step_avg:178.58ms step:1464/5100 train_loss:3.7474 train_time:259649ms step_avg:178.58ms step:1465/5100 train_loss:3.6401 train_time:259826ms step_avg:178.57ms step:1466/5100 train_loss:3.6484 train_time:260003ms step_avg:178.57ms step:1467/5100 train_loss:3.5671 train_time:260179ms step_avg:178.57ms step:1468/5100 train_loss:3.7245 train_time:260355ms step_avg:178.57ms step:1469/5100 train_loss:3.5875 train_time:260531ms step_avg:178.57ms step:1470/5100 train_loss:3.5571 train_time:260708ms step_avg:178.57ms step:1471/5100 train_loss:3.6081 train_time:260885ms step_avg:178.57ms step:1472/5100 train_loss:3.5365 train_time:261061ms step_avg:178.56ms step:1473/5100 train_loss:3.6260 train_time:261238ms step_avg:178.56ms step:1474/5100 train_loss:3.7156 train_time:261414ms step_avg:178.56ms step:1475/5100 train_loss:3.5980 train_time:261591ms step_avg:178.56ms step:1476/5100 train_loss:3.4210 train_time:261767ms step_avg:178.56ms step:1477/5100 train_loss:3.5456 train_time:261944ms step_avg:178.56ms step:1478/5100 train_loss:3.5224 train_time:262120ms step_avg:178.56ms step:1479/5100 train_loss:3.6058 train_time:262296ms step_avg:178.55ms step:1480/5100 train_loss:3.6911 train_time:262472ms step_avg:178.55ms step:1481/5100 train_loss:3.5620 train_time:262649ms step_avg:178.55ms step:1482/5100 train_loss:3.7380 train_time:262825ms step_avg:178.55ms step:1483/5100 train_loss:3.6587 train_time:263002ms step_avg:178.55ms step:1484/5100 train_loss:3.5582 train_time:263179ms step_avg:178.55ms step:1485/5100 train_loss:3.5576 train_time:263355ms step_avg:178.55ms step:1486/5100 train_loss:3.5522 train_time:263532ms step_avg:178.54ms step:1487/5100 train_loss:3.5281 train_time:263708ms step_avg:178.54ms step:1488/5100 train_loss:3.6145 train_time:263885ms step_avg:178.54ms step:1489/5100 train_loss:3.5234 train_time:264061ms step_avg:178.54ms step:1490/5100 train_loss:3.6153 train_time:264238ms step_avg:178.54ms step:1491/5100 train_loss:3.5494 train_time:264413ms step_avg:178.54ms step:1492/5100 train_loss:3.4770 train_time:264589ms step_avg:178.54ms step:1493/5100 train_loss:3.5474 train_time:264766ms step_avg:178.53ms step:1494/5100 train_loss:3.7292 train_time:264943ms step_avg:178.53ms step:1495/5100 train_loss:3.5766 train_time:265118ms step_avg:178.53ms step:1496/5100 train_loss:3.3383 train_time:265294ms step_avg:178.53ms step:1497/5100 train_loss:3.6401 train_time:265470ms step_avg:178.53ms step:1498/5100 train_loss:3.5997 train_time:265646ms step_avg:178.53ms step:1499/5100 train_loss:3.6450 train_time:265823ms step_avg:178.52ms step:1500/5100 train_loss:3.6002 train_time:266000ms step_avg:178.52ms step:1500/5100 val_loss:3.5783 train_time:266000ms step_avg:178.52ms step:1501/5100 train_loss:3.5840 train_time:266186ms step_avg:178.53ms step:1502/5100 train_loss:3.3782 train_time:266362ms step_avg:178.53ms step:1503/5100 train_loss:3.6553 train_time:266540ms step_avg:178.53ms step:1504/5100 train_loss:3.5283 train_time:266716ms step_avg:178.52ms step:1505/5100 train_loss:3.5382 train_time:266892ms step_avg:178.52ms step:1506/5100 train_loss:3.4951 train_time:267068ms step_avg:178.52ms step:1507/5100 train_loss:3.5801 train_time:267244ms step_avg:178.52ms step:1508/5100 train_loss:3.4946 train_time:267421ms step_avg:178.52ms step:1509/5100 train_loss:3.8093 train_time:267597ms step_avg:178.52ms step:1510/5100 train_loss:3.5451 train_time:267774ms step_avg:178.52ms step:1511/5100 train_loss:3.5598 train_time:267951ms step_avg:178.52ms step:1512/5100 train_loss:3.6850 train_time:268357ms step_avg:178.67ms step:1513/5100 train_loss:3.7155 train_time:268534ms step_avg:178.67ms step:1514/5100 train_loss:3.5712 train_time:268709ms step_avg:178.66ms step:1515/5100 train_loss:3.4004 train_time:268885ms step_avg:178.66ms step:1516/5100 train_loss:3.5275 train_time:269061ms step_avg:178.66ms step:1517/5100 train_loss:3.5361 train_time:269239ms step_avg:178.66ms step:1518/5100 train_loss:3.6153 train_time:269415ms step_avg:178.66ms step:1519/5100 train_loss:3.5019 train_time:269591ms step_avg:178.66ms step:1520/5100 train_loss:3.8002 train_time:270014ms step_avg:178.82ms step:1521/5100 train_loss:3.4610 train_time:270191ms step_avg:178.82ms step:1522/5100 train_loss:3.5236 train_time:270367ms step_avg:178.81ms step:1523/5100 train_loss:3.6636 train_time:270543ms step_avg:178.81ms step:1524/5100 train_loss:3.5266 train_time:270720ms step_avg:178.81ms step:1525/5100 train_loss:3.6184 train_time:270896ms step_avg:178.81ms step:1526/5100 train_loss:3.6118 train_time:271073ms step_avg:178.81ms step:1527/5100 train_loss:3.5740 train_time:271248ms step_avg:178.81ms step:1528/5100 train_loss:3.5757 train_time:271424ms step_avg:178.80ms step:1529/5100 train_loss:3.7269 train_time:271601ms step_avg:178.80ms step:1530/5100 train_loss:3.6978 train_time:271778ms step_avg:178.80ms step:1531/5100 train_loss:3.5244 train_time:271955ms step_avg:178.80ms step:1532/5100 train_loss:3.4850 train_time:272131ms step_avg:178.80ms step:1533/5100 train_loss:3.6392 train_time:272308ms step_avg:178.80ms step:1534/5100 train_loss:3.5958 train_time:272484ms step_avg:178.80ms step:1535/5100 train_loss:3.5809 train_time:272661ms step_avg:178.79ms step:1536/5100 train_loss:3.5803 train_time:272837ms step_avg:178.79ms step:1537/5100 train_loss:3.5120 train_time:273014ms step_avg:178.79ms step:1538/5100 train_loss:3.5765 train_time:273190ms step_avg:178.79ms step:1539/5100 train_loss:3.7475 train_time:273366ms step_avg:178.79ms step:1540/5100 train_loss:3.6846 train_time:273542ms step_avg:178.79ms step:1541/5100 train_loss:3.5905 train_time:273719ms step_avg:178.78ms step:1542/5100 train_loss:3.5417 train_time:273895ms step_avg:178.78ms step:1543/5100 train_loss:3.5399 train_time:274072ms step_avg:178.78ms step:1544/5100 train_loss:3.5033 train_time:274248ms step_avg:178.78ms step:1545/5100 train_loss:3.5936 train_time:274424ms step_avg:178.78ms step:1546/5100 train_loss:3.5607 train_time:274600ms step_avg:178.78ms step:1547/5100 train_loss:3.5433 train_time:274776ms step_avg:178.77ms step:1548/5100 train_loss:3.5005 train_time:274953ms step_avg:178.77ms step:1549/5100 train_loss:3.5357 train_time:275129ms step_avg:178.77ms step:1550/5100 train_loss:3.6445 train_time:275305ms step_avg:178.77ms step:1551/5100 train_loss:3.5754 train_time:275483ms step_avg:178.77ms step:1552/5100 train_loss:3.5091 train_time:275661ms step_avg:178.77ms step:1553/5100 train_loss:3.5105 train_time:275837ms step_avg:178.77ms step:1554/5100 train_loss:3.5003 train_time:276014ms step_avg:178.77ms step:1555/5100 train_loss:3.6297 train_time:276190ms step_avg:178.76ms step:1556/5100 train_loss:3.6359 train_time:276366ms step_avg:178.76ms step:1557/5100 train_loss:3.5669 train_time:276542ms step_avg:178.76ms step:1558/5100 train_loss:3.6225 train_time:276718ms step_avg:178.76ms step:1559/5100 train_loss:3.5421 train_time:276895ms step_avg:178.76ms step:1560/5100 train_loss:3.4622 train_time:277071ms step_avg:178.76ms step:1561/5100 train_loss:3.7054 train_time:277247ms step_avg:178.75ms step:1562/5100 train_loss:3.5289 train_time:277423ms step_avg:178.75ms step:1563/5100 train_loss:3.5026 train_time:277599ms step_avg:178.75ms step:1564/5100 train_loss:3.6279 train_time:277776ms step_avg:178.75ms step:1565/5100 train_loss:3.4550 train_time:277952ms step_avg:178.75ms step:1566/5100 train_loss:3.5173 train_time:278128ms step_avg:178.75ms step:1567/5100 train_loss:3.6609 train_time:278304ms step_avg:178.74ms step:1568/5100 train_loss:3.5367 train_time:278481ms step_avg:178.74ms step:1569/5100 train_loss:3.5211 train_time:278657ms step_avg:178.74ms step:1570/5100 train_loss:3.6261 train_time:278834ms step_avg:178.74ms step:1571/5100 train_loss:3.6302 train_time:279011ms step_avg:178.74ms step:1572/5100 train_loss:3.4562 train_time:279187ms step_avg:178.74ms step:1573/5100 train_loss:3.4883 train_time:279363ms step_avg:178.74ms step:1574/5100 train_loss:3.6077 train_time:279540ms step_avg:178.73ms step:1575/5100 train_loss:3.4748 train_time:279716ms step_avg:178.73ms step:1576/5100 train_loss:3.6241 train_time:279892ms step_avg:178.73ms step:1577/5100 train_loss:3.5203 train_time:280068ms step_avg:178.73ms step:1578/5100 train_loss:3.5763 train_time:280243ms step_avg:178.73ms step:1579/5100 train_loss:3.5532 train_time:280420ms step_avg:178.73ms step:1580/5100 train_loss:3.5212 train_time:280596ms step_avg:178.72ms step:1581/5100 train_loss:3.4949 train_time:280772ms step_avg:178.72ms step:1582/5100 train_loss:3.7312 train_time:280947ms step_avg:178.72ms step:1583/5100 train_loss:3.5075 train_time:281124ms step_avg:178.72ms step:1584/5100 train_loss:3.6601 train_time:281300ms step_avg:178.72ms step:1585/5100 train_loss:3.5028 train_time:281476ms step_avg:178.71ms step:1586/5100 train_loss:3.6594 train_time:281652ms step_avg:178.71ms step:1587/5100 train_loss:3.4415 train_time:281829ms step_avg:178.71ms step:1588/5100 train_loss:3.6310 train_time:282005ms step_avg:178.71ms step:1589/5100 train_loss:3.5430 train_time:282182ms step_avg:178.71ms step:1590/5100 train_loss:3.7052 train_time:282359ms step_avg:178.71ms step:1591/5100 train_loss:3.5187 train_time:282536ms step_avg:178.71ms step:1592/5100 train_loss:3.5357 train_time:282712ms step_avg:178.71ms step:1593/5100 train_loss:3.6061 train_time:282888ms step_avg:178.70ms step:1594/5100 train_loss:3.5822 train_time:283066ms step_avg:178.70ms step:1595/5100 train_loss:3.5594 train_time:283242ms step_avg:178.70ms step:1596/5100 train_loss:3.7001 train_time:283418ms step_avg:178.70ms step:1597/5100 train_loss:3.4320 train_time:283594ms step_avg:178.70ms step:1598/5100 train_loss:3.5917 train_time:283770ms step_avg:178.70ms step:1599/5100 train_loss:3.6319 train_time:283946ms step_avg:178.69ms step:1600/5100 train_loss:3.6754 train_time:284122ms step_avg:178.69ms step:1601/5100 train_loss:3.5293 train_time:284299ms step_avg:178.69ms step:1602/5100 train_loss:3.8279 train_time:284475ms step_avg:178.69ms step:1603/5100 train_loss:3.7083 train_time:284652ms step_avg:178.69ms step:1604/5100 train_loss:3.4916 train_time:284829ms step_avg:178.69ms step:1605/5100 train_loss:3.5298 train_time:285005ms step_avg:178.69ms step:1606/5100 train_loss:3.4150 train_time:285181ms step_avg:178.68ms step:1607/5100 train_loss:3.7401 train_time:285358ms step_avg:178.68ms step:1608/5100 train_loss:3.5368 train_time:285534ms step_avg:178.68ms step:1609/5100 train_loss:3.5661 train_time:285710ms step_avg:178.68ms step:1610/5100 train_loss:3.5113 train_time:285886ms step_avg:178.68ms step:1611/5100 train_loss:4.1190 train_time:286063ms step_avg:178.68ms step:1612/5100 train_loss:3.7462 train_time:286238ms step_avg:178.68ms step:1613/5100 train_loss:3.6595 train_time:286415ms step_avg:178.67ms step:1614/5100 train_loss:3.5264 train_time:286591ms step_avg:178.67ms step:1615/5100 train_loss:3.5736 train_time:286768ms step_avg:178.67ms step:1616/5100 train_loss:3.5631 train_time:286944ms step_avg:178.67ms step:1617/5100 train_loss:3.5239 train_time:287120ms step_avg:178.67ms step:1618/5100 train_loss:3.6006 train_time:287296ms step_avg:178.67ms step:1619/5100 train_loss:3.5594 train_time:287473ms step_avg:178.67ms step:1620/5100 train_loss:3.4501 train_time:287649ms step_avg:178.66ms step:1621/5100 train_loss:3.7196 train_time:287825ms step_avg:178.66ms step:1622/5100 train_loss:3.6239 train_time:288001ms step_avg:178.66ms step:1623/5100 train_loss:3.4122 train_time:288179ms step_avg:178.66ms step:1624/5100 train_loss:3.5350 train_time:288355ms step_avg:178.66ms step:1625/5100 train_loss:3.4876 train_time:288532ms step_avg:178.66ms step:1625/5100 val_loss:3.5626 train_time:288532ms step_avg:178.66ms step:1626/5100 train_loss:3.5727 train_time:288714ms step_avg:178.66ms step:1627/5100 train_loss:3.5351 train_time:288891ms step_avg:178.66ms step:1628/5100 train_loss:3.4995 train_time:289067ms step_avg:178.66ms step:1629/5100 train_loss:3.6091 train_time:289244ms step_avg:178.66ms step:1630/5100 train_loss:3.5059 train_time:289420ms step_avg:178.65ms step:1631/5100 train_loss:3.5622 train_time:289597ms step_avg:178.65ms step:1632/5100 train_loss:3.4383 train_time:289774ms step_avg:178.65ms step:1633/5100 train_loss:3.4081 train_time:289950ms step_avg:178.65ms step:1634/5100 train_loss:3.5747 train_time:290126ms step_avg:178.65ms step:1635/5100 train_loss:3.5567 train_time:290302ms step_avg:178.65ms step:1636/5100 train_loss:3.4997 train_time:290479ms step_avg:178.65ms step:1637/5100 train_loss:3.5878 train_time:290655ms step_avg:178.64ms step:1638/5100 train_loss:3.6359 train_time:290830ms step_avg:178.64ms step:1639/5100 train_loss:3.6679 train_time:291006ms step_avg:178.64ms step:1640/5100 train_loss:3.8256 train_time:291184ms step_avg:178.64ms step:1641/5100 train_loss:3.6529 train_time:291359ms step_avg:178.64ms step:1642/5100 train_loss:3.5628 train_time:291536ms step_avg:178.64ms step:1643/5100 train_loss:3.6427 train_time:291712ms step_avg:178.64ms step:1644/5100 train_loss:3.5451 train_time:291888ms step_avg:178.63ms step:1645/5100 train_loss:3.5606 train_time:292065ms step_avg:178.63ms step:1646/5100 train_loss:3.5689 train_time:292241ms step_avg:178.63ms step:1647/5100 train_loss:3.3375 train_time:292416ms step_avg:178.63ms step:1648/5100 train_loss:3.5989 train_time:292593ms step_avg:178.63ms step:1649/5100 train_loss:3.4685 train_time:292769ms step_avg:178.63ms step:1650/5100 train_loss:3.5431 train_time:292946ms step_avg:178.63ms step:1651/5100 train_loss:3.5163 train_time:293121ms step_avg:178.62ms step:1652/5100 train_loss:3.5916 train_time:293298ms step_avg:178.62ms step:1653/5100 train_loss:3.5162 train_time:293474ms step_avg:178.62ms step:1654/5100 train_loss:3.6433 train_time:293650ms step_avg:178.62ms step:1655/5100 train_loss:3.6367 train_time:293826ms step_avg:178.62ms step:1656/5100 train_loss:3.4530 train_time:294002ms step_avg:178.62ms step:1657/5100 train_loss:3.6036 train_time:294178ms step_avg:178.61ms step:1658/5100 train_loss:3.5021 train_time:294355ms step_avg:178.61ms step:1659/5100 train_loss:3.4872 train_time:294531ms step_avg:178.61ms step:1660/5100 train_loss:3.5697 train_time:294707ms step_avg:178.61ms step:1661/5100 train_loss:3.5926 train_time:294884ms step_avg:178.61ms step:1662/5100 train_loss:3.5086 train_time:295060ms step_avg:178.61ms step:1663/5100 train_loss:3.6057 train_time:295236ms step_avg:178.61ms step:1664/5100 train_loss:3.6110 train_time:295412ms step_avg:178.60ms step:1665/5100 train_loss:3.6420 train_time:295588ms step_avg:178.60ms step:1666/5100 train_loss:3.6178 train_time:295764ms step_avg:178.60ms step:1667/5100 train_loss:3.7541 train_time:295941ms step_avg:178.60ms step:1668/5100 train_loss:3.4714 train_time:296118ms step_avg:178.60ms step:1669/5100 train_loss:3.5493 train_time:296294ms step_avg:178.60ms step:1670/5100 train_loss:3.4738 train_time:296470ms step_avg:178.60ms step:1671/5100 train_loss:3.4733 train_time:296646ms step_avg:178.60ms step:1672/5100 train_loss:3.6361 train_time:296823ms step_avg:178.59ms step:1673/5100 train_loss:3.8152 train_time:296999ms step_avg:178.59ms step:1674/5100 train_loss:3.5362 train_time:297175ms step_avg:178.59ms step:1675/5100 train_loss:3.5205 train_time:297351ms step_avg:178.59ms step:1676/5100 train_loss:3.4058 train_time:297527ms step_avg:178.59ms step:1677/5100 train_loss:3.6094 train_time:297703ms step_avg:178.59ms step:1678/5100 train_loss:3.5280 train_time:297879ms step_avg:178.58ms step:1679/5100 train_loss:3.5522 train_time:298055ms step_avg:178.58ms step:1680/5100 train_loss:3.5349 train_time:298232ms step_avg:178.58ms step:1681/5100 train_loss:3.3578 train_time:298409ms step_avg:178.58ms step:1682/5100 train_loss:3.5436 train_time:298586ms step_avg:178.58ms step:1683/5100 train_loss:3.5575 train_time:298762ms step_avg:178.58ms step:1684/5100 train_loss:3.6008 train_time:298939ms step_avg:178.58ms step:1685/5100 train_loss:3.6018 train_time:299115ms step_avg:178.58ms step:1686/5100 train_loss:3.5097 train_time:299290ms step_avg:178.57ms step:1687/5100 train_loss:3.6132 train_time:299467ms step_avg:178.57ms step:1688/5100 train_loss:3.4927 train_time:299645ms step_avg:178.57ms step:1689/5100 train_loss:3.5788 train_time:299821ms step_avg:178.57ms step:1690/5100 train_loss:3.4945 train_time:299997ms step_avg:178.57ms step:1691/5100 train_loss:3.3952 train_time:300173ms step_avg:178.57ms step:1692/5100 train_loss:3.5490 train_time:300349ms step_avg:178.57ms step:1693/5100 train_loss:3.5396 train_time:300526ms step_avg:178.57ms step:1694/5100 train_loss:3.4593 train_time:300702ms step_avg:178.56ms step:1695/5100 train_loss:3.9071 train_time:300879ms step_avg:178.56ms step:1696/5100 train_loss:3.6191 train_time:301056ms step_avg:178.56ms step:1697/5100 train_loss:3.5940 train_time:301232ms step_avg:178.56ms step:1698/5100 train_loss:3.5038 train_time:301408ms step_avg:178.56ms step:1699/5100 train_loss:3.4132 train_time:301584ms step_avg:178.56ms step:1700/5100 train_loss:3.5119 train_time:301760ms step_avg:178.56ms step:1701/5100 train_loss:3.5047 train_time:302167ms step_avg:178.69ms step:1702/5100 train_loss:3.5767 train_time:302345ms step_avg:178.69ms step:1703/5100 train_loss:3.5015 train_time:302521ms step_avg:178.69ms step:1704/5100 train_loss:3.7002 train_time:302699ms step_avg:178.69ms step:1705/5100 train_loss:3.4662 train_time:302875ms step_avg:178.69ms step:1706/5100 train_loss:3.6922 train_time:303052ms step_avg:178.69ms step:1707/5100 train_loss:3.5365 train_time:303228ms step_avg:178.68ms step:1708/5100 train_loss:3.3079 train_time:303404ms step_avg:178.68ms step:1709/5100 train_loss:3.6496 train_time:303581ms step_avg:178.68ms step:1710/5100 train_loss:3.5593 train_time:304005ms step_avg:178.83ms step:1711/5100 train_loss:3.5517 train_time:304182ms step_avg:178.83ms step:1712/5100 train_loss:3.5391 train_time:304358ms step_avg:178.82ms step:1713/5100 train_loss:3.5779 train_time:304534ms step_avg:178.82ms step:1714/5100 train_loss:3.6084 train_time:304709ms step_avg:178.82ms step:1715/5100 train_loss:3.5241 train_time:304886ms step_avg:178.82ms step:1716/5100 train_loss:3.5225 train_time:305062ms step_avg:178.82ms step:1717/5100 train_loss:3.3676 train_time:305238ms step_avg:178.82ms step:1718/5100 train_loss:3.5081 train_time:305414ms step_avg:178.81ms step:1719/5100 train_loss:3.5233 train_time:305590ms step_avg:178.81ms step:1720/5100 train_loss:3.4757 train_time:305767ms step_avg:178.81ms step:1721/5100 train_loss:3.6251 train_time:305943ms step_avg:178.81ms step:1722/5100 train_loss:3.4332 train_time:306119ms step_avg:178.81ms step:1723/5100 train_loss:3.5774 train_time:306295ms step_avg:178.81ms step:1724/5100 train_loss:3.6619 train_time:306472ms step_avg:178.81ms step:1725/5100 train_loss:3.5138 train_time:306648ms step_avg:178.80ms step:1726/5100 train_loss:3.7447 train_time:306826ms step_avg:178.80ms step:1727/5100 train_loss:3.5335 train_time:307002ms step_avg:178.80ms step:1728/5100 train_loss:3.5859 train_time:307179ms step_avg:178.80ms step:1729/5100 train_loss:3.5647 train_time:307355ms step_avg:178.80ms step:1730/5100 train_loss:3.5684 train_time:307530ms step_avg:178.80ms step:1731/5100 train_loss:3.9363 train_time:307706ms step_avg:178.80ms step:1732/5100 train_loss:3.5549 train_time:307882ms step_avg:178.79ms step:1733/5100 train_loss:3.6822 train_time:308058ms step_avg:178.79ms step:1734/5100 train_loss:3.4591 train_time:308234ms step_avg:178.79ms step:1735/5100 train_loss:3.4995 train_time:308410ms step_avg:178.79ms step:1736/5100 train_loss:3.5241 train_time:308586ms step_avg:178.79ms step:1737/5100 train_loss:3.5041 train_time:308762ms step_avg:178.79ms step:1738/5100 train_loss:3.6516 train_time:308938ms step_avg:178.78ms step:1739/5100 train_loss:3.5110 train_time:309114ms step_avg:178.78ms step:1740/5100 train_loss:3.5741 train_time:309291ms step_avg:178.78ms step:1741/5100 train_loss:3.6216 train_time:309467ms step_avg:178.78ms step:1742/5100 train_loss:3.4253 train_time:309644ms step_avg:178.78ms step:1743/5100 train_loss:3.3193 train_time:309820ms step_avg:178.78ms step:1744/5100 train_loss:3.2647 train_time:309996ms step_avg:178.78ms step:1745/5100 train_loss:3.5490 train_time:310173ms step_avg:178.77ms step:1746/5100 train_loss:3.5617 train_time:310348ms step_avg:178.77ms step:1747/5100 train_loss:3.5346 train_time:310525ms step_avg:178.77ms step:1748/5100 train_loss:3.5360 train_time:310701ms step_avg:178.77ms step:1749/5100 train_loss:3.7859 train_time:310879ms step_avg:178.77ms step:1750/5100 train_loss:3.4957 train_time:311053ms step_avg:178.77ms step:1750/5100 val_loss:3.5427 train_time:311054ms step_avg:178.77ms step:1751/5100 train_loss:3.5648 train_time:311239ms step_avg:178.77ms step:1752/5100 train_loss:3.5500 train_time:311416ms step_avg:178.77ms step:1753/5100 train_loss:3.1818 train_time:311592ms step_avg:178.77ms step:1754/5100 train_loss:3.3040 train_time:311769ms step_avg:178.77ms step:1755/5100 train_loss:3.4097 train_time:311945ms step_avg:178.76ms step:1756/5100 train_loss:3.3598 train_time:312129ms step_avg:178.77ms step:1757/5100 train_loss:3.5148 train_time:312306ms step_avg:178.77ms step:1758/5100 train_loss:3.3989 train_time:312481ms step_avg:178.77ms step:1759/5100 train_loss:3.3911 train_time:312658ms step_avg:178.76ms step:1760/5100 train_loss:4.4481 train_time:312834ms step_avg:178.76ms step:1761/5100 train_loss:3.5186 train_time:313011ms step_avg:178.76ms step:1762/5100 train_loss:3.5641 train_time:313188ms step_avg:178.76ms step:1763/5100 train_loss:3.5610 train_time:313367ms step_avg:178.76ms step:1764/5100 train_loss:3.5843 train_time:313543ms step_avg:178.76ms step:1765/5100 train_loss:3.4918 train_time:313719ms step_avg:178.76ms step:1766/5100 train_loss:3.5324 train_time:313895ms step_avg:178.76ms step:1767/5100 train_loss:3.5538 train_time:314072ms step_avg:178.75ms step:1768/5100 train_loss:3.8009 train_time:314248ms step_avg:178.75ms step:1769/5100 train_loss:3.5296 train_time:314424ms step_avg:178.75ms step:1770/5100 train_loss:3.5935 train_time:314600ms step_avg:178.75ms step:1771/5100 train_loss:3.9556 train_time:314777ms step_avg:178.75ms step:1772/5100 train_loss:3.5282 train_time:314954ms step_avg:178.75ms step:1773/5100 train_loss:3.4318 train_time:315130ms step_avg:178.75ms step:1774/5100 train_loss:3.6923 train_time:315306ms step_avg:178.74ms step:1775/5100 train_loss:3.4260 train_time:315482ms step_avg:178.74ms step:1776/5100 train_loss:3.5899 train_time:315658ms step_avg:178.74ms step:1777/5100 train_loss:3.6514 train_time:315833ms step_avg:178.74ms step:1778/5100 train_loss:3.7285 train_time:316010ms step_avg:178.74ms step:1779/5100 train_loss:3.5401 train_time:316187ms step_avg:178.74ms step:1780/5100 train_loss:3.8367 train_time:316363ms step_avg:178.74ms step:1781/5100 train_loss:3.6106 train_time:316539ms step_avg:178.73ms step:1782/5100 train_loss:3.6241 train_time:316716ms step_avg:178.73ms step:1783/5100 train_loss:3.4107 train_time:316893ms step_avg:178.73ms step:1784/5100 train_loss:3.4922 train_time:317069ms step_avg:178.73ms step:1785/5100 train_loss:3.6315 train_time:317245ms step_avg:178.73ms step:1786/5100 train_loss:3.5255 train_time:317421ms step_avg:178.73ms step:1787/5100 train_loss:3.6910 train_time:317597ms step_avg:178.73ms step:1788/5100 train_loss:3.5100 train_time:317774ms step_avg:178.73ms step:1789/5100 train_loss:3.4780 train_time:317950ms step_avg:178.72ms step:1790/5100 train_loss:3.6300 train_time:318127ms step_avg:178.72ms step:1791/5100 train_loss:3.5212 train_time:318303ms step_avg:178.72ms step:1792/5100 train_loss:3.4779 train_time:318479ms step_avg:178.72ms step:1793/5100 train_loss:3.6071 train_time:318655ms step_avg:178.72ms step:1794/5100 train_loss:3.4859 train_time:318832ms step_avg:178.72ms step:1795/5100 train_loss:3.4779 train_time:319009ms step_avg:178.72ms step:1796/5100 train_loss:3.5305 train_time:319185ms step_avg:178.72ms step:1797/5100 train_loss:3.4954 train_time:319361ms step_avg:178.71ms step:1798/5100 train_loss:3.6290 train_time:319537ms step_avg:178.71ms step:1799/5100 train_loss:3.5157 train_time:319713ms step_avg:178.71ms step:1800/5100 train_loss:3.5936 train_time:319890ms step_avg:178.71ms step:1801/5100 train_loss:3.5308 train_time:320066ms step_avg:178.71ms step:1802/5100 train_loss:3.5600 train_time:320242ms step_avg:178.71ms step:1803/5100 train_loss:3.4683 train_time:320418ms step_avg:178.71ms step:1804/5100 train_loss:3.4003 train_time:320595ms step_avg:178.70ms step:1805/5100 train_loss:3.6484 train_time:320771ms step_avg:178.70ms step:1806/5100 train_loss:3.5743 train_time:320948ms step_avg:178.70ms step:1807/5100 train_loss:3.5811 train_time:321123ms step_avg:178.70ms step:1808/5100 train_loss:3.6973 train_time:321299ms step_avg:178.70ms step:1809/5100 train_loss:3.4901 train_time:321476ms step_avg:178.70ms step:1810/5100 train_loss:3.5870 train_time:321653ms step_avg:178.70ms step:1811/5100 train_loss:3.7265 train_time:321830ms step_avg:178.69ms step:1812/5100 train_loss:3.5795 train_time:322006ms step_avg:178.69ms step:1813/5100 train_loss:3.6183 train_time:322182ms step_avg:178.69ms step:1814/5100 train_loss:3.6442 train_time:322358ms step_avg:178.69ms step:1815/5100 train_loss:3.5886 train_time:322534ms step_avg:178.69ms step:1816/5100 train_loss:3.6298 train_time:322711ms step_avg:178.69ms step:1817/5100 train_loss:3.5784 train_time:322887ms step_avg:178.69ms step:1818/5100 train_loss:3.6335 train_time:323063ms step_avg:178.69ms step:1819/5100 train_loss:3.5541 train_time:323239ms step_avg:178.68ms step:1820/5100 train_loss:3.5456 train_time:323416ms step_avg:178.68ms step:1821/5100 train_loss:3.4987 train_time:323592ms step_avg:178.68ms step:1822/5100 train_loss:3.4749 train_time:323768ms step_avg:178.68ms step:1823/5100 train_loss:3.4049 train_time:323945ms step_avg:178.68ms step:1824/5100 train_loss:3.5626 train_time:324121ms step_avg:178.68ms step:1825/5100 train_loss:3.6811 train_time:324297ms step_avg:178.68ms step:1826/5100 train_loss:3.6381 train_time:324474ms step_avg:178.68ms step:1827/5100 train_loss:3.6158 train_time:324651ms step_avg:178.67ms step:1828/5100 train_loss:3.4859 train_time:324827ms step_avg:178.67ms step:1829/5100 train_loss:3.5117 train_time:325003ms step_avg:178.67ms step:1830/5100 train_loss:3.6437 train_time:325179ms step_avg:178.67ms step:1831/5100 train_loss:3.4182 train_time:325355ms step_avg:178.67ms step:1832/5100 train_loss:3.5748 train_time:325531ms step_avg:178.67ms step:1833/5100 train_loss:3.4463 train_time:325707ms step_avg:178.67ms step:1834/5100 train_loss:3.7724 train_time:325883ms step_avg:178.66ms step:1835/5100 train_loss:3.6038 train_time:326060ms step_avg:178.66ms step:1836/5100 train_loss:3.5878 train_time:326237ms step_avg:178.66ms step:1837/5100 train_loss:3.7139 train_time:326414ms step_avg:178.66ms step:1838/5100 train_loss:3.5715 train_time:326590ms step_avg:178.66ms step:1839/5100 train_loss:3.4514 train_time:326766ms step_avg:178.66ms step:1840/5100 train_loss:3.5646 train_time:326943ms step_avg:178.66ms step:1841/5100 train_loss:3.4535 train_time:327119ms step_avg:178.66ms step:1842/5100 train_loss:3.5687 train_time:327297ms step_avg:178.66ms step:1843/5100 train_loss:3.6213 train_time:327473ms step_avg:178.65ms step:1844/5100 train_loss:3.3679 train_time:327649ms step_avg:178.65ms step:1845/5100 train_loss:3.4920 train_time:327825ms step_avg:178.65ms step:1846/5100 train_loss:3.5561 train_time:328001ms step_avg:178.65ms step:1847/5100 train_loss:3.4960 train_time:328177ms step_avg:178.65ms step:1848/5100 train_loss:3.3935 train_time:328355ms step_avg:178.65ms step:1849/5100 train_loss:3.6573 train_time:328530ms step_avg:178.65ms step:1850/5100 train_loss:3.4239 train_time:328708ms step_avg:178.65ms step:1851/5100 train_loss:3.5152 train_time:328884ms step_avg:178.64ms step:1852/5100 train_loss:3.4717 train_time:329060ms step_avg:178.64ms step:1853/5100 train_loss:3.6727 train_time:329236ms step_avg:178.64ms step:1854/5100 train_loss:3.6430 train_time:329413ms step_avg:178.64ms step:1855/5100 train_loss:3.5197 train_time:329590ms step_avg:178.64ms step:1856/5100 train_loss:3.4723 train_time:329768ms step_avg:178.64ms step:1857/5100 train_loss:3.5038 train_time:329945ms step_avg:178.64ms step:1858/5100 train_loss:3.7537 train_time:330121ms step_avg:178.64ms step:1859/5100 train_loss:3.5820 train_time:330297ms step_avg:178.64ms step:1860/5100 train_loss:3.5282 train_time:330474ms step_avg:178.63ms step:1861/5100 train_loss:3.5747 train_time:330651ms step_avg:178.63ms step:1862/5100 train_loss:3.4668 train_time:330828ms step_avg:178.63ms step:1863/5100 train_loss:3.4571 train_time:331004ms step_avg:178.63ms step:1864/5100 train_loss:3.5374 train_time:331180ms step_avg:178.63ms step:1865/5100 train_loss:3.5726 train_time:331356ms step_avg:178.63ms step:1866/5100 train_loss:3.3363 train_time:331532ms step_avg:178.63ms step:1867/5100 train_loss:3.4646 train_time:331709ms step_avg:178.63ms step:1868/5100 train_loss:3.4189 train_time:331885ms step_avg:178.63ms step:1869/5100 train_loss:3.4188 train_time:332061ms step_avg:178.62ms step:1870/5100 train_loss:3.5800 train_time:332238ms step_avg:178.62ms step:1871/5100 train_loss:3.5668 train_time:332416ms step_avg:178.62ms step:1872/5100 train_loss:3.5138 train_time:332593ms step_avg:178.62ms step:1873/5100 train_loss:3.5205 train_time:332769ms step_avg:178.62ms step:1874/5100 train_loss:3.4555 train_time:332946ms step_avg:178.62ms step:1875/5100 train_loss:3.5574 train_time:333123ms step_avg:178.62ms step:1875/5100 val_loss:3.5287 train_time:333124ms step_avg:178.62ms step:1876/5100 train_loss:3.5533 train_time:333311ms step_avg:178.62ms step:1877/5100 train_loss:3.4800 train_time:333487ms step_avg:178.62ms step:1878/5100 train_loss:3.5238 train_time:333664ms step_avg:178.62ms step:1879/5100 train_loss:3.6278 train_time:333840ms step_avg:178.62ms step:1880/5100 train_loss:3.5161 train_time:334016ms step_avg:178.62ms step:1881/5100 train_loss:3.5686 train_time:334192ms step_avg:178.62ms step:1882/5100 train_loss:3.4825 train_time:334368ms step_avg:178.62ms step:1883/5100 train_loss:3.5591 train_time:334545ms step_avg:178.61ms step:1884/5100 train_loss:3.5519 train_time:334721ms step_avg:178.61ms step:1885/5100 train_loss:3.3048 train_time:334898ms step_avg:178.61ms step:1886/5100 train_loss:3.7025 train_time:335073ms step_avg:178.61ms step:1887/5100 train_loss:3.4350 train_time:335250ms step_avg:178.61ms step:1888/5100 train_loss:3.4575 train_time:335426ms step_avg:178.61ms step:1889/5100 train_loss:3.5279 train_time:335603ms step_avg:178.61ms step:1890/5100 train_loss:3.5694 train_time:336009ms step_avg:178.73ms step:1891/5100 train_loss:3.3955 train_time:336186ms step_avg:178.73ms step:1892/5100 train_loss:3.6692 train_time:336361ms step_avg:178.73ms step:1893/5100 train_loss:3.4273 train_time:336538ms step_avg:178.72ms step:1894/5100 train_loss:3.5497 train_time:336714ms step_avg:178.72ms step:1895/5100 train_loss:3.5946 train_time:336891ms step_avg:178.72ms step:1896/5100 train_loss:3.3920 train_time:337067ms step_avg:178.72ms step:1897/5100 train_loss:3.5555 train_time:337244ms step_avg:178.72ms step:1898/5100 train_loss:3.5225 train_time:337421ms step_avg:178.72ms step:1899/5100 train_loss:3.5978 train_time:337597ms step_avg:178.72ms step:1900/5100 train_loss:3.3768 train_time:338018ms step_avg:178.85ms step:1901/5100 train_loss:3.6139 train_time:338195ms step_avg:178.84ms step:1902/5100 train_loss:3.5053 train_time:338371ms step_avg:178.84ms step:1903/5100 train_loss:3.6625 train_time:338547ms step_avg:178.84ms step:1904/5100 train_loss:3.4701 train_time:338724ms step_avg:178.84ms step:1905/5100 train_loss:3.7531 train_time:338900ms step_avg:178.84ms step:1906/5100 train_loss:3.4817 train_time:339075ms step_avg:178.84ms step:1907/5100 train_loss:3.4768 train_time:339250ms step_avg:178.84ms step:1908/5100 train_loss:3.5510 train_time:339428ms step_avg:178.83ms step:1909/5100 train_loss:3.4331 train_time:339604ms step_avg:178.83ms step:1910/5100 train_loss:3.5038 train_time:339782ms step_avg:178.83ms step:1911/5100 train_loss:3.5960 train_time:339958ms step_avg:178.83ms step:1912/5100 train_loss:3.5176 train_time:340134ms step_avg:178.83ms step:1913/5100 train_loss:3.3970 train_time:340310ms step_avg:178.83ms step:1914/5100 train_loss:3.2651 train_time:340486ms step_avg:178.83ms step:1915/5100 train_loss:3.4666 train_time:340662ms step_avg:178.83ms step:1916/5100 train_loss:3.6877 train_time:340838ms step_avg:178.82ms step:1917/5100 train_loss:3.6865 train_time:341015ms step_avg:178.82ms step:1918/5100 train_loss:3.6327 train_time:341190ms step_avg:178.82ms step:1919/5100 train_loss:3.4643 train_time:341367ms step_avg:178.82ms step:1920/5100 train_loss:3.7081 train_time:341543ms step_avg:178.82ms step:1921/5100 train_loss:3.5302 train_time:341720ms step_avg:178.82ms step:1922/5100 train_loss:3.4610 train_time:341895ms step_avg:178.82ms step:1923/5100 train_loss:3.6404 train_time:342071ms step_avg:178.81ms step:1924/5100 train_loss:3.6056 train_time:342248ms step_avg:178.81ms step:1925/5100 train_loss:3.4403 train_time:342425ms step_avg:178.81ms step:1926/5100 train_loss:3.4816 train_time:342600ms step_avg:178.81ms step:1927/5100 train_loss:3.3910 train_time:342776ms step_avg:178.81ms step:1928/5100 train_loss:3.4974 train_time:342952ms step_avg:178.81ms step:1929/5100 train_loss:3.3532 train_time:343128ms step_avg:178.81ms step:1930/5100 train_loss:3.4659 train_time:343304ms step_avg:178.80ms step:1931/5100 train_loss:3.6006 train_time:343481ms step_avg:178.80ms step:1932/5100 train_loss:3.4734 train_time:343657ms step_avg:178.80ms step:1933/5100 train_loss:3.6112 train_time:343833ms step_avg:178.80ms step:1934/5100 train_loss:3.4859 train_time:344009ms step_avg:178.80ms step:1935/5100 train_loss:3.5333 train_time:344186ms step_avg:178.80ms step:1936/5100 train_loss:3.5690 train_time:344363ms step_avg:178.80ms step:1937/5100 train_loss:3.5227 train_time:344539ms step_avg:178.80ms step:1938/5100 train_loss:3.5510 train_time:344716ms step_avg:178.79ms step:1939/5100 train_loss:3.4752 train_time:344892ms step_avg:178.79ms step:1940/5100 train_loss:3.5724 train_time:345068ms step_avg:178.79ms step:1941/5100 train_loss:3.6016 train_time:345245ms step_avg:178.79ms step:1942/5100 train_loss:3.4431 train_time:345422ms step_avg:178.79ms step:1943/5100 train_loss:3.4797 train_time:345599ms step_avg:178.79ms step:1944/5100 train_loss:3.5485 train_time:345775ms step_avg:178.79ms step:1945/5100 train_loss:3.3866 train_time:345951ms step_avg:178.79ms step:1946/5100 train_loss:3.6594 train_time:346127ms step_avg:178.78ms step:1947/5100 train_loss:3.5373 train_time:346304ms step_avg:178.78ms step:1948/5100 train_loss:3.5162 train_time:346480ms step_avg:178.78ms step:1949/5100 train_loss:3.5166 train_time:346655ms step_avg:178.78ms step:1950/5100 train_loss:3.3923 train_time:346831ms step_avg:178.78ms step:1951/5100 train_loss:3.5169 train_time:347007ms step_avg:178.78ms step:1952/5100 train_loss:3.3632 train_time:347183ms step_avg:178.78ms step:1953/5100 train_loss:3.5714 train_time:347359ms step_avg:178.77ms step:1954/5100 train_loss:3.5670 train_time:347536ms step_avg:178.77ms step:1955/5100 train_loss:3.5155 train_time:347712ms step_avg:178.77ms step:1956/5100 train_loss:3.4080 train_time:347890ms step_avg:178.77ms step:1957/5100 train_loss:3.4946 train_time:348065ms step_avg:178.77ms step:1958/5100 train_loss:3.6836 train_time:348241ms step_avg:178.77ms step:1959/5100 train_loss:3.6064 train_time:348418ms step_avg:178.77ms step:1960/5100 train_loss:3.6224 train_time:348594ms step_avg:178.77ms step:1961/5100 train_loss:3.4193 train_time:348770ms step_avg:178.76ms step:1962/5100 train_loss:3.5462 train_time:348946ms step_avg:178.76ms step:1963/5100 train_loss:3.5921 train_time:349123ms step_avg:178.76ms step:1964/5100 train_loss:3.5375 train_time:349300ms step_avg:178.76ms step:1965/5100 train_loss:3.4448 train_time:349476ms step_avg:178.76ms step:1966/5100 train_loss:3.8454 train_time:349653ms step_avg:178.76ms step:1967/5100 train_loss:3.4585 train_time:349829ms step_avg:178.76ms step:1968/5100 train_loss:3.4970 train_time:350005ms step_avg:178.76ms step:1969/5100 train_loss:3.5495 train_time:350182ms step_avg:178.76ms step:1970/5100 train_loss:3.5073 train_time:350358ms step_avg:178.75ms step:1971/5100 train_loss:3.4005 train_time:350534ms step_avg:178.75ms step:1972/5100 train_loss:3.3853 train_time:350710ms step_avg:178.75ms step:1973/5100 train_loss:3.5017 train_time:350887ms step_avg:178.75ms step:1974/5100 train_loss:3.4711 train_time:351062ms step_avg:178.75ms step:1975/5100 train_loss:3.4464 train_time:351239ms step_avg:178.75ms step:1976/5100 train_loss:3.6019 train_time:351414ms step_avg:178.75ms step:1977/5100 train_loss:3.4712 train_time:351591ms step_avg:178.74ms step:1978/5100 train_loss:3.8415 train_time:351766ms step_avg:178.74ms step:1979/5100 train_loss:3.5271 train_time:351942ms step_avg:178.74ms step:1980/5100 train_loss:3.5231 train_time:352119ms step_avg:178.74ms step:1981/5100 train_loss:3.5278 train_time:352295ms step_avg:178.74ms step:1982/5100 train_loss:3.5500 train_time:352471ms step_avg:178.74ms step:1983/5100 train_loss:3.4835 train_time:352647ms step_avg:178.74ms step:1984/5100 train_loss:3.4485 train_time:352823ms step_avg:178.74ms step:1985/5100 train_loss:3.5003 train_time:353000ms step_avg:178.73ms step:1986/5100 train_loss:3.5684 train_time:353175ms step_avg:178.73ms step:1987/5100 train_loss:3.5402 train_time:353352ms step_avg:178.73ms step:1988/5100 train_loss:3.5045 train_time:353529ms step_avg:178.73ms step:1989/5100 train_loss:3.5972 train_time:353706ms step_avg:178.73ms step:1990/5100 train_loss:3.6314 train_time:353883ms step_avg:178.73ms step:1991/5100 train_loss:3.4015 train_time:354059ms step_avg:178.73ms step:1992/5100 train_loss:3.3997 train_time:354235ms step_avg:178.73ms step:1993/5100 train_loss:3.5818 train_time:354412ms step_avg:178.72ms step:1994/5100 train_loss:3.4093 train_time:354588ms step_avg:178.72ms step:1995/5100 train_loss:3.4934 train_time:354764ms step_avg:178.72ms step:1996/5100 train_loss:3.5696 train_time:354941ms step_avg:178.72ms step:1997/5100 train_loss:3.4334 train_time:355117ms step_avg:178.72ms step:1998/5100 train_loss:3.5364 train_time:355293ms step_avg:178.72ms step:1999/5100 train_loss:3.5405 train_time:355470ms step_avg:178.72ms step:2000/5100 train_loss:3.4596 train_time:355647ms step_avg:178.72ms step:2000/5100 val_loss:3.5143 train_time:355648ms step_avg:178.72ms step:2001/5100 train_loss:3.6116 train_time:355833ms step_avg:178.72ms step:2002/5100 train_loss:3.5520 train_time:356010ms step_avg:178.72ms step:2003/5100 train_loss:3.6410 train_time:356187ms step_avg:178.72ms step:2004/5100 train_loss:3.5513 train_time:356363ms step_avg:178.72ms step:2005/5100 train_loss:3.5635 train_time:356540ms step_avg:178.72ms step:2006/5100 train_loss:3.4540 train_time:356717ms step_avg:178.72ms step:2007/5100 train_loss:3.4821 train_time:356893ms step_avg:178.71ms step:2008/5100 train_loss:3.5246 train_time:357070ms step_avg:178.71ms step:2009/5100 train_loss:3.5693 train_time:357246ms step_avg:178.71ms step:2010/5100 train_loss:3.4660 train_time:357422ms step_avg:178.71ms step:2011/5100 train_loss:3.5500 train_time:357599ms step_avg:178.71ms step:2012/5100 train_loss:3.5242 train_time:357775ms step_avg:178.71ms step:2013/5100 train_loss:3.5283 train_time:357952ms step_avg:178.71ms step:2014/5100 train_loss:3.4468 train_time:358127ms step_avg:178.71ms step:2015/5100 train_loss:3.4936 train_time:358304ms step_avg:178.71ms step:2016/5100 train_loss:3.5033 train_time:358479ms step_avg:178.70ms step:2017/5100 train_loss:3.6440 train_time:358656ms step_avg:178.70ms step:2018/5100 train_loss:3.4887 train_time:358832ms step_avg:178.70ms step:2019/5100 train_loss:3.6452 train_time:359008ms step_avg:178.70ms step:2020/5100 train_loss:3.6475 train_time:359184ms step_avg:178.70ms step:2021/5100 train_loss:3.3650 train_time:359360ms step_avg:178.70ms step:2022/5100 train_loss:3.5957 train_time:359537ms step_avg:178.70ms step:2023/5100 train_loss:3.5179 train_time:359714ms step_avg:178.70ms step:2024/5100 train_loss:3.6100 train_time:359890ms step_avg:178.69ms step:2025/5100 train_loss:3.6559 train_time:360068ms step_avg:178.69ms step:2026/5100 train_loss:3.4362 train_time:360245ms step_avg:178.69ms step:2027/5100 train_loss:3.4791 train_time:360422ms step_avg:178.69ms step:2028/5100 train_loss:3.3816 train_time:360598ms step_avg:178.69ms step:2029/5100 train_loss:3.4964 train_time:360774ms step_avg:178.69ms step:2030/5100 train_loss:3.4203 train_time:360951ms step_avg:178.69ms step:2031/5100 train_loss:3.5057 train_time:361126ms step_avg:178.69ms step:2032/5100 train_loss:3.5024 train_time:361303ms step_avg:178.69ms step:2033/5100 train_loss:3.5108 train_time:361480ms step_avg:178.68ms step:2034/5100 train_loss:3.4102 train_time:361657ms step_avg:178.68ms step:2035/5100 train_loss:3.5806 train_time:361833ms step_avg:178.68ms step:2036/5100 train_loss:3.5695 train_time:362009ms step_avg:178.68ms step:2037/5100 train_loss:3.5581 train_time:362185ms step_avg:178.68ms step:2038/5100 train_loss:3.4318 train_time:362362ms step_avg:178.68ms step:2039/5100 train_loss:3.6908 train_time:362538ms step_avg:178.68ms step:2040/5100 train_loss:3.5255 train_time:362714ms step_avg:178.68ms step:2041/5100 train_loss:3.5441 train_time:362891ms step_avg:178.68ms step:2042/5100 train_loss:3.4907 train_time:363066ms step_avg:178.67ms step:2043/5100 train_loss:3.3874 train_time:363243ms step_avg:178.67ms step:2044/5100 train_loss:3.5203 train_time:363419ms step_avg:178.67ms step:2045/5100 train_loss:3.5173 train_time:363596ms step_avg:178.67ms step:2046/5100 train_loss:3.3800 train_time:363773ms step_avg:178.67ms step:2047/5100 train_loss:3.4554 train_time:363949ms step_avg:178.67ms step:2048/5100 train_loss:3.5304 train_time:364125ms step_avg:178.67ms step:2049/5100 train_loss:3.4811 train_time:364301ms step_avg:178.67ms step:2050/5100 train_loss:3.5251 train_time:364477ms step_avg:178.67ms step:2051/5100 train_loss:3.6738 train_time:364654ms step_avg:178.66ms step:2052/5100 train_loss:3.5395 train_time:364830ms step_avg:178.66ms step:2053/5100 train_loss:3.4921 train_time:365006ms step_avg:178.66ms step:2054/5100 train_loss:3.4679 train_time:365183ms step_avg:178.66ms step:2055/5100 train_loss:3.3322 train_time:365360ms step_avg:178.66ms step:2056/5100 train_loss:3.4509 train_time:365537ms step_avg:178.66ms step:2057/5100 train_loss:3.6325 train_time:365714ms step_avg:178.66ms step:2058/5100 train_loss:3.6491 train_time:365890ms step_avg:178.66ms step:2059/5100 train_loss:3.5100 train_time:366066ms step_avg:178.66ms step:2060/5100 train_loss:3.5494 train_time:366242ms step_avg:178.65ms step:2061/5100 train_loss:3.5394 train_time:366419ms step_avg:178.65ms step:2062/5100 train_loss:3.4959 train_time:366596ms step_avg:178.65ms step:2063/5100 train_loss:3.4008 train_time:366773ms step_avg:178.65ms step:2064/5100 train_loss:3.7117 train_time:366950ms step_avg:178.65ms step:2065/5100 train_loss:3.5743 train_time:367126ms step_avg:178.65ms step:2066/5100 train_loss:3.5266 train_time:367302ms step_avg:178.65ms step:2067/5100 train_loss:3.5660 train_time:367478ms step_avg:178.65ms step:2068/5100 train_loss:3.4679 train_time:367655ms step_avg:178.65ms step:2069/5100 train_loss:3.5258 train_time:367831ms step_avg:178.65ms step:2070/5100 train_loss:3.6538 train_time:368007ms step_avg:178.64ms step:2071/5100 train_loss:3.6559 train_time:368184ms step_avg:178.64ms step:2072/5100 train_loss:3.5118 train_time:368362ms step_avg:178.64ms step:2073/5100 train_loss:3.5428 train_time:368537ms step_avg:178.64ms step:2074/5100 train_loss:3.4301 train_time:368714ms step_avg:178.64ms step:2075/5100 train_loss:3.9629 train_time:368891ms step_avg:178.64ms step:2076/5100 train_loss:3.3869 train_time:369067ms step_avg:178.64ms step:2077/5100 train_loss:3.5513 train_time:369243ms step_avg:178.64ms step:2078/5100 train_loss:3.4488 train_time:369420ms step_avg:178.64ms step:2079/5100 train_loss:3.4190 train_time:369837ms step_avg:178.75ms step:2080/5100 train_loss:3.5112 train_time:370012ms step_avg:178.75ms step:2081/5100 train_loss:3.7675 train_time:370188ms step_avg:178.75ms step:2082/5100 train_loss:3.3918 train_time:370365ms step_avg:178.75ms step:2083/5100 train_loss:3.7323 train_time:370541ms step_avg:178.75ms step:2084/5100 train_loss:3.4374 train_time:370717ms step_avg:178.74ms step:2085/5100 train_loss:3.4184 train_time:370894ms step_avg:178.74ms step:2086/5100 train_loss:3.6687 train_time:371070ms step_avg:178.74ms step:2087/5100 train_loss:3.5937 train_time:371245ms step_avg:178.74ms step:2088/5100 train_loss:3.5745 train_time:371422ms step_avg:178.74ms step:2089/5100 train_loss:3.6346 train_time:371598ms step_avg:178.74ms step:2090/5100 train_loss:3.5540 train_time:372016ms step_avg:178.85ms step:2091/5100 train_loss:3.5514 train_time:372193ms step_avg:178.85ms step:2092/5100 train_loss:3.5016 train_time:372370ms step_avg:178.85ms step:2093/5100 train_loss:3.5720 train_time:372545ms step_avg:178.85ms step:2094/5100 train_loss:3.4676 train_time:372722ms step_avg:178.85ms step:2095/5100 train_loss:3.2637 train_time:372898ms step_avg:178.85ms step:2096/5100 train_loss:3.4951 train_time:373073ms step_avg:178.85ms step:2097/5100 train_loss:3.6643 train_time:373249ms step_avg:178.84ms step:2098/5100 train_loss:3.4932 train_time:373428ms step_avg:178.84ms step:2099/5100 train_loss:3.3862 train_time:373603ms step_avg:178.84ms step:2100/5100 train_loss:3.4824 train_time:373780ms step_avg:178.84ms step:2101/5100 train_loss:3.4460 train_time:373957ms step_avg:178.84ms step:2102/5100 train_loss:3.5881 train_time:374132ms step_avg:178.84ms step:2103/5100 train_loss:3.4184 train_time:374309ms step_avg:178.84ms step:2104/5100 train_loss:3.3832 train_time:374485ms step_avg:178.84ms step:2105/5100 train_loss:3.6457 train_time:374661ms step_avg:178.84ms step:2106/5100 train_loss:3.3767 train_time:374839ms step_avg:178.84ms step:2107/5100 train_loss:3.7709 train_time:375015ms step_avg:178.83ms step:2108/5100 train_loss:3.6103 train_time:375191ms step_avg:178.83ms step:2109/5100 train_loss:3.5090 train_time:375367ms step_avg:178.83ms step:2110/5100 train_loss:3.5322 train_time:375544ms step_avg:178.83ms step:2111/5100 train_loss:3.3566 train_time:375721ms step_avg:178.83ms step:2112/5100 train_loss:3.8385 train_time:375897ms step_avg:178.83ms step:2113/5100 train_loss:3.5252 train_time:376074ms step_avg:178.83ms step:2114/5100 train_loss:3.4594 train_time:376250ms step_avg:178.83ms step:2115/5100 train_loss:3.5715 train_time:376426ms step_avg:178.82ms step:2116/5100 train_loss:3.5300 train_time:376602ms step_avg:178.82ms step:2117/5100 train_loss:3.5145 train_time:376779ms step_avg:178.82ms step:2118/5100 train_loss:3.5693 train_time:376956ms step_avg:178.82ms step:2119/5100 train_loss:3.4316 train_time:377132ms step_avg:178.82ms step:2120/5100 train_loss:3.4929 train_time:377308ms step_avg:178.82ms step:2121/5100 train_loss:3.2025 train_time:377484ms step_avg:178.82ms step:2122/5100 train_loss:3.3916 train_time:377661ms step_avg:178.82ms step:2123/5100 train_loss:3.5606 train_time:377837ms step_avg:178.82ms step:2124/5100 train_loss:3.4761 train_time:378013ms step_avg:178.81ms step:2125/5100 train_loss:3.6261 train_time:378190ms step_avg:178.81ms step:2125/5100 val_loss:3.5030 train_time:378190ms step_avg:178.81ms step:2126/5100 train_loss:3.4902 train_time:378376ms step_avg:178.82ms step:2127/5100 train_loss:3.6076 train_time:378552ms step_avg:178.82ms step:2128/5100 train_loss:3.5856 train_time:378728ms step_avg:178.81ms step:2129/5100 train_loss:3.4467 train_time:378905ms step_avg:178.81ms step:2130/5100 train_loss:3.4310 train_time:379081ms step_avg:178.81ms step:2131/5100 train_loss:3.4541 train_time:379257ms step_avg:178.81ms step:2132/5100 train_loss:3.6070 train_time:379434ms step_avg:178.81ms step:2133/5100 train_loss:3.4935 train_time:379611ms step_avg:178.81ms step:2134/5100 train_loss:3.3940 train_time:379788ms step_avg:178.81ms step:2135/5100 train_loss:3.4563 train_time:379965ms step_avg:178.81ms step:2136/5100 train_loss:3.5803 train_time:380140ms step_avg:178.81ms step:2137/5100 train_loss:3.5920 train_time:380317ms step_avg:178.80ms step:2138/5100 train_loss:3.5398 train_time:380493ms step_avg:178.80ms step:2139/5100 train_loss:3.5318 train_time:380669ms step_avg:178.80ms step:2140/5100 train_loss:3.5164 train_time:380846ms step_avg:178.80ms step:2141/5100 train_loss:3.5961 train_time:381022ms step_avg:178.80ms step:2142/5100 train_loss:3.9100 train_time:381198ms step_avg:178.80ms step:2143/5100 train_loss:3.4296 train_time:381374ms step_avg:178.80ms step:2144/5100 train_loss:3.4591 train_time:381550ms step_avg:178.80ms step:2145/5100 train_loss:3.5038 train_time:381727ms step_avg:178.79ms step:2146/5100 train_loss:3.6349 train_time:381903ms step_avg:178.79ms step:2147/5100 train_loss:3.5593 train_time:382079ms step_avg:178.79ms step:2148/5100 train_loss:3.9696 train_time:382256ms step_avg:178.79ms step:2149/5100 train_loss:3.4888 train_time:382433ms step_avg:178.79ms step:2150/5100 train_loss:3.4600 train_time:382610ms step_avg:178.79ms step:2151/5100 train_loss:3.5216 train_time:382786ms step_avg:178.79ms step:2152/5100 train_loss:3.5593 train_time:382963ms step_avg:178.79ms step:2153/5100 train_loss:3.5089 train_time:383139ms step_avg:178.79ms step:2154/5100 train_loss:3.4445 train_time:383316ms step_avg:178.79ms step:2155/5100 train_loss:3.6552 train_time:383494ms step_avg:178.78ms step:2156/5100 train_loss:3.2825 train_time:383671ms step_avg:178.78ms step:2157/5100 train_loss:3.4406 train_time:383847ms step_avg:178.78ms step:2158/5100 train_loss:3.5732 train_time:384023ms step_avg:178.78ms step:2159/5100 train_loss:3.5122 train_time:384199ms step_avg:178.78ms step:2160/5100 train_loss:3.6721 train_time:384376ms step_avg:178.78ms step:2161/5100 train_loss:3.5897 train_time:384551ms step_avg:178.78ms step:2162/5100 train_loss:3.5086 train_time:384728ms step_avg:178.78ms step:2163/5100 train_loss:3.4823 train_time:384904ms step_avg:178.78ms step:2164/5100 train_loss:3.4860 train_time:385080ms step_avg:178.77ms step:2165/5100 train_loss:3.5636 train_time:385256ms step_avg:178.77ms step:2166/5100 train_loss:3.5854 train_time:385432ms step_avg:178.77ms step:2167/5100 train_loss:3.5157 train_time:385609ms step_avg:178.77ms step:2168/5100 train_loss:3.4113 train_time:385786ms step_avg:178.77ms step:2169/5100 train_loss:3.4917 train_time:385962ms step_avg:178.77ms step:2170/5100 train_loss:3.5397 train_time:386138ms step_avg:178.77ms step:2171/5100 train_loss:3.6585 train_time:386315ms step_avg:178.77ms step:2172/5100 train_loss:3.4573 train_time:386491ms step_avg:178.77ms step:2173/5100 train_loss:3.4478 train_time:386668ms step_avg:178.76ms step:2174/5100 train_loss:3.4474 train_time:386844ms step_avg:178.76ms step:2175/5100 train_loss:3.5077 train_time:387020ms step_avg:178.76ms step:2176/5100 train_loss:3.4674 train_time:387197ms step_avg:178.76ms step:2177/5100 train_loss:3.4334 train_time:387373ms step_avg:178.76ms step:2178/5100 train_loss:3.6553 train_time:387549ms step_avg:178.76ms step:2179/5100 train_loss:3.4777 train_time:387726ms step_avg:178.76ms step:2180/5100 train_loss:3.4930 train_time:387902ms step_avg:178.76ms step:2181/5100 train_loss:3.5495 train_time:388077ms step_avg:178.76ms step:2182/5100 train_loss:3.5263 train_time:388254ms step_avg:178.75ms step:2183/5100 train_loss:3.5019 train_time:388430ms step_avg:178.75ms step:2184/5100 train_loss:3.3892 train_time:388607ms step_avg:178.75ms step:2185/5100 train_loss:3.5664 train_time:388783ms step_avg:178.75ms step:2186/5100 train_loss:3.7454 train_time:388959ms step_avg:178.75ms step:2187/5100 train_loss:3.3776 train_time:389135ms step_avg:178.75ms step:2188/5100 train_loss:3.4242 train_time:389313ms step_avg:178.75ms step:2189/5100 train_loss:3.2771 train_time:389489ms step_avg:178.75ms step:2190/5100 train_loss:3.4262 train_time:389666ms step_avg:178.75ms step:2191/5100 train_loss:3.5634 train_time:389841ms step_avg:178.74ms step:2192/5100 train_loss:3.5046 train_time:390018ms step_avg:178.74ms step:2193/5100 train_loss:3.7398 train_time:390194ms step_avg:178.74ms step:2194/5100 train_loss:3.5049 train_time:390370ms step_avg:178.74ms step:2195/5100 train_loss:3.5708 train_time:390546ms step_avg:178.74ms step:2196/5100 train_loss:3.5026 train_time:390722ms step_avg:178.74ms step:2197/5100 train_loss:3.4279 train_time:390899ms step_avg:178.74ms step:2198/5100 train_loss:3.5086 train_time:391076ms step_avg:178.74ms step:2199/5100 train_loss:3.4536 train_time:391251ms step_avg:178.74ms step:2200/5100 train_loss:3.4513 train_time:391428ms step_avg:178.73ms step:2201/5100 train_loss:3.5093 train_time:391604ms step_avg:178.73ms step:2202/5100 train_loss:3.4900 train_time:391780ms step_avg:178.73ms step:2203/5100 train_loss:3.4692 train_time:391956ms step_avg:178.73ms step:2204/5100 train_loss:3.9696 train_time:392134ms step_avg:178.73ms step:2205/5100 train_loss:3.3809 train_time:392311ms step_avg:178.73ms step:2206/5100 train_loss:3.5037 train_time:392488ms step_avg:178.73ms step:2207/5100 train_loss:3.5175 train_time:392663ms step_avg:178.73ms step:2208/5100 train_loss:3.5337 train_time:392839ms step_avg:178.73ms step:2209/5100 train_loss:3.4302 train_time:393015ms step_avg:178.72ms step:2210/5100 train_loss:3.5089 train_time:393192ms step_avg:178.72ms step:2211/5100 train_loss:3.5217 train_time:393368ms step_avg:178.72ms step:2212/5100 train_loss:3.5169 train_time:393544ms step_avg:178.72ms step:2213/5100 train_loss:3.5450 train_time:393724ms step_avg:178.72ms step:2214/5100 train_loss:3.3981 train_time:393900ms step_avg:178.72ms step:2215/5100 train_loss:3.4648 train_time:394076ms step_avg:178.72ms step:2216/5100 train_loss:3.6039 train_time:394252ms step_avg:178.72ms step:2217/5100 train_loss:3.5586 train_time:394429ms step_avg:178.72ms step:2218/5100 train_loss:3.5099 train_time:394605ms step_avg:178.72ms step:2219/5100 train_loss:3.5212 train_time:394781ms step_avg:178.71ms step:2220/5100 train_loss:3.4351 train_time:394957ms step_avg:178.71ms step:2221/5100 train_loss:3.6944 train_time:395134ms step_avg:178.71ms step:2222/5100 train_loss:3.5789 train_time:395310ms step_avg:178.71ms step:2223/5100 train_loss:3.6015 train_time:395486ms step_avg:178.71ms step:2224/5100 train_loss:3.4993 train_time:395662ms step_avg:178.71ms step:2225/5100 train_loss:3.6078 train_time:395838ms step_avg:178.71ms step:2226/5100 train_loss:3.3543 train_time:396015ms step_avg:178.71ms step:2227/5100 train_loss:3.6321 train_time:396192ms step_avg:178.71ms step:2228/5100 train_loss:3.5643 train_time:396368ms step_avg:178.71ms step:2229/5100 train_loss:3.3644 train_time:396545ms step_avg:178.70ms step:2230/5100 train_loss:3.7185 train_time:396722ms step_avg:178.70ms step:2231/5100 train_loss:3.4087 train_time:396898ms step_avg:178.70ms step:2232/5100 train_loss:3.8752 train_time:397074ms step_avg:178.70ms step:2233/5100 train_loss:3.5628 train_time:397250ms step_avg:178.70ms step:2234/5100 train_loss:3.5027 train_time:397428ms step_avg:178.70ms step:2235/5100 train_loss:3.5379 train_time:397604ms step_avg:178.70ms step:2236/5100 train_loss:3.3261 train_time:397781ms step_avg:178.70ms step:2237/5100 train_loss:3.3194 train_time:397958ms step_avg:178.70ms step:2238/5100 train_loss:3.5470 train_time:398135ms step_avg:178.70ms step:2239/5100 train_loss:3.6435 train_time:398312ms step_avg:178.70ms step:2240/5100 train_loss:3.3664 train_time:398488ms step_avg:178.69ms step:2241/5100 train_loss:3.4390 train_time:398665ms step_avg:178.69ms step:2242/5100 train_loss:3.6122 train_time:398841ms step_avg:178.69ms step:2243/5100 train_loss:3.5815 train_time:399017ms step_avg:178.69ms step:2244/5100 train_loss:3.4392 train_time:399194ms step_avg:178.69ms step:2245/5100 train_loss:3.5047 train_time:399370ms step_avg:178.69ms step:2246/5100 train_loss:3.5276 train_time:399546ms step_avg:178.69ms step:2247/5100 train_loss:3.3661 train_time:399722ms step_avg:178.69ms step:2248/5100 train_loss:3.3838 train_time:399898ms step_avg:178.69ms step:2249/5100 train_loss:3.6316 train_time:400075ms step_avg:178.68ms step:2250/5100 train_loss:3.3715 train_time:400251ms step_avg:178.68ms step:2250/5100 val_loss:3.4929 train_time:400252ms step_avg:178.68ms step:2251/5100 train_loss:3.3646 train_time:400437ms step_avg:178.69ms step:2252/5100 train_loss:3.4418 train_time:400613ms step_avg:178.69ms step:2253/5100 train_loss:3.4200 train_time:400789ms step_avg:178.68ms step:2254/5100 train_loss:3.4644 train_time:400967ms step_avg:178.68ms step:2255/5100 train_loss:3.5214 train_time:401144ms step_avg:178.68ms step:2256/5100 train_loss:3.3987 train_time:401321ms step_avg:178.68ms step:2257/5100 train_loss:3.6828 train_time:401497ms step_avg:178.68ms step:2258/5100 train_loss:3.5636 train_time:401673ms step_avg:178.68ms step:2259/5100 train_loss:3.8706 train_time:401850ms step_avg:178.68ms step:2260/5100 train_loss:3.5627 train_time:402026ms step_avg:178.68ms step:2261/5100 train_loss:3.6151 train_time:402202ms step_avg:178.68ms step:2262/5100 train_loss:3.5213 train_time:402378ms step_avg:178.68ms step:2263/5100 train_loss:3.5243 train_time:402554ms step_avg:178.67ms step:2264/5100 train_loss:3.2844 train_time:402730ms step_avg:178.67ms step:2265/5100 train_loss:3.4059 train_time:402906ms step_avg:178.67ms step:2266/5100 train_loss:3.6188 train_time:403083ms step_avg:178.67ms step:2267/5100 train_loss:3.3534 train_time:403260ms step_avg:178.67ms step:2268/5100 train_loss:3.4271 train_time:403668ms step_avg:178.77ms step:2269/5100 train_loss:3.4037 train_time:403843ms step_avg:178.77ms step:2270/5100 train_loss:3.3720 train_time:404019ms step_avg:178.77ms step:2271/5100 train_loss:3.7692 train_time:404195ms step_avg:178.77ms step:2272/5100 train_loss:3.4212 train_time:404372ms step_avg:178.77ms step:2273/5100 train_loss:3.4344 train_time:404549ms step_avg:178.77ms step:2274/5100 train_loss:3.5145 train_time:404726ms step_avg:178.77ms step:2275/5100 train_loss:3.4589 train_time:404902ms step_avg:178.76ms step:2276/5100 train_loss:3.4780 train_time:405078ms step_avg:178.76ms step:2277/5100 train_loss:3.3571 train_time:405255ms step_avg:178.76ms step:2278/5100 train_loss:3.4684 train_time:405431ms step_avg:178.76ms step:2279/5100 train_loss:3.5892 train_time:405607ms step_avg:178.76ms step:2280/5100 train_loss:3.3915 train_time:406027ms step_avg:178.87ms step:2281/5100 train_loss:3.4554 train_time:406204ms step_avg:178.87ms step:2282/5100 train_loss:3.4641 train_time:406381ms step_avg:178.86ms step:2283/5100 train_loss:3.6012 train_time:406559ms step_avg:178.86ms step:2284/5100 train_loss:3.4850 train_time:406735ms step_avg:178.86ms step:2285/5100 train_loss:3.5050 train_time:406911ms step_avg:178.86ms step:2286/5100 train_loss:3.5068 train_time:407087ms step_avg:178.86ms step:2287/5100 train_loss:3.5043 train_time:407263ms step_avg:178.86ms step:2288/5100 train_loss:3.4563 train_time:407440ms step_avg:178.86ms step:2289/5100 train_loss:3.5868 train_time:407616ms step_avg:178.86ms step:2290/5100 train_loss:3.5656 train_time:407791ms step_avg:178.86ms step:2291/5100 train_loss:3.4422 train_time:407968ms step_avg:178.85ms step:2292/5100 train_loss:3.7839 train_time:408144ms step_avg:178.85ms step:2293/5100 train_loss:3.4454 train_time:408320ms step_avg:178.85ms step:2294/5100 train_loss:3.3896 train_time:408496ms step_avg:178.85ms step:2295/5100 train_loss:3.5807 train_time:408672ms step_avg:178.85ms step:2296/5100 train_loss:3.5253 train_time:408849ms step_avg:178.85ms step:2297/5100 train_loss:3.4917 train_time:409024ms step_avg:178.85ms step:2298/5100 train_loss:3.8740 train_time:409201ms step_avg:178.85ms step:2299/5100 train_loss:3.3952 train_time:409378ms step_avg:178.85ms step:2300/5100 train_loss:3.3941 train_time:409554ms step_avg:178.84ms step:2301/5100 train_loss:3.7352 train_time:409730ms step_avg:178.84ms step:2302/5100 train_loss:3.4598 train_time:409907ms step_avg:178.84ms step:2303/5100 train_loss:3.4758 train_time:410084ms step_avg:178.84ms step:2304/5100 train_loss:3.4569 train_time:410260ms step_avg:178.84ms step:2305/5100 train_loss:3.3945 train_time:410436ms step_avg:178.84ms step:2306/5100 train_loss:3.5615 train_time:410612ms step_avg:178.84ms step:2307/5100 train_loss:3.4156 train_time:410788ms step_avg:178.84ms step:2308/5100 train_loss:3.4358 train_time:410965ms step_avg:178.84ms step:2309/5100 train_loss:3.5650 train_time:411142ms step_avg:178.83ms step:2310/5100 train_loss:3.5187 train_time:411317ms step_avg:178.83ms step:2311/5100 train_loss:3.3919 train_time:411493ms step_avg:178.83ms step:2312/5100 train_loss:3.5105 train_time:411670ms step_avg:178.83ms step:2313/5100 train_loss:3.6359 train_time:411846ms step_avg:178.83ms step:2314/5100 train_loss:3.4495 train_time:412022ms step_avg:178.83ms step:2315/5100 train_loss:3.3816 train_time:412198ms step_avg:178.83ms step:2316/5100 train_loss:3.4675 train_time:412375ms step_avg:178.83ms step:2317/5100 train_loss:3.3544 train_time:412551ms step_avg:178.83ms step:2318/5100 train_loss:3.4605 train_time:412727ms step_avg:178.82ms step:2319/5100 train_loss:3.4800 train_time:412904ms step_avg:178.82ms step:2320/5100 train_loss:3.3217 train_time:413080ms step_avg:178.82ms step:2321/5100 train_loss:3.4596 train_time:413256ms step_avg:178.82ms step:2322/5100 train_loss:3.5106 train_time:413433ms step_avg:178.82ms step:2323/5100 train_loss:3.4257 train_time:413609ms step_avg:178.82ms step:2324/5100 train_loss:3.4717 train_time:413786ms step_avg:178.82ms step:2325/5100 train_loss:3.3958 train_time:413962ms step_avg:178.82ms step:2326/5100 train_loss:3.5308 train_time:414138ms step_avg:178.82ms step:2327/5100 train_loss:3.5424 train_time:414314ms step_avg:178.81ms step:2328/5100 train_loss:3.3184 train_time:414489ms step_avg:178.81ms step:2329/5100 train_loss:3.4309 train_time:414666ms step_avg:178.81ms step:2330/5100 train_loss:3.4537 train_time:414842ms step_avg:178.81ms step:2331/5100 train_loss:3.4204 train_time:415018ms step_avg:178.81ms step:2332/5100 train_loss:3.5963 train_time:415194ms step_avg:178.81ms step:2333/5100 train_loss:3.4886 train_time:415369ms step_avg:178.81ms step:2334/5100 train_loss:3.4645 train_time:415546ms step_avg:178.81ms step:2335/5100 train_loss:3.5494 train_time:415722ms step_avg:178.81ms step:2336/5100 train_loss:3.3900 train_time:415899ms step_avg:178.80ms step:2337/5100 train_loss:3.5412 train_time:416074ms step_avg:178.80ms step:2338/5100 train_loss:3.4968 train_time:416250ms step_avg:178.80ms step:2339/5100 train_loss:3.4456 train_time:416426ms step_avg:178.80ms step:2340/5100 train_loss:3.5247 train_time:416603ms step_avg:178.80ms step:2341/5100 train_loss:3.5729 train_time:416780ms step_avg:178.80ms step:2342/5100 train_loss:3.4389 train_time:416956ms step_avg:178.80ms step:2343/5100 train_loss:3.4516 train_time:417132ms step_avg:178.80ms step:2344/5100 train_loss:3.5151 train_time:417308ms step_avg:178.80ms step:2345/5100 train_loss:3.4595 train_time:417484ms step_avg:178.79ms step:2346/5100 train_loss:3.5840 train_time:417661ms step_avg:178.79ms step:2347/5100 train_loss:3.4867 train_time:417837ms step_avg:178.79ms step:2348/5100 train_loss:3.5970 train_time:418013ms step_avg:178.79ms step:2349/5100 train_loss:3.5543 train_time:418190ms step_avg:178.79ms step:2350/5100 train_loss:3.5963 train_time:418367ms step_avg:178.79ms step:2351/5100 train_loss:3.2808 train_time:418543ms step_avg:178.79ms step:2352/5100 train_loss:3.4064 train_time:418720ms step_avg:178.79ms step:2353/5100 train_loss:3.4033 train_time:418896ms step_avg:178.79ms step:2354/5100 train_loss:3.6178 train_time:419072ms step_avg:178.78ms step:2355/5100 train_loss:3.4089 train_time:419250ms step_avg:178.78ms step:2356/5100 train_loss:3.4056 train_time:419426ms step_avg:178.78ms step:2357/5100 train_loss:3.5575 train_time:419603ms step_avg:178.78ms step:2358/5100 train_loss:3.4108 train_time:419779ms step_avg:178.78ms step:2359/5100 train_loss:3.5177 train_time:419956ms step_avg:178.78ms step:2360/5100 train_loss:3.4171 train_time:420133ms step_avg:178.78ms step:2361/5100 train_loss:3.4333 train_time:420308ms step_avg:178.78ms step:2362/5100 train_loss:3.4681 train_time:420485ms step_avg:178.78ms step:2363/5100 train_loss:3.5323 train_time:420662ms step_avg:178.78ms step:2364/5100 train_loss:3.4751 train_time:420838ms step_avg:178.78ms step:2365/5100 train_loss:3.9097 train_time:421013ms step_avg:178.77ms step:2366/5100 train_loss:3.5317 train_time:421189ms step_avg:178.77ms step:2367/5100 train_loss:3.6797 train_time:421365ms step_avg:178.77ms step:2368/5100 train_loss:3.5037 train_time:421541ms step_avg:178.77ms step:2369/5100 train_loss:3.5021 train_time:421717ms step_avg:178.77ms step:2370/5100 train_loss:3.5353 train_time:421893ms step_avg:178.77ms step:2371/5100 train_loss:3.4189 train_time:422069ms step_avg:178.77ms step:2372/5100 train_loss:3.6458 train_time:422246ms step_avg:178.77ms step:2373/5100 train_loss:3.4969 train_time:422422ms step_avg:178.76ms step:2374/5100 train_loss:4.0562 train_time:422598ms step_avg:178.76ms step:2375/5100 train_loss:3.4757 train_time:422774ms step_avg:178.76ms step:2375/5100 val_loss:3.4823 train_time:422774ms step_avg:178.76ms step:2376/5100 train_loss:3.3850 train_time:422960ms step_avg:178.77ms step:2377/5100 train_loss:3.5467 train_time:423136ms step_avg:178.76ms step:2378/5100 train_loss:3.5116 train_time:423314ms step_avg:178.76ms step:2379/5100 train_loss:3.5305 train_time:423490ms step_avg:178.76ms step:2380/5100 train_loss:3.5091 train_time:423666ms step_avg:178.76ms step:2381/5100 train_loss:3.4054 train_time:423843ms step_avg:178.76ms step:2382/5100 train_loss:3.5080 train_time:424019ms step_avg:178.76ms step:2383/5100 train_loss:3.5234 train_time:424195ms step_avg:178.76ms step:2384/5100 train_loss:3.4758 train_time:424372ms step_avg:178.76ms step:2385/5100 train_loss:3.4007 train_time:424549ms step_avg:178.76ms step:2386/5100 train_loss:3.5134 train_time:424726ms step_avg:178.76ms step:2387/5100 train_loss:3.4658 train_time:424903ms step_avg:178.76ms step:2388/5100 train_loss:3.4760 train_time:425079ms step_avg:178.75ms step:2389/5100 train_loss:3.5074 train_time:425254ms step_avg:178.75ms step:2390/5100 train_loss:3.4865 train_time:425430ms step_avg:178.75ms step:2391/5100 train_loss:3.4896 train_time:425607ms step_avg:178.75ms step:2392/5100 train_loss:3.3645 train_time:425784ms step_avg:178.75ms step:2393/5100 train_loss:3.5867 train_time:425959ms step_avg:178.75ms step:2394/5100 train_loss:3.4260 train_time:426137ms step_avg:178.75ms step:2395/5100 train_loss:3.5247 train_time:426315ms step_avg:178.75ms step:2396/5100 train_loss:3.6387 train_time:426491ms step_avg:178.75ms step:2397/5100 train_loss:3.6519 train_time:426667ms step_avg:178.75ms step:2398/5100 train_loss:3.6013 train_time:426843ms step_avg:178.74ms step:2399/5100 train_loss:3.5718 train_time:427019ms step_avg:178.74ms step:2400/5100 train_loss:3.4435 train_time:427195ms step_avg:178.74ms step:2401/5100 train_loss:3.4475 train_time:427372ms step_avg:178.74ms step:2402/5100 train_loss:3.5487 train_time:427548ms step_avg:178.74ms step:2403/5100 train_loss:3.3912 train_time:427724ms step_avg:178.74ms step:2404/5100 train_loss:3.5261 train_time:427901ms step_avg:178.74ms step:2405/5100 train_loss:3.7424 train_time:428078ms step_avg:178.74ms step:2406/5100 train_loss:3.4564 train_time:428254ms step_avg:178.74ms step:2407/5100 train_loss:3.6092 train_time:428431ms step_avg:178.74ms step:2408/5100 train_loss:3.4691 train_time:428607ms step_avg:178.74ms step:2409/5100 train_loss:3.4081 train_time:428782ms step_avg:178.73ms step:2410/5100 train_loss:3.5350 train_time:428958ms step_avg:178.73ms step:2411/5100 train_loss:3.3346 train_time:429134ms step_avg:178.73ms step:2412/5100 train_loss:3.7696 train_time:429310ms step_avg:178.73ms step:2413/5100 train_loss:3.4467 train_time:429486ms step_avg:178.73ms step:2414/5100 train_loss:3.5218 train_time:429662ms step_avg:178.73ms step:2415/5100 train_loss:3.4405 train_time:429839ms step_avg:178.73ms step:2416/5100 train_loss:3.5134 train_time:430016ms step_avg:178.73ms step:2417/5100 train_loss:3.3278 train_time:430191ms step_avg:178.73ms step:2418/5100 train_loss:3.2633 train_time:430367ms step_avg:178.72ms step:2419/5100 train_loss:3.5567 train_time:430543ms step_avg:178.72ms step:2420/5100 train_loss:3.4344 train_time:430719ms step_avg:178.72ms step:2421/5100 train_loss:3.4668 train_time:430895ms step_avg:178.72ms step:2422/5100 train_loss:3.5699 train_time:431072ms step_avg:178.72ms step:2423/5100 train_loss:3.6120 train_time:431248ms step_avg:178.72ms step:2424/5100 train_loss:3.4374 train_time:431425ms step_avg:178.72ms step:2425/5100 train_loss:3.5254 train_time:431601ms step_avg:178.72ms step:2426/5100 train_loss:3.5285 train_time:431778ms step_avg:178.72ms step:2427/5100 train_loss:3.4547 train_time:431954ms step_avg:178.72ms step:2428/5100 train_loss:3.4017 train_time:432131ms step_avg:178.71ms step:2429/5100 train_loss:3.5312 train_time:432308ms step_avg:178.71ms step:2430/5100 train_loss:3.4301 train_time:432483ms step_avg:178.71ms step:2431/5100 train_loss:3.4843 train_time:432659ms step_avg:178.71ms step:2432/5100 train_loss:3.5392 train_time:432836ms step_avg:178.71ms step:2433/5100 train_loss:3.5051 train_time:433012ms step_avg:178.71ms step:2434/5100 train_loss:3.3796 train_time:433188ms step_avg:178.71ms step:2435/5100 train_loss:3.3398 train_time:433363ms step_avg:178.71ms step:2436/5100 train_loss:3.5024 train_time:433540ms step_avg:178.71ms step:2437/5100 train_loss:3.3655 train_time:433717ms step_avg:178.70ms step:2438/5100 train_loss:3.4387 train_time:433893ms step_avg:178.70ms step:2439/5100 train_loss:3.5335 train_time:434069ms step_avg:178.70ms step:2440/5100 train_loss:3.4495 train_time:434246ms step_avg:178.70ms step:2441/5100 train_loss:3.5371 train_time:434422ms step_avg:178.70ms step:2442/5100 train_loss:3.4257 train_time:434598ms step_avg:178.70ms step:2443/5100 train_loss:3.4853 train_time:434775ms step_avg:178.70ms step:2444/5100 train_loss:3.3669 train_time:434952ms step_avg:178.70ms step:2445/5100 train_loss:3.3784 train_time:435128ms step_avg:178.70ms step:2446/5100 train_loss:3.5422 train_time:435304ms step_avg:178.70ms step:2447/5100 train_loss:3.4009 train_time:435480ms step_avg:178.70ms step:2448/5100 train_loss:3.4695 train_time:435656ms step_avg:178.69ms step:2449/5100 train_loss:3.6337 train_time:435832ms step_avg:178.69ms step:2450/5100 train_loss:3.4622 train_time:436009ms step_avg:178.69ms step:2451/5100 train_loss:3.5348 train_time:436184ms step_avg:178.69ms step:2452/5100 train_loss:3.4375 train_time:436361ms step_avg:178.69ms step:2453/5100 train_loss:3.5474 train_time:436537ms step_avg:178.69ms step:2454/5100 train_loss:3.4349 train_time:436713ms step_avg:178.69ms step:2455/5100 train_loss:3.5729 train_time:436891ms step_avg:178.69ms step:2456/5100 train_loss:3.4981 train_time:437068ms step_avg:178.69ms step:2457/5100 train_loss:3.4193 train_time:437475ms step_avg:178.78ms step:2458/5100 train_loss:3.3508 train_time:437650ms step_avg:178.78ms step:2459/5100 train_loss:3.4750 train_time:437827ms step_avg:178.78ms step:2460/5100 train_loss:4.0809 train_time:438003ms step_avg:178.78ms step:2461/5100 train_loss:3.5398 train_time:438179ms step_avg:178.78ms step:2462/5100 train_loss:3.3575 train_time:438356ms step_avg:178.77ms step:2463/5100 train_loss:3.5541 train_time:438532ms step_avg:178.77ms step:2464/5100 train_loss:3.4685 train_time:438709ms step_avg:178.77ms step:2465/5100 train_loss:3.6697 train_time:438884ms step_avg:178.77ms step:2466/5100 train_loss:3.8625 train_time:439061ms step_avg:178.77ms step:2467/5100 train_loss:3.5777 train_time:439239ms step_avg:178.77ms step:2468/5100 train_loss:3.4563 train_time:439415ms step_avg:178.77ms step:2469/5100 train_loss:3.5681 train_time:439592ms step_avg:178.77ms step:2470/5100 train_loss:3.5820 train_time:440024ms step_avg:178.87ms step:2471/5100 train_loss:3.3862 train_time:440200ms step_avg:178.87ms step:2472/5100 train_loss:3.4794 train_time:440377ms step_avg:178.87ms step:2473/5100 train_loss:3.4721 train_time:440553ms step_avg:178.87ms step:2474/5100 train_loss:3.6188 train_time:440730ms step_avg:178.87ms step:2475/5100 train_loss:3.7510 train_time:440906ms step_avg:178.87ms step:2476/5100 train_loss:3.3383 train_time:441083ms step_avg:178.87ms step:2477/5100 train_loss:3.5457 train_time:441258ms step_avg:178.86ms step:2478/5100 train_loss:3.5043 train_time:441436ms step_avg:178.86ms step:2479/5100 train_loss:3.3481 train_time:441612ms step_avg:178.86ms step:2480/5100 train_loss:3.3395 train_time:441788ms step_avg:178.86ms step:2481/5100 train_loss:3.4889 train_time:441964ms step_avg:178.86ms step:2482/5100 train_loss:3.5132 train_time:442140ms step_avg:178.86ms step:2483/5100 train_loss:3.5187 train_time:442316ms step_avg:178.86ms step:2484/5100 train_loss:3.4782 train_time:442493ms step_avg:178.86ms step:2485/5100 train_loss:3.4852 train_time:442670ms step_avg:178.86ms step:2486/5100 train_loss:3.3706 train_time:442846ms step_avg:178.86ms step:2487/5100 train_loss:3.5680 train_time:443022ms step_avg:178.85ms step:2488/5100 train_loss:3.5209 train_time:443199ms step_avg:178.85ms step:2489/5100 train_loss:3.4320 train_time:443375ms step_avg:178.85ms step:2490/5100 train_loss:3.5339 train_time:443553ms step_avg:178.85ms step:2491/5100 train_loss:3.5871 train_time:443728ms step_avg:178.85ms step:2492/5100 train_loss:3.6698 train_time:443904ms step_avg:178.85ms step:2493/5100 train_loss:3.5179 train_time:444080ms step_avg:178.85ms step:2494/5100 train_loss:3.4408 train_time:444257ms step_avg:178.85ms step:2495/5100 train_loss:3.5706 train_time:444433ms step_avg:178.85ms step:2496/5100 train_loss:3.5158 train_time:444609ms step_avg:178.85ms step:2497/5100 train_loss:3.4286 train_time:444786ms step_avg:178.84ms step:2498/5100 train_loss:3.5246 train_time:444961ms step_avg:178.84ms step:2499/5100 train_loss:3.5822 train_time:445138ms step_avg:178.84ms step:2500/5100 train_loss:3.5991 train_time:445315ms step_avg:178.84ms step:2500/5100 val_loss:3.4737 train_time:445315ms step_avg:178.84ms step:2501/5100 train_loss:3.5441 train_time:445499ms step_avg:178.84ms step:2502/5100 train_loss:3.5005 train_time:445675ms step_avg:178.84ms step:2503/5100 train_loss:3.5158 train_time:445851ms step_avg:178.84ms step:2504/5100 train_loss:3.3905 train_time:446027ms step_avg:178.84ms step:2505/5100 train_loss:3.5740 train_time:446204ms step_avg:178.84ms step:2506/5100 train_loss:3.5245 train_time:446380ms step_avg:178.84ms step:2507/5100 train_loss:3.4800 train_time:446556ms step_avg:178.84ms step:2508/5100 train_loss:3.4814 train_time:446732ms step_avg:178.84ms step:2509/5100 train_loss:3.4449 train_time:446908ms step_avg:178.83ms step:2510/5100 train_loss:3.6095 train_time:447085ms step_avg:178.83ms step:2511/5100 train_loss:3.4362 train_time:447262ms step_avg:178.83ms step:2512/5100 train_loss:3.4256 train_time:447439ms step_avg:178.83ms step:2513/5100 train_loss:3.5034 train_time:447615ms step_avg:178.83ms step:2514/5100 train_loss:3.5285 train_time:447791ms step_avg:178.83ms step:2515/5100 train_loss:3.4357 train_time:447967ms step_avg:178.83ms step:2516/5100 train_loss:3.5204 train_time:448143ms step_avg:178.83ms step:2517/5100 train_loss:3.5180 train_time:448319ms step_avg:178.83ms step:2518/5100 train_loss:3.3966 train_time:448495ms step_avg:178.83ms step:2519/5100 train_loss:3.4274 train_time:448671ms step_avg:178.82ms step:2520/5100 train_loss:3.5443 train_time:448847ms step_avg:178.82ms step:2521/5100 train_loss:3.5339 train_time:449023ms step_avg:178.82ms step:2522/5100 train_loss:3.4169 train_time:449200ms step_avg:178.82ms step:2523/5100 train_loss:3.3992 train_time:449376ms step_avg:178.82ms step:2524/5100 train_loss:3.5030 train_time:449552ms step_avg:178.82ms step:2525/5100 train_loss:3.3395 train_time:449728ms step_avg:178.82ms step:2526/5100 train_loss:3.5591 train_time:449906ms step_avg:178.82ms step:2527/5100 train_loss:3.4696 train_time:450082ms step_avg:178.82ms step:2528/5100 train_loss:3.4731 train_time:450259ms step_avg:178.82ms step:2529/5100 train_loss:3.4588 train_time:450435ms step_avg:178.81ms step:2530/5100 train_loss:3.4870 train_time:450610ms step_avg:178.81ms step:2531/5100 train_loss:3.5199 train_time:450786ms step_avg:178.81ms step:2532/5100 train_loss:3.3380 train_time:450962ms step_avg:178.81ms step:2533/5100 train_loss:3.5003 train_time:451138ms step_avg:178.81ms step:2534/5100 train_loss:3.4004 train_time:451314ms step_avg:178.81ms step:2535/5100 train_loss:3.4248 train_time:451491ms step_avg:178.81ms step:2536/5100 train_loss:3.4903 train_time:451667ms step_avg:178.81ms step:2537/5100 train_loss:3.5005 train_time:451843ms step_avg:178.81ms step:2538/5100 train_loss:3.3250 train_time:452020ms step_avg:178.81ms step:2539/5100 train_loss:3.6345 train_time:452196ms step_avg:178.80ms step:2540/5100 train_loss:3.3185 train_time:452372ms step_avg:178.80ms step:2541/5100 train_loss:3.4952 train_time:452548ms step_avg:178.80ms step:2542/5100 train_loss:3.2777 train_time:452725ms step_avg:178.80ms step:2543/5100 train_loss:3.7052 train_time:452901ms step_avg:178.80ms step:2544/5100 train_loss:3.4710 train_time:453077ms step_avg:178.80ms step:2545/5100 train_loss:3.6302 train_time:453252ms step_avg:178.80ms step:2546/5100 train_loss:3.4611 train_time:453428ms step_avg:178.80ms step:2547/5100 train_loss:3.4498 train_time:453604ms step_avg:178.80ms step:2548/5100 train_loss:3.4456 train_time:453781ms step_avg:178.79ms step:2549/5100 train_loss:3.6099 train_time:453957ms step_avg:178.79ms step:2550/5100 train_loss:3.4711 train_time:454133ms step_avg:178.79ms step:2551/5100 train_loss:3.4713 train_time:454309ms step_avg:178.79ms step:2552/5100 train_loss:3.4942 train_time:454486ms step_avg:178.79ms step:2553/5100 train_loss:3.5087 train_time:454662ms step_avg:178.79ms step:2554/5100 train_loss:3.4321 train_time:454842ms step_avg:178.79ms step:2555/5100 train_loss:3.5322 train_time:455018ms step_avg:178.79ms step:2556/5100 train_loss:3.5901 train_time:455194ms step_avg:178.79ms step:2557/5100 train_loss:3.5736 train_time:455370ms step_avg:178.79ms step:2558/5100 train_loss:3.4132 train_time:455546ms step_avg:178.79ms step:2559/5100 train_loss:3.4213 train_time:455722ms step_avg:178.78ms step:2560/5100 train_loss:3.4334 train_time:455899ms step_avg:178.78ms step:2561/5100 train_loss:3.5483 train_time:456075ms step_avg:178.78ms step:2562/5100 train_loss:3.5852 train_time:456251ms step_avg:178.78ms step:2563/5100 train_loss:3.4703 train_time:456427ms step_avg:178.78ms step:2564/5100 train_loss:3.4945 train_time:456604ms step_avg:178.78ms step:2565/5100 train_loss:3.4117 train_time:456780ms step_avg:178.78ms step:2566/5100 train_loss:3.4252 train_time:456957ms step_avg:178.78ms step:2567/5100 train_loss:3.4225 train_time:457132ms step_avg:178.78ms step:2568/5100 train_loss:3.4649 train_time:457308ms step_avg:178.78ms step:2569/5100 train_loss:3.6141 train_time:457484ms step_avg:178.77ms step:2570/5100 train_loss:3.5150 train_time:457660ms step_avg:178.77ms step:2571/5100 train_loss:3.6010 train_time:457836ms step_avg:178.77ms step:2572/5100 train_loss:3.3498 train_time:458012ms step_avg:178.77ms step:2573/5100 train_loss:3.4638 train_time:458188ms step_avg:178.77ms step:2574/5100 train_loss:3.1324 train_time:458365ms step_avg:178.77ms step:2575/5100 train_loss:3.3683 train_time:458540ms step_avg:178.77ms step:2576/5100 train_loss:3.3029 train_time:458716ms step_avg:178.77ms step:2577/5100 train_loss:3.4296 train_time:458892ms step_avg:178.77ms step:2578/5100 train_loss:3.4782 train_time:459068ms step_avg:178.76ms step:2579/5100 train_loss:3.3801 train_time:459245ms step_avg:178.76ms step:2580/5100 train_loss:3.4403 train_time:459421ms step_avg:178.76ms step:2581/5100 train_loss:3.3943 train_time:459598ms step_avg:178.76ms step:2582/5100 train_loss:3.4939 train_time:459774ms step_avg:178.76ms step:2583/5100 train_loss:3.3715 train_time:459949ms step_avg:178.76ms step:2584/5100 train_loss:3.5676 train_time:460125ms step_avg:178.76ms step:2585/5100 train_loss:3.4797 train_time:460302ms step_avg:178.76ms step:2586/5100 train_loss:3.4929 train_time:460479ms step_avg:178.76ms step:2587/5100 train_loss:3.6161 train_time:460654ms step_avg:178.76ms step:2588/5100 train_loss:3.5049 train_time:460831ms step_avg:178.76ms step:2589/5100 train_loss:3.3589 train_time:461007ms step_avg:178.75ms step:2590/5100 train_loss:3.5284 train_time:461184ms step_avg:178.75ms step:2591/5100 train_loss:3.4360 train_time:461360ms step_avg:178.75ms step:2592/5100 train_loss:3.6403 train_time:461537ms step_avg:178.75ms step:2593/5100 train_loss:3.5095 train_time:461713ms step_avg:178.75ms step:2594/5100 train_loss:3.3284 train_time:461889ms step_avg:178.75ms step:2595/5100 train_loss:3.3992 train_time:462066ms step_avg:178.75ms step:2596/5100 train_loss:3.8655 train_time:462243ms step_avg:178.75ms step:2597/5100 train_loss:3.4911 train_time:462419ms step_avg:178.75ms step:2598/5100 train_loss:3.4900 train_time:462596ms step_avg:178.75ms step:2599/5100 train_loss:3.3398 train_time:462771ms step_avg:178.75ms step:2600/5100 train_loss:3.5870 train_time:462947ms step_avg:178.74ms step:2601/5100 train_loss:3.7564 train_time:463124ms step_avg:178.74ms step:2602/5100 train_loss:3.3285 train_time:463301ms step_avg:178.74ms step:2603/5100 train_loss:3.4717 train_time:463477ms step_avg:178.74ms step:2604/5100 train_loss:3.3038 train_time:463653ms step_avg:178.74ms step:2605/5100 train_loss:3.5960 train_time:463829ms step_avg:178.74ms step:2606/5100 train_loss:3.4710 train_time:464006ms step_avg:178.74ms step:2607/5100 train_loss:3.3696 train_time:464185ms step_avg:178.74ms step:2608/5100 train_loss:3.3137 train_time:464362ms step_avg:178.74ms step:2609/5100 train_loss:3.4358 train_time:464538ms step_avg:178.74ms step:2610/5100 train_loss:3.6173 train_time:464714ms step_avg:178.74ms step:2611/5100 train_loss:3.4905 train_time:464890ms step_avg:178.74ms step:2612/5100 train_loss:3.3176 train_time:465067ms step_avg:178.73ms step:2613/5100 train_loss:3.4123 train_time:465243ms step_avg:178.73ms step:2614/5100 train_loss:3.5231 train_time:465419ms step_avg:178.73ms step:2615/5100 train_loss:3.4596 train_time:465595ms step_avg:178.73ms step:2616/5100 train_loss:3.4568 train_time:465772ms step_avg:178.73ms step:2617/5100 train_loss:3.4883 train_time:465948ms step_avg:178.73ms step:2618/5100 train_loss:3.5383 train_time:466124ms step_avg:178.73ms step:2619/5100 train_loss:3.3761 train_time:466301ms step_avg:178.73ms step:2620/5100 train_loss:3.5575 train_time:466476ms step_avg:178.73ms step:2621/5100 train_loss:3.5189 train_time:466652ms step_avg:178.73ms step:2622/5100 train_loss:3.6418 train_time:466828ms step_avg:178.72ms step:2623/5100 train_loss:3.5559 train_time:467004ms step_avg:178.72ms step:2624/5100 train_loss:3.4744 train_time:467180ms step_avg:178.72ms step:2625/5100 train_loss:3.4360 train_time:467357ms step_avg:178.72ms step:2625/5100 val_loss:3.4637 train_time:467357ms step_avg:178.72ms step:2626/5100 train_loss:3.4579 train_time:467543ms step_avg:178.72ms step:2627/5100 train_loss:3.5186 train_time:467720ms step_avg:178.72ms step:2628/5100 train_loss:3.3416 train_time:467897ms step_avg:178.72ms step:2629/5100 train_loss:3.6054 train_time:468074ms step_avg:178.72ms step:2630/5100 train_loss:3.4808 train_time:468250ms step_avg:178.72ms step:2631/5100 train_loss:3.5340 train_time:468426ms step_avg:178.72ms step:2632/5100 train_loss:3.7734 train_time:468615ms step_avg:178.72ms step:2633/5100 train_loss:3.5116 train_time:468789ms step_avg:178.72ms step:2634/5100 train_loss:3.4263 train_time:468965ms step_avg:178.72ms step:2635/5100 train_loss:3.3998 train_time:469141ms step_avg:178.72ms step:2636/5100 train_loss:3.4428 train_time:469318ms step_avg:178.72ms step:2637/5100 train_loss:3.2333 train_time:469494ms step_avg:178.72ms step:2638/5100 train_loss:3.5349 train_time:469670ms step_avg:178.72ms step:2639/5100 train_loss:3.5167 train_time:469846ms step_avg:178.72ms step:2640/5100 train_loss:3.4117 train_time:470022ms step_avg:178.72ms step:2641/5100 train_loss:3.4934 train_time:470198ms step_avg:178.71ms step:2642/5100 train_loss:3.5275 train_time:470376ms step_avg:178.71ms step:2643/5100 train_loss:3.3119 train_time:470552ms step_avg:178.71ms step:2644/5100 train_loss:3.4375 train_time:470729ms step_avg:178.71ms step:2645/5100 train_loss:3.5105 train_time:470905ms step_avg:178.71ms step:2646/5100 train_loss:3.4768 train_time:471311ms step_avg:178.80ms step:2647/5100 train_loss:3.3634 train_time:471488ms step_avg:178.80ms step:2648/5100 train_loss:3.5917 train_time:471664ms step_avg:178.80ms step:2649/5100 train_loss:3.8491 train_time:471840ms step_avg:178.79ms step:2650/5100 train_loss:3.4794 train_time:472016ms step_avg:178.79ms step:2651/5100 train_loss:3.4438 train_time:472192ms step_avg:178.79ms step:2652/5100 train_loss:3.5824 train_time:472370ms step_avg:178.79ms step:2653/5100 train_loss:3.4212 train_time:472546ms step_avg:178.79ms step:2654/5100 train_loss:3.4008 train_time:472723ms step_avg:178.79ms step:2655/5100 train_loss:3.4698 train_time:472899ms step_avg:178.79ms step:2656/5100 train_loss:3.3942 train_time:473076ms step_avg:178.79ms step:2657/5100 train_loss:3.4301 train_time:473253ms step_avg:178.79ms step:2658/5100 train_loss:3.3995 train_time:473430ms step_avg:178.79ms step:2659/5100 train_loss:3.4790 train_time:473606ms step_avg:178.79ms step:2660/5100 train_loss:3.6201 train_time:474025ms step_avg:178.88ms step:2661/5100 train_loss:3.4177 train_time:474202ms step_avg:178.88ms step:2662/5100 train_loss:3.5708 train_time:474378ms step_avg:178.88ms step:2663/5100 train_loss:3.4353 train_time:474554ms step_avg:178.87ms step:2664/5100 train_loss:3.4323 train_time:474731ms step_avg:178.87ms step:2665/5100 train_loss:3.3579 train_time:474908ms step_avg:178.87ms step:2666/5100 train_loss:3.4131 train_time:475083ms step_avg:178.87ms step:2667/5100 train_loss:3.4509 train_time:475259ms step_avg:178.87ms step:2668/5100 train_loss:3.4911 train_time:475437ms step_avg:178.87ms step:2669/5100 train_loss:3.4045 train_time:475614ms step_avg:178.87ms step:2670/5100 train_loss:3.4653 train_time:475790ms step_avg:178.87ms step:2671/5100 train_loss:3.3493 train_time:475966ms step_avg:178.87ms step:2672/5100 train_loss:3.4196 train_time:476141ms step_avg:178.87ms step:2673/5100 train_loss:3.4092 train_time:476318ms step_avg:178.87ms step:2674/5100 train_loss:3.4666 train_time:476494ms step_avg:178.86ms step:2675/5100 train_loss:3.4916 train_time:476671ms step_avg:178.86ms step:2676/5100 train_loss:3.4627 train_time:476848ms step_avg:178.86ms step:2677/5100 train_loss:3.4521 train_time:477025ms step_avg:178.86ms step:2678/5100 train_loss:3.4833 train_time:477201ms step_avg:178.86ms step:2679/5100 train_loss:3.5298 train_time:477378ms step_avg:178.86ms step:2680/5100 train_loss:3.4379 train_time:477554ms step_avg:178.86ms step:2681/5100 train_loss:3.3676 train_time:477730ms step_avg:178.86ms step:2682/5100 train_loss:3.4099 train_time:477906ms step_avg:178.86ms step:2683/5100 train_loss:3.8822 train_time:478082ms step_avg:178.86ms step:2684/5100 train_loss:3.4680 train_time:478258ms step_avg:178.85ms step:2685/5100 train_loss:3.4966 train_time:478437ms step_avg:178.85ms step:2686/5100 train_loss:3.5444 train_time:478614ms step_avg:178.85ms step:2687/5100 train_loss:3.4627 train_time:478789ms step_avg:178.85ms step:2688/5100 train_loss:3.5471 train_time:478966ms step_avg:178.85ms step:2689/5100 train_loss:3.4669 train_time:479141ms step_avg:178.85ms step:2690/5100 train_loss:3.4547 train_time:479318ms step_avg:178.85ms step:2691/5100 train_loss:3.4847 train_time:479494ms step_avg:178.85ms step:2692/5100 train_loss:3.5540 train_time:479671ms step_avg:178.85ms step:2693/5100 train_loss:3.3642 train_time:479847ms step_avg:178.85ms step:2694/5100 train_loss:3.7335 train_time:480023ms step_avg:178.85ms step:2695/5100 train_loss:3.5395 train_time:480199ms step_avg:178.84ms step:2696/5100 train_loss:3.3825 train_time:480379ms step_avg:178.85ms step:2697/5100 train_loss:3.5233 train_time:480554ms step_avg:178.84ms step:2698/5100 train_loss:3.4849 train_time:480730ms step_avg:178.84ms step:2699/5100 train_loss:3.4346 train_time:480906ms step_avg:178.84ms step:2700/5100 train_loss:3.5405 train_time:481081ms step_avg:178.84ms step:2701/5100 train_loss:3.5058 train_time:481257ms step_avg:178.84ms step:2702/5100 train_loss:3.4142 train_time:481434ms step_avg:178.84ms step:2703/5100 train_loss:3.4373 train_time:481611ms step_avg:178.84ms step:2704/5100 train_loss:3.4497 train_time:481788ms step_avg:178.84ms step:2705/5100 train_loss:3.4157 train_time:481964ms step_avg:178.84ms step:2706/5100 train_loss:3.5967 train_time:482140ms step_avg:178.84ms step:2707/5100 train_loss:3.5548 train_time:482317ms step_avg:178.83ms step:2708/5100 train_loss:3.4560 train_time:482493ms step_avg:178.83ms step:2709/5100 train_loss:3.4567 train_time:482670ms step_avg:178.83ms step:2710/5100 train_loss:3.5572 train_time:482848ms step_avg:178.83ms step:2711/5100 train_loss:3.4360 train_time:483024ms step_avg:178.83ms step:2712/5100 train_loss:3.5475 train_time:483200ms step_avg:178.83ms step:2713/5100 train_loss:3.2851 train_time:483376ms step_avg:178.83ms step:2714/5100 train_loss:3.4785 train_time:483552ms step_avg:178.83ms step:2715/5100 train_loss:3.3783 train_time:483728ms step_avg:178.83ms step:2716/5100 train_loss:3.3822 train_time:483904ms step_avg:178.83ms step:2717/5100 train_loss:3.5751 train_time:484081ms step_avg:178.83ms step:2718/5100 train_loss:3.4717 train_time:484257ms step_avg:178.82ms step:2719/5100 train_loss:3.7069 train_time:484433ms step_avg:178.82ms step:2720/5100 train_loss:3.4481 train_time:484610ms step_avg:178.82ms step:2721/5100 train_loss:3.4439 train_time:484786ms step_avg:178.82ms step:2722/5100 train_loss:3.6682 train_time:484962ms step_avg:178.82ms step:2723/5100 train_loss:3.4383 train_time:485139ms step_avg:178.82ms step:2724/5100 train_loss:3.6164 train_time:485315ms step_avg:178.82ms step:2725/5100 train_loss:3.4894 train_time:485491ms step_avg:178.82ms step:2726/5100 train_loss:3.4571 train_time:485667ms step_avg:178.82ms step:2727/5100 train_loss:3.4560 train_time:485843ms step_avg:178.82ms step:2728/5100 train_loss:3.8028 train_time:486021ms step_avg:178.82ms step:2729/5100 train_loss:3.5305 train_time:486198ms step_avg:178.81ms step:2730/5100 train_loss:3.3978 train_time:486375ms step_avg:178.81ms step:2731/5100 train_loss:3.5002 train_time:486552ms step_avg:178.81ms step:2732/5100 train_loss:3.4084 train_time:486728ms step_avg:178.81ms step:2733/5100 train_loss:3.2980 train_time:486904ms step_avg:178.81ms step:2734/5100 train_loss:3.4099 train_time:487080ms step_avg:178.81ms step:2735/5100 train_loss:3.4836 train_time:487256ms step_avg:178.81ms step:2736/5100 train_loss:3.3794 train_time:487433ms step_avg:178.81ms step:2737/5100 train_loss:3.7772 train_time:487609ms step_avg:178.81ms step:2738/5100 train_loss:3.5197 train_time:487784ms step_avg:178.81ms step:2739/5100 train_loss:3.7200 train_time:487961ms step_avg:178.81ms step:2740/5100 train_loss:3.4726 train_time:488137ms step_avg:178.80ms step:2741/5100 train_loss:3.4646 train_time:488313ms step_avg:178.80ms step:2742/5100 train_loss:3.3977 train_time:488489ms step_avg:178.80ms step:2743/5100 train_loss:3.4720 train_time:488665ms step_avg:178.80ms step:2744/5100 train_loss:3.4851 train_time:488841ms step_avg:178.80ms step:2745/5100 train_loss:3.5947 train_time:489017ms step_avg:178.80ms step:2746/5100 train_loss:3.3554 train_time:489193ms step_avg:178.80ms step:2747/5100 train_loss:3.4454 train_time:489370ms step_avg:178.80ms step:2748/5100 train_loss:3.4867 train_time:489546ms step_avg:178.80ms step:2749/5100 train_loss:3.6038 train_time:489721ms step_avg:178.80ms step:2750/5100 train_loss:3.4390 train_time:489898ms step_avg:178.79ms step:2750/5100 val_loss:3.4569 train_time:489898ms step_avg:178.79ms step:2751/5100 train_loss:3.5078 train_time:490088ms step_avg:178.80ms step:2752/5100 train_loss:3.5702 train_time:490266ms step_avg:178.80ms step:2753/5100 train_loss:3.4783 train_time:490442ms step_avg:178.80ms step:2754/5100 train_loss:3.4069 train_time:490618ms step_avg:178.80ms step:2755/5100 train_loss:3.4017 train_time:490794ms step_avg:178.80ms step:2756/5100 train_loss:3.4893 train_time:490970ms step_avg:178.79ms step:2757/5100 train_loss:3.4293 train_time:491147ms step_avg:178.79ms step:2758/5100 train_loss:3.3045 train_time:491322ms step_avg:178.79ms step:2759/5100 train_loss:3.6985 train_time:491498ms step_avg:178.79ms step:2760/5100 train_loss:3.5054 train_time:491675ms step_avg:178.79ms step:2761/5100 train_loss:3.4705 train_time:491851ms step_avg:178.79ms step:2762/5100 train_loss:3.4360 train_time:492027ms step_avg:178.79ms step:2763/5100 train_loss:3.3583 train_time:492203ms step_avg:178.79ms step:2764/5100 train_loss:3.5256 train_time:492380ms step_avg:178.79ms step:2765/5100 train_loss:3.4468 train_time:492556ms step_avg:178.79ms step:2766/5100 train_loss:3.3437 train_time:492731ms step_avg:178.78ms step:2767/5100 train_loss:3.4276 train_time:492908ms step_avg:178.78ms step:2768/5100 train_loss:3.5149 train_time:493084ms step_avg:178.78ms step:2769/5100 train_loss:3.3961 train_time:493260ms step_avg:178.78ms step:2770/5100 train_loss:3.4720 train_time:493436ms step_avg:178.78ms step:2771/5100 train_loss:3.4468 train_time:493612ms step_avg:178.78ms step:2772/5100 train_loss:3.8853 train_time:493790ms step_avg:178.78ms step:2773/5100 train_loss:3.3518 train_time:493966ms step_avg:178.78ms step:2774/5100 train_loss:3.4869 train_time:494143ms step_avg:178.78ms step:2775/5100 train_loss:3.5508 train_time:494320ms step_avg:178.78ms step:2776/5100 train_loss:3.5167 train_time:494496ms step_avg:178.78ms step:2777/5100 train_loss:3.5935 train_time:494673ms step_avg:178.78ms step:2778/5100 train_loss:3.5977 train_time:494849ms step_avg:178.77ms step:2779/5100 train_loss:3.4607 train_time:495026ms step_avg:178.77ms step:2780/5100 train_loss:3.3304 train_time:495202ms step_avg:178.77ms step:2781/5100 train_loss:3.4793 train_time:495378ms step_avg:178.77ms step:2782/5100 train_loss:3.5022 train_time:495555ms step_avg:178.77ms step:2783/5100 train_loss:3.3602 train_time:495732ms step_avg:178.77ms step:2784/5100 train_loss:3.4777 train_time:495909ms step_avg:178.77ms step:2785/5100 train_loss:3.5265 train_time:496085ms step_avg:178.77ms step:2786/5100 train_loss:3.3998 train_time:496261ms step_avg:178.77ms step:2787/5100 train_loss:3.5349 train_time:496437ms step_avg:178.77ms step:2788/5100 train_loss:3.4908 train_time:496613ms step_avg:178.77ms step:2789/5100 train_loss:3.4267 train_time:496790ms step_avg:178.77ms step:2790/5100 train_loss:3.5075 train_time:496965ms step_avg:178.76ms step:2791/5100 train_loss:3.4335 train_time:497142ms step_avg:178.76ms step:2792/5100 train_loss:3.3346 train_time:497318ms step_avg:178.76ms step:2793/5100 train_loss:3.4378 train_time:497495ms step_avg:178.76ms step:2794/5100 train_loss:3.4726 train_time:497671ms step_avg:178.76ms step:2795/5100 train_loss:3.3890 train_time:497847ms step_avg:178.76ms step:2796/5100 train_loss:3.4307 train_time:498023ms step_avg:178.76ms step:2797/5100 train_loss:3.3528 train_time:498199ms step_avg:178.76ms step:2798/5100 train_loss:3.4633 train_time:498376ms step_avg:178.76ms step:2799/5100 train_loss:3.4119 train_time:498552ms step_avg:178.76ms step:2800/5100 train_loss:3.5826 train_time:498729ms step_avg:178.76ms step:2801/5100 train_loss:3.5387 train_time:498905ms step_avg:178.75ms step:2802/5100 train_loss:3.4979 train_time:499081ms step_avg:178.75ms step:2803/5100 train_loss:3.4460 train_time:499259ms step_avg:178.75ms step:2804/5100 train_loss:3.6281 train_time:499435ms step_avg:178.75ms step:2805/5100 train_loss:3.5957 train_time:499612ms step_avg:178.75ms step:2806/5100 train_loss:3.3219 train_time:499788ms step_avg:178.75ms step:2807/5100 train_loss:3.7234 train_time:499964ms step_avg:178.75ms step:2808/5100 train_loss:3.4706 train_time:500140ms step_avg:178.75ms step:2809/5100 train_loss:3.3919 train_time:500316ms step_avg:178.75ms step:2810/5100 train_loss:3.4283 train_time:500493ms step_avg:178.75ms step:2811/5100 train_loss:3.5787 train_time:500669ms step_avg:178.75ms step:2812/5100 train_loss:3.5657 train_time:500845ms step_avg:178.75ms step:2813/5100 train_loss:3.3133 train_time:501021ms step_avg:178.74ms step:2814/5100 train_loss:3.5338 train_time:501197ms step_avg:178.74ms step:2815/5100 train_loss:3.6064 train_time:501373ms step_avg:178.74ms step:2816/5100 train_loss:3.4137 train_time:501549ms step_avg:178.74ms step:2817/5100 train_loss:3.0837 train_time:501726ms step_avg:178.74ms step:2818/5100 train_loss:3.4399 train_time:501902ms step_avg:178.74ms step:2819/5100 train_loss:3.4197 train_time:502078ms step_avg:178.74ms step:2820/5100 train_loss:3.6037 train_time:502255ms step_avg:178.74ms step:2821/5100 train_loss:3.5667 train_time:502432ms step_avg:178.74ms step:2822/5100 train_loss:3.5268 train_time:502608ms step_avg:178.74ms step:2823/5100 train_loss:3.4651 train_time:502784ms step_avg:178.74ms step:2824/5100 train_loss:3.4253 train_time:502961ms step_avg:178.74ms step:2825/5100 train_loss:3.3233 train_time:503136ms step_avg:178.73ms step:2826/5100 train_loss:3.5886 train_time:503313ms step_avg:178.73ms step:2827/5100 train_loss:3.4893 train_time:503490ms step_avg:178.73ms step:2828/5100 train_loss:3.3639 train_time:503665ms step_avg:178.73ms step:2829/5100 train_loss:3.4976 train_time:503842ms step_avg:178.73ms step:2830/5100 train_loss:3.4987 train_time:504018ms step_avg:178.73ms step:2831/5100 train_loss:3.4340 train_time:504194ms step_avg:178.73ms step:2832/5100 train_loss:3.5831 train_time:504370ms step_avg:178.73ms step:2833/5100 train_loss:3.4927 train_time:504546ms step_avg:178.73ms step:2834/5100 train_loss:3.4862 train_time:504724ms step_avg:178.73ms step:2835/5100 train_loss:3.2971 train_time:505133ms step_avg:178.81ms step:2836/5100 train_loss:3.5180 train_time:505309ms step_avg:178.81ms step:2837/5100 train_loss:3.4446 train_time:505487ms step_avg:178.81ms step:2838/5100 train_loss:3.7504 train_time:505664ms step_avg:178.81ms step:2839/5100 train_loss:3.3983 train_time:505840ms step_avg:178.81ms step:2840/5100 train_loss:3.4070 train_time:506016ms step_avg:178.80ms step:2841/5100 train_loss:3.4720 train_time:506193ms step_avg:178.80ms step:2842/5100 train_loss:3.3963 train_time:506369ms step_avg:178.80ms step:2843/5100 train_loss:3.3987 train_time:506546ms step_avg:178.80ms step:2844/5100 train_loss:3.5706 train_time:506722ms step_avg:178.80ms step:2845/5100 train_loss:3.4603 train_time:506898ms step_avg:178.80ms step:2846/5100 train_loss:3.4820 train_time:507074ms step_avg:178.80ms step:2847/5100 train_loss:3.4387 train_time:507251ms step_avg:178.80ms step:2848/5100 train_loss:3.7156 train_time:507426ms step_avg:178.80ms step:2849/5100 train_loss:3.3806 train_time:507602ms step_avg:178.80ms step:2850/5100 train_loss:3.4087 train_time:508021ms step_avg:178.88ms step:2851/5100 train_loss:3.5103 train_time:508197ms step_avg:178.88ms step:2852/5100 train_loss:3.4768 train_time:508374ms step_avg:178.88ms step:2853/5100 train_loss:3.4441 train_time:508550ms step_avg:178.88ms step:2854/5100 train_loss:3.5238 train_time:508726ms step_avg:178.88ms step:2855/5100 train_loss:3.3447 train_time:508903ms step_avg:178.88ms step:2856/5100 train_loss:3.3611 train_time:509081ms step_avg:178.88ms step:2857/5100 train_loss:3.4525 train_time:509257ms step_avg:178.88ms step:2858/5100 train_loss:3.4635 train_time:509433ms step_avg:178.87ms step:2859/5100 train_loss:3.3390 train_time:509610ms step_avg:178.87ms step:2860/5100 train_loss:3.4446 train_time:509786ms step_avg:178.87ms step:2861/5100 train_loss:3.4105 train_time:509962ms step_avg:178.87ms step:2862/5100 train_loss:3.4476 train_time:510138ms step_avg:178.87ms step:2863/5100 train_loss:3.4901 train_time:510315ms step_avg:178.87ms step:2864/5100 train_loss:3.7540 train_time:510491ms step_avg:178.87ms step:2865/5100 train_loss:3.5690 train_time:510668ms step_avg:178.87ms step:2866/5100 train_loss:3.4476 train_time:510845ms step_avg:178.87ms step:2867/5100 train_loss:3.3494 train_time:511020ms step_avg:178.87ms step:2868/5100 train_loss:3.5392 train_time:511196ms step_avg:178.87ms step:2869/5100 train_loss:3.4885 train_time:511373ms step_avg:178.86ms step:2870/5100 train_loss:3.4547 train_time:511549ms step_avg:178.86ms step:2871/5100 train_loss:3.5861 train_time:511725ms step_avg:178.86ms step:2872/5100 train_loss:3.3736 train_time:511901ms step_avg:178.86ms step:2873/5100 train_loss:3.4248 train_time:512077ms step_avg:178.86ms step:2874/5100 train_loss:3.2946 train_time:512255ms step_avg:178.86ms step:2875/5100 train_loss:3.4489 train_time:512431ms step_avg:178.86ms step:2875/5100 val_loss:3.4505 train_time:512432ms step_avg:178.86ms step:2876/5100 train_loss:3.3693 train_time:512617ms step_avg:178.86ms step:2877/5100 train_loss:3.3503 train_time:512793ms step_avg:178.86ms step:2878/5100 train_loss:3.4406 train_time:512969ms step_avg:178.86ms step:2879/5100 train_loss:3.5625 train_time:513146ms step_avg:178.86ms step:2880/5100 train_loss:3.5100 train_time:513324ms step_avg:178.86ms step:2881/5100 train_loss:3.4511 train_time:513501ms step_avg:178.86ms step:2882/5100 train_loss:3.4383 train_time:513677ms step_avg:178.86ms step:2883/5100 train_loss:3.5630 train_time:513854ms step_avg:178.86ms step:2884/5100 train_loss:3.3497 train_time:514030ms step_avg:178.86ms step:2885/5100 train_loss:3.3701 train_time:514207ms step_avg:178.85ms step:2886/5100 train_loss:3.4051 train_time:514383ms step_avg:178.85ms step:2887/5100 train_loss:3.4094 train_time:514559ms step_avg:178.85ms step:2888/5100 train_loss:3.4230 train_time:514735ms step_avg:178.85ms step:2889/5100 train_loss:3.4423 train_time:514911ms step_avg:178.85ms step:2890/5100 train_loss:3.6285 train_time:515088ms step_avg:178.85ms step:2891/5100 train_loss:3.4745 train_time:515264ms step_avg:178.85ms step:2892/5100 train_loss:3.3105 train_time:515439ms step_avg:178.85ms step:2893/5100 train_loss:3.2413 train_time:515615ms step_avg:178.85ms step:2894/5100 train_loss:3.3776 train_time:515792ms step_avg:178.85ms step:2895/5100 train_loss:3.2566 train_time:515970ms step_avg:178.85ms step:2896/5100 train_loss:3.4457 train_time:516146ms step_avg:178.84ms step:2897/5100 train_loss:3.5666 train_time:516323ms step_avg:178.84ms step:2898/5100 train_loss:3.3903 train_time:516499ms step_avg:178.84ms step:2899/5100 train_loss:3.4920 train_time:516675ms step_avg:178.84ms step:2900/5100 train_loss:3.3727 train_time:516851ms step_avg:178.84ms step:2901/5100 train_loss:3.5605 train_time:517029ms step_avg:178.84ms step:2902/5100 train_loss:3.5440 train_time:517205ms step_avg:178.84ms step:2903/5100 train_loss:3.5935 train_time:517383ms step_avg:178.84ms step:2904/5100 train_loss:3.3012 train_time:517560ms step_avg:178.84ms step:2905/5100 train_loss:3.4433 train_time:517736ms step_avg:178.84ms step:2906/5100 train_loss:3.4203 train_time:517911ms step_avg:178.84ms step:2907/5100 train_loss:3.5047 train_time:518088ms step_avg:178.84ms step:2908/5100 train_loss:3.4424 train_time:518264ms step_avg:178.84ms step:2909/5100 train_loss:3.3997 train_time:518441ms step_avg:178.83ms step:2910/5100 train_loss:3.7389 train_time:518616ms step_avg:178.83ms step:2911/5100 train_loss:3.4547 train_time:518793ms step_avg:178.83ms step:2912/5100 train_loss:3.3518 train_time:518969ms step_avg:178.83ms step:2913/5100 train_loss:3.3447 train_time:519145ms step_avg:178.83ms step:2914/5100 train_loss:3.8363 train_time:519322ms step_avg:178.83ms step:2915/5100 train_loss:3.4206 train_time:519498ms step_avg:178.83ms step:2916/5100 train_loss:3.3671 train_time:519674ms step_avg:178.83ms step:2917/5100 train_loss:3.3555 train_time:519850ms step_avg:178.83ms step:2918/5100 train_loss:3.6438 train_time:520027ms step_avg:178.83ms step:2919/5100 train_loss:3.1506 train_time:520203ms step_avg:178.83ms step:2920/5100 train_loss:3.3531 train_time:520379ms step_avg:178.82ms step:2921/5100 train_loss:3.3635 train_time:520556ms step_avg:178.82ms step:2922/5100 train_loss:3.4542 train_time:520731ms step_avg:178.82ms step:2923/5100 train_loss:3.5040 train_time:520908ms step_avg:178.82ms step:2924/5100 train_loss:3.5352 train_time:521085ms step_avg:178.82ms step:2925/5100 train_loss:3.5436 train_time:521260ms step_avg:178.82ms step:2926/5100 train_loss:3.4197 train_time:521436ms step_avg:178.82ms step:2927/5100 train_loss:3.4380 train_time:521611ms step_avg:178.82ms step:2928/5100 train_loss:3.4271 train_time:521789ms step_avg:178.82ms step:2929/5100 train_loss:3.4274 train_time:521965ms step_avg:178.82ms step:2930/5100 train_loss:3.3843 train_time:522141ms step_avg:178.82ms step:2931/5100 train_loss:3.4192 train_time:522316ms step_avg:178.81ms step:2932/5100 train_loss:3.5533 train_time:522492ms step_avg:178.81ms step:2933/5100 train_loss:3.5786 train_time:522669ms step_avg:178.81ms step:2934/5100 train_loss:3.5509 train_time:522847ms step_avg:178.81ms step:2935/5100 train_loss:3.4016 train_time:523023ms step_avg:178.81ms step:2936/5100 train_loss:3.4526 train_time:523199ms step_avg:178.81ms step:2937/5100 train_loss:3.3820 train_time:523375ms step_avg:178.81ms step:2938/5100 train_loss:3.4216 train_time:523552ms step_avg:178.81ms step:2939/5100 train_loss:3.4403 train_time:523728ms step_avg:178.81ms step:2940/5100 train_loss:3.4881 train_time:523906ms step_avg:178.81ms step:2941/5100 train_loss:3.5289 train_time:524081ms step_avg:178.81ms step:2942/5100 train_loss:3.5226 train_time:524257ms step_avg:178.81ms step:2943/5100 train_loss:3.4443 train_time:524433ms step_avg:178.80ms step:2944/5100 train_loss:3.3263 train_time:524609ms step_avg:178.80ms step:2945/5100 train_loss:3.2630 train_time:524785ms step_avg:178.80ms step:2946/5100 train_loss:3.4670 train_time:524961ms step_avg:178.80ms step:2947/5100 train_loss:3.5346 train_time:525137ms step_avg:178.80ms step:2948/5100 train_loss:3.4654 train_time:525313ms step_avg:178.80ms step:2949/5100 train_loss:3.6448 train_time:525490ms step_avg:178.80ms step:2950/5100 train_loss:3.4641 train_time:525665ms step_avg:178.80ms step:2951/5100 train_loss:3.4722 train_time:525841ms step_avg:178.80ms step:2952/5100 train_loss:3.8933 train_time:526018ms step_avg:178.80ms step:2953/5100 train_loss:3.5591 train_time:526197ms step_avg:178.80ms step:2954/5100 train_loss:3.4939 train_time:526372ms step_avg:178.79ms step:2955/5100 train_loss:3.5017 train_time:526549ms step_avg:178.79ms step:2956/5100 train_loss:3.4402 train_time:526725ms step_avg:178.79ms step:2957/5100 train_loss:3.4629 train_time:526902ms step_avg:178.79ms step:2958/5100 train_loss:3.3287 train_time:527078ms step_avg:178.79ms step:2959/5100 train_loss:3.4217 train_time:527255ms step_avg:178.79ms step:2960/5100 train_loss:3.5620 train_time:527431ms step_avg:178.79ms step:2961/5100 train_loss:3.3668 train_time:527608ms step_avg:178.79ms step:2962/5100 train_loss:3.5041 train_time:527784ms step_avg:178.79ms step:2963/5100 train_loss:3.3551 train_time:527961ms step_avg:178.79ms step:2964/5100 train_loss:3.4154 train_time:528136ms step_avg:178.79ms step:2965/5100 train_loss:3.4009 train_time:528312ms step_avg:178.79ms step:2966/5100 train_loss:3.5137 train_time:528489ms step_avg:178.79ms step:2967/5100 train_loss:3.3834 train_time:528665ms step_avg:178.78ms step:2968/5100 train_loss:3.6319 train_time:528841ms step_avg:178.78ms step:2969/5100 train_loss:3.4874 train_time:529017ms step_avg:178.78ms step:2970/5100 train_loss:3.5016 train_time:529193ms step_avg:178.78ms step:2971/5100 train_loss:3.4819 train_time:529369ms step_avg:178.78ms step:2972/5100 train_loss:3.5613 train_time:529545ms step_avg:178.78ms step:2973/5100 train_loss:3.3850 train_time:529723ms step_avg:178.78ms step:2974/5100 train_loss:3.3973 train_time:529898ms step_avg:178.78ms step:2975/5100 train_loss:3.3104 train_time:530074ms step_avg:178.78ms step:2976/5100 train_loss:3.3847 train_time:530251ms step_avg:178.78ms step:2977/5100 train_loss:3.3853 train_time:530428ms step_avg:178.78ms step:2978/5100 train_loss:3.4011 train_time:530605ms step_avg:178.78ms step:2979/5100 train_loss:3.6728 train_time:530781ms step_avg:178.77ms step:2980/5100 train_loss:3.4874 train_time:530957ms step_avg:178.77ms step:2981/5100 train_loss:3.5252 train_time:531133ms step_avg:178.77ms step:2982/5100 train_loss:3.5387 train_time:531310ms step_avg:178.77ms step:2983/5100 train_loss:3.6240 train_time:531487ms step_avg:178.77ms step:2984/5100 train_loss:3.4238 train_time:531664ms step_avg:178.77ms step:2985/5100 train_loss:3.5166 train_time:531840ms step_avg:178.77ms step:2986/5100 train_loss:3.5228 train_time:532016ms step_avg:178.77ms step:2987/5100 train_loss:3.4715 train_time:532192ms step_avg:178.77ms step:2988/5100 train_loss:3.5943 train_time:532369ms step_avg:178.77ms step:2989/5100 train_loss:3.1906 train_time:532545ms step_avg:178.77ms step:2990/5100 train_loss:3.5280 train_time:532721ms step_avg:178.77ms step:2991/5100 train_loss:3.4903 train_time:532897ms step_avg:178.76ms step:2992/5100 train_loss:3.4660 train_time:533073ms step_avg:178.76ms step:2993/5100 train_loss:3.3814 train_time:533250ms step_avg:178.76ms step:2994/5100 train_loss:3.5188 train_time:533425ms step_avg:178.76ms step:2995/5100 train_loss:3.3355 train_time:533602ms step_avg:178.76ms step:2996/5100 train_loss:3.3712 train_time:533778ms step_avg:178.76ms step:2997/5100 train_loss:3.4394 train_time:533955ms step_avg:178.76ms step:2998/5100 train_loss:3.3846 train_time:534131ms step_avg:178.76ms step:2999/5100 train_loss:3.5057 train_time:534309ms step_avg:178.76ms step:3000/5100 train_loss:3.4084 train_time:534486ms step_avg:178.76ms step:3000/5100 val_loss:3.4414 train_time:534486ms step_avg:178.76ms step:3001/5100 train_loss:3.3924 train_time:534670ms step_avg:178.76ms step:3002/5100 train_loss:3.3401 train_time:534847ms step_avg:178.76ms step:3003/5100 train_loss:3.3833 train_time:535023ms step_avg:178.76ms step:3004/5100 train_loss:3.5112 train_time:535200ms step_avg:178.76ms step:3005/5100 train_loss:3.8586 train_time:535377ms step_avg:178.76ms step:3006/5100 train_loss:3.4280 train_time:535553ms step_avg:178.76ms step:3007/5100 train_loss:3.4934 train_time:535730ms step_avg:178.76ms step:3008/5100 train_loss:3.3078 train_time:535906ms step_avg:178.75ms step:3009/5100 train_loss:3.5229 train_time:536082ms step_avg:178.75ms step:3010/5100 train_loss:3.4145 train_time:536258ms step_avg:178.75ms step:3011/5100 train_loss:3.4787 train_time:536435ms step_avg:178.75ms step:3012/5100 train_loss:3.4724 train_time:536612ms step_avg:178.75ms step:3013/5100 train_loss:3.3705 train_time:536788ms step_avg:178.75ms step:3014/5100 train_loss:3.5649 train_time:536963ms step_avg:178.75ms step:3015/5100 train_loss:3.5243 train_time:537140ms step_avg:178.75ms step:3016/5100 train_loss:3.3965 train_time:537316ms step_avg:178.75ms step:3017/5100 train_loss:3.4290 train_time:537493ms step_avg:178.75ms step:3018/5100 train_loss:3.4749 train_time:537669ms step_avg:178.75ms step:3019/5100 train_loss:3.5052 train_time:537845ms step_avg:178.75ms step:3020/5100 train_loss:3.3034 train_time:538021ms step_avg:178.74ms step:3021/5100 train_loss:3.5920 train_time:538198ms step_avg:178.74ms step:3022/5100 train_loss:3.4286 train_time:538374ms step_avg:178.74ms step:3023/5100 train_loss:3.3475 train_time:538551ms step_avg:178.74ms step:3024/5100 train_loss:3.4392 train_time:538959ms step_avg:178.82ms step:3025/5100 train_loss:3.4220 train_time:539135ms step_avg:178.82ms step:3026/5100 train_loss:3.4735 train_time:539311ms step_avg:178.82ms step:3027/5100 train_loss:3.5020 train_time:539487ms step_avg:178.82ms step:3028/5100 train_loss:3.4109 train_time:539663ms step_avg:178.81ms step:3029/5100 train_loss:3.2099 train_time:539839ms step_avg:178.81ms step:3030/5100 train_loss:3.5555 train_time:540016ms step_avg:178.81ms step:3031/5100 train_loss:3.3115 train_time:540192ms step_avg:178.81ms step:3032/5100 train_loss:3.3048 train_time:540369ms step_avg:178.81ms step:3033/5100 train_loss:3.6470 train_time:540544ms step_avg:178.81ms step:3034/5100 train_loss:3.6422 train_time:540720ms step_avg:178.81ms step:3035/5100 train_loss:3.4084 train_time:540897ms step_avg:178.81ms step:3036/5100 train_loss:3.4854 train_time:541074ms step_avg:178.81ms step:3037/5100 train_loss:3.4432 train_time:541251ms step_avg:178.81ms step:3038/5100 train_loss:3.3382 train_time:541427ms step_avg:178.81ms step:3039/5100 train_loss:3.3929 train_time:541603ms step_avg:178.81ms step:3040/5100 train_loss:3.4919 train_time:542035ms step_avg:178.89ms step:3041/5100 train_loss:3.4847 train_time:542210ms step_avg:178.89ms step:3042/5100 train_loss:3.2893 train_time:542386ms step_avg:178.89ms step:3043/5100 train_loss:3.4332 train_time:542563ms step_avg:178.89ms step:3044/5100 train_loss:3.4629 train_time:542739ms step_avg:178.89ms step:3045/5100 train_loss:3.4675 train_time:542915ms step_avg:178.88ms step:3046/5100 train_loss:3.5498 train_time:543091ms step_avg:178.88ms step:3047/5100 train_loss:3.3641 train_time:543268ms step_avg:178.88ms step:3048/5100 train_loss:3.4866 train_time:543445ms step_avg:178.88ms step:3049/5100 train_loss:3.4364 train_time:543621ms step_avg:178.88ms step:3050/5100 train_loss:3.3651 train_time:543798ms step_avg:178.88ms step:3051/5100 train_loss:3.4899 train_time:543975ms step_avg:178.88ms step:3052/5100 train_loss:3.3387 train_time:544151ms step_avg:178.88ms step:3053/5100 train_loss:3.5826 train_time:544329ms step_avg:178.88ms step:3054/5100 train_loss:3.5248 train_time:544505ms step_avg:178.88ms step:3055/5100 train_loss:3.5019 train_time:544683ms step_avg:178.88ms step:3056/5100 train_loss:3.5008 train_time:544860ms step_avg:178.88ms step:3057/5100 train_loss:3.3899 train_time:545037ms step_avg:178.88ms step:3058/5100 train_loss:3.4161 train_time:545213ms step_avg:178.88ms step:3059/5100 train_loss:3.4964 train_time:545389ms step_avg:178.87ms step:3060/5100 train_loss:3.3961 train_time:545565ms step_avg:178.87ms step:3061/5100 train_loss:3.4480 train_time:545742ms step_avg:178.87ms step:3062/5100 train_loss:3.4530 train_time:545917ms step_avg:178.87ms step:3063/5100 train_loss:3.3870 train_time:546094ms step_avg:178.87ms step:3064/5100 train_loss:3.3636 train_time:546270ms step_avg:178.87ms step:3065/5100 train_loss:3.3810 train_time:546446ms step_avg:178.87ms step:3066/5100 train_loss:3.3609 train_time:546622ms step_avg:178.87ms step:3067/5100 train_loss:3.3501 train_time:546799ms step_avg:178.87ms step:3068/5100 train_loss:3.3053 train_time:546975ms step_avg:178.87ms step:3069/5100 train_loss:3.3514 train_time:547152ms step_avg:178.87ms step:3070/5100 train_loss:3.3486 train_time:547329ms step_avg:178.87ms step:3071/5100 train_loss:3.5334 train_time:547505ms step_avg:178.86ms step:3072/5100 train_loss:3.4612 train_time:547681ms step_avg:178.86ms step:3073/5100 train_loss:3.5057 train_time:547857ms step_avg:178.86ms step:3074/5100 train_loss:3.4885 train_time:548034ms step_avg:178.86ms step:3075/5100 train_loss:3.4401 train_time:548210ms step_avg:178.86ms step:3076/5100 train_loss:3.4860 train_time:548387ms step_avg:178.86ms step:3077/5100 train_loss:3.5445 train_time:548564ms step_avg:178.86ms step:3078/5100 train_loss:3.3503 train_time:548740ms step_avg:178.86ms step:3079/5100 train_loss:3.8774 train_time:548916ms step_avg:178.86ms step:3080/5100 train_loss:3.4408 train_time:549092ms step_avg:178.86ms step:3081/5100 train_loss:3.4008 train_time:549268ms step_avg:178.86ms step:3082/5100 train_loss:3.5442 train_time:549444ms step_avg:178.86ms step:3083/5100 train_loss:3.3563 train_time:549622ms step_avg:178.86ms step:3084/5100 train_loss:3.3874 train_time:549800ms step_avg:178.85ms step:3085/5100 train_loss:3.4335 train_time:549977ms step_avg:178.85ms step:3086/5100 train_loss:3.5260 train_time:550153ms step_avg:178.85ms step:3087/5100 train_loss:3.4405 train_time:550329ms step_avg:178.85ms step:3088/5100 train_loss:3.3574 train_time:550505ms step_avg:178.85ms step:3089/5100 train_loss:3.5050 train_time:550680ms step_avg:178.85ms step:3090/5100 train_loss:3.3710 train_time:550857ms step_avg:178.85ms step:3091/5100 train_loss:3.6309 train_time:551033ms step_avg:178.85ms step:3092/5100 train_loss:4.2060 train_time:551209ms step_avg:178.85ms step:3093/5100 train_loss:3.4696 train_time:551385ms step_avg:178.85ms step:3094/5100 train_loss:3.3576 train_time:551561ms step_avg:178.85ms step:3095/5100 train_loss:3.3116 train_time:551738ms step_avg:178.85ms step:3096/5100 train_loss:3.4821 train_time:551914ms step_avg:178.84ms step:3097/5100 train_loss:3.6095 train_time:552090ms step_avg:178.84ms step:3098/5100 train_loss:3.3834 train_time:552266ms step_avg:178.84ms step:3099/5100 train_loss:3.4233 train_time:552442ms step_avg:178.84ms step:3100/5100 train_loss:3.5983 train_time:552619ms step_avg:178.84ms step:3101/5100 train_loss:3.4992 train_time:552795ms step_avg:178.84ms step:3102/5100 train_loss:3.4944 train_time:552972ms step_avg:178.84ms step:3103/5100 train_loss:3.4011 train_time:553150ms step_avg:178.84ms step:3104/5100 train_loss:3.6543 train_time:553326ms step_avg:178.84ms step:3105/5100 train_loss:3.4711 train_time:553502ms step_avg:178.84ms step:3106/5100 train_loss:3.3307 train_time:553680ms step_avg:178.84ms step:3107/5100 train_loss:3.3600 train_time:553856ms step_avg:178.84ms step:3108/5100 train_loss:3.3228 train_time:554034ms step_avg:178.84ms step:3109/5100 train_loss:3.5350 train_time:554211ms step_avg:178.84ms step:3110/5100 train_loss:3.4244 train_time:554387ms step_avg:178.83ms step:3111/5100 train_loss:3.4648 train_time:554563ms step_avg:178.83ms step:3112/5100 train_loss:3.4402 train_time:554740ms step_avg:178.83ms step:3113/5100 train_loss:3.4986 train_time:554926ms step_avg:178.84ms step:3114/5100 train_loss:3.4473 train_time:555102ms step_avg:178.83ms step:3115/5100 train_loss:3.4626 train_time:555278ms step_avg:178.83ms step:3116/5100 train_loss:3.4896 train_time:555455ms step_avg:178.83ms step:3117/5100 train_loss:3.3542 train_time:555631ms step_avg:178.83ms step:3118/5100 train_loss:3.3557 train_time:555807ms step_avg:178.83ms step:3119/5100 train_loss:3.5518 train_time:555983ms step_avg:178.83ms step:3120/5100 train_loss:3.5258 train_time:556159ms step_avg:178.83ms step:3121/5100 train_loss:3.3155 train_time:556336ms step_avg:178.83ms step:3122/5100 train_loss:3.5063 train_time:556512ms step_avg:178.83ms step:3123/5100 train_loss:3.5639 train_time:556688ms step_avg:178.83ms step:3124/5100 train_loss:3.5332 train_time:556864ms step_avg:178.83ms step:3125/5100 train_loss:3.3233 train_time:557040ms step_avg:178.83ms step:3125/5100 val_loss:3.4343 train_time:557041ms step_avg:178.83ms step:3126/5100 train_loss:3.4032 train_time:557227ms step_avg:178.83ms step:3127/5100 train_loss:3.4428 train_time:557403ms step_avg:178.83ms step:3128/5100 train_loss:3.5280 train_time:557581ms step_avg:178.83ms step:3129/5100 train_loss:3.6034 train_time:557757ms step_avg:178.83ms step:3130/5100 train_loss:3.3089 train_time:557933ms step_avg:178.82ms step:3131/5100 train_loss:3.4718 train_time:558110ms step_avg:178.82ms step:3132/5100 train_loss:3.4696 train_time:558286ms step_avg:178.82ms step:3133/5100 train_loss:3.4897 train_time:558463ms step_avg:178.82ms step:3134/5100 train_loss:3.3954 train_time:558639ms step_avg:178.82ms step:3135/5100 train_loss:3.5126 train_time:558815ms step_avg:178.82ms step:3136/5100 train_loss:3.4271 train_time:558992ms step_avg:178.82ms step:3137/5100 train_loss:3.4889 train_time:559169ms step_avg:178.82ms step:3138/5100 train_loss:3.6788 train_time:559346ms step_avg:178.82ms step:3139/5100 train_loss:3.6510 train_time:559523ms step_avg:178.82ms step:3140/5100 train_loss:3.4116 train_time:559699ms step_avg:178.82ms step:3141/5100 train_loss:3.4333 train_time:559875ms step_avg:178.82ms step:3142/5100 train_loss:3.3495 train_time:560053ms step_avg:178.82ms step:3143/5100 train_loss:3.4468 train_time:560230ms step_avg:178.82ms step:3144/5100 train_loss:3.2521 train_time:560406ms step_avg:178.81ms step:3145/5100 train_loss:3.4866 train_time:560582ms step_avg:178.81ms step:3146/5100 train_loss:3.3997 train_time:560758ms step_avg:178.81ms step:3147/5100 train_loss:3.4218 train_time:560935ms step_avg:178.81ms step:3148/5100 train_loss:3.5912 train_time:561112ms step_avg:178.81ms step:3149/5100 train_loss:3.6812 train_time:561288ms step_avg:178.81ms step:3150/5100 train_loss:3.5413 train_time:561465ms step_avg:178.81ms step:3151/5100 train_loss:3.3579 train_time:561642ms step_avg:178.81ms step:3152/5100 train_loss:3.4014 train_time:561818ms step_avg:178.81ms step:3153/5100 train_loss:3.3794 train_time:561995ms step_avg:178.81ms step:3154/5100 train_loss:3.4998 train_time:562171ms step_avg:178.81ms step:3155/5100 train_loss:3.3192 train_time:562348ms step_avg:178.81ms step:3156/5100 train_loss:3.4560 train_time:562525ms step_avg:178.81ms step:3157/5100 train_loss:3.4106 train_time:562701ms step_avg:178.81ms step:3158/5100 train_loss:3.5222 train_time:562877ms step_avg:178.80ms step:3159/5100 train_loss:3.5872 train_time:563053ms step_avg:178.80ms step:3160/5100 train_loss:3.4293 train_time:563230ms step_avg:178.80ms step:3161/5100 train_loss:3.4936 train_time:563407ms step_avg:178.80ms step:3162/5100 train_loss:3.5680 train_time:563583ms step_avg:178.80ms step:3163/5100 train_loss:3.4709 train_time:563760ms step_avg:178.80ms step:3164/5100 train_loss:3.5273 train_time:563936ms step_avg:178.80ms step:3165/5100 train_loss:3.3549 train_time:564113ms step_avg:178.80ms step:3166/5100 train_loss:3.3347 train_time:564289ms step_avg:178.80ms step:3167/5100 train_loss:3.3700 train_time:564466ms step_avg:178.80ms step:3168/5100 train_loss:3.1991 train_time:564642ms step_avg:178.80ms step:3169/5100 train_loss:3.3641 train_time:564818ms step_avg:178.80ms step:3170/5100 train_loss:3.5113 train_time:564995ms step_avg:178.80ms step:3171/5100 train_loss:3.5461 train_time:565171ms step_avg:178.80ms step:3172/5100 train_loss:3.4989 train_time:565349ms step_avg:178.79ms step:3173/5100 train_loss:3.4683 train_time:565525ms step_avg:178.79ms step:3174/5100 train_loss:3.4452 train_time:565702ms step_avg:178.79ms step:3175/5100 train_loss:3.4424 train_time:565879ms step_avg:178.79ms step:3176/5100 train_loss:3.4439 train_time:566055ms step_avg:178.79ms step:3177/5100 train_loss:3.3746 train_time:566232ms step_avg:178.79ms step:3178/5100 train_loss:3.4969 train_time:566409ms step_avg:178.79ms step:3179/5100 train_loss:3.5757 train_time:566586ms step_avg:178.79ms step:3180/5100 train_loss:3.4215 train_time:566764ms step_avg:178.79ms step:3181/5100 train_loss:3.4012 train_time:566941ms step_avg:178.79ms step:3182/5100 train_loss:3.4559 train_time:567117ms step_avg:178.79ms step:3183/5100 train_loss:3.5536 train_time:567294ms step_avg:178.79ms step:3184/5100 train_loss:3.5680 train_time:567478ms step_avg:178.79ms step:3185/5100 train_loss:3.4732 train_time:567654ms step_avg:178.79ms step:3186/5100 train_loss:3.5382 train_time:567831ms step_avg:178.79ms step:3187/5100 train_loss:3.5216 train_time:568008ms step_avg:178.79ms step:3188/5100 train_loss:3.3151 train_time:568185ms step_avg:178.79ms step:3189/5100 train_loss:3.5386 train_time:568361ms step_avg:178.79ms step:3190/5100 train_loss:3.4266 train_time:568537ms step_avg:178.79ms step:3191/5100 train_loss:3.4539 train_time:568715ms step_avg:178.79ms step:3192/5100 train_loss:3.4119 train_time:568891ms step_avg:178.78ms step:3193/5100 train_loss:3.3480 train_time:569067ms step_avg:178.78ms step:3194/5100 train_loss:4.3705 train_time:569244ms step_avg:178.78ms step:3195/5100 train_loss:3.4642 train_time:569421ms step_avg:178.78ms step:3196/5100 train_loss:3.2488 train_time:569597ms step_avg:178.78ms step:3197/5100 train_loss:3.4097 train_time:569773ms step_avg:178.78ms step:3198/5100 train_loss:3.2948 train_time:569950ms step_avg:178.78ms step:3199/5100 train_loss:3.4019 train_time:570127ms step_avg:178.78ms step:3200/5100 train_loss:3.3314 train_time:570304ms step_avg:178.78ms step:3201/5100 train_loss:3.4162 train_time:570481ms step_avg:178.78ms step:3202/5100 train_loss:3.5077 train_time:570657ms step_avg:178.78ms step:3203/5100 train_loss:3.3628 train_time:570834ms step_avg:178.78ms step:3204/5100 train_loss:3.3999 train_time:571010ms step_avg:178.78ms step:3205/5100 train_loss:3.4880 train_time:571187ms step_avg:178.78ms step:3206/5100 train_loss:3.6487 train_time:571363ms step_avg:178.77ms step:3207/5100 train_loss:3.2485 train_time:571539ms step_avg:178.77ms step:3208/5100 train_loss:3.5973 train_time:571716ms step_avg:178.77ms step:3209/5100 train_loss:3.4411 train_time:571892ms step_avg:178.77ms step:3210/5100 train_loss:3.5197 train_time:572069ms step_avg:178.77ms step:3211/5100 train_loss:3.6073 train_time:572246ms step_avg:178.77ms step:3212/5100 train_loss:3.2866 train_time:572423ms step_avg:178.77ms step:3213/5100 train_loss:3.3376 train_time:572831ms step_avg:178.84ms step:3214/5100 train_loss:3.5551 train_time:573007ms step_avg:178.84ms step:3215/5100 train_loss:3.3419 train_time:573183ms step_avg:178.84ms step:3216/5100 train_loss:3.4137 train_time:573360ms step_avg:178.84ms step:3217/5100 train_loss:3.3123 train_time:573536ms step_avg:178.84ms step:3218/5100 train_loss:3.4450 train_time:573712ms step_avg:178.84ms step:3219/5100 train_loss:3.4884 train_time:573889ms step_avg:178.84ms step:3220/5100 train_loss:3.5348 train_time:574066ms step_avg:178.84ms step:3221/5100 train_loss:3.4840 train_time:574242ms step_avg:178.84ms step:3222/5100 train_loss:3.4759 train_time:574419ms step_avg:178.84ms step:3223/5100 train_loss:3.3506 train_time:574594ms step_avg:178.83ms step:3224/5100 train_loss:3.3670 train_time:574771ms step_avg:178.83ms step:3225/5100 train_loss:3.3702 train_time:574947ms step_avg:178.83ms step:3226/5100 train_loss:3.4078 train_time:575124ms step_avg:178.83ms step:3227/5100 train_loss:3.3427 train_time:575301ms step_avg:178.83ms step:3228/5100 train_loss:3.2558 train_time:575477ms step_avg:178.83ms step:3229/5100 train_loss:3.3811 train_time:575653ms step_avg:178.83ms step:3230/5100 train_loss:3.1581 train_time:576072ms step_avg:178.90ms step:3231/5100 train_loss:3.3333 train_time:576249ms step_avg:178.90ms step:3232/5100 train_loss:3.3262 train_time:576426ms step_avg:178.90ms step:3233/5100 train_loss:3.5716 train_time:576604ms step_avg:178.90ms step:3234/5100 train_loss:3.5519 train_time:576780ms step_avg:178.90ms step:3235/5100 train_loss:3.5160 train_time:576957ms step_avg:178.90ms step:3236/5100 train_loss:3.4037 train_time:577132ms step_avg:178.90ms step:3237/5100 train_loss:3.5687 train_time:577309ms step_avg:178.90ms step:3238/5100 train_loss:3.4302 train_time:577485ms step_avg:178.90ms step:3239/5100 train_loss:3.5501 train_time:577662ms step_avg:178.90ms step:3240/5100 train_loss:3.5208 train_time:577838ms step_avg:178.90ms step:3241/5100 train_loss:3.4198 train_time:578014ms step_avg:178.90ms step:3242/5100 train_loss:3.3760 train_time:578191ms step_avg:178.90ms step:3243/5100 train_loss:3.6042 train_time:578367ms step_avg:178.89ms step:3244/5100 train_loss:3.4730 train_time:578544ms step_avg:178.89ms step:3245/5100 train_loss:3.5184 train_time:578720ms step_avg:178.89ms step:3246/5100 train_loss:3.4041 train_time:578895ms step_avg:178.89ms step:3247/5100 train_loss:3.5428 train_time:579071ms step_avg:178.89ms step:3248/5100 train_loss:3.4612 train_time:579248ms step_avg:178.89ms step:3249/5100 train_loss:3.4136 train_time:579425ms step_avg:178.89ms step:3250/5100 train_loss:3.2820 train_time:579602ms step_avg:178.89ms step:3250/5100 val_loss:3.4255 train_time:579602ms step_avg:178.89ms step:3251/5100 train_loss:3.4818 train_time:579789ms step_avg:178.89ms step:3252/5100 train_loss:3.4798 train_time:579965ms step_avg:178.89ms step:3253/5100 train_loss:3.4399 train_time:580143ms step_avg:178.89ms step:3254/5100 train_loss:3.3604 train_time:580320ms step_avg:178.89ms step:3255/5100 train_loss:3.5120 train_time:580496ms step_avg:178.89ms step:3256/5100 train_loss:3.5429 train_time:580672ms step_avg:178.89ms step:3257/5100 train_loss:3.4764 train_time:580848ms step_avg:178.89ms step:3258/5100 train_loss:3.5093 train_time:581024ms step_avg:178.89ms step:3259/5100 train_loss:3.3419 train_time:581201ms step_avg:178.89ms step:3260/5100 train_loss:3.4344 train_time:581377ms step_avg:178.89ms step:3261/5100 train_loss:3.3004 train_time:581554ms step_avg:178.88ms step:3262/5100 train_loss:3.3375 train_time:581730ms step_avg:178.88ms step:3263/5100 train_loss:3.3739 train_time:581906ms step_avg:178.88ms step:3264/5100 train_loss:3.5241 train_time:582083ms step_avg:178.88ms step:3265/5100 train_loss:3.4051 train_time:582259ms step_avg:178.88ms step:3266/5100 train_loss:3.4599 train_time:582438ms step_avg:178.88ms step:3267/5100 train_loss:3.4773 train_time:582614ms step_avg:178.88ms step:3268/5100 train_loss:3.5582 train_time:582789ms step_avg:178.88ms step:3269/5100 train_loss:3.3732 train_time:582965ms step_avg:178.88ms step:3270/5100 train_loss:3.4917 train_time:583143ms step_avg:178.88ms step:3271/5100 train_loss:3.3679 train_time:583319ms step_avg:178.88ms step:3272/5100 train_loss:3.2659 train_time:583497ms step_avg:178.88ms step:3273/5100 train_loss:3.3793 train_time:583673ms step_avg:178.88ms step:3274/5100 train_loss:3.5200 train_time:583849ms step_avg:178.88ms step:3275/5100 train_loss:3.3159 train_time:584025ms step_avg:178.87ms step:3276/5100 train_loss:3.4630 train_time:584202ms step_avg:178.87ms step:3277/5100 train_loss:3.4611 train_time:584378ms step_avg:178.87ms step:3278/5100 train_loss:3.4538 train_time:584554ms step_avg:178.87ms step:3279/5100 train_loss:3.4254 train_time:584730ms step_avg:178.87ms step:3280/5100 train_loss:3.5781 train_time:584907ms step_avg:178.87ms step:3281/5100 train_loss:3.4264 train_time:585084ms step_avg:178.87ms step:3282/5100 train_loss:3.4721 train_time:585260ms step_avg:178.87ms step:3283/5100 train_loss:3.3300 train_time:585436ms step_avg:178.87ms step:3284/5100 train_loss:3.4591 train_time:585612ms step_avg:178.87ms step:3285/5100 train_loss:3.5114 train_time:585788ms step_avg:178.87ms step:3286/5100 train_loss:3.5002 train_time:585965ms step_avg:178.87ms step:3287/5100 train_loss:3.5340 train_time:586141ms step_avg:178.87ms step:3288/5100 train_loss:3.4080 train_time:586317ms step_avg:178.86ms step:3289/5100 train_loss:3.5181 train_time:586493ms step_avg:178.86ms step:3290/5100 train_loss:3.4433 train_time:586670ms step_avg:178.86ms step:3291/5100 train_loss:3.3222 train_time:586846ms step_avg:178.86ms step:3292/5100 train_loss:3.4525 train_time:587022ms step_avg:178.86ms step:3293/5100 train_loss:3.4851 train_time:587199ms step_avg:178.86ms step:3294/5100 train_loss:3.4664 train_time:587377ms step_avg:178.86ms step:3295/5100 train_loss:3.3452 train_time:587552ms step_avg:178.86ms step:3296/5100 train_loss:3.4120 train_time:587727ms step_avg:178.86ms step:3297/5100 train_loss:3.4502 train_time:587904ms step_avg:178.86ms step:3298/5100 train_loss:3.4410 train_time:588081ms step_avg:178.86ms step:3299/5100 train_loss:3.4338 train_time:588257ms step_avg:178.86ms step:3300/5100 train_loss:3.4898 train_time:588433ms step_avg:178.86ms step:3301/5100 train_loss:3.3931 train_time:588609ms step_avg:178.85ms step:3302/5100 train_loss:3.4643 train_time:588785ms step_avg:178.85ms step:3303/5100 train_loss:3.4054 train_time:588962ms step_avg:178.85ms step:3304/5100 train_loss:3.4214 train_time:589137ms step_avg:178.85ms step:3305/5100 train_loss:3.4207 train_time:589313ms step_avg:178.85ms step:3306/5100 train_loss:3.5234 train_time:589489ms step_avg:178.85ms step:3307/5100 train_loss:3.4434 train_time:589666ms step_avg:178.85ms step:3308/5100 train_loss:3.4098 train_time:589842ms step_avg:178.85ms step:3309/5100 train_loss:3.5261 train_time:590018ms step_avg:178.85ms step:3310/5100 train_loss:3.4049 train_time:590195ms step_avg:178.85ms step:3311/5100 train_loss:3.3489 train_time:590371ms step_avg:178.85ms step:3312/5100 train_loss:3.4596 train_time:590547ms step_avg:178.85ms step:3313/5100 train_loss:3.4299 train_time:590724ms step_avg:178.84ms step:3314/5100 train_loss:3.6332 train_time:590900ms step_avg:178.84ms step:3315/5100 train_loss:3.4559 train_time:591077ms step_avg:178.84ms step:3316/5100 train_loss:3.4187 train_time:591253ms step_avg:178.84ms step:3317/5100 train_loss:3.0533 train_time:591429ms step_avg:178.84ms step:3318/5100 train_loss:3.5610 train_time:591606ms step_avg:178.84ms step:3319/5100 train_loss:3.3993 train_time:591783ms step_avg:178.84ms step:3320/5100 train_loss:3.4781 train_time:591959ms step_avg:178.84ms step:3321/5100 train_loss:3.4080 train_time:592136ms step_avg:178.84ms step:3322/5100 train_loss:3.4816 train_time:592312ms step_avg:178.84ms step:3323/5100 train_loss:3.4131 train_time:592488ms step_avg:178.84ms step:3324/5100 train_loss:3.3356 train_time:592665ms step_avg:178.84ms step:3325/5100 train_loss:3.2776 train_time:592841ms step_avg:178.84ms step:3326/5100 train_loss:3.4365 train_time:593018ms step_avg:178.84ms step:3327/5100 train_loss:3.4019 train_time:593194ms step_avg:178.83ms step:3328/5100 train_loss:3.3244 train_time:593370ms step_avg:178.83ms step:3329/5100 train_loss:3.3599 train_time:593546ms step_avg:178.83ms step:3330/5100 train_loss:3.3187 train_time:593722ms step_avg:178.83ms step:3331/5100 train_loss:3.5685 train_time:593898ms step_avg:178.83ms step:3332/5100 train_loss:3.4537 train_time:594074ms step_avg:178.83ms step:3333/5100 train_loss:3.4493 train_time:594250ms step_avg:178.83ms step:3334/5100 train_loss:3.3055 train_time:594425ms step_avg:178.83ms step:3335/5100 train_loss:3.3739 train_time:594602ms step_avg:178.83ms step:3336/5100 train_loss:3.4829 train_time:594778ms step_avg:178.83ms step:3337/5100 train_loss:3.4547 train_time:594954ms step_avg:178.83ms step:3338/5100 train_loss:3.4880 train_time:595130ms step_avg:178.83ms step:3339/5100 train_loss:3.4142 train_time:595306ms step_avg:178.82ms step:3340/5100 train_loss:3.4380 train_time:595483ms step_avg:178.82ms step:3341/5100 train_loss:3.4544 train_time:595658ms step_avg:178.82ms step:3342/5100 train_loss:3.4610 train_time:595834ms step_avg:178.82ms step:3343/5100 train_loss:3.4596 train_time:596010ms step_avg:178.82ms step:3344/5100 train_loss:3.3840 train_time:596187ms step_avg:178.82ms step:3345/5100 train_loss:3.2977 train_time:596363ms step_avg:178.82ms step:3346/5100 train_loss:3.6247 train_time:596540ms step_avg:178.82ms step:3347/5100 train_loss:3.3769 train_time:596717ms step_avg:178.82ms step:3348/5100 train_loss:3.5441 train_time:596893ms step_avg:178.82ms step:3349/5100 train_loss:3.4153 train_time:597069ms step_avg:178.82ms step:3350/5100 train_loss:3.4962 train_time:597246ms step_avg:178.82ms step:3351/5100 train_loss:3.2316 train_time:597422ms step_avg:178.82ms step:3352/5100 train_loss:3.2639 train_time:597598ms step_avg:178.81ms step:3353/5100 train_loss:3.4443 train_time:597776ms step_avg:178.81ms step:3354/5100 train_loss:3.3114 train_time:597952ms step_avg:178.81ms step:3355/5100 train_loss:3.4683 train_time:598128ms step_avg:178.81ms step:3356/5100 train_loss:3.3287 train_time:598304ms step_avg:178.81ms step:3357/5100 train_loss:3.5036 train_time:598482ms step_avg:178.81ms step:3358/5100 train_loss:3.3529 train_time:598658ms step_avg:178.81ms step:3359/5100 train_loss:3.5182 train_time:598834ms step_avg:178.81ms step:3360/5100 train_loss:3.3321 train_time:599010ms step_avg:178.81ms step:3361/5100 train_loss:4.0179 train_time:599187ms step_avg:178.81ms step:3362/5100 train_loss:3.5003 train_time:599363ms step_avg:178.81ms step:3363/5100 train_loss:3.5179 train_time:599540ms step_avg:178.81ms step:3364/5100 train_loss:3.3993 train_time:599716ms step_avg:178.81ms step:3365/5100 train_loss:3.5243 train_time:599893ms step_avg:178.81ms step:3366/5100 train_loss:3.4162 train_time:600069ms step_avg:178.80ms step:3367/5100 train_loss:3.5862 train_time:600245ms step_avg:178.80ms step:3368/5100 train_loss:3.3961 train_time:600422ms step_avg:178.80ms step:3369/5100 train_loss:3.4165 train_time:600598ms step_avg:178.80ms step:3370/5100 train_loss:3.3855 train_time:600774ms step_avg:178.80ms step:3371/5100 train_loss:3.3408 train_time:600950ms step_avg:178.80ms step:3372/5100 train_loss:3.3467 train_time:601126ms step_avg:178.80ms step:3373/5100 train_loss:3.4021 train_time:601302ms step_avg:178.80ms step:3374/5100 train_loss:3.4426 train_time:601479ms step_avg:178.80ms step:3375/5100 train_loss:3.4120 train_time:601655ms step_avg:178.80ms step:3375/5100 val_loss:3.4212 train_time:601655ms step_avg:178.80ms step:3376/5100 train_loss:3.4572 train_time:601841ms step_avg:178.80ms step:3377/5100 train_loss:3.4569 train_time:602017ms step_avg:178.80ms step:3378/5100 train_loss:3.5395 train_time:602194ms step_avg:178.80ms step:3379/5100 train_loss:3.3931 train_time:602370ms step_avg:178.80ms step:3380/5100 train_loss:3.3993 train_time:602547ms step_avg:178.80ms step:3381/5100 train_loss:3.4061 train_time:602723ms step_avg:178.80ms step:3382/5100 train_loss:3.5112 train_time:602899ms step_avg:178.80ms step:3383/5100 train_loss:3.3498 train_time:603076ms step_avg:178.80ms step:3384/5100 train_loss:3.5235 train_time:603252ms step_avg:178.79ms step:3385/5100 train_loss:3.3763 train_time:603428ms step_avg:178.79ms step:3386/5100 train_loss:3.4037 train_time:603604ms step_avg:178.79ms step:3387/5100 train_loss:3.3422 train_time:603780ms step_avg:178.79ms step:3388/5100 train_loss:3.5183 train_time:603958ms step_avg:178.79ms step:3389/5100 train_loss:3.4831 train_time:604134ms step_avg:178.79ms step:3390/5100 train_loss:3.4929 train_time:604311ms step_avg:178.79ms step:3391/5100 train_loss:3.4772 train_time:604488ms step_avg:178.79ms step:3392/5100 train_loss:3.4184 train_time:604664ms step_avg:178.79ms step:3393/5100 train_loss:3.5416 train_time:604841ms step_avg:178.79ms step:3394/5100 train_loss:3.4993 train_time:605017ms step_avg:178.79ms step:3395/5100 train_loss:3.5896 train_time:605194ms step_avg:178.79ms step:3396/5100 train_loss:3.4439 train_time:605371ms step_avg:178.79ms step:3397/5100 train_loss:3.4285 train_time:605548ms step_avg:178.79ms step:3398/5100 train_loss:3.3888 train_time:605724ms step_avg:178.79ms step:3399/5100 train_loss:3.4477 train_time:605900ms step_avg:178.78ms step:3400/5100 train_loss:3.4372 train_time:606077ms step_avg:178.78ms step:3401/5100 train_loss:3.5304 train_time:606254ms step_avg:178.78ms step:3402/5100 train_loss:3.3950 train_time:606662ms step_avg:178.85ms step:3403/5100 train_loss:3.5840 train_time:606839ms step_avg:178.85ms step:3404/5100 train_loss:3.4099 train_time:607016ms step_avg:178.85ms step:3405/5100 train_loss:3.4230 train_time:607194ms step_avg:178.85ms step:3406/5100 train_loss:3.3665 train_time:607371ms step_avg:178.85ms step:3407/5100 train_loss:3.4304 train_time:607547ms step_avg:178.85ms step:3408/5100 train_loss:3.4381 train_time:607723ms step_avg:178.85ms step:3409/5100 train_loss:3.4196 train_time:607899ms step_avg:178.85ms step:3410/5100 train_loss:3.4334 train_time:608076ms step_avg:178.85ms step:3411/5100 train_loss:3.3971 train_time:608253ms step_avg:178.85ms step:3412/5100 train_loss:3.4281 train_time:608429ms step_avg:178.84ms step:3413/5100 train_loss:3.3577 train_time:608605ms step_avg:178.84ms step:3414/5100 train_loss:3.5736 train_time:608781ms step_avg:178.84ms step:3415/5100 train_loss:3.3261 train_time:608957ms step_avg:178.84ms step:3416/5100 train_loss:3.4846 train_time:609134ms step_avg:178.84ms step:3417/5100 train_loss:3.3548 train_time:609311ms step_avg:178.84ms step:3418/5100 train_loss:3.4617 train_time:609488ms step_avg:178.84ms step:3419/5100 train_loss:3.4645 train_time:609665ms step_avg:178.84ms step:3420/5100 train_loss:3.4899 train_time:610083ms step_avg:178.91ms step:3421/5100 train_loss:3.3686 train_time:610260ms step_avg:178.91ms step:3422/5100 train_loss:3.4075 train_time:610436ms step_avg:178.91ms step:3423/5100 train_loss:3.3467 train_time:610613ms step_avg:178.91ms step:3424/5100 train_loss:3.6646 train_time:610789ms step_avg:178.91ms step:3425/5100 train_loss:3.5521 train_time:610965ms step_avg:178.91ms step:3426/5100 train_loss:3.4206 train_time:611141ms step_avg:178.91ms step:3427/5100 train_loss:3.3753 train_time:611318ms step_avg:178.90ms step:3428/5100 train_loss:3.3568 train_time:611494ms step_avg:178.90ms step:3429/5100 train_loss:3.3501 train_time:611670ms step_avg:178.90ms step:3430/5100 train_loss:3.4149 train_time:611847ms step_avg:178.90ms step:3431/5100 train_loss:3.4385 train_time:612023ms step_avg:178.90ms step:3432/5100 train_loss:3.5409 train_time:612199ms step_avg:178.90ms step:3433/5100 train_loss:3.3544 train_time:612376ms step_avg:178.90ms step:3434/5100 train_loss:3.5767 train_time:612552ms step_avg:178.90ms step:3435/5100 train_loss:3.4927 train_time:612728ms step_avg:178.90ms step:3436/5100 train_loss:3.3388 train_time:612904ms step_avg:178.90ms step:3437/5100 train_loss:3.3841 train_time:613080ms step_avg:178.90ms step:3438/5100 train_loss:3.4257 train_time:613257ms step_avg:178.90ms step:3439/5100 train_loss:3.5250 train_time:613433ms step_avg:178.90ms step:3440/5100 train_loss:3.3012 train_time:613609ms step_avg:178.89ms step:3441/5100 train_loss:3.4756 train_time:613786ms step_avg:178.89ms step:3442/5100 train_loss:3.3794 train_time:613961ms step_avg:178.89ms step:3443/5100 train_loss:3.5588 train_time:614138ms step_avg:178.89ms step:3444/5100 train_loss:3.4259 train_time:614314ms step_avg:178.89ms step:3445/5100 train_loss:3.3087 train_time:614490ms step_avg:178.89ms step:3446/5100 train_loss:3.5234 train_time:614666ms step_avg:178.89ms step:3447/5100 train_loss:3.5954 train_time:614842ms step_avg:178.89ms step:3448/5100 train_loss:3.4164 train_time:615019ms step_avg:178.89ms step:3449/5100 train_loss:3.4277 train_time:615195ms step_avg:178.89ms step:3450/5100 train_loss:3.5172 train_time:615372ms step_avg:178.89ms step:3451/5100 train_loss:3.5173 train_time:615548ms step_avg:178.89ms step:3452/5100 train_loss:3.5168 train_time:615724ms step_avg:178.89ms step:3453/5100 train_loss:3.3293 train_time:615900ms step_avg:178.88ms step:3454/5100 train_loss:3.4473 train_time:616077ms step_avg:178.88ms step:3455/5100 train_loss:3.3406 train_time:616253ms step_avg:178.88ms step:3456/5100 train_loss:3.6161 train_time:616429ms step_avg:178.88ms step:3457/5100 train_loss:3.2991 train_time:616606ms step_avg:178.88ms step:3458/5100 train_loss:3.4419 train_time:616781ms step_avg:178.88ms step:3459/5100 train_loss:3.3844 train_time:616958ms step_avg:178.88ms step:3460/5100 train_loss:3.3859 train_time:617134ms step_avg:178.88ms step:3461/5100 train_loss:3.3784 train_time:617312ms step_avg:178.88ms step:3462/5100 train_loss:3.3924 train_time:617487ms step_avg:178.88ms step:3463/5100 train_loss:3.4933 train_time:617663ms step_avg:178.88ms step:3464/5100 train_loss:3.3680 train_time:617839ms step_avg:178.88ms step:3465/5100 train_loss:3.3791 train_time:618016ms step_avg:178.88ms step:3466/5100 train_loss:3.3500 train_time:618192ms step_avg:178.88ms step:3467/5100 train_loss:3.5116 train_time:618369ms step_avg:178.87ms step:3468/5100 train_loss:3.3993 train_time:618545ms step_avg:178.87ms step:3469/5100 train_loss:3.4162 train_time:618720ms step_avg:178.87ms step:3470/5100 train_loss:3.6060 train_time:618897ms step_avg:178.87ms step:3471/5100 train_loss:3.5037 train_time:619074ms step_avg:178.87ms step:3472/5100 train_loss:3.5448 train_time:619250ms step_avg:178.87ms step:3473/5100 train_loss:4.1772 train_time:619426ms step_avg:178.87ms step:3474/5100 train_loss:3.4220 train_time:619602ms step_avg:178.87ms step:3475/5100 train_loss:3.4294 train_time:619778ms step_avg:178.87ms step:3476/5100 train_loss:3.4081 train_time:619955ms step_avg:178.87ms step:3477/5100 train_loss:3.3577 train_time:620131ms step_avg:178.87ms step:3478/5100 train_loss:3.4298 train_time:620307ms step_avg:178.87ms step:3479/5100 train_loss:3.4201 train_time:620484ms step_avg:178.87ms step:3480/5100 train_loss:3.3247 train_time:620660ms step_avg:178.86ms step:3481/5100 train_loss:3.6206 train_time:620835ms step_avg:178.86ms step:3482/5100 train_loss:3.4908 train_time:621011ms step_avg:178.86ms step:3483/5100 train_loss:3.4416 train_time:621188ms step_avg:178.86ms step:3484/5100 train_loss:3.4618 train_time:621365ms step_avg:178.86ms step:3485/5100 train_loss:3.4294 train_time:621541ms step_avg:178.86ms step:3486/5100 train_loss:3.6091 train_time:621718ms step_avg:178.86ms step:3487/5100 train_loss:3.6418 train_time:621893ms step_avg:178.86ms step:3488/5100 train_loss:3.5007 train_time:622070ms step_avg:178.86ms step:3489/5100 train_loss:3.3559 train_time:622246ms step_avg:178.86ms step:3490/5100 train_loss:3.5258 train_time:622421ms step_avg:178.86ms step:3491/5100 train_loss:3.4345 train_time:622597ms step_avg:178.86ms step:3492/5100 train_loss:3.4713 train_time:622775ms step_avg:178.86ms step:3493/5100 train_loss:3.3111 train_time:622952ms step_avg:178.85ms step:3494/5100 train_loss:3.4457 train_time:623128ms step_avg:178.85ms step:3495/5100 train_loss:3.4015 train_time:623304ms step_avg:178.85ms step:3496/5100 train_loss:3.4220 train_time:623481ms step_avg:178.85ms step:3497/5100 train_loss:3.5803 train_time:623657ms step_avg:178.85ms step:3498/5100 train_loss:3.4202 train_time:623833ms step_avg:178.85ms step:3499/5100 train_loss:3.4478 train_time:624009ms step_avg:178.85ms step:3500/5100 train_loss:3.4483 train_time:624185ms step_avg:178.85ms step:3500/5100 val_loss:3.4140 train_time:624185ms step_avg:178.85ms step:3501/5100 train_loss:3.4900 train_time:624370ms step_avg:178.85ms step:3502/5100 train_loss:3.5613 train_time:624547ms step_avg:178.85ms step:3503/5100 train_loss:3.2693 train_time:624724ms step_avg:178.85ms step:3504/5100 train_loss:3.4302 train_time:624900ms step_avg:178.85ms step:3505/5100 train_loss:3.4550 train_time:625077ms step_avg:178.85ms step:3506/5100 train_loss:3.4784 train_time:625260ms step_avg:178.85ms step:3507/5100 train_loss:3.3503 train_time:625436ms step_avg:178.85ms step:3508/5100 train_loss:3.5014 train_time:625612ms step_avg:178.85ms step:3509/5100 train_loss:3.3942 train_time:625788ms step_avg:178.85ms step:3510/5100 train_loss:3.6071 train_time:625965ms step_avg:178.85ms step:3511/5100 train_loss:3.4146 train_time:626141ms step_avg:178.85ms step:3512/5100 train_loss:3.3696 train_time:626319ms step_avg:178.85ms step:3513/5100 train_loss:3.4253 train_time:626496ms step_avg:178.85ms step:3514/5100 train_loss:3.3921 train_time:626672ms step_avg:178.84ms step:3515/5100 train_loss:3.4435 train_time:626848ms step_avg:178.84ms step:3516/5100 train_loss:3.4371 train_time:627025ms step_avg:178.84ms step:3517/5100 train_loss:3.4099 train_time:627201ms step_avg:178.84ms step:3518/5100 train_loss:3.4325 train_time:627377ms step_avg:178.84ms step:3519/5100 train_loss:3.4339 train_time:627553ms step_avg:178.84ms step:3520/5100 train_loss:3.4532 train_time:627729ms step_avg:178.84ms step:3521/5100 train_loss:3.5130 train_time:627905ms step_avg:178.84ms step:3522/5100 train_loss:3.4307 train_time:628083ms step_avg:178.84ms step:3523/5100 train_loss:3.3500 train_time:628260ms step_avg:178.84ms step:3524/5100 train_loss:3.3996 train_time:628435ms step_avg:178.84ms step:3525/5100 train_loss:3.3944 train_time:628612ms step_avg:178.84ms step:3526/5100 train_loss:3.3939 train_time:628788ms step_avg:178.84ms step:3527/5100 train_loss:3.5229 train_time:628964ms step_avg:178.84ms step:3528/5100 train_loss:3.3563 train_time:629140ms step_avg:178.83ms step:3529/5100 train_loss:3.2409 train_time:629316ms step_avg:178.83ms step:3530/5100 train_loss:3.5328 train_time:629492ms step_avg:178.83ms step:3531/5100 train_loss:3.3407 train_time:629669ms step_avg:178.83ms step:3532/5100 train_loss:3.3761 train_time:629845ms step_avg:178.83ms step:3533/5100 train_loss:3.2917 train_time:630023ms step_avg:178.83ms step:3534/5100 train_loss:3.3617 train_time:630199ms step_avg:178.83ms step:3535/5100 train_loss:3.3249 train_time:630376ms step_avg:178.83ms step:3536/5100 train_loss:3.5006 train_time:630552ms step_avg:178.83ms step:3537/5100 train_loss:3.4561 train_time:630727ms step_avg:178.83ms step:3538/5100 train_loss:3.4854 train_time:630903ms step_avg:178.83ms step:3539/5100 train_loss:3.3838 train_time:631081ms step_avg:178.83ms step:3540/5100 train_loss:3.3561 train_time:631257ms step_avg:178.83ms step:3541/5100 train_loss:3.4918 train_time:631433ms step_avg:178.83ms step:3542/5100 train_loss:3.3423 train_time:631609ms step_avg:178.82ms step:3543/5100 train_loss:3.5612 train_time:631785ms step_avg:178.82ms step:3544/5100 train_loss:3.6766 train_time:631963ms step_avg:178.82ms step:3545/5100 train_loss:3.4919 train_time:632138ms step_avg:178.82ms step:3546/5100 train_loss:3.5294 train_time:632314ms step_avg:178.82ms step:3547/5100 train_loss:3.2886 train_time:632490ms step_avg:178.82ms step:3548/5100 train_loss:3.3743 train_time:632667ms step_avg:178.82ms step:3549/5100 train_loss:3.4004 train_time:632843ms step_avg:178.82ms step:3550/5100 train_loss:3.4901 train_time:633020ms step_avg:178.82ms step:3551/5100 train_loss:3.4481 train_time:633194ms step_avg:178.82ms step:3552/5100 train_loss:3.3803 train_time:633370ms step_avg:178.82ms step:3553/5100 train_loss:3.4946 train_time:633546ms step_avg:178.82ms step:3554/5100 train_loss:3.4263 train_time:633723ms step_avg:178.82ms step:3555/5100 train_loss:3.3858 train_time:633900ms step_avg:178.82ms step:3556/5100 train_loss:3.3245 train_time:634078ms step_avg:178.81ms step:3557/5100 train_loss:3.2807 train_time:634254ms step_avg:178.81ms step:3558/5100 train_loss:3.3942 train_time:634430ms step_avg:178.81ms step:3559/5100 train_loss:3.4184 train_time:634606ms step_avg:178.81ms step:3560/5100 train_loss:3.6234 train_time:634783ms step_avg:178.81ms step:3561/5100 train_loss:3.4926 train_time:634960ms step_avg:178.81ms step:3562/5100 train_loss:3.3958 train_time:635136ms step_avg:178.81ms step:3563/5100 train_loss:3.2765 train_time:635312ms step_avg:178.81ms step:3564/5100 train_loss:3.7825 train_time:635488ms step_avg:178.81ms step:3565/5100 train_loss:3.3574 train_time:635665ms step_avg:178.81ms step:3566/5100 train_loss:3.2992 train_time:635841ms step_avg:178.81ms step:3567/5100 train_loss:3.3224 train_time:636017ms step_avg:178.81ms step:3568/5100 train_loss:3.4603 train_time:636193ms step_avg:178.81ms step:3569/5100 train_loss:3.3971 train_time:636369ms step_avg:178.81ms step:3570/5100 train_loss:3.5152 train_time:636546ms step_avg:178.80ms step:3571/5100 train_loss:3.4302 train_time:636722ms step_avg:178.80ms step:3572/5100 train_loss:3.7199 train_time:636898ms step_avg:178.80ms step:3573/5100 train_loss:3.3434 train_time:637074ms step_avg:178.80ms step:3574/5100 train_loss:3.4240 train_time:637251ms step_avg:178.80ms step:3575/5100 train_loss:3.5858 train_time:637427ms step_avg:178.80ms step:3576/5100 train_loss:3.4598 train_time:637603ms step_avg:178.80ms step:3577/5100 train_loss:3.3934 train_time:637779ms step_avg:178.80ms step:3578/5100 train_loss:3.3616 train_time:637955ms step_avg:178.80ms step:3579/5100 train_loss:3.4413 train_time:638131ms step_avg:178.80ms step:3580/5100 train_loss:3.3935 train_time:638308ms step_avg:178.80ms step:3581/5100 train_loss:3.2896 train_time:638485ms step_avg:178.80ms step:3582/5100 train_loss:3.3637 train_time:638661ms step_avg:178.80ms step:3583/5100 train_loss:3.3183 train_time:638838ms step_avg:178.80ms step:3584/5100 train_loss:3.4213 train_time:639014ms step_avg:178.80ms step:3585/5100 train_loss:3.5138 train_time:639189ms step_avg:178.79ms step:3586/5100 train_loss:3.3693 train_time:639366ms step_avg:178.79ms step:3587/5100 train_loss:3.4139 train_time:639543ms step_avg:178.79ms step:3588/5100 train_loss:3.4173 train_time:639719ms step_avg:178.79ms step:3589/5100 train_loss:3.3990 train_time:639895ms step_avg:178.79ms step:3590/5100 train_loss:3.4002 train_time:640071ms step_avg:178.79ms step:3591/5100 train_loss:3.5162 train_time:640479ms step_avg:178.85ms step:3592/5100 train_loss:3.3888 train_time:640656ms step_avg:178.85ms step:3593/5100 train_loss:3.4785 train_time:640833ms step_avg:178.85ms step:3594/5100 train_loss:3.4639 train_time:641009ms step_avg:178.85ms step:3595/5100 train_loss:3.4180 train_time:641186ms step_avg:178.85ms step:3596/5100 train_loss:3.3471 train_time:641362ms step_avg:178.85ms step:3597/5100 train_loss:3.3536 train_time:641539ms step_avg:178.85ms step:3598/5100 train_loss:3.6242 train_time:641715ms step_avg:178.85ms step:3599/5100 train_loss:3.3957 train_time:641891ms step_avg:178.85ms step:3600/5100 train_loss:3.4166 train_time:642067ms step_avg:178.85ms step:3601/5100 train_loss:3.2774 train_time:642243ms step_avg:178.85ms step:3602/5100 train_loss:3.4517 train_time:642419ms step_avg:178.85ms step:3603/5100 train_loss:3.3987 train_time:642597ms step_avg:178.85ms step:3604/5100 train_loss:3.5410 train_time:642773ms step_avg:178.85ms step:3605/5100 train_loss:3.5943 train_time:642949ms step_avg:178.85ms step:3606/5100 train_loss:3.3988 train_time:643125ms step_avg:178.84ms step:3607/5100 train_loss:3.4389 train_time:643301ms step_avg:178.84ms step:3608/5100 train_loss:3.6950 train_time:643477ms step_avg:178.84ms step:3609/5100 train_loss:3.3923 train_time:643653ms step_avg:178.84ms step:3610/5100 train_loss:3.5373 train_time:644072ms step_avg:178.91ms step:3611/5100 train_loss:3.2870 train_time:644248ms step_avg:178.91ms step:3612/5100 train_loss:3.3939 train_time:644425ms step_avg:178.91ms step:3613/5100 train_loss:3.4611 train_time:644601ms step_avg:178.91ms step:3614/5100 train_loss:3.6621 train_time:644777ms step_avg:178.91ms step:3615/5100 train_loss:3.6786 train_time:644953ms step_avg:178.91ms step:3616/5100 train_loss:3.3252 train_time:645130ms step_avg:178.90ms step:3617/5100 train_loss:3.4152 train_time:645307ms step_avg:178.90ms step:3618/5100 train_loss:3.3936 train_time:645485ms step_avg:178.90ms step:3619/5100 train_loss:3.5161 train_time:645661ms step_avg:178.90ms step:3620/5100 train_loss:3.4495 train_time:645837ms step_avg:178.90ms step:3621/5100 train_loss:3.2786 train_time:646013ms step_avg:178.90ms step:3622/5100 train_loss:3.4218 train_time:646189ms step_avg:178.90ms step:3623/5100 train_loss:3.4227 train_time:646364ms step_avg:178.90ms step:3624/5100 train_loss:3.3609 train_time:646541ms step_avg:178.90ms step:3625/5100 train_loss:3.4925 train_time:646718ms step_avg:178.90ms step:3625/5100 val_loss:3.4103 train_time:646718ms step_avg:178.90ms step:3626/5100 train_loss:3.5507 train_time:646907ms step_avg:178.90ms step:3627/5100 train_loss:3.5594 train_time:647083ms step_avg:178.90ms step:3628/5100 train_loss:3.4405 train_time:647261ms step_avg:178.90ms step:3629/5100 train_loss:3.5944 train_time:647438ms step_avg:178.90ms step:3630/5100 train_loss:3.4247 train_time:647614ms step_avg:178.90ms step:3631/5100 train_loss:3.4208 train_time:647790ms step_avg:178.90ms step:3632/5100 train_loss:3.4975 train_time:647967ms step_avg:178.90ms step:3633/5100 train_loss:3.4816 train_time:648143ms step_avg:178.90ms step:3634/5100 train_loss:3.4082 train_time:648320ms step_avg:178.90ms step:3635/5100 train_loss:3.4143 train_time:648496ms step_avg:178.90ms step:3636/5100 train_loss:3.4550 train_time:648673ms step_avg:178.89ms step:3637/5100 train_loss:3.6406 train_time:648850ms step_avg:178.89ms step:3638/5100 train_loss:3.4329 train_time:649025ms step_avg:178.89ms step:3639/5100 train_loss:3.4016 train_time:649202ms step_avg:178.89ms step:3640/5100 train_loss:3.3933 train_time:649378ms step_avg:178.89ms step:3641/5100 train_loss:3.6823 train_time:649555ms step_avg:178.89ms step:3642/5100 train_loss:3.4118 train_time:649732ms step_avg:178.89ms step:3643/5100 train_loss:3.4629 train_time:649908ms step_avg:178.89ms step:3644/5100 train_loss:3.4363 train_time:650085ms step_avg:178.89ms step:3645/5100 train_loss:3.3565 train_time:650261ms step_avg:178.89ms step:3646/5100 train_loss:3.5523 train_time:650438ms step_avg:178.89ms step:3647/5100 train_loss:3.3357 train_time:650614ms step_avg:178.89ms step:3648/5100 train_loss:3.4087 train_time:650791ms step_avg:178.89ms step:3649/5100 train_loss:3.4733 train_time:650967ms step_avg:178.89ms step:3650/5100 train_loss:3.4276 train_time:651143ms step_avg:178.89ms step:3651/5100 train_loss:3.4675 train_time:651320ms step_avg:178.88ms step:3652/5100 train_loss:3.5243 train_time:651496ms step_avg:178.88ms step:3653/5100 train_loss:3.3466 train_time:651672ms step_avg:178.88ms step:3654/5100 train_loss:3.4504 train_time:651848ms step_avg:178.88ms step:3655/5100 train_loss:3.4767 train_time:652024ms step_avg:178.88ms step:3656/5100 train_loss:4.1543 train_time:652201ms step_avg:178.88ms step:3657/5100 train_loss:3.5306 train_time:652377ms step_avg:178.88ms step:3658/5100 train_loss:3.4462 train_time:652553ms step_avg:178.88ms step:3659/5100 train_loss:3.4357 train_time:652730ms step_avg:178.88ms step:3660/5100 train_loss:3.3161 train_time:652906ms step_avg:178.88ms step:3661/5100 train_loss:3.4457 train_time:653082ms step_avg:178.88ms step:3662/5100 train_loss:3.3269 train_time:653260ms step_avg:178.88ms step:3663/5100 train_loss:3.4750 train_time:653437ms step_avg:178.88ms step:3664/5100 train_loss:3.4781 train_time:653614ms step_avg:178.88ms step:3665/5100 train_loss:3.3175 train_time:653791ms step_avg:178.88ms step:3666/5100 train_loss:3.2538 train_time:653968ms step_avg:178.88ms step:3667/5100 train_loss:3.6815 train_time:654144ms step_avg:178.87ms step:3668/5100 train_loss:3.4608 train_time:654320ms step_avg:178.87ms step:3669/5100 train_loss:3.4919 train_time:654498ms step_avg:178.87ms step:3670/5100 train_loss:3.4098 train_time:654674ms step_avg:178.87ms step:3671/5100 train_loss:3.4811 train_time:654850ms step_avg:178.87ms step:3672/5100 train_loss:3.3735 train_time:655025ms step_avg:178.87ms step:3673/5100 train_loss:3.3747 train_time:655202ms step_avg:178.87ms step:3674/5100 train_loss:3.2792 train_time:655378ms step_avg:178.87ms step:3675/5100 train_loss:3.3504 train_time:655554ms step_avg:178.87ms step:3676/5100 train_loss:3.5116 train_time:655731ms step_avg:178.87ms step:3677/5100 train_loss:3.3120 train_time:655907ms step_avg:178.87ms step:3678/5100 train_loss:3.4766 train_time:656083ms step_avg:178.87ms step:3679/5100 train_loss:3.4617 train_time:656259ms step_avg:178.87ms step:3680/5100 train_loss:3.3666 train_time:656437ms step_avg:178.87ms step:3681/5100 train_loss:3.4317 train_time:656613ms step_avg:178.86ms step:3682/5100 train_loss:3.4955 train_time:656790ms step_avg:178.86ms step:3683/5100 train_loss:3.5890 train_time:656966ms step_avg:178.86ms step:3684/5100 train_loss:3.3372 train_time:657142ms step_avg:178.86ms step:3685/5100 train_loss:3.4164 train_time:657319ms step_avg:178.86ms step:3686/5100 train_loss:3.5627 train_time:657496ms step_avg:178.86ms step:3687/5100 train_loss:3.3478 train_time:657672ms step_avg:178.86ms step:3688/5100 train_loss:3.5536 train_time:657849ms step_avg:178.86ms step:3689/5100 train_loss:3.2813 train_time:658024ms step_avg:178.86ms step:3690/5100 train_loss:3.3724 train_time:658201ms step_avg:178.86ms step:3691/5100 train_loss:3.5075 train_time:658377ms step_avg:178.86ms step:3692/5100 train_loss:3.2951 train_time:658554ms step_avg:178.86ms step:3693/5100 train_loss:3.4454 train_time:658730ms step_avg:178.86ms step:3694/5100 train_loss:3.4296 train_time:658907ms step_avg:178.86ms step:3695/5100 train_loss:3.4258 train_time:659083ms step_avg:178.86ms step:3696/5100 train_loss:3.4656 train_time:659260ms step_avg:178.86ms step:3697/5100 train_loss:3.3005 train_time:659437ms step_avg:178.85ms step:3698/5100 train_loss:3.4390 train_time:659613ms step_avg:178.85ms step:3699/5100 train_loss:3.4556 train_time:659791ms step_avg:178.85ms step:3700/5100 train_loss:3.4222 train_time:659967ms step_avg:178.85ms step:3701/5100 train_loss:3.4862 train_time:660143ms step_avg:178.85ms step:3702/5100 train_loss:3.4522 train_time:660319ms step_avg:178.85ms step:3703/5100 train_loss:3.3685 train_time:660496ms step_avg:178.85ms step:3704/5100 train_loss:3.3452 train_time:660672ms step_avg:178.85ms step:3705/5100 train_loss:3.4861 train_time:660849ms step_avg:178.85ms step:3706/5100 train_loss:3.4869 train_time:661026ms step_avg:178.85ms step:3707/5100 train_loss:3.4896 train_time:661203ms step_avg:178.85ms step:3708/5100 train_loss:3.4512 train_time:661379ms step_avg:178.85ms step:3709/5100 train_loss:3.3168 train_time:661556ms step_avg:178.85ms step:3710/5100 train_loss:3.6245 train_time:661732ms step_avg:178.85ms step:3711/5100 train_loss:3.2053 train_time:661911ms step_avg:178.85ms step:3712/5100 train_loss:3.4799 train_time:662088ms step_avg:178.85ms step:3713/5100 train_loss:3.3658 train_time:662264ms step_avg:178.85ms step:3714/5100 train_loss:3.3939 train_time:662441ms step_avg:178.84ms step:3715/5100 train_loss:3.7706 train_time:662618ms step_avg:178.84ms step:3716/5100 train_loss:3.6197 train_time:662794ms step_avg:178.84ms step:3717/5100 train_loss:3.8956 train_time:662970ms step_avg:178.84ms step:3718/5100 train_loss:3.3963 train_time:663146ms step_avg:178.84ms step:3719/5100 train_loss:3.3078 train_time:663323ms step_avg:178.84ms step:3720/5100 train_loss:3.5682 train_time:663499ms step_avg:178.84ms step:3721/5100 train_loss:3.3311 train_time:663675ms step_avg:178.84ms step:3722/5100 train_loss:3.4297 train_time:663852ms step_avg:178.84ms step:3723/5100 train_loss:3.2923 train_time:664027ms step_avg:178.84ms step:3724/5100 train_loss:3.2830 train_time:664203ms step_avg:178.84ms step:3725/5100 train_loss:3.4080 train_time:664380ms step_avg:178.84ms step:3726/5100 train_loss:3.3603 train_time:664557ms step_avg:178.84ms step:3727/5100 train_loss:3.6349 train_time:664732ms step_avg:178.84ms step:3728/5100 train_loss:3.3595 train_time:664910ms step_avg:178.84ms step:3729/5100 train_loss:3.3493 train_time:665085ms step_avg:178.83ms step:3730/5100 train_loss:3.7070 train_time:665262ms step_avg:178.83ms step:3731/5100 train_loss:3.4677 train_time:665438ms step_avg:178.83ms step:3732/5100 train_loss:3.3723 train_time:665615ms step_avg:178.83ms step:3733/5100 train_loss:3.2189 train_time:665792ms step_avg:178.83ms step:3734/5100 train_loss:3.4752 train_time:665968ms step_avg:178.83ms step:3735/5100 train_loss:3.3421 train_time:666144ms step_avg:178.83ms step:3736/5100 train_loss:3.4365 train_time:666320ms step_avg:178.83ms step:3737/5100 train_loss:3.3450 train_time:666497ms step_avg:178.83ms step:3738/5100 train_loss:3.4483 train_time:666673ms step_avg:178.83ms step:3739/5100 train_loss:3.3547 train_time:666850ms step_avg:178.83ms step:3740/5100 train_loss:3.3996 train_time:667026ms step_avg:178.83ms step:3741/5100 train_loss:3.7019 train_time:667202ms step_avg:178.83ms step:3742/5100 train_loss:3.3562 train_time:667378ms step_avg:178.83ms step:3743/5100 train_loss:3.3995 train_time:667555ms step_avg:178.83ms step:3744/5100 train_loss:3.6092 train_time:667732ms step_avg:178.82ms step:3745/5100 train_loss:3.3378 train_time:667908ms step_avg:178.82ms step:3746/5100 train_loss:3.2722 train_time:668084ms step_avg:178.82ms step:3747/5100 train_loss:3.4526 train_time:668260ms step_avg:178.82ms step:3748/5100 train_loss:3.3069 train_time:668437ms step_avg:178.82ms step:3749/5100 train_loss:3.3450 train_time:668614ms step_avg:178.82ms step:3750/5100 train_loss:3.5397 train_time:668794ms step_avg:178.82ms step:3750/5100 val_loss:3.3997 train_time:668795ms step_avg:178.82ms step:3751/5100 train_loss:3.4275 train_time:668979ms step_avg:178.82ms step:3752/5100 train_loss:3.6638 train_time:669157ms step_avg:178.82ms step:3753/5100 train_loss:3.3805 train_time:669334ms step_avg:178.82ms step:3754/5100 train_loss:3.3888 train_time:669510ms step_avg:178.82ms step:3755/5100 train_loss:3.3443 train_time:669687ms step_avg:178.82ms step:3756/5100 train_loss:3.4490 train_time:669863ms step_avg:178.82ms step:3757/5100 train_loss:3.3982 train_time:670039ms step_avg:178.82ms step:3758/5100 train_loss:3.4028 train_time:670215ms step_avg:178.82ms step:3759/5100 train_loss:3.5889 train_time:670393ms step_avg:178.82ms step:3760/5100 train_loss:3.4740 train_time:670569ms step_avg:178.82ms step:3761/5100 train_loss:3.5866 train_time:670744ms step_avg:178.82ms step:3762/5100 train_loss:3.3529 train_time:670921ms step_avg:178.82ms step:3763/5100 train_loss:3.3658 train_time:671096ms step_avg:178.82ms step:3764/5100 train_loss:3.5309 train_time:671274ms step_avg:178.82ms step:3765/5100 train_loss:3.2939 train_time:671450ms step_avg:178.81ms step:3766/5100 train_loss:3.3760 train_time:671626ms step_avg:178.81ms step:3767/5100 train_loss:3.4733 train_time:671802ms step_avg:178.81ms step:3768/5100 train_loss:3.2736 train_time:671977ms step_avg:178.81ms step:3769/5100 train_loss:3.5502 train_time:672155ms step_avg:178.81ms step:3770/5100 train_loss:3.3659 train_time:672332ms step_avg:178.81ms step:3771/5100 train_loss:3.2396 train_time:672508ms step_avg:178.81ms step:3772/5100 train_loss:3.4855 train_time:672684ms step_avg:178.81ms step:3773/5100 train_loss:3.4098 train_time:672860ms step_avg:178.81ms step:3774/5100 train_loss:3.4022 train_time:673036ms step_avg:178.81ms step:3775/5100 train_loss:3.3961 train_time:673212ms step_avg:178.81ms step:3776/5100 train_loss:3.4472 train_time:673388ms step_avg:178.81ms step:3777/5100 train_loss:3.2835 train_time:673565ms step_avg:178.81ms step:3778/5100 train_loss:3.3930 train_time:673741ms step_avg:178.81ms step:3779/5100 train_loss:3.5122 train_time:673917ms step_avg:178.81ms step:3780/5100 train_loss:3.4860 train_time:674325ms step_avg:178.87ms step:3781/5100 train_loss:3.4830 train_time:674504ms step_avg:178.87ms step:3782/5100 train_loss:3.4307 train_time:674680ms step_avg:178.87ms step:3783/5100 train_loss:3.4365 train_time:674856ms step_avg:178.86ms step:3784/5100 train_loss:3.4074 train_time:675032ms step_avg:178.86ms step:3785/5100 train_loss:3.2895 train_time:675210ms step_avg:178.86ms step:3786/5100 train_loss:3.3648 train_time:675386ms step_avg:178.86ms step:3787/5100 train_loss:3.4096 train_time:675562ms step_avg:178.86ms step:3788/5100 train_loss:3.3986 train_time:675738ms step_avg:178.86ms step:3789/5100 train_loss:3.3483 train_time:675915ms step_avg:178.86ms step:3790/5100 train_loss:3.3763 train_time:676091ms step_avg:178.86ms step:3791/5100 train_loss:3.2450 train_time:676267ms step_avg:178.86ms step:3792/5100 train_loss:3.4644 train_time:676443ms step_avg:178.86ms step:3793/5100 train_loss:3.4575 train_time:676619ms step_avg:178.86ms step:3794/5100 train_loss:3.3760 train_time:676797ms step_avg:178.86ms step:3795/5100 train_loss:3.3357 train_time:676973ms step_avg:178.86ms step:3796/5100 train_loss:3.1876 train_time:677150ms step_avg:178.86ms step:3797/5100 train_loss:3.3750 train_time:677327ms step_avg:178.86ms step:3798/5100 train_loss:3.3764 train_time:677502ms step_avg:178.85ms step:3799/5100 train_loss:3.4395 train_time:677678ms step_avg:178.85ms step:3800/5100 train_loss:3.3462 train_time:678098ms step_avg:178.92ms step:3801/5100 train_loss:3.3287 train_time:678275ms step_avg:178.92ms step:3802/5100 train_loss:3.2828 train_time:678452ms step_avg:178.92ms step:3803/5100 train_loss:3.5968 train_time:678629ms step_avg:178.92ms step:3804/5100 train_loss:3.4435 train_time:678806ms step_avg:178.92ms step:3805/5100 train_loss:3.2992 train_time:678982ms step_avg:178.91ms step:3806/5100 train_loss:3.5333 train_time:679159ms step_avg:178.91ms step:3807/5100 train_loss:3.4986 train_time:679336ms step_avg:178.91ms step:3808/5100 train_loss:3.3710 train_time:679512ms step_avg:178.91ms step:3809/5100 train_loss:3.4327 train_time:679689ms step_avg:178.91ms step:3810/5100 train_loss:3.3409 train_time:679865ms step_avg:178.91ms step:3811/5100 train_loss:3.4185 train_time:680042ms step_avg:178.91ms step:3812/5100 train_loss:3.3875 train_time:680218ms step_avg:178.91ms step:3813/5100 train_loss:3.4192 train_time:680395ms step_avg:178.91ms step:3814/5100 train_loss:3.4236 train_time:680571ms step_avg:178.91ms step:3815/5100 train_loss:3.3084 train_time:680747ms step_avg:178.91ms step:3816/5100 train_loss:3.6447 train_time:680923ms step_avg:178.91ms step:3817/5100 train_loss:3.2419 train_time:681099ms step_avg:178.91ms step:3818/5100 train_loss:3.4291 train_time:681275ms step_avg:178.91ms step:3819/5100 train_loss:3.4038 train_time:681452ms step_avg:178.91ms step:3820/5100 train_loss:3.3819 train_time:681628ms step_avg:178.90ms step:3821/5100 train_loss:3.3153 train_time:681804ms step_avg:178.90ms step:3822/5100 train_loss:3.4861 train_time:681981ms step_avg:178.90ms step:3823/5100 train_loss:3.2230 train_time:682157ms step_avg:178.90ms step:3824/5100 train_loss:3.3359 train_time:682333ms step_avg:178.90ms step:3825/5100 train_loss:3.3810 train_time:682510ms step_avg:178.90ms step:3826/5100 train_loss:3.5286 train_time:682686ms step_avg:178.90ms step:3827/5100 train_loss:3.4669 train_time:682864ms step_avg:178.90ms step:3828/5100 train_loss:3.8427 train_time:683040ms step_avg:178.90ms step:3829/5100 train_loss:3.4642 train_time:683217ms step_avg:178.90ms step:3830/5100 train_loss:3.2892 train_time:683394ms step_avg:178.90ms step:3831/5100 train_loss:3.3329 train_time:683570ms step_avg:178.90ms step:3832/5100 train_loss:3.5362 train_time:683746ms step_avg:178.90ms step:3833/5100 train_loss:3.3601 train_time:683923ms step_avg:178.90ms step:3834/5100 train_loss:3.4833 train_time:684099ms step_avg:178.90ms step:3835/5100 train_loss:3.4137 train_time:684278ms step_avg:178.90ms step:3836/5100 train_loss:3.2174 train_time:684455ms step_avg:178.90ms step:3837/5100 train_loss:3.5014 train_time:684632ms step_avg:178.90ms step:3838/5100 train_loss:3.4738 train_time:684810ms step_avg:178.89ms step:3839/5100 train_loss:3.4309 train_time:684986ms step_avg:178.89ms step:3840/5100 train_loss:3.4920 train_time:685163ms step_avg:178.89ms step:3841/5100 train_loss:3.6137 train_time:685339ms step_avg:178.89ms step:3842/5100 train_loss:3.3644 train_time:685515ms step_avg:178.89ms step:3843/5100 train_loss:3.4189 train_time:685691ms step_avg:178.89ms step:3844/5100 train_loss:3.5553 train_time:685867ms step_avg:178.89ms step:3845/5100 train_loss:3.3626 train_time:686043ms step_avg:178.89ms step:3846/5100 train_loss:3.2298 train_time:686220ms step_avg:178.89ms step:3847/5100 train_loss:3.4711 train_time:686395ms step_avg:178.89ms step:3848/5100 train_loss:3.3878 train_time:686572ms step_avg:178.89ms step:3849/5100 train_loss:3.4301 train_time:686748ms step_avg:178.89ms step:3850/5100 train_loss:3.3170 train_time:686925ms step_avg:178.89ms step:3851/5100 train_loss:3.3122 train_time:687100ms step_avg:178.89ms step:3852/5100 train_loss:3.4764 train_time:687277ms step_avg:178.89ms step:3853/5100 train_loss:3.3220 train_time:687454ms step_avg:178.88ms step:3854/5100 train_loss:3.2948 train_time:687631ms step_avg:178.88ms step:3855/5100 train_loss:3.3682 train_time:687808ms step_avg:178.88ms step:3856/5100 train_loss:3.4118 train_time:687985ms step_avg:178.88ms step:3857/5100 train_loss:3.3864 train_time:688161ms step_avg:178.88ms step:3858/5100 train_loss:3.4096 train_time:688337ms step_avg:178.88ms step:3859/5100 train_loss:3.3848 train_time:688513ms step_avg:178.88ms step:3860/5100 train_loss:3.3973 train_time:688690ms step_avg:178.88ms step:3861/5100 train_loss:3.5566 train_time:688866ms step_avg:178.88ms step:3862/5100 train_loss:3.3639 train_time:689043ms step_avg:178.88ms step:3863/5100 train_loss:3.4853 train_time:689220ms step_avg:178.88ms step:3864/5100 train_loss:3.4380 train_time:689396ms step_avg:178.88ms step:3865/5100 train_loss:3.4929 train_time:689573ms step_avg:178.88ms step:3866/5100 train_loss:3.4559 train_time:689749ms step_avg:178.88ms step:3867/5100 train_loss:3.3968 train_time:689926ms step_avg:178.88ms step:3868/5100 train_loss:3.4876 train_time:690101ms step_avg:178.88ms step:3869/5100 train_loss:3.6454 train_time:690277ms step_avg:178.87ms step:3870/5100 train_loss:3.4902 train_time:690454ms step_avg:178.87ms step:3871/5100 train_loss:3.3776 train_time:690631ms step_avg:178.87ms step:3872/5100 train_loss:3.5253 train_time:690807ms step_avg:178.87ms step:3873/5100 train_loss:3.4263 train_time:690984ms step_avg:178.87ms step:3874/5100 train_loss:3.3672 train_time:691161ms step_avg:178.87ms step:3875/5100 train_loss:3.4719 train_time:691337ms step_avg:178.87ms step:3875/5100 val_loss:3.3860 train_time:691338ms step_avg:178.87ms step:3876/5100 train_loss:3.9974 train_time:691526ms step_avg:178.87ms step:3877/5100 train_loss:3.4072 train_time:691702ms step_avg:178.87ms step:3878/5100 train_loss:3.4044 train_time:691879ms step_avg:178.87ms step:3879/5100 train_loss:3.3818 train_time:692055ms step_avg:178.87ms step:3880/5100 train_loss:3.5883 train_time:692231ms step_avg:178.87ms step:3881/5100 train_loss:3.3944 train_time:692407ms step_avg:178.87ms step:3882/5100 train_loss:3.4630 train_time:692584ms step_avg:178.87ms step:3883/5100 train_loss:3.5054 train_time:692761ms step_avg:178.87ms step:3884/5100 train_loss:3.3258 train_time:692937ms step_avg:178.87ms step:3885/5100 train_loss:3.3193 train_time:693117ms step_avg:178.87ms step:3886/5100 train_loss:3.3564 train_time:693293ms step_avg:178.87ms step:3887/5100 train_loss:3.3939 train_time:693469ms step_avg:178.87ms step:3888/5100 train_loss:3.5655 train_time:693646ms step_avg:178.87ms step:3889/5100 train_loss:3.4187 train_time:693822ms step_avg:178.87ms step:3890/5100 train_loss:3.3540 train_time:693999ms step_avg:178.87ms step:3891/5100 train_loss:3.4980 train_time:694176ms step_avg:178.87ms step:3892/5100 train_loss:3.3575 train_time:694352ms step_avg:178.86ms step:3893/5100 train_loss:3.6028 train_time:694529ms step_avg:178.86ms step:3894/5100 train_loss:3.3434 train_time:694705ms step_avg:178.86ms step:3895/5100 train_loss:3.3544 train_time:694882ms step_avg:178.86ms step:3896/5100 train_loss:3.4353 train_time:695058ms step_avg:178.86ms step:3897/5100 train_loss:3.6821 train_time:695234ms step_avg:178.86ms step:3898/5100 train_loss:3.2364 train_time:695410ms step_avg:178.86ms step:3899/5100 train_loss:3.3535 train_time:695587ms step_avg:178.86ms step:3900/5100 train_loss:3.4939 train_time:695763ms step_avg:178.86ms step:3901/5100 train_loss:3.4212 train_time:695939ms step_avg:178.86ms step:3902/5100 train_loss:3.4640 train_time:696116ms step_avg:178.86ms step:3903/5100 train_loss:3.7480 train_time:696293ms step_avg:178.86ms step:3904/5100 train_loss:3.3452 train_time:696469ms step_avg:178.86ms step:3905/5100 train_loss:3.3694 train_time:696646ms step_avg:178.86ms step:3906/5100 train_loss:3.3180 train_time:696822ms step_avg:178.86ms step:3907/5100 train_loss:3.4736 train_time:696998ms step_avg:178.86ms step:3908/5100 train_loss:3.4889 train_time:697174ms step_avg:178.85ms step:3909/5100 train_loss:3.4823 train_time:697351ms step_avg:178.85ms step:3910/5100 train_loss:3.4268 train_time:697527ms step_avg:178.85ms step:3911/5100 train_loss:3.3607 train_time:697703ms step_avg:178.85ms step:3912/5100 train_loss:3.3886 train_time:697880ms step_avg:178.85ms step:3913/5100 train_loss:3.3747 train_time:698056ms step_avg:178.85ms step:3914/5100 train_loss:3.4970 train_time:698232ms step_avg:178.85ms step:3915/5100 train_loss:3.3318 train_time:698408ms step_avg:178.85ms step:3916/5100 train_loss:3.3135 train_time:698585ms step_avg:178.85ms step:3917/5100 train_loss:3.2996 train_time:698762ms step_avg:178.85ms step:3918/5100 train_loss:3.4225 train_time:698938ms step_avg:178.85ms step:3919/5100 train_loss:3.5456 train_time:699114ms step_avg:178.85ms step:3920/5100 train_loss:3.3247 train_time:699290ms step_avg:178.85ms step:3921/5100 train_loss:3.2995 train_time:699466ms step_avg:178.85ms step:3922/5100 train_loss:3.3832 train_time:699643ms step_avg:178.85ms step:3923/5100 train_loss:3.3711 train_time:699819ms step_avg:178.84ms step:3924/5100 train_loss:3.3913 train_time:699995ms step_avg:178.84ms step:3925/5100 train_loss:3.4677 train_time:700171ms step_avg:178.84ms step:3926/5100 train_loss:3.4287 train_time:700347ms step_avg:178.84ms step:3927/5100 train_loss:3.5292 train_time:700524ms step_avg:178.84ms step:3928/5100 train_loss:3.4155 train_time:700701ms step_avg:178.84ms step:3929/5100 train_loss:3.2707 train_time:700878ms step_avg:178.84ms step:3930/5100 train_loss:3.5940 train_time:701054ms step_avg:178.84ms step:3931/5100 train_loss:3.3772 train_time:701230ms step_avg:178.84ms step:3932/5100 train_loss:3.4237 train_time:701407ms step_avg:178.84ms step:3933/5100 train_loss:3.4695 train_time:701584ms step_avg:178.84ms step:3934/5100 train_loss:3.3420 train_time:701760ms step_avg:178.84ms step:3935/5100 train_loss:3.4718 train_time:701937ms step_avg:178.84ms step:3936/5100 train_loss:3.4761 train_time:702113ms step_avg:178.84ms step:3937/5100 train_loss:3.4023 train_time:702290ms step_avg:178.84ms step:3938/5100 train_loss:3.4612 train_time:702466ms step_avg:178.84ms step:3939/5100 train_loss:3.3850 train_time:702643ms step_avg:178.84ms step:3940/5100 train_loss:3.1414 train_time:702820ms step_avg:178.83ms step:3941/5100 train_loss:3.3584 train_time:702996ms step_avg:178.83ms step:3942/5100 train_loss:3.4649 train_time:703172ms step_avg:178.83ms step:3943/5100 train_loss:3.5599 train_time:703348ms step_avg:178.83ms step:3944/5100 train_loss:3.5898 train_time:703526ms step_avg:178.83ms step:3945/5100 train_loss:3.4381 train_time:703703ms step_avg:178.83ms step:3946/5100 train_loss:3.3466 train_time:703880ms step_avg:178.83ms step:3947/5100 train_loss:3.3655 train_time:704056ms step_avg:178.83ms step:3948/5100 train_loss:3.4411 train_time:704233ms step_avg:178.83ms step:3949/5100 train_loss:3.2436 train_time:704409ms step_avg:178.83ms step:3950/5100 train_loss:3.4426 train_time:704586ms step_avg:178.83ms step:3951/5100 train_loss:3.3801 train_time:704763ms step_avg:178.83ms step:3952/5100 train_loss:3.1914 train_time:704940ms step_avg:178.83ms step:3953/5100 train_loss:3.2571 train_time:705116ms step_avg:178.83ms step:3954/5100 train_loss:3.5231 train_time:705293ms step_avg:178.83ms step:3955/5100 train_loss:3.4375 train_time:705468ms step_avg:178.83ms step:3956/5100 train_loss:3.3773 train_time:705645ms step_avg:178.83ms step:3957/5100 train_loss:3.4336 train_time:705823ms step_avg:178.83ms step:3958/5100 train_loss:3.1538 train_time:706000ms step_avg:178.82ms step:3959/5100 train_loss:3.4447 train_time:706177ms step_avg:178.82ms step:3960/5100 train_loss:3.3932 train_time:706353ms step_avg:178.82ms step:3961/5100 train_loss:3.3594 train_time:706529ms step_avg:178.82ms step:3962/5100 train_loss:3.3820 train_time:706706ms step_avg:178.82ms step:3963/5100 train_loss:3.3988 train_time:706883ms step_avg:178.82ms step:3964/5100 train_loss:3.4291 train_time:707059ms step_avg:178.82ms step:3965/5100 train_loss:3.2780 train_time:707235ms step_avg:178.82ms step:3966/5100 train_loss:3.3986 train_time:707411ms step_avg:178.82ms step:3967/5100 train_loss:3.4722 train_time:707588ms step_avg:178.82ms step:3968/5100 train_loss:3.3900 train_time:707764ms step_avg:178.82ms step:3969/5100 train_loss:3.4746 train_time:708175ms step_avg:178.88ms step:3970/5100 train_loss:3.3616 train_time:708351ms step_avg:178.88ms step:3971/5100 train_loss:3.5454 train_time:708527ms step_avg:178.88ms step:3972/5100 train_loss:3.4749 train_time:708704ms step_avg:178.88ms step:3973/5100 train_loss:3.4279 train_time:708883ms step_avg:178.88ms step:3974/5100 train_loss:3.3174 train_time:709059ms step_avg:178.87ms step:3975/5100 train_loss:3.3701 train_time:709236ms step_avg:178.87ms step:3976/5100 train_loss:3.4376 train_time:709411ms step_avg:178.87ms step:3977/5100 train_loss:3.3441 train_time:709587ms step_avg:178.87ms step:3978/5100 train_loss:3.4050 train_time:709764ms step_avg:178.87ms step:3979/5100 train_loss:3.4862 train_time:709941ms step_avg:178.87ms step:3980/5100 train_loss:3.4234 train_time:710117ms step_avg:178.87ms step:3981/5100 train_loss:3.4418 train_time:710292ms step_avg:178.87ms step:3982/5100 train_loss:3.6319 train_time:710468ms step_avg:178.87ms step:3983/5100 train_loss:3.3698 train_time:710645ms step_avg:178.87ms step:3984/5100 train_loss:3.4360 train_time:710822ms step_avg:178.87ms step:3985/5100 train_loss:3.3701 train_time:710999ms step_avg:178.87ms step:3986/5100 train_loss:3.2930 train_time:711175ms step_avg:178.87ms step:3987/5100 train_loss:3.3327 train_time:711351ms step_avg:178.87ms step:3988/5100 train_loss:3.3599 train_time:711527ms step_avg:178.87ms step:3989/5100 train_loss:3.0907 train_time:711704ms step_avg:178.87ms step:3990/5100 train_loss:3.4116 train_time:712128ms step_avg:178.93ms step:3991/5100 train_loss:3.3851 train_time:712304ms step_avg:178.93ms step:3992/5100 train_loss:3.2287 train_time:712481ms step_avg:178.93ms step:3993/5100 train_loss:3.3418 train_time:712659ms step_avg:178.93ms step:3994/5100 train_loss:3.5335 train_time:712835ms step_avg:178.92ms step:3995/5100 train_loss:3.3581 train_time:713011ms step_avg:178.92ms step:3996/5100 train_loss:3.2740 train_time:713187ms step_avg:178.92ms step:3997/5100 train_loss:3.4221 train_time:713364ms step_avg:178.92ms step:3998/5100 train_loss:3.3377 train_time:713540ms step_avg:178.92ms step:3999/5100 train_loss:3.2932 train_time:713717ms step_avg:178.92ms step:4000/5100 train_loss:3.3623 train_time:713893ms step_avg:178.92ms step:4000/5100 val_loss:3.3709 train_time:713893ms step_avg:178.92ms step:4001/5100 train_loss:3.4916 train_time:714078ms step_avg:178.92ms step:4002/5100 train_loss:3.5594 train_time:714255ms step_avg:178.92ms step:4003/5100 train_loss:3.2315 train_time:714432ms step_avg:178.92ms step:4004/5100 train_loss:3.4284 train_time:714608ms step_avg:178.92ms step:4005/5100 train_loss:3.3296 train_time:714785ms step_avg:178.92ms step:4006/5100 train_loss:3.3834 train_time:714961ms step_avg:178.92ms step:4007/5100 train_loss:3.3643 train_time:715138ms step_avg:178.92ms step:4008/5100 train_loss:3.5670 train_time:715315ms step_avg:178.92ms step:4009/5100 train_loss:3.1523 train_time:715492ms step_avg:178.92ms step:4010/5100 train_loss:3.3549 train_time:715668ms step_avg:178.92ms step:4011/5100 train_loss:3.3375 train_time:715844ms step_avg:178.92ms step:4012/5100 train_loss:3.3074 train_time:716021ms step_avg:178.92ms step:4013/5100 train_loss:3.4750 train_time:716197ms step_avg:178.91ms step:4014/5100 train_loss:3.3358 train_time:716375ms step_avg:178.91ms step:4015/5100 train_loss:3.4312 train_time:716551ms step_avg:178.91ms step:4016/5100 train_loss:3.5234 train_time:716727ms step_avg:178.91ms step:4017/5100 train_loss:3.5155 train_time:716906ms step_avg:178.91ms step:4018/5100 train_loss:3.2604 train_time:717083ms step_avg:178.91ms step:4019/5100 train_loss:3.3984 train_time:717259ms step_avg:178.91ms step:4020/5100 train_loss:3.3243 train_time:717436ms step_avg:178.91ms step:4021/5100 train_loss:3.5912 train_time:717612ms step_avg:178.91ms step:4022/5100 train_loss:3.4664 train_time:717788ms step_avg:178.91ms step:4023/5100 train_loss:3.4516 train_time:717966ms step_avg:178.91ms step:4024/5100 train_loss:3.4236 train_time:718141ms step_avg:178.91ms step:4025/5100 train_loss:3.4619 train_time:718319ms step_avg:178.91ms step:4026/5100 train_loss:3.2151 train_time:718495ms step_avg:178.91ms step:4027/5100 train_loss:3.4228 train_time:718672ms step_avg:178.91ms step:4028/5100 train_loss:3.3785 train_time:718848ms step_avg:178.91ms step:4029/5100 train_loss:3.2644 train_time:719026ms step_avg:178.91ms step:4030/5100 train_loss:3.2926 train_time:719202ms step_avg:178.91ms step:4031/5100 train_loss:3.3451 train_time:719379ms step_avg:178.91ms step:4032/5100 train_loss:3.4300 train_time:719556ms step_avg:178.90ms step:4033/5100 train_loss:3.4043 train_time:719732ms step_avg:178.90ms step:4034/5100 train_loss:3.3824 train_time:719909ms step_avg:178.90ms step:4035/5100 train_loss:3.3589 train_time:720085ms step_avg:178.90ms step:4036/5100 train_loss:3.3093 train_time:720262ms step_avg:178.90ms step:4037/5100 train_loss:3.4820 train_time:720438ms step_avg:178.90ms step:4038/5100 train_loss:3.4148 train_time:720614ms step_avg:178.90ms step:4039/5100 train_loss:3.3921 train_time:720791ms step_avg:178.90ms step:4040/5100 train_loss:3.3796 train_time:720967ms step_avg:178.90ms step:4041/5100 train_loss:3.4300 train_time:721144ms step_avg:178.90ms step:4042/5100 train_loss:3.6227 train_time:721320ms step_avg:178.90ms step:4043/5100 train_loss:3.5194 train_time:721497ms step_avg:178.90ms step:4044/5100 train_loss:3.3029 train_time:721674ms step_avg:178.90ms step:4045/5100 train_loss:3.4724 train_time:721851ms step_avg:178.90ms step:4046/5100 train_loss:3.1736 train_time:722036ms step_avg:178.90ms step:4047/5100 train_loss:3.4427 train_time:722213ms step_avg:178.90ms step:4048/5100 train_loss:3.5249 train_time:722389ms step_avg:178.90ms step:4049/5100 train_loss:3.3977 train_time:722566ms step_avg:178.90ms step:4050/5100 train_loss:3.3282 train_time:722742ms step_avg:178.90ms step:4051/5100 train_loss:3.3681 train_time:722918ms step_avg:178.90ms step:4052/5100 train_loss:3.3013 train_time:723095ms step_avg:178.90ms step:4053/5100 train_loss:3.5062 train_time:723272ms step_avg:178.89ms step:4054/5100 train_loss:3.3679 train_time:723448ms step_avg:178.89ms step:4055/5100 train_loss:3.4492 train_time:723625ms step_avg:178.89ms step:4056/5100 train_loss:3.4324 train_time:723801ms step_avg:178.89ms step:4057/5100 train_loss:3.4010 train_time:723978ms step_avg:178.89ms step:4058/5100 train_loss:3.2702 train_time:724155ms step_avg:178.89ms step:4059/5100 train_loss:3.4208 train_time:724332ms step_avg:178.89ms step:4060/5100 train_loss:3.2805 train_time:724508ms step_avg:178.89ms step:4061/5100 train_loss:3.3661 train_time:724685ms step_avg:178.89ms step:4062/5100 train_loss:3.4848 train_time:724862ms step_avg:178.89ms step:4063/5100 train_loss:3.6558 train_time:725037ms step_avg:178.89ms step:4064/5100 train_loss:3.0501 train_time:725214ms step_avg:178.89ms step:4065/5100 train_loss:3.4045 train_time:725390ms step_avg:178.89ms step:4066/5100 train_loss:3.2955 train_time:725567ms step_avg:178.89ms step:4067/5100 train_loss:3.4488 train_time:725743ms step_avg:178.89ms step:4068/5100 train_loss:3.4550 train_time:725920ms step_avg:178.89ms step:4069/5100 train_loss:3.2560 train_time:726096ms step_avg:178.89ms step:4070/5100 train_loss:3.4222 train_time:726273ms step_avg:178.88ms step:4071/5100 train_loss:3.2379 train_time:726449ms step_avg:178.88ms step:4072/5100 train_loss:3.4255 train_time:726626ms step_avg:178.88ms step:4073/5100 train_loss:3.5229 train_time:726802ms step_avg:178.88ms step:4074/5100 train_loss:3.4553 train_time:726979ms step_avg:178.88ms step:4075/5100 train_loss:3.3655 train_time:727156ms step_avg:178.88ms step:4076/5100 train_loss:3.3667 train_time:727333ms step_avg:178.88ms step:4077/5100 train_loss:3.2361 train_time:727509ms step_avg:178.88ms step:4078/5100 train_loss:3.4045 train_time:727685ms step_avg:178.88ms step:4079/5100 train_loss:3.4228 train_time:727861ms step_avg:178.88ms step:4080/5100 train_loss:3.2112 train_time:728038ms step_avg:178.88ms step:4081/5100 train_loss:3.3815 train_time:728214ms step_avg:178.88ms step:4082/5100 train_loss:3.3328 train_time:728391ms step_avg:178.88ms step:4083/5100 train_loss:3.3930 train_time:728568ms step_avg:178.88ms step:4084/5100 train_loss:3.3875 train_time:728744ms step_avg:178.88ms step:4085/5100 train_loss:3.4274 train_time:728921ms step_avg:178.88ms step:4086/5100 train_loss:3.3824 train_time:729098ms step_avg:178.88ms step:4087/5100 train_loss:3.3618 train_time:729274ms step_avg:178.88ms step:4088/5100 train_loss:3.4767 train_time:729450ms step_avg:178.87ms step:4089/5100 train_loss:3.3122 train_time:729627ms step_avg:178.87ms step:4090/5100 train_loss:3.3343 train_time:729803ms step_avg:178.87ms step:4091/5100 train_loss:3.3592 train_time:729980ms step_avg:178.87ms step:4092/5100 train_loss:3.3035 train_time:730156ms step_avg:178.87ms step:4093/5100 train_loss:3.2981 train_time:730333ms step_avg:178.87ms step:4094/5100 train_loss:3.4799 train_time:730510ms step_avg:178.87ms step:4095/5100 train_loss:3.4625 train_time:730686ms step_avg:178.87ms step:4096/5100 train_loss:3.3697 train_time:730862ms step_avg:178.87ms step:4097/5100 train_loss:3.4396 train_time:731039ms step_avg:178.87ms step:4098/5100 train_loss:3.2171 train_time:731216ms step_avg:178.87ms step:4099/5100 train_loss:3.3559 train_time:731393ms step_avg:178.87ms step:4100/5100 train_loss:3.3345 train_time:731570ms step_avg:178.87ms step:4101/5100 train_loss:3.1220 train_time:731745ms step_avg:178.87ms step:4102/5100 train_loss:3.4067 train_time:731921ms step_avg:178.87ms step:4103/5100 train_loss:3.3686 train_time:732098ms step_avg:178.87ms step:4104/5100 train_loss:3.2003 train_time:732275ms step_avg:178.87ms step:4105/5100 train_loss:3.2855 train_time:732450ms step_avg:178.86ms step:4106/5100 train_loss:3.4438 train_time:732627ms step_avg:178.86ms step:4107/5100 train_loss:3.4962 train_time:732804ms step_avg:178.86ms step:4108/5100 train_loss:3.3875 train_time:732980ms step_avg:178.86ms step:4109/5100 train_loss:3.4753 train_time:733156ms step_avg:178.86ms step:4110/5100 train_loss:3.4685 train_time:733333ms step_avg:178.86ms step:4111/5100 train_loss:3.6388 train_time:733510ms step_avg:178.86ms step:4112/5100 train_loss:3.2874 train_time:733687ms step_avg:178.86ms step:4113/5100 train_loss:3.4260 train_time:733864ms step_avg:178.86ms step:4114/5100 train_loss:3.3192 train_time:734040ms step_avg:178.86ms step:4115/5100 train_loss:3.4341 train_time:734217ms step_avg:178.86ms step:4116/5100 train_loss:3.4290 train_time:734393ms step_avg:178.86ms step:4117/5100 train_loss:3.6659 train_time:734571ms step_avg:178.86ms step:4118/5100 train_loss:3.2392 train_time:734746ms step_avg:178.86ms step:4119/5100 train_loss:3.4112 train_time:734922ms step_avg:178.86ms step:4120/5100 train_loss:3.3213 train_time:735099ms step_avg:178.86ms step:4121/5100 train_loss:3.4328 train_time:735275ms step_avg:178.86ms step:4122/5100 train_loss:3.4221 train_time:735451ms step_avg:178.85ms step:4123/5100 train_loss:3.4053 train_time:735627ms step_avg:178.85ms step:4124/5100 train_loss:3.2485 train_time:735804ms step_avg:178.85ms step:4125/5100 train_loss:3.2556 train_time:735981ms step_avg:178.85ms step:4125/5100 val_loss:3.3576 train_time:735981ms step_avg:178.85ms step:4126/5100 train_loss:3.3816 train_time:736168ms step_avg:178.86ms step:4127/5100 train_loss:3.3153 train_time:736346ms step_avg:178.85ms step:4128/5100 train_loss:3.3814 train_time:736523ms step_avg:178.85ms step:4129/5100 train_loss:3.3786 train_time:736701ms step_avg:178.85ms step:4130/5100 train_loss:3.1625 train_time:736877ms step_avg:178.85ms step:4131/5100 train_loss:3.4518 train_time:737053ms step_avg:178.85ms step:4132/5100 train_loss:3.4067 train_time:737229ms step_avg:178.85ms step:4133/5100 train_loss:3.3354 train_time:737406ms step_avg:178.85ms step:4134/5100 train_loss:3.5367 train_time:737582ms step_avg:178.85ms step:4135/5100 train_loss:3.3565 train_time:737759ms step_avg:178.85ms step:4136/5100 train_loss:3.3425 train_time:737935ms step_avg:178.85ms step:4137/5100 train_loss:3.4914 train_time:738111ms step_avg:178.85ms step:4138/5100 train_loss:3.3327 train_time:738288ms step_avg:178.85ms step:4139/5100 train_loss:3.3959 train_time:738464ms step_avg:178.85ms step:4140/5100 train_loss:3.4876 train_time:738641ms step_avg:178.85ms step:4141/5100 train_loss:3.5082 train_time:738817ms step_avg:178.85ms step:4142/5100 train_loss:3.4742 train_time:738995ms step_avg:178.85ms step:4143/5100 train_loss:3.4505 train_time:739171ms step_avg:178.85ms step:4144/5100 train_loss:3.3580 train_time:739347ms step_avg:178.85ms step:4145/5100 train_loss:3.3256 train_time:739525ms step_avg:178.85ms step:4146/5100 train_loss:3.4323 train_time:739701ms step_avg:178.84ms step:4147/5100 train_loss:2.9978 train_time:739878ms step_avg:178.84ms step:4148/5100 train_loss:3.3504 train_time:740054ms step_avg:178.84ms step:4149/5100 train_loss:3.3854 train_time:740230ms step_avg:178.84ms step:4150/5100 train_loss:3.1972 train_time:740407ms step_avg:178.84ms step:4151/5100 train_loss:3.2357 train_time:740584ms step_avg:178.84ms step:4152/5100 train_loss:3.2696 train_time:740761ms step_avg:178.84ms step:4153/5100 train_loss:3.3233 train_time:740937ms step_avg:178.84ms step:4154/5100 train_loss:3.3857 train_time:741114ms step_avg:178.84ms step:4155/5100 train_loss:3.4944 train_time:741290ms step_avg:178.84ms step:4156/5100 train_loss:3.3000 train_time:741467ms step_avg:178.84ms step:4157/5100 train_loss:3.2471 train_time:741644ms step_avg:178.84ms step:4158/5100 train_loss:3.3685 train_time:742052ms step_avg:178.89ms step:4159/5100 train_loss:3.3757 train_time:742227ms step_avg:178.89ms step:4160/5100 train_loss:3.2929 train_time:742403ms step_avg:178.89ms step:4161/5100 train_loss:3.3790 train_time:742579ms step_avg:178.89ms step:4162/5100 train_loss:3.3120 train_time:742755ms step_avg:178.89ms step:4163/5100 train_loss:3.5388 train_time:742931ms step_avg:178.89ms step:4164/5100 train_loss:3.2198 train_time:743107ms step_avg:178.89ms step:4165/5100 train_loss:3.3226 train_time:743284ms step_avg:178.89ms step:4166/5100 train_loss:3.3101 train_time:743460ms step_avg:178.89ms step:4167/5100 train_loss:3.3532 train_time:743636ms step_avg:178.89ms step:4168/5100 train_loss:3.3432 train_time:743812ms step_avg:178.89ms step:4169/5100 train_loss:3.3752 train_time:743988ms step_avg:178.89ms step:4170/5100 train_loss:3.2102 train_time:744165ms step_avg:178.89ms step:4171/5100 train_loss:3.3164 train_time:744343ms step_avg:178.89ms step:4172/5100 train_loss:3.4396 train_time:744518ms step_avg:178.88ms step:4173/5100 train_loss:3.5067 train_time:744694ms step_avg:178.88ms step:4174/5100 train_loss:3.8818 train_time:744870ms step_avg:178.88ms step:4175/5100 train_loss:3.3178 train_time:745047ms step_avg:178.88ms step:4176/5100 train_loss:3.4686 train_time:745224ms step_avg:178.88ms step:4177/5100 train_loss:3.2719 train_time:745400ms step_avg:178.88ms step:4178/5100 train_loss:3.2984 train_time:745575ms step_avg:178.88ms step:4179/5100 train_loss:3.4543 train_time:745752ms step_avg:178.88ms step:4180/5100 train_loss:3.4040 train_time:746172ms step_avg:178.94ms step:4181/5100 train_loss:3.3944 train_time:746349ms step_avg:178.94ms step:4182/5100 train_loss:3.3935 train_time:746526ms step_avg:178.94ms step:4183/5100 train_loss:3.4280 train_time:746702ms step_avg:178.94ms step:4184/5100 train_loss:3.8644 train_time:746879ms step_avg:178.94ms step:4185/5100 train_loss:3.3722 train_time:747054ms step_avg:178.94ms step:4186/5100 train_loss:3.4212 train_time:747231ms step_avg:178.93ms step:4187/5100 train_loss:3.4646 train_time:747408ms step_avg:178.93ms step:4188/5100 train_loss:3.4568 train_time:747585ms step_avg:178.93ms step:4189/5100 train_loss:3.0892 train_time:747762ms step_avg:178.93ms step:4190/5100 train_loss:3.4526 train_time:747940ms step_avg:178.93ms step:4191/5100 train_loss:3.4646 train_time:748117ms step_avg:178.93ms step:4192/5100 train_loss:3.4339 train_time:748293ms step_avg:178.93ms step:4193/5100 train_loss:3.3671 train_time:748469ms step_avg:178.93ms step:4194/5100 train_loss:3.3859 train_time:748645ms step_avg:178.93ms step:4195/5100 train_loss:3.3630 train_time:748822ms step_avg:178.93ms step:4196/5100 train_loss:3.3129 train_time:748999ms step_avg:178.93ms step:4197/5100 train_loss:3.6567 train_time:749178ms step_avg:178.93ms step:4198/5100 train_loss:3.1064 train_time:749354ms step_avg:178.93ms step:4199/5100 train_loss:3.5097 train_time:749529ms step_avg:178.93ms step:4200/5100 train_loss:3.3800 train_time:749706ms step_avg:178.93ms step:4201/5100 train_loss:3.2652 train_time:749882ms step_avg:178.93ms step:4202/5100 train_loss:3.4100 train_time:750059ms step_avg:178.93ms step:4203/5100 train_loss:3.2712 train_time:750235ms step_avg:178.93ms step:4204/5100 train_loss:3.2874 train_time:750411ms step_avg:178.92ms step:4205/5100 train_loss:3.2956 train_time:750588ms step_avg:178.92ms step:4206/5100 train_loss:3.2949 train_time:750766ms step_avg:178.92ms step:4207/5100 train_loss:3.7545 train_time:750942ms step_avg:178.92ms step:4208/5100 train_loss:3.3281 train_time:751120ms step_avg:178.92ms step:4209/5100 train_loss:3.4554 train_time:751296ms step_avg:178.92ms step:4210/5100 train_loss:3.3505 train_time:751473ms step_avg:178.92ms step:4211/5100 train_loss:3.7441 train_time:751649ms step_avg:178.92ms step:4212/5100 train_loss:3.4028 train_time:751826ms step_avg:178.92ms step:4213/5100 train_loss:3.3967 train_time:752003ms step_avg:178.92ms step:4214/5100 train_loss:3.2744 train_time:752180ms step_avg:178.92ms step:4215/5100 train_loss:3.3278 train_time:752356ms step_avg:178.92ms step:4216/5100 train_loss:3.4102 train_time:752532ms step_avg:178.92ms step:4217/5100 train_loss:3.2676 train_time:752708ms step_avg:178.92ms step:4218/5100 train_loss:3.3275 train_time:752886ms step_avg:178.92ms step:4219/5100 train_loss:3.3860 train_time:753062ms step_avg:178.92ms step:4220/5100 train_loss:3.1962 train_time:753239ms step_avg:178.92ms step:4221/5100 train_loss:3.3601 train_time:753416ms step_avg:178.92ms step:4222/5100 train_loss:3.3835 train_time:753594ms step_avg:178.92ms step:4223/5100 train_loss:3.3523 train_time:753770ms step_avg:178.92ms step:4224/5100 train_loss:3.5582 train_time:753947ms step_avg:178.91ms step:4225/5100 train_loss:3.4364 train_time:754124ms step_avg:178.91ms step:4226/5100 train_loss:3.4882 train_time:754301ms step_avg:178.91ms step:4227/5100 train_loss:3.2785 train_time:754477ms step_avg:178.91ms step:4228/5100 train_loss:3.3449 train_time:754653ms step_avg:178.91ms step:4229/5100 train_loss:3.3798 train_time:754829ms step_avg:178.91ms step:4230/5100 train_loss:3.2842 train_time:755006ms step_avg:178.91ms step:4231/5100 train_loss:3.4780 train_time:755184ms step_avg:178.91ms step:4232/5100 train_loss:3.4769 train_time:755360ms step_avg:178.91ms step:4233/5100 train_loss:3.4661 train_time:755539ms step_avg:178.91ms step:4234/5100 train_loss:3.5781 train_time:755715ms step_avg:178.91ms step:4235/5100 train_loss:3.4329 train_time:755892ms step_avg:178.91ms step:4236/5100 train_loss:3.3825 train_time:756069ms step_avg:178.91ms step:4237/5100 train_loss:3.2230 train_time:756245ms step_avg:178.91ms step:4238/5100 train_loss:3.4381 train_time:756421ms step_avg:178.91ms step:4239/5100 train_loss:3.3543 train_time:756597ms step_avg:178.91ms step:4240/5100 train_loss:3.2717 train_time:756774ms step_avg:178.91ms step:4241/5100 train_loss:3.3093 train_time:756950ms step_avg:178.91ms step:4242/5100 train_loss:3.2598 train_time:757127ms step_avg:178.91ms step:4243/5100 train_loss:3.3407 train_time:757304ms step_avg:178.90ms step:4244/5100 train_loss:3.2633 train_time:757479ms step_avg:178.90ms step:4245/5100 train_loss:3.1693 train_time:757657ms step_avg:178.90ms step:4246/5100 train_loss:3.4782 train_time:757833ms step_avg:178.90ms step:4247/5100 train_loss:3.2787 train_time:758009ms step_avg:178.90ms step:4248/5100 train_loss:3.2105 train_time:758185ms step_avg:178.90ms step:4249/5100 train_loss:3.4205 train_time:758362ms step_avg:178.90ms step:4250/5100 train_loss:3.6986 train_time:758539ms step_avg:178.90ms step:4250/5100 val_loss:3.3452 train_time:758539ms step_avg:178.90ms step:4251/5100 train_loss:3.3286 train_time:758726ms step_avg:178.90ms step:4252/5100 train_loss:3.5730 train_time:758903ms step_avg:178.90ms step:4253/5100 train_loss:3.4177 train_time:759079ms step_avg:178.90ms step:4254/5100 train_loss:3.2254 train_time:759255ms step_avg:178.90ms step:4255/5100 train_loss:3.3027 train_time:759431ms step_avg:178.90ms step:4256/5100 train_loss:3.2376 train_time:759608ms step_avg:178.90ms step:4257/5100 train_loss:3.4680 train_time:759784ms step_avg:178.90ms step:4258/5100 train_loss:3.3586 train_time:759960ms step_avg:178.90ms step:4259/5100 train_loss:3.4045 train_time:760137ms step_avg:178.90ms step:4260/5100 train_loss:3.2274 train_time:760314ms step_avg:178.90ms step:4261/5100 train_loss:3.5280 train_time:760491ms step_avg:178.90ms step:4262/5100 train_loss:3.3562 train_time:760667ms step_avg:178.90ms step:4263/5100 train_loss:3.3701 train_time:760844ms step_avg:178.90ms step:4264/5100 train_loss:3.4123 train_time:761020ms step_avg:178.90ms step:4265/5100 train_loss:3.3597 train_time:761197ms step_avg:178.89ms step:4266/5100 train_loss:3.3685 train_time:761373ms step_avg:178.89ms step:4267/5100 train_loss:3.4839 train_time:761550ms step_avg:178.89ms step:4268/5100 train_loss:3.3181 train_time:761729ms step_avg:178.89ms step:4269/5100 train_loss:3.8521 train_time:761905ms step_avg:178.89ms step:4270/5100 train_loss:3.2985 train_time:762081ms step_avg:178.89ms step:4271/5100 train_loss:3.3947 train_time:762258ms step_avg:178.89ms step:4272/5100 train_loss:3.3227 train_time:762434ms step_avg:178.89ms step:4273/5100 train_loss:3.5309 train_time:762610ms step_avg:178.89ms step:4274/5100 train_loss:3.4501 train_time:762787ms step_avg:178.89ms step:4275/5100 train_loss:3.3084 train_time:762963ms step_avg:178.89ms step:4276/5100 train_loss:3.3602 train_time:763140ms step_avg:178.89ms step:4277/5100 train_loss:3.2863 train_time:763316ms step_avg:178.89ms step:4278/5100 train_loss:3.3196 train_time:763493ms step_avg:178.89ms step:4279/5100 train_loss:3.3280 train_time:763670ms step_avg:178.89ms step:4280/5100 train_loss:3.3968 train_time:763847ms step_avg:178.89ms step:4281/5100 train_loss:3.3818 train_time:764023ms step_avg:178.89ms step:4282/5100 train_loss:3.3918 train_time:764199ms step_avg:178.89ms step:4283/5100 train_loss:3.3241 train_time:764376ms step_avg:178.89ms step:4284/5100 train_loss:3.3619 train_time:764553ms step_avg:178.88ms step:4285/5100 train_loss:3.4355 train_time:764728ms step_avg:178.88ms step:4286/5100 train_loss:3.3823 train_time:764904ms step_avg:178.88ms step:4287/5100 train_loss:3.2679 train_time:765080ms step_avg:178.88ms step:4288/5100 train_loss:3.3100 train_time:765257ms step_avg:178.88ms step:4289/5100 train_loss:3.4025 train_time:765433ms step_avg:178.88ms step:4290/5100 train_loss:3.3623 train_time:765610ms step_avg:178.88ms step:4291/5100 train_loss:3.2623 train_time:765786ms step_avg:178.88ms step:4292/5100 train_loss:3.2949 train_time:765962ms step_avg:178.88ms step:4293/5100 train_loss:3.3620 train_time:766138ms step_avg:178.88ms step:4294/5100 train_loss:3.1411 train_time:766316ms step_avg:178.88ms step:4295/5100 train_loss:3.4958 train_time:766493ms step_avg:178.88ms step:4296/5100 train_loss:3.3859 train_time:766669ms step_avg:178.88ms step:4297/5100 train_loss:3.3372 train_time:766846ms step_avg:178.88ms step:4298/5100 train_loss:3.5002 train_time:767022ms step_avg:178.88ms step:4299/5100 train_loss:3.4268 train_time:767198ms step_avg:178.88ms step:4300/5100 train_loss:3.2585 train_time:767374ms step_avg:178.88ms step:4301/5100 train_loss:3.2519 train_time:767554ms step_avg:178.88ms step:4302/5100 train_loss:3.4025 train_time:767728ms step_avg:178.87ms step:4303/5100 train_loss:3.2484 train_time:767905ms step_avg:178.87ms step:4304/5100 train_loss:3.3866 train_time:768081ms step_avg:178.87ms step:4305/5100 train_loss:3.4745 train_time:768257ms step_avg:178.87ms step:4306/5100 train_loss:3.2248 train_time:768433ms step_avg:178.87ms step:4307/5100 train_loss:3.7501 train_time:768610ms step_avg:178.87ms step:4308/5100 train_loss:3.3573 train_time:768788ms step_avg:178.87ms step:4309/5100 train_loss:3.2828 train_time:768966ms step_avg:178.87ms step:4310/5100 train_loss:3.2963 train_time:769142ms step_avg:178.87ms step:4311/5100 train_loss:3.5868 train_time:769318ms step_avg:178.87ms step:4312/5100 train_loss:3.4153 train_time:769495ms step_avg:178.87ms step:4313/5100 train_loss:3.2743 train_time:769672ms step_avg:178.87ms step:4314/5100 train_loss:3.4869 train_time:769849ms step_avg:178.87ms step:4315/5100 train_loss:3.4107 train_time:770025ms step_avg:178.87ms step:4316/5100 train_loss:3.3219 train_time:770203ms step_avg:178.87ms step:4317/5100 train_loss:3.3716 train_time:770379ms step_avg:178.87ms step:4318/5100 train_loss:3.3176 train_time:770556ms step_avg:178.87ms step:4319/5100 train_loss:3.4399 train_time:770732ms step_avg:178.87ms step:4320/5100 train_loss:3.4889 train_time:770909ms step_avg:178.87ms step:4321/5100 train_loss:3.3146 train_time:771085ms step_avg:178.86ms step:4322/5100 train_loss:3.4805 train_time:771261ms step_avg:178.86ms step:4323/5100 train_loss:3.3398 train_time:771438ms step_avg:178.86ms step:4324/5100 train_loss:3.2598 train_time:771614ms step_avg:178.86ms step:4325/5100 train_loss:3.2067 train_time:771792ms step_avg:178.86ms step:4326/5100 train_loss:3.3020 train_time:771969ms step_avg:178.86ms step:4327/5100 train_loss:3.1820 train_time:772145ms step_avg:178.86ms step:4328/5100 train_loss:3.3114 train_time:772321ms step_avg:178.86ms step:4329/5100 train_loss:3.3363 train_time:772497ms step_avg:178.86ms step:4330/5100 train_loss:3.2797 train_time:772674ms step_avg:178.86ms step:4331/5100 train_loss:3.5323 train_time:772850ms step_avg:178.86ms step:4332/5100 train_loss:3.3314 train_time:773026ms step_avg:178.86ms step:4333/5100 train_loss:3.4502 train_time:773202ms step_avg:178.86ms step:4334/5100 train_loss:3.8132 train_time:773378ms step_avg:178.86ms step:4335/5100 train_loss:3.3469 train_time:773554ms step_avg:178.86ms step:4336/5100 train_loss:3.4506 train_time:773729ms step_avg:178.86ms step:4337/5100 train_loss:3.3454 train_time:773906ms step_avg:178.85ms step:4338/5100 train_loss:3.2271 train_time:774082ms step_avg:178.85ms step:4339/5100 train_loss:3.3682 train_time:774258ms step_avg:178.85ms step:4340/5100 train_loss:3.2686 train_time:774434ms step_avg:178.85ms step:4341/5100 train_loss:3.3664 train_time:774611ms step_avg:178.85ms step:4342/5100 train_loss:3.3826 train_time:774787ms step_avg:178.85ms step:4343/5100 train_loss:3.3809 train_time:774963ms step_avg:178.85ms step:4344/5100 train_loss:3.3743 train_time:775140ms step_avg:178.85ms step:4345/5100 train_loss:4.0096 train_time:775317ms step_avg:178.85ms step:4346/5100 train_loss:3.4494 train_time:775494ms step_avg:178.85ms step:4347/5100 train_loss:3.2458 train_time:775903ms step_avg:178.90ms step:4348/5100 train_loss:3.3816 train_time:776079ms step_avg:178.90ms step:4349/5100 train_loss:3.3298 train_time:776255ms step_avg:178.90ms step:4350/5100 train_loss:3.2471 train_time:776431ms step_avg:178.90ms step:4351/5100 train_loss:3.4100 train_time:776607ms step_avg:178.90ms step:4352/5100 train_loss:3.3578 train_time:776783ms step_avg:178.90ms step:4353/5100 train_loss:3.4333 train_time:776960ms step_avg:178.90ms step:4354/5100 train_loss:3.4737 train_time:777136ms step_avg:178.90ms step:4355/5100 train_loss:3.2936 train_time:777313ms step_avg:178.90ms step:4356/5100 train_loss:3.2466 train_time:777489ms step_avg:178.90ms step:4357/5100 train_loss:3.3814 train_time:777667ms step_avg:178.90ms step:4358/5100 train_loss:3.3230 train_time:777843ms step_avg:178.90ms step:4359/5100 train_loss:3.5110 train_time:778020ms step_avg:178.90ms step:4360/5100 train_loss:3.3724 train_time:778196ms step_avg:178.90ms step:4361/5100 train_loss:3.4530 train_time:778372ms step_avg:178.90ms step:4362/5100 train_loss:3.6102 train_time:778548ms step_avg:178.89ms step:4363/5100 train_loss:3.3970 train_time:778725ms step_avg:178.89ms step:4364/5100 train_loss:3.3708 train_time:778901ms step_avg:178.89ms step:4365/5100 train_loss:3.5717 train_time:779078ms step_avg:178.89ms step:4366/5100 train_loss:3.4609 train_time:779255ms step_avg:178.89ms step:4367/5100 train_loss:3.2850 train_time:779432ms step_avg:178.89ms step:4368/5100 train_loss:3.3089 train_time:779609ms step_avg:178.89ms step:4369/5100 train_loss:3.4002 train_time:779785ms step_avg:178.89ms step:4370/5100 train_loss:3.4036 train_time:780203ms step_avg:178.95ms step:4371/5100 train_loss:3.5463 train_time:780380ms step_avg:178.95ms step:4372/5100 train_loss:3.2409 train_time:780556ms step_avg:178.94ms step:4373/5100 train_loss:3.2364 train_time:780732ms step_avg:178.94ms step:4374/5100 train_loss:3.3928 train_time:780909ms step_avg:178.94ms step:4375/5100 train_loss:3.4194 train_time:781084ms step_avg:178.94ms step:4375/5100 val_loss:3.3310 train_time:781085ms step_avg:178.94ms step:4376/5100 train_loss:3.4846 train_time:781271ms step_avg:178.94ms step:4377/5100 train_loss:3.2685 train_time:781449ms step_avg:178.94ms step:4378/5100 train_loss:3.3234 train_time:781625ms step_avg:178.94ms step:4379/5100 train_loss:3.3325 train_time:781802ms step_avg:178.94ms step:4380/5100 train_loss:3.3726 train_time:781978ms step_avg:178.94ms step:4381/5100 train_loss:3.2736 train_time:782154ms step_avg:178.94ms step:4382/5100 train_loss:3.5161 train_time:782330ms step_avg:178.94ms step:4383/5100 train_loss:3.4269 train_time:782506ms step_avg:178.94ms step:4384/5100 train_loss:3.4240 train_time:782683ms step_avg:178.94ms step:4385/5100 train_loss:3.3028 train_time:782860ms step_avg:178.94ms step:4386/5100 train_loss:3.4275 train_time:783035ms step_avg:178.94ms step:4387/5100 train_loss:3.3252 train_time:783211ms step_avg:178.94ms step:4388/5100 train_loss:3.4550 train_time:783388ms step_avg:178.94ms step:4389/5100 train_loss:3.2725 train_time:783564ms step_avg:178.94ms step:4390/5100 train_loss:3.4007 train_time:783739ms step_avg:178.94ms step:4391/5100 train_loss:3.4236 train_time:783915ms step_avg:178.94ms step:4392/5100 train_loss:3.2404 train_time:784092ms step_avg:178.93ms step:4393/5100 train_loss:3.9192 train_time:784267ms step_avg:178.93ms step:4394/5100 train_loss:3.3148 train_time:784444ms step_avg:178.93ms step:4395/5100 train_loss:3.4989 train_time:784620ms step_avg:178.93ms step:4396/5100 train_loss:3.2740 train_time:784796ms step_avg:178.93ms step:4397/5100 train_loss:3.4020 train_time:784972ms step_avg:178.93ms step:4398/5100 train_loss:3.1992 train_time:785148ms step_avg:178.93ms step:4399/5100 train_loss:3.4231 train_time:785325ms step_avg:178.93ms step:4400/5100 train_loss:3.2522 train_time:785501ms step_avg:178.93ms step:4401/5100 train_loss:3.3189 train_time:785677ms step_avg:178.93ms step:4402/5100 train_loss:3.3898 train_time:785853ms step_avg:178.93ms step:4403/5100 train_loss:3.2207 train_time:786030ms step_avg:178.93ms step:4404/5100 train_loss:3.2771 train_time:786206ms step_avg:178.93ms step:4405/5100 train_loss:3.4725 train_time:786383ms step_avg:178.93ms step:4406/5100 train_loss:3.3313 train_time:786560ms step_avg:178.93ms step:4407/5100 train_loss:3.3511 train_time:786737ms step_avg:178.93ms step:4408/5100 train_loss:3.3160 train_time:786913ms step_avg:178.93ms step:4409/5100 train_loss:3.3909 train_time:787089ms step_avg:178.92ms step:4410/5100 train_loss:3.3777 train_time:787267ms step_avg:178.92ms step:4411/5100 train_loss:3.5001 train_time:787443ms step_avg:178.92ms step:4412/5100 train_loss:3.3406 train_time:787619ms step_avg:178.92ms step:4413/5100 train_loss:3.3704 train_time:787795ms step_avg:178.92ms step:4414/5100 train_loss:3.3528 train_time:787971ms step_avg:178.92ms step:4415/5100 train_loss:3.4055 train_time:788148ms step_avg:178.92ms step:4416/5100 train_loss:3.3533 train_time:788324ms step_avg:178.92ms step:4417/5100 train_loss:3.4199 train_time:788500ms step_avg:178.92ms step:4418/5100 train_loss:3.3477 train_time:788676ms step_avg:178.92ms step:4419/5100 train_loss:3.2572 train_time:788852ms step_avg:178.92ms step:4420/5100 train_loss:3.3074 train_time:789028ms step_avg:178.92ms step:4421/5100 train_loss:3.5393 train_time:789204ms step_avg:178.92ms step:4422/5100 train_loss:3.3439 train_time:789381ms step_avg:178.92ms step:4423/5100 train_loss:3.2614 train_time:789559ms step_avg:178.92ms step:4424/5100 train_loss:3.2825 train_time:789734ms step_avg:178.92ms step:4425/5100 train_loss:3.4618 train_time:789911ms step_avg:178.92ms step:4426/5100 train_loss:3.4181 train_time:790088ms step_avg:178.91ms step:4427/5100 train_loss:3.3225 train_time:790264ms step_avg:178.91ms step:4428/5100 train_loss:3.5352 train_time:790441ms step_avg:178.91ms step:4429/5100 train_loss:3.4468 train_time:790617ms step_avg:178.91ms step:4430/5100 train_loss:3.2416 train_time:790793ms step_avg:178.91ms step:4431/5100 train_loss:3.2383 train_time:790969ms step_avg:178.91ms step:4432/5100 train_loss:3.3789 train_time:791146ms step_avg:178.91ms step:4433/5100 train_loss:3.2579 train_time:791322ms step_avg:178.91ms step:4434/5100 train_loss:3.3731 train_time:791499ms step_avg:178.91ms step:4435/5100 train_loss:3.4242 train_time:791675ms step_avg:178.91ms step:4436/5100 train_loss:3.3184 train_time:791851ms step_avg:178.91ms step:4437/5100 train_loss:3.2433 train_time:792027ms step_avg:178.91ms step:4438/5100 train_loss:3.4880 train_time:792203ms step_avg:178.91ms step:4439/5100 train_loss:3.4106 train_time:792379ms step_avg:178.91ms step:4440/5100 train_loss:3.3094 train_time:792556ms step_avg:178.91ms step:4441/5100 train_loss:3.4124 train_time:792731ms step_avg:178.91ms step:4442/5100 train_loss:3.4482 train_time:792907ms step_avg:178.91ms step:4443/5100 train_loss:3.5009 train_time:793085ms step_avg:178.90ms step:4444/5100 train_loss:3.3861 train_time:793262ms step_avg:178.90ms step:4445/5100 train_loss:3.2063 train_time:793439ms step_avg:178.90ms step:4446/5100 train_loss:3.4773 train_time:793615ms step_avg:178.90ms step:4447/5100 train_loss:3.3573 train_time:793792ms step_avg:178.90ms step:4448/5100 train_loss:3.2441 train_time:793968ms step_avg:178.90ms step:4449/5100 train_loss:3.3837 train_time:794145ms step_avg:178.90ms step:4450/5100 train_loss:3.3742 train_time:794321ms step_avg:178.90ms step:4451/5100 train_loss:3.3851 train_time:794498ms step_avg:178.90ms step:4452/5100 train_loss:3.4172 train_time:794673ms step_avg:178.90ms step:4453/5100 train_loss:3.2831 train_time:794851ms step_avg:178.90ms step:4454/5100 train_loss:3.3034 train_time:795026ms step_avg:178.90ms step:4455/5100 train_loss:3.3128 train_time:795204ms step_avg:178.90ms step:4456/5100 train_loss:3.2053 train_time:795380ms step_avg:178.90ms step:4457/5100 train_loss:3.4218 train_time:795558ms step_avg:178.90ms step:4458/5100 train_loss:3.2841 train_time:795734ms step_avg:178.90ms step:4459/5100 train_loss:3.2581 train_time:795910ms step_avg:178.90ms step:4460/5100 train_loss:3.3747 train_time:796087ms step_avg:178.90ms step:4461/5100 train_loss:3.8663 train_time:796264ms step_avg:178.90ms step:4462/5100 train_loss:3.3631 train_time:796443ms step_avg:178.90ms step:4463/5100 train_loss:3.4811 train_time:796620ms step_avg:178.90ms step:4464/5100 train_loss:3.3852 train_time:796796ms step_avg:178.89ms step:4465/5100 train_loss:3.3610 train_time:796971ms step_avg:178.89ms step:4466/5100 train_loss:3.4330 train_time:797148ms step_avg:178.89ms step:4467/5100 train_loss:3.2402 train_time:797325ms step_avg:178.89ms step:4468/5100 train_loss:3.2843 train_time:797501ms step_avg:178.89ms step:4469/5100 train_loss:3.4127 train_time:797677ms step_avg:178.89ms step:4470/5100 train_loss:3.3955 train_time:797853ms step_avg:178.89ms step:4471/5100 train_loss:3.3175 train_time:798029ms step_avg:178.89ms step:4472/5100 train_loss:3.2856 train_time:798206ms step_avg:178.89ms step:4473/5100 train_loss:3.3598 train_time:798384ms step_avg:178.89ms step:4474/5100 train_loss:3.2106 train_time:798561ms step_avg:178.89ms step:4475/5100 train_loss:3.2878 train_time:798737ms step_avg:178.89ms step:4476/5100 train_loss:3.3088 train_time:798913ms step_avg:178.89ms step:4477/5100 train_loss:3.4808 train_time:799090ms step_avg:178.89ms step:4478/5100 train_loss:3.2085 train_time:799266ms step_avg:178.89ms step:4479/5100 train_loss:3.3324 train_time:799443ms step_avg:178.89ms step:4480/5100 train_loss:3.3718 train_time:799620ms step_avg:178.89ms step:4481/5100 train_loss:3.3389 train_time:799796ms step_avg:178.89ms step:4482/5100 train_loss:3.3352 train_time:799972ms step_avg:178.88ms step:4483/5100 train_loss:3.1493 train_time:800148ms step_avg:178.88ms step:4484/5100 train_loss:3.2961 train_time:800325ms step_avg:178.88ms step:4485/5100 train_loss:3.2478 train_time:800501ms step_avg:178.88ms step:4486/5100 train_loss:3.3813 train_time:800677ms step_avg:178.88ms step:4487/5100 train_loss:3.2586 train_time:800853ms step_avg:178.88ms step:4488/5100 train_loss:3.3293 train_time:801030ms step_avg:178.88ms step:4489/5100 train_loss:3.4680 train_time:801206ms step_avg:178.88ms step:4490/5100 train_loss:3.4365 train_time:801383ms step_avg:178.88ms step:4491/5100 train_loss:3.3139 train_time:801559ms step_avg:178.88ms step:4492/5100 train_loss:3.2730 train_time:801735ms step_avg:178.88ms step:4493/5100 train_loss:3.3245 train_time:801911ms step_avg:178.88ms step:4494/5100 train_loss:3.3493 train_time:802088ms step_avg:178.88ms step:4495/5100 train_loss:3.3455 train_time:802267ms step_avg:178.88ms step:4496/5100 train_loss:3.2914 train_time:802444ms step_avg:178.88ms step:4497/5100 train_loss:3.4427 train_time:802620ms step_avg:178.88ms step:4498/5100 train_loss:3.3299 train_time:802796ms step_avg:178.88ms step:4499/5100 train_loss:3.1723 train_time:802972ms step_avg:178.88ms step:4500/5100 train_loss:3.4708 train_time:803149ms step_avg:178.88ms step:4500/5100 val_loss:3.3195 train_time:803150ms step_avg:178.88ms step:4501/5100 train_loss:3.2757 train_time:803335ms step_avg:178.88ms step:4502/5100 train_loss:3.2387 train_time:803512ms step_avg:178.88ms step:4503/5100 train_loss:3.4269 train_time:803689ms step_avg:178.88ms step:4504/5100 train_loss:3.3029 train_time:803864ms step_avg:178.87ms step:4505/5100 train_loss:3.4004 train_time:804040ms step_avg:178.87ms step:4506/5100 train_loss:3.3185 train_time:804218ms step_avg:178.87ms step:4507/5100 train_loss:3.3899 train_time:804395ms step_avg:178.87ms step:4508/5100 train_loss:3.1134 train_time:804570ms step_avg:178.87ms step:4509/5100 train_loss:3.3921 train_time:804745ms step_avg:178.87ms step:4510/5100 train_loss:3.2350 train_time:804922ms step_avg:178.87ms step:4511/5100 train_loss:3.3056 train_time:805099ms step_avg:178.87ms step:4512/5100 train_loss:3.2516 train_time:805275ms step_avg:178.87ms step:4513/5100 train_loss:3.2471 train_time:805451ms step_avg:178.87ms step:4514/5100 train_loss:3.2071 train_time:805627ms step_avg:178.87ms step:4515/5100 train_loss:3.3499 train_time:805803ms step_avg:178.87ms step:4516/5100 train_loss:3.2102 train_time:805979ms step_avg:178.87ms step:4517/5100 train_loss:3.3084 train_time:806157ms step_avg:178.87ms step:4518/5100 train_loss:3.3144 train_time:806333ms step_avg:178.87ms step:4519/5100 train_loss:3.3277 train_time:806510ms step_avg:178.87ms step:4520/5100 train_loss:3.2471 train_time:806686ms step_avg:178.87ms step:4521/5100 train_loss:3.4297 train_time:806862ms step_avg:178.87ms step:4522/5100 train_loss:3.5037 train_time:807038ms step_avg:178.86ms step:4523/5100 train_loss:3.8446 train_time:807216ms step_avg:178.86ms step:4524/5100 train_loss:3.5783 train_time:807392ms step_avg:178.86ms step:4525/5100 train_loss:3.3401 train_time:807568ms step_avg:178.86ms step:4526/5100 train_loss:3.2974 train_time:807744ms step_avg:178.86ms step:4527/5100 train_loss:3.3591 train_time:807921ms step_avg:178.86ms step:4528/5100 train_loss:3.3215 train_time:808097ms step_avg:178.86ms step:4529/5100 train_loss:3.2260 train_time:808273ms step_avg:178.86ms step:4530/5100 train_loss:3.9350 train_time:808450ms step_avg:178.86ms step:4531/5100 train_loss:3.4057 train_time:808626ms step_avg:178.86ms step:4532/5100 train_loss:3.1315 train_time:808803ms step_avg:178.86ms step:4533/5100 train_loss:3.2377 train_time:808983ms step_avg:178.86ms step:4534/5100 train_loss:3.3515 train_time:809159ms step_avg:178.86ms step:4535/5100 train_loss:3.5672 train_time:809336ms step_avg:178.86ms step:4536/5100 train_loss:3.5691 train_time:809745ms step_avg:178.91ms step:4537/5100 train_loss:3.2996 train_time:809922ms step_avg:178.91ms step:4538/5100 train_loss:3.2987 train_time:810098ms step_avg:178.91ms step:4539/5100 train_loss:3.3311 train_time:810277ms step_avg:178.91ms step:4540/5100 train_loss:3.9032 train_time:810453ms step_avg:178.91ms step:4541/5100 train_loss:3.3889 train_time:810631ms step_avg:178.91ms step:4542/5100 train_loss:3.2980 train_time:810807ms step_avg:178.91ms step:4543/5100 train_loss:3.4623 train_time:810983ms step_avg:178.91ms step:4544/5100 train_loss:3.2506 train_time:811158ms step_avg:178.91ms step:4545/5100 train_loss:3.3565 train_time:811336ms step_avg:178.91ms step:4546/5100 train_loss:3.5344 train_time:811512ms step_avg:178.90ms step:4547/5100 train_loss:3.3922 train_time:811689ms step_avg:178.90ms step:4548/5100 train_loss:3.3368 train_time:811865ms step_avg:178.90ms step:4549/5100 train_loss:3.3270 train_time:812042ms step_avg:178.90ms step:4550/5100 train_loss:3.2764 train_time:812218ms step_avg:178.90ms step:4551/5100 train_loss:3.2539 train_time:812395ms step_avg:178.90ms step:4552/5100 train_loss:3.2307 train_time:812572ms step_avg:178.90ms step:4553/5100 train_loss:3.3457 train_time:812748ms step_avg:178.90ms step:4554/5100 train_loss:3.5412 train_time:812923ms step_avg:178.90ms step:4555/5100 train_loss:3.4198 train_time:813100ms step_avg:178.90ms step:4556/5100 train_loss:3.1689 train_time:813279ms step_avg:178.90ms step:4557/5100 train_loss:3.3724 train_time:813456ms step_avg:178.90ms step:4558/5100 train_loss:3.3844 train_time:813631ms step_avg:178.90ms step:4559/5100 train_loss:3.3705 train_time:813808ms step_avg:178.90ms step:4560/5100 train_loss:3.4735 train_time:814230ms step_avg:178.95ms step:4561/5100 train_loss:3.3080 train_time:814407ms step_avg:178.95ms step:4562/5100 train_loss:3.3096 train_time:814583ms step_avg:178.95ms step:4563/5100 train_loss:3.3343 train_time:814759ms step_avg:178.95ms step:4564/5100 train_loss:3.3745 train_time:814936ms step_avg:178.95ms step:4565/5100 train_loss:3.4592 train_time:815112ms step_avg:178.95ms step:4566/5100 train_loss:3.5144 train_time:815289ms step_avg:178.95ms step:4567/5100 train_loss:3.3691 train_time:815465ms step_avg:178.95ms step:4568/5100 train_loss:3.2392 train_time:815641ms step_avg:178.95ms step:4569/5100 train_loss:3.3454 train_time:815818ms step_avg:178.95ms step:4570/5100 train_loss:3.2242 train_time:815995ms step_avg:178.95ms step:4571/5100 train_loss:3.2523 train_time:816172ms step_avg:178.95ms step:4572/5100 train_loss:3.4596 train_time:816349ms step_avg:178.95ms step:4573/5100 train_loss:3.1615 train_time:816525ms step_avg:178.94ms step:4574/5100 train_loss:3.2472 train_time:816701ms step_avg:178.94ms step:4575/5100 train_loss:3.3698 train_time:816878ms step_avg:178.94ms step:4576/5100 train_loss:3.4058 train_time:817055ms step_avg:178.94ms step:4577/5100 train_loss:3.3603 train_time:817231ms step_avg:178.94ms step:4578/5100 train_loss:3.3288 train_time:817407ms step_avg:178.94ms step:4579/5100 train_loss:3.3479 train_time:817583ms step_avg:178.94ms step:4580/5100 train_loss:3.4465 train_time:817760ms step_avg:178.94ms step:4581/5100 train_loss:3.2801 train_time:817937ms step_avg:178.94ms step:4582/5100 train_loss:3.3017 train_time:818114ms step_avg:178.94ms step:4583/5100 train_loss:3.4049 train_time:818290ms step_avg:178.94ms step:4584/5100 train_loss:3.2664 train_time:818466ms step_avg:178.94ms step:4585/5100 train_loss:3.3787 train_time:818642ms step_avg:178.94ms step:4586/5100 train_loss:3.3517 train_time:818819ms step_avg:178.94ms step:4587/5100 train_loss:3.3339 train_time:818996ms step_avg:178.94ms step:4588/5100 train_loss:3.1884 train_time:819172ms step_avg:178.94ms step:4589/5100 train_loss:3.3204 train_time:819349ms step_avg:178.94ms step:4590/5100 train_loss:3.5085 train_time:819525ms step_avg:178.94ms step:4591/5100 train_loss:3.3360 train_time:819703ms step_avg:178.94ms step:4592/5100 train_loss:3.3296 train_time:819879ms step_avg:178.93ms step:4593/5100 train_loss:3.2889 train_time:820056ms step_avg:178.93ms step:4594/5100 train_loss:3.4577 train_time:820232ms step_avg:178.93ms step:4595/5100 train_loss:3.3345 train_time:820411ms step_avg:178.93ms step:4596/5100 train_loss:3.2414 train_time:820587ms step_avg:178.93ms step:4597/5100 train_loss:3.2285 train_time:820763ms step_avg:178.93ms step:4598/5100 train_loss:3.4093 train_time:820939ms step_avg:178.93ms step:4599/5100 train_loss:3.3368 train_time:821116ms step_avg:178.93ms step:4600/5100 train_loss:3.4582 train_time:821293ms step_avg:178.93ms step:4601/5100 train_loss:3.3617 train_time:821469ms step_avg:178.93ms step:4602/5100 train_loss:3.1646 train_time:821645ms step_avg:178.93ms step:4603/5100 train_loss:3.2819 train_time:821822ms step_avg:178.93ms step:4604/5100 train_loss:3.3659 train_time:822002ms step_avg:178.93ms step:4605/5100 train_loss:3.3741 train_time:822176ms step_avg:178.93ms step:4606/5100 train_loss:3.2955 train_time:822353ms step_avg:178.93ms step:4607/5100 train_loss:3.4129 train_time:822529ms step_avg:178.93ms step:4608/5100 train_loss:3.2874 train_time:822705ms step_avg:178.93ms step:4609/5100 train_loss:3.3942 train_time:822881ms step_avg:178.93ms step:4610/5100 train_loss:3.3373 train_time:823058ms step_avg:178.93ms step:4611/5100 train_loss:3.3796 train_time:823235ms step_avg:178.93ms step:4612/5100 train_loss:3.5181 train_time:823411ms step_avg:178.92ms step:4613/5100 train_loss:3.2130 train_time:823588ms step_avg:178.92ms step:4614/5100 train_loss:3.0622 train_time:823764ms step_avg:178.92ms step:4615/5100 train_loss:3.2774 train_time:823941ms step_avg:178.92ms step:4616/5100 train_loss:3.1982 train_time:824117ms step_avg:178.92ms step:4617/5100 train_loss:3.3000 train_time:824295ms step_avg:178.92ms step:4618/5100 train_loss:3.1711 train_time:824471ms step_avg:178.92ms step:4619/5100 train_loss:3.3884 train_time:824647ms step_avg:178.92ms step:4620/5100 train_loss:3.4287 train_time:824823ms step_avg:178.92ms step:4621/5100 train_loss:3.4717 train_time:825000ms step_avg:178.92ms step:4622/5100 train_loss:3.2606 train_time:825190ms step_avg:178.92ms step:4623/5100 train_loss:3.2650 train_time:825366ms step_avg:178.92ms step:4624/5100 train_loss:3.3020 train_time:825542ms step_avg:178.92ms step:4625/5100 train_loss:3.2078 train_time:825721ms step_avg:178.92ms step:4625/5100 val_loss:3.3077 train_time:825721ms step_avg:178.92ms step:4626/5100 train_loss:3.3854 train_time:825909ms step_avg:178.92ms step:4627/5100 train_loss:3.2681 train_time:826086ms step_avg:178.92ms step:4628/5100 train_loss:3.3289 train_time:826263ms step_avg:178.92ms step:4629/5100 train_loss:3.5223 train_time:826439ms step_avg:178.92ms step:4630/5100 train_loss:3.3688 train_time:826616ms step_avg:178.92ms step:4631/5100 train_loss:3.4498 train_time:826792ms step_avg:178.92ms step:4632/5100 train_loss:3.2621 train_time:826969ms step_avg:178.92ms step:4633/5100 train_loss:3.4471 train_time:827145ms step_avg:178.92ms step:4634/5100 train_loss:3.3065 train_time:827322ms step_avg:178.92ms step:4635/5100 train_loss:3.3589 train_time:827498ms step_avg:178.92ms step:4636/5100 train_loss:3.3756 train_time:827674ms step_avg:178.92ms step:4637/5100 train_loss:3.2045 train_time:827850ms step_avg:178.92ms step:4638/5100 train_loss:3.3748 train_time:828027ms step_avg:178.92ms step:4639/5100 train_loss:3.3266 train_time:828204ms step_avg:178.92ms step:4640/5100 train_loss:3.3303 train_time:828381ms step_avg:178.92ms step:4641/5100 train_loss:3.2730 train_time:828557ms step_avg:178.92ms step:4642/5100 train_loss:3.2977 train_time:828733ms step_avg:178.91ms step:4643/5100 train_loss:3.3193 train_time:828909ms step_avg:178.91ms step:4644/5100 train_loss:3.5348 train_time:829086ms step_avg:178.91ms step:4645/5100 train_loss:3.4036 train_time:829262ms step_avg:178.91ms step:4646/5100 train_loss:3.4410 train_time:829440ms step_avg:178.91ms step:4647/5100 train_loss:3.2645 train_time:829616ms step_avg:178.91ms step:4648/5100 train_loss:3.3730 train_time:829792ms step_avg:178.91ms step:4649/5100 train_loss:3.3261 train_time:829969ms step_avg:178.91ms step:4650/5100 train_loss:3.3689 train_time:830145ms step_avg:178.91ms step:4651/5100 train_loss:3.4927 train_time:830321ms step_avg:178.91ms step:4652/5100 train_loss:3.3182 train_time:830497ms step_avg:178.91ms step:4653/5100 train_loss:3.4234 train_time:830673ms step_avg:178.91ms step:4654/5100 train_loss:3.2754 train_time:830850ms step_avg:178.91ms step:4655/5100 train_loss:3.3171 train_time:831027ms step_avg:178.91ms step:4656/5100 train_loss:3.3412 train_time:831203ms step_avg:178.91ms step:4657/5100 train_loss:3.2860 train_time:831379ms step_avg:178.91ms step:4658/5100 train_loss:3.2183 train_time:831555ms step_avg:178.91ms step:4659/5100 train_loss:3.2554 train_time:831731ms step_avg:178.91ms step:4660/5100 train_loss:3.1830 train_time:831908ms step_avg:178.90ms step:4661/5100 train_loss:3.3576 train_time:832084ms step_avg:178.90ms step:4662/5100 train_loss:3.3425 train_time:832261ms step_avg:178.90ms step:4663/5100 train_loss:3.3028 train_time:832437ms step_avg:178.90ms step:4664/5100 train_loss:3.2293 train_time:832612ms step_avg:178.90ms step:4665/5100 train_loss:3.2364 train_time:832789ms step_avg:178.90ms step:4666/5100 train_loss:3.2817 train_time:832966ms step_avg:178.90ms step:4667/5100 train_loss:3.3822 train_time:833142ms step_avg:178.90ms step:4668/5100 train_loss:3.2947 train_time:833318ms step_avg:178.90ms step:4669/5100 train_loss:3.2690 train_time:833495ms step_avg:178.90ms step:4670/5100 train_loss:3.3337 train_time:833671ms step_avg:178.90ms step:4671/5100 train_loss:3.4177 train_time:833848ms step_avg:178.90ms step:4672/5100 train_loss:3.3129 train_time:834024ms step_avg:178.90ms step:4673/5100 train_loss:3.3609 train_time:834200ms step_avg:178.90ms step:4674/5100 train_loss:3.2958 train_time:834376ms step_avg:178.90ms step:4675/5100 train_loss:3.3281 train_time:834553ms step_avg:178.90ms step:4676/5100 train_loss:3.4134 train_time:834729ms step_avg:178.90ms step:4677/5100 train_loss:3.1398 train_time:834906ms step_avg:178.90ms step:4678/5100 train_loss:3.2009 train_time:835082ms step_avg:178.90ms step:4679/5100 train_loss:3.3246 train_time:835259ms step_avg:178.89ms step:4680/5100 train_loss:3.2911 train_time:835437ms step_avg:178.89ms step:4681/5100 train_loss:3.3145 train_time:835612ms step_avg:178.89ms step:4682/5100 train_loss:3.2999 train_time:835789ms step_avg:178.89ms step:4683/5100 train_loss:3.2332 train_time:835966ms step_avg:178.89ms step:4684/5100 train_loss:3.2104 train_time:836142ms step_avg:178.89ms step:4685/5100 train_loss:3.4612 train_time:836319ms step_avg:178.89ms step:4686/5100 train_loss:3.5376 train_time:836495ms step_avg:178.89ms step:4687/5100 train_loss:3.2340 train_time:836671ms step_avg:178.89ms step:4688/5100 train_loss:3.2377 train_time:836849ms step_avg:178.89ms step:4689/5100 train_loss:3.4314 train_time:837025ms step_avg:178.89ms step:4690/5100 train_loss:3.2564 train_time:837202ms step_avg:178.89ms step:4691/5100 train_loss:3.1173 train_time:837378ms step_avg:178.89ms step:4692/5100 train_loss:3.2190 train_time:837555ms step_avg:178.89ms step:4693/5100 train_loss:3.2298 train_time:837731ms step_avg:178.89ms step:4694/5100 train_loss:3.2546 train_time:837908ms step_avg:178.89ms step:4695/5100 train_loss:3.2642 train_time:838084ms step_avg:178.89ms step:4696/5100 train_loss:3.2787 train_time:838261ms step_avg:178.89ms step:4697/5100 train_loss:3.3451 train_time:838437ms step_avg:178.89ms step:4698/5100 train_loss:3.2534 train_time:838613ms step_avg:178.89ms step:4699/5100 train_loss:3.2834 train_time:838790ms step_avg:178.88ms step:4700/5100 train_loss:3.3779 train_time:838973ms step_avg:178.89ms step:4701/5100 train_loss:3.3054 train_time:839149ms step_avg:178.88ms step:4702/5100 train_loss:3.2940 train_time:839326ms step_avg:178.88ms step:4703/5100 train_loss:3.2474 train_time:839503ms step_avg:178.88ms step:4704/5100 train_loss:3.3292 train_time:839679ms step_avg:178.88ms step:4705/5100 train_loss:3.2839 train_time:839855ms step_avg:178.88ms step:4706/5100 train_loss:3.2137 train_time:840031ms step_avg:178.88ms step:4707/5100 train_loss:3.3401 train_time:840208ms step_avg:178.88ms step:4708/5100 train_loss:3.4348 train_time:840384ms step_avg:178.88ms step:4709/5100 train_loss:3.2397 train_time:840562ms step_avg:178.88ms step:4710/5100 train_loss:3.2200 train_time:840739ms step_avg:178.88ms step:4711/5100 train_loss:3.2445 train_time:840915ms step_avg:178.88ms step:4712/5100 train_loss:3.2620 train_time:841091ms step_avg:178.88ms step:4713/5100 train_loss:3.3781 train_time:841268ms step_avg:178.88ms step:4714/5100 train_loss:3.2250 train_time:841445ms step_avg:178.88ms step:4715/5100 train_loss:3.2935 train_time:841622ms step_avg:178.88ms step:4716/5100 train_loss:3.2222 train_time:841798ms step_avg:178.88ms step:4717/5100 train_loss:3.2925 train_time:841974ms step_avg:178.88ms step:4718/5100 train_loss:3.2233 train_time:842151ms step_avg:178.88ms step:4719/5100 train_loss:3.1837 train_time:842328ms step_avg:178.88ms step:4720/5100 train_loss:3.3517 train_time:842504ms step_avg:178.88ms step:4721/5100 train_loss:3.3413 train_time:842681ms step_avg:178.88ms step:4722/5100 train_loss:3.3415 train_time:842858ms step_avg:178.87ms step:4723/5100 train_loss:3.1863 train_time:843033ms step_avg:178.87ms step:4724/5100 train_loss:3.3753 train_time:843211ms step_avg:178.87ms step:4725/5100 train_loss:3.2557 train_time:843619ms step_avg:178.92ms step:4726/5100 train_loss:3.5330 train_time:843795ms step_avg:178.92ms step:4727/5100 train_loss:3.3845 train_time:843971ms step_avg:178.92ms step:4728/5100 train_loss:3.2570 train_time:844147ms step_avg:178.92ms step:4729/5100 train_loss:3.1890 train_time:844324ms step_avg:178.92ms step:4730/5100 train_loss:3.1456 train_time:844501ms step_avg:178.92ms step:4731/5100 train_loss:3.2389 train_time:844677ms step_avg:178.92ms step:4732/5100 train_loss:3.3163 train_time:844853ms step_avg:178.92ms step:4733/5100 train_loss:3.2152 train_time:845029ms step_avg:178.92ms step:4734/5100 train_loss:3.0926 train_time:845206ms step_avg:178.92ms step:4735/5100 train_loss:3.3874 train_time:845383ms step_avg:178.92ms step:4736/5100 train_loss:3.2609 train_time:845559ms step_avg:178.92ms step:4737/5100 train_loss:3.4352 train_time:845735ms step_avg:178.92ms step:4738/5100 train_loss:3.3515 train_time:845911ms step_avg:178.92ms step:4739/5100 train_loss:3.3054 train_time:846088ms step_avg:178.91ms step:4740/5100 train_loss:3.2726 train_time:846264ms step_avg:178.91ms step:4741/5100 train_loss:3.2945 train_time:846441ms step_avg:178.91ms step:4742/5100 train_loss:3.2830 train_time:846618ms step_avg:178.91ms step:4743/5100 train_loss:3.1665 train_time:846794ms step_avg:178.91ms step:4744/5100 train_loss:3.3011 train_time:846970ms step_avg:178.91ms step:4745/5100 train_loss:3.2635 train_time:847146ms step_avg:178.91ms step:4746/5100 train_loss:3.2673 train_time:847324ms step_avg:178.91ms step:4747/5100 train_loss:3.2375 train_time:847500ms step_avg:178.91ms step:4748/5100 train_loss:3.4275 train_time:847677ms step_avg:178.91ms step:4749/5100 train_loss:3.2721 train_time:847853ms step_avg:178.91ms step:4750/5100 train_loss:3.3754 train_time:848273ms step_avg:178.96ms step:4750/5100 val_loss:3.2960 train_time:848273ms step_avg:178.96ms step:4751/5100 train_loss:3.1911 train_time:848462ms step_avg:178.96ms step:4752/5100 train_loss:3.1111 train_time:848639ms step_avg:178.96ms step:4753/5100 train_loss:3.1946 train_time:848815ms step_avg:178.96ms step:4754/5100 train_loss:3.3971 train_time:848991ms step_avg:178.96ms step:4755/5100 train_loss:3.2770 train_time:849167ms step_avg:178.96ms step:4756/5100 train_loss:3.5196 train_time:849343ms step_avg:178.96ms step:4757/5100 train_loss:3.3828 train_time:849520ms step_avg:178.96ms step:4758/5100 train_loss:3.2917 train_time:849696ms step_avg:178.96ms step:4759/5100 train_loss:3.3461 train_time:849872ms step_avg:178.96ms step:4760/5100 train_loss:3.3284 train_time:850050ms step_avg:178.96ms step:4761/5100 train_loss:3.2670 train_time:850225ms step_avg:178.96ms step:4762/5100 train_loss:3.3037 train_time:850401ms step_avg:178.96ms step:4763/5100 train_loss:3.2825 train_time:850577ms step_avg:178.96ms step:4764/5100 train_loss:3.1391 train_time:850753ms step_avg:178.96ms step:4765/5100 train_loss:3.1572 train_time:850929ms step_avg:178.95ms step:4766/5100 train_loss:3.1497 train_time:851105ms step_avg:178.95ms step:4767/5100 train_loss:3.3884 train_time:851281ms step_avg:178.95ms step:4768/5100 train_loss:3.6520 train_time:851468ms step_avg:178.95ms step:4769/5100 train_loss:3.3652 train_time:851644ms step_avg:178.95ms step:4770/5100 train_loss:3.2696 train_time:851820ms step_avg:178.95ms step:4771/5100 train_loss:3.3336 train_time:851996ms step_avg:178.95ms step:4772/5100 train_loss:3.2871 train_time:852173ms step_avg:178.95ms step:4773/5100 train_loss:3.2684 train_time:852349ms step_avg:178.95ms step:4774/5100 train_loss:3.4547 train_time:852526ms step_avg:178.95ms step:4775/5100 train_loss:3.2708 train_time:852704ms step_avg:178.95ms step:4776/5100 train_loss:3.4079 train_time:852880ms step_avg:178.95ms step:4777/5100 train_loss:3.3179 train_time:853056ms step_avg:178.95ms step:4778/5100 train_loss:3.1575 train_time:853233ms step_avg:178.95ms step:4779/5100 train_loss:3.3371 train_time:853410ms step_avg:178.95ms step:4780/5100 train_loss:3.2666 train_time:853587ms step_avg:178.95ms step:4781/5100 train_loss:3.3384 train_time:853763ms step_avg:178.95ms step:4782/5100 train_loss:3.2594 train_time:853939ms step_avg:178.95ms step:4783/5100 train_loss:3.2057 train_time:854115ms step_avg:178.95ms step:4784/5100 train_loss:3.2609 train_time:854292ms step_avg:178.95ms step:4785/5100 train_loss:3.1943 train_time:854469ms step_avg:178.95ms step:4786/5100 train_loss:3.5220 train_time:854645ms step_avg:178.95ms step:4787/5100 train_loss:3.4040 train_time:854821ms step_avg:178.95ms step:4788/5100 train_loss:3.3273 train_time:854997ms step_avg:178.94ms step:4789/5100 train_loss:3.3148 train_time:855173ms step_avg:178.94ms step:4790/5100 train_loss:3.2324 train_time:855349ms step_avg:178.94ms step:4791/5100 train_loss:3.3359 train_time:855526ms step_avg:178.94ms step:4792/5100 train_loss:3.3466 train_time:855703ms step_avg:178.94ms step:4793/5100 train_loss:3.2655 train_time:855879ms step_avg:178.94ms step:4794/5100 train_loss:3.3352 train_time:856056ms step_avg:178.94ms step:4795/5100 train_loss:3.1746 train_time:856232ms step_avg:178.94ms step:4796/5100 train_loss:3.3317 train_time:856409ms step_avg:178.94ms step:4797/5100 train_loss:3.4041 train_time:856585ms step_avg:178.94ms step:4798/5100 train_loss:3.0625 train_time:856762ms step_avg:178.94ms step:4799/5100 train_loss:3.2462 train_time:856938ms step_avg:178.94ms step:4800/5100 train_loss:3.2215 train_time:857115ms step_avg:178.94ms step:4801/5100 train_loss:3.3277 train_time:857292ms step_avg:178.94ms step:4802/5100 train_loss:3.1499 train_time:857469ms step_avg:178.94ms step:4803/5100 train_loss:3.1798 train_time:857646ms step_avg:178.94ms step:4804/5100 train_loss:3.3846 train_time:857821ms step_avg:178.94ms step:4805/5100 train_loss:3.3286 train_time:857998ms step_avg:178.94ms step:4806/5100 train_loss:3.4008 train_time:858175ms step_avg:178.94ms step:4807/5100 train_loss:3.4115 train_time:858352ms step_avg:178.94ms step:4808/5100 train_loss:3.1829 train_time:858528ms step_avg:178.93ms step:4809/5100 train_loss:3.2929 train_time:858704ms step_avg:178.93ms step:4810/5100 train_loss:3.2424 train_time:858880ms step_avg:178.93ms step:4811/5100 train_loss:3.4771 train_time:859056ms step_avg:178.93ms step:4812/5100 train_loss:3.2734 train_time:859232ms step_avg:178.93ms step:4813/5100 train_loss:3.3117 train_time:859408ms step_avg:178.93ms step:4814/5100 train_loss:3.2078 train_time:859584ms step_avg:178.93ms step:4815/5100 train_loss:3.2556 train_time:859760ms step_avg:178.93ms step:4816/5100 train_loss:3.6901 train_time:859936ms step_avg:178.93ms step:4817/5100 train_loss:3.3663 train_time:860112ms step_avg:178.93ms step:4818/5100 train_loss:3.3051 train_time:860288ms step_avg:178.93ms step:4819/5100 train_loss:3.1560 train_time:860465ms step_avg:178.93ms step:4820/5100 train_loss:3.2817 train_time:860641ms step_avg:178.93ms step:4821/5100 train_loss:3.2919 train_time:860817ms step_avg:178.93ms step:4822/5100 train_loss:3.3446 train_time:860994ms step_avg:178.93ms step:4823/5100 train_loss:3.4002 train_time:861171ms step_avg:178.93ms step:4824/5100 train_loss:3.2800 train_time:861348ms step_avg:178.93ms step:4825/5100 train_loss:3.2609 train_time:861523ms step_avg:178.92ms step:4826/5100 train_loss:3.1880 train_time:861699ms step_avg:178.92ms step:4827/5100 train_loss:3.1471 train_time:861876ms step_avg:178.92ms step:4828/5100 train_loss:3.3346 train_time:862053ms step_avg:178.92ms step:4829/5100 train_loss:3.2113 train_time:862229ms step_avg:178.92ms step:4830/5100 train_loss:3.3400 train_time:862406ms step_avg:178.92ms step:4831/5100 train_loss:3.5006 train_time:862582ms step_avg:178.92ms step:4832/5100 train_loss:3.2365 train_time:862757ms step_avg:178.92ms step:4833/5100 train_loss:3.3191 train_time:862934ms step_avg:178.92ms step:4834/5100 train_loss:3.2808 train_time:863111ms step_avg:178.92ms step:4835/5100 train_loss:3.4596 train_time:863288ms step_avg:178.92ms step:4836/5100 train_loss:3.2799 train_time:863464ms step_avg:178.92ms step:4837/5100 train_loss:3.5404 train_time:863640ms step_avg:178.92ms step:4838/5100 train_loss:3.4890 train_time:863817ms step_avg:178.92ms step:4839/5100 train_loss:3.3134 train_time:863993ms step_avg:178.92ms step:4840/5100 train_loss:3.3109 train_time:864170ms step_avg:178.92ms step:4841/5100 train_loss:3.2982 train_time:864347ms step_avg:178.92ms step:4842/5100 train_loss:3.3371 train_time:864524ms step_avg:178.92ms step:4843/5100 train_loss:3.3332 train_time:864699ms step_avg:178.92ms step:4844/5100 train_loss:3.1903 train_time:864876ms step_avg:178.92ms step:4845/5100 train_loss:3.2271 train_time:865052ms step_avg:178.91ms step:4846/5100 train_loss:3.1982 train_time:865228ms step_avg:178.91ms step:4847/5100 train_loss:3.3597 train_time:865405ms step_avg:178.91ms step:4848/5100 train_loss:3.2281 train_time:865581ms step_avg:178.91ms step:4849/5100 train_loss:3.2609 train_time:865757ms step_avg:178.91ms step:4850/5100 train_loss:3.3956 train_time:865933ms step_avg:178.91ms step:4851/5100 train_loss:3.2784 train_time:866109ms step_avg:178.91ms step:4852/5100 train_loss:3.0887 train_time:866286ms step_avg:178.91ms step:4853/5100 train_loss:3.1795 train_time:866464ms step_avg:178.91ms step:4854/5100 train_loss:3.3180 train_time:866640ms step_avg:178.91ms step:4855/5100 train_loss:3.2755 train_time:866818ms step_avg:178.91ms step:4856/5100 train_loss:3.4083 train_time:866995ms step_avg:178.91ms step:4857/5100 train_loss:3.2489 train_time:867172ms step_avg:178.91ms step:4858/5100 train_loss:3.2860 train_time:867349ms step_avg:178.91ms step:4859/5100 train_loss:3.2315 train_time:867526ms step_avg:178.91ms step:4860/5100 train_loss:3.3638 train_time:867704ms step_avg:178.91ms step:4861/5100 train_loss:3.2282 train_time:867880ms step_avg:178.91ms step:4862/5100 train_loss:3.2861 train_time:868057ms step_avg:178.91ms step:4863/5100 train_loss:3.2932 train_time:868233ms step_avg:178.91ms step:4864/5100 train_loss:3.2523 train_time:868410ms step_avg:178.91ms step:4865/5100 train_loss:3.3326 train_time:868585ms step_avg:178.91ms step:4866/5100 train_loss:2.9808 train_time:868761ms step_avg:178.90ms step:4867/5100 train_loss:3.1964 train_time:868937ms step_avg:178.90ms step:4868/5100 train_loss:3.2515 train_time:869113ms step_avg:178.90ms step:4869/5100 train_loss:3.2735 train_time:869290ms step_avg:178.90ms step:4870/5100 train_loss:3.2800 train_time:869467ms step_avg:178.90ms step:4871/5100 train_loss:3.2850 train_time:869643ms step_avg:178.90ms step:4872/5100 train_loss:3.3958 train_time:869819ms step_avg:178.90ms step:4873/5100 train_loss:3.3834 train_time:869994ms step_avg:178.90ms step:4874/5100 train_loss:3.4110 train_time:870172ms step_avg:178.90ms step:4875/5100 train_loss:3.4896 train_time:870348ms step_avg:178.90ms step:4875/5100 val_loss:3.2857 train_time:870349ms step_avg:178.90ms step:4876/5100 train_loss:3.3060 train_time:870533ms step_avg:178.90ms step:4877/5100 train_loss:3.2161 train_time:870711ms step_avg:178.90ms step:4878/5100 train_loss:3.1843 train_time:870888ms step_avg:178.90ms step:4879/5100 train_loss:3.2351 train_time:871066ms step_avg:178.90ms step:4880/5100 train_loss:3.3678 train_time:871243ms step_avg:178.90ms step:4881/5100 train_loss:3.2121 train_time:871419ms step_avg:178.90ms step:4882/5100 train_loss:3.3367 train_time:871596ms step_avg:178.90ms step:4883/5100 train_loss:3.3471 train_time:871772ms step_avg:178.90ms step:4884/5100 train_loss:3.2576 train_time:871950ms step_avg:178.90ms step:4885/5100 train_loss:3.2574 train_time:872128ms step_avg:178.90ms step:4886/5100 train_loss:3.3715 train_time:872304ms step_avg:178.90ms step:4887/5100 train_loss:3.3830 train_time:872480ms step_avg:178.90ms step:4888/5100 train_loss:3.2613 train_time:872656ms step_avg:178.90ms step:4889/5100 train_loss:3.2303 train_time:872832ms step_avg:178.90ms step:4890/5100 train_loss:3.3191 train_time:873009ms step_avg:178.90ms step:4891/5100 train_loss:3.2239 train_time:873185ms step_avg:178.89ms step:4892/5100 train_loss:3.3188 train_time:873362ms step_avg:178.89ms step:4893/5100 train_loss:3.3143 train_time:873539ms step_avg:178.89ms step:4894/5100 train_loss:3.3427 train_time:873715ms step_avg:178.89ms step:4895/5100 train_loss:3.4076 train_time:873892ms step_avg:178.89ms step:4896/5100 train_loss:3.3070 train_time:874069ms step_avg:178.89ms step:4897/5100 train_loss:3.2599 train_time:874245ms step_avg:178.89ms step:4898/5100 train_loss:3.4519 train_time:874421ms step_avg:178.89ms step:4899/5100 train_loss:3.2149 train_time:874597ms step_avg:178.89ms step:4900/5100 train_loss:3.2706 train_time:874772ms step_avg:178.89ms step:4901/5100 train_loss:3.1795 train_time:874949ms step_avg:178.89ms step:4902/5100 train_loss:3.1480 train_time:875125ms step_avg:178.89ms step:4903/5100 train_loss:3.2817 train_time:875301ms step_avg:178.89ms step:4904/5100 train_loss:3.2319 train_time:875479ms step_avg:178.89ms step:4905/5100 train_loss:3.3038 train_time:875655ms step_avg:178.89ms step:4906/5100 train_loss:3.3433 train_time:875832ms step_avg:178.89ms step:4907/5100 train_loss:3.2190 train_time:876008ms step_avg:178.89ms step:4908/5100 train_loss:3.3135 train_time:876185ms step_avg:178.89ms step:4909/5100 train_loss:3.2099 train_time:876361ms step_avg:178.89ms step:4910/5100 train_loss:3.3210 train_time:876537ms step_avg:178.89ms step:4911/5100 train_loss:3.3636 train_time:876713ms step_avg:178.88ms step:4912/5100 train_loss:3.2766 train_time:876890ms step_avg:178.88ms step:4913/5100 train_loss:3.2298 train_time:877067ms step_avg:178.88ms step:4914/5100 train_loss:3.2356 train_time:877475ms step_avg:178.93ms step:4915/5100 train_loss:3.1584 train_time:877651ms step_avg:178.93ms step:4916/5100 train_loss:3.3544 train_time:877828ms step_avg:178.93ms step:4917/5100 train_loss:3.3301 train_time:878003ms step_avg:178.93ms step:4918/5100 train_loss:3.2572 train_time:878179ms step_avg:178.93ms step:4919/5100 train_loss:3.2667 train_time:878355ms step_avg:178.93ms step:4920/5100 train_loss:3.2746 train_time:878532ms step_avg:178.93ms step:4921/5100 train_loss:3.3513 train_time:878709ms step_avg:178.93ms step:4922/5100 train_loss:3.5177 train_time:878885ms step_avg:178.93ms step:4923/5100 train_loss:3.3672 train_time:879061ms step_avg:178.93ms step:4924/5100 train_loss:3.2344 train_time:879237ms step_avg:178.92ms step:4925/5100 train_loss:3.5416 train_time:879413ms step_avg:178.92ms step:4926/5100 train_loss:3.2817 train_time:879590ms step_avg:178.92ms step:4927/5100 train_loss:3.2607 train_time:879767ms step_avg:178.92ms step:4928/5100 train_loss:3.1832 train_time:879944ms step_avg:178.92ms step:4929/5100 train_loss:3.1906 train_time:880120ms step_avg:178.92ms step:4930/5100 train_loss:3.3395 train_time:880298ms step_avg:178.92ms step:4931/5100 train_loss:3.5824 train_time:880473ms step_avg:178.92ms step:4932/5100 train_loss:3.1887 train_time:880650ms step_avg:178.92ms step:4933/5100 train_loss:3.2912 train_time:880826ms step_avg:178.92ms step:4934/5100 train_loss:3.3526 train_time:881002ms step_avg:178.92ms step:4935/5100 train_loss:3.1437 train_time:881179ms step_avg:178.92ms step:4936/5100 train_loss:3.2934 train_time:881355ms step_avg:178.92ms step:4937/5100 train_loss:3.3484 train_time:881532ms step_avg:178.92ms step:4938/5100 train_loss:3.3312 train_time:881708ms step_avg:178.92ms step:4939/5100 train_loss:3.3336 train_time:881885ms step_avg:178.92ms step:4940/5100 train_loss:3.4448 train_time:882304ms step_avg:178.97ms step:4941/5100 train_loss:3.2867 train_time:882480ms step_avg:178.97ms step:4942/5100 train_loss:3.3035 train_time:882657ms step_avg:178.97ms step:4943/5100 train_loss:3.0410 train_time:882833ms step_avg:178.96ms step:4944/5100 train_loss:3.5316 train_time:883009ms step_avg:178.96ms step:4945/5100 train_loss:3.4968 train_time:883185ms step_avg:178.96ms step:4946/5100 train_loss:3.1140 train_time:883362ms step_avg:178.96ms step:4947/5100 train_loss:3.3672 train_time:883538ms step_avg:178.96ms step:4948/5100 train_loss:3.3962 train_time:883715ms step_avg:178.96ms step:4949/5100 train_loss:3.2330 train_time:883893ms step_avg:178.96ms step:4950/5100 train_loss:3.3550 train_time:884068ms step_avg:178.96ms step:4951/5100 train_loss:3.2178 train_time:884244ms step_avg:178.96ms step:4952/5100 train_loss:3.3425 train_time:884421ms step_avg:178.96ms step:4953/5100 train_loss:3.3035 train_time:884599ms step_avg:178.96ms step:4954/5100 train_loss:3.1911 train_time:884774ms step_avg:178.96ms step:4955/5100 train_loss:3.3346 train_time:884950ms step_avg:178.96ms step:4956/5100 train_loss:3.1592 train_time:885128ms step_avg:178.96ms step:4957/5100 train_loss:3.2567 train_time:885304ms step_avg:178.96ms step:4958/5100 train_loss:3.2358 train_time:885480ms step_avg:178.96ms step:4959/5100 train_loss:3.2534 train_time:885656ms step_avg:178.96ms step:4960/5100 train_loss:3.2895 train_time:885832ms step_avg:178.96ms step:4961/5100 train_loss:3.4398 train_time:886009ms step_avg:178.96ms step:4962/5100 train_loss:3.1990 train_time:886185ms step_avg:178.95ms step:4963/5100 train_loss:3.3363 train_time:886361ms step_avg:178.95ms step:4964/5100 train_loss:3.1756 train_time:886536ms step_avg:178.95ms step:4965/5100 train_loss:3.9021 train_time:886713ms step_avg:178.95ms step:4966/5100 train_loss:3.1709 train_time:886890ms step_avg:178.95ms step:4967/5100 train_loss:3.3125 train_time:887067ms step_avg:178.95ms step:4968/5100 train_loss:3.1307 train_time:887244ms step_avg:178.95ms step:4969/5100 train_loss:3.8425 train_time:887420ms step_avg:178.95ms step:4970/5100 train_loss:3.3720 train_time:887596ms step_avg:178.95ms step:4971/5100 train_loss:3.2921 train_time:887772ms step_avg:178.95ms step:4972/5100 train_loss:3.2383 train_time:887949ms step_avg:178.95ms step:4973/5100 train_loss:3.3218 train_time:888126ms step_avg:178.95ms step:4974/5100 train_loss:3.1955 train_time:888303ms step_avg:178.95ms step:4975/5100 train_loss:3.1928 train_time:888481ms step_avg:178.95ms step:4976/5100 train_loss:3.3405 train_time:888656ms step_avg:178.95ms step:4977/5100 train_loss:3.2734 train_time:888835ms step_avg:178.95ms step:4978/5100 train_loss:3.2240 train_time:889010ms step_avg:178.95ms step:4979/5100 train_loss:3.2850 train_time:889187ms step_avg:178.95ms step:4980/5100 train_loss:3.2181 train_time:889363ms step_avg:178.95ms step:4981/5100 train_loss:3.3599 train_time:889539ms step_avg:178.95ms step:4982/5100 train_loss:3.3312 train_time:889715ms step_avg:178.95ms step:4983/5100 train_loss:3.1358 train_time:889892ms step_avg:178.94ms step:4984/5100 train_loss:3.1619 train_time:890069ms step_avg:178.94ms step:4985/5100 train_loss:3.4481 train_time:890246ms step_avg:178.94ms step:4986/5100 train_loss:3.3304 train_time:890423ms step_avg:178.94ms step:4987/5100 train_loss:3.2494 train_time:890600ms step_avg:178.94ms step:4988/5100 train_loss:3.2821 train_time:890776ms step_avg:178.94ms step:4989/5100 train_loss:3.2667 train_time:890952ms step_avg:178.94ms step:4990/5100 train_loss:3.2460 train_time:891129ms step_avg:178.94ms step:4991/5100 train_loss:3.2974 train_time:891304ms step_avg:178.94ms step:4992/5100 train_loss:3.3405 train_time:891481ms step_avg:178.94ms step:4993/5100 train_loss:3.1636 train_time:891657ms step_avg:178.94ms step:4994/5100 train_loss:3.2937 train_time:891833ms step_avg:178.94ms step:4995/5100 train_loss:3.2152 train_time:892010ms step_avg:178.94ms step:4996/5100 train_loss:3.3653 train_time:892186ms step_avg:178.94ms step:4997/5100 train_loss:3.2369 train_time:892363ms step_avg:178.94ms step:4998/5100 train_loss:3.4063 train_time:892540ms step_avg:178.94ms step:4999/5100 train_loss:3.2632 train_time:892717ms step_avg:178.94ms step:5000/5100 train_loss:3.3880 train_time:892893ms step_avg:178.94ms step:5000/5100 val_loss:3.2776 train_time:892894ms step_avg:178.94ms step:5001/5100 train_loss:3.3106 train_time:893080ms step_avg:178.94ms step:5002/5100 train_loss:3.3108 train_time:893256ms step_avg:178.94ms step:5003/5100 train_loss:3.2026 train_time:893432ms step_avg:178.94ms step:5004/5100 train_loss:3.2753 train_time:893609ms step_avg:178.94ms step:5005/5100 train_loss:3.2867 train_time:893786ms step_avg:178.94ms step:5006/5100 train_loss:3.1625 train_time:893962ms step_avg:178.94ms step:5007/5100 train_loss:3.3899 train_time:894139ms step_avg:178.94ms step:5008/5100 train_loss:3.2428 train_time:894316ms step_avg:178.93ms step:5009/5100 train_loss:3.2694 train_time:894493ms step_avg:178.93ms step:5010/5100 train_loss:3.2368 train_time:894670ms step_avg:178.93ms step:5011/5100 train_loss:3.4429 train_time:894846ms step_avg:178.93ms step:5012/5100 train_loss:3.2462 train_time:895022ms step_avg:178.93ms step:5013/5100 train_loss:3.2288 train_time:895198ms step_avg:178.93ms step:5014/5100 train_loss:3.1826 train_time:895374ms step_avg:178.93ms step:5015/5100 train_loss:3.2906 train_time:895551ms step_avg:178.93ms step:5016/5100 train_loss:3.2679 train_time:895727ms step_avg:178.93ms step:5017/5100 train_loss:3.3082 train_time:895903ms step_avg:178.93ms step:5018/5100 train_loss:3.3264 train_time:896080ms step_avg:178.93ms step:5019/5100 train_loss:3.2831 train_time:896259ms step_avg:178.93ms step:5020/5100 train_loss:3.7962 train_time:896435ms step_avg:178.93ms step:5021/5100 train_loss:3.2377 train_time:896612ms step_avg:178.93ms step:5022/5100 train_loss:3.3340 train_time:896788ms step_avg:178.93ms step:5023/5100 train_loss:3.2624 train_time:896963ms step_avg:178.93ms step:5024/5100 train_loss:3.4036 train_time:897139ms step_avg:178.93ms step:5025/5100 train_loss:3.1962 train_time:897317ms step_avg:178.93ms step:5026/5100 train_loss:3.3593 train_time:897492ms step_avg:178.93ms step:5027/5100 train_loss:3.2123 train_time:897669ms step_avg:178.93ms step:5028/5100 train_loss:3.4234 train_time:897845ms step_avg:178.92ms step:5029/5100 train_loss:3.3229 train_time:898021ms step_avg:178.92ms step:5030/5100 train_loss:3.3554 train_time:898198ms step_avg:178.92ms step:5031/5100 train_loss:3.2103 train_time:898375ms step_avg:178.92ms step:5032/5100 train_loss:3.2551 train_time:898551ms step_avg:178.92ms step:5033/5100 train_loss:3.1932 train_time:898728ms step_avg:178.92ms step:5034/5100 train_loss:3.3872 train_time:898904ms step_avg:178.92ms step:5035/5100 train_loss:3.3779 train_time:899080ms step_avg:178.92ms step:5036/5100 train_loss:3.2378 train_time:899257ms step_avg:178.92ms step:5037/5100 train_loss:3.1610 train_time:899434ms step_avg:178.92ms step:5038/5100 train_loss:3.1999 train_time:899610ms step_avg:178.92ms step:5039/5100 train_loss:3.3295 train_time:899786ms step_avg:178.92ms step:5040/5100 train_loss:3.2590 train_time:899962ms step_avg:178.92ms step:5041/5100 train_loss:3.4360 train_time:900138ms step_avg:178.92ms step:5042/5100 train_loss:3.2262 train_time:900314ms step_avg:178.92ms step:5043/5100 train_loss:3.4122 train_time:900491ms step_avg:178.92ms step:5044/5100 train_loss:3.3233 train_time:900667ms step_avg:178.92ms step:5045/5100 train_loss:3.3841 train_time:900842ms step_avg:178.92ms step:5046/5100 train_loss:3.2091 train_time:901019ms step_avg:178.92ms step:5047/5100 train_loss:3.3491 train_time:901196ms step_avg:178.92ms step:5048/5100 train_loss:3.1006 train_time:901373ms step_avg:178.91ms step:5049/5100 train_loss:3.2642 train_time:901549ms step_avg:178.91ms step:5050/5100 train_loss:3.2667 train_time:901725ms step_avg:178.91ms step:5051/5100 train_loss:3.2116 train_time:901901ms step_avg:178.91ms step:5052/5100 train_loss:3.2473 train_time:902078ms step_avg:178.91ms step:5053/5100 train_loss:3.2921 train_time:902254ms step_avg:178.91ms step:5054/5100 train_loss:3.3338 train_time:902431ms step_avg:178.91ms step:5055/5100 train_loss:3.4050 train_time:902606ms step_avg:178.91ms step:5056/5100 train_loss:3.3543 train_time:902783ms step_avg:178.91ms step:5057/5100 train_loss:3.2368 train_time:902959ms step_avg:178.91ms step:5058/5100 train_loss:3.1402 train_time:903136ms step_avg:178.91ms step:5059/5100 train_loss:3.0586 train_time:903313ms step_avg:178.91ms step:5060/5100 train_loss:3.2752 train_time:903490ms step_avg:178.91ms step:5061/5100 train_loss:3.3575 train_time:903667ms step_avg:178.91ms step:5062/5100 train_loss:3.2861 train_time:903842ms step_avg:178.91ms step:5063/5100 train_loss:3.4098 train_time:904018ms step_avg:178.91ms step:5064/5100 train_loss:3.4153 train_time:904195ms step_avg:178.91ms step:5065/5100 train_loss:3.2649 train_time:904371ms step_avg:178.91ms step:5066/5100 train_loss:3.3762 train_time:904547ms step_avg:178.91ms step:5067/5100 train_loss:3.5709 train_time:904723ms step_avg:178.90ms step:5068/5100 train_loss:3.2275 train_time:904900ms step_avg:178.90ms step:5069/5100 train_loss:3.5736 train_time:905076ms step_avg:178.90ms step:5070/5100 train_loss:3.2782 train_time:905253ms step_avg:178.90ms step:5071/5100 train_loss:3.7045 train_time:905429ms step_avg:178.90ms step:5072/5100 train_loss:3.2251 train_time:905604ms step_avg:178.90ms step:5073/5100 train_loss:3.2856 train_time:905781ms step_avg:178.90ms step:5074/5100 train_loss:3.4247 train_time:905958ms step_avg:178.90ms step:5075/5100 train_loss:3.2599 train_time:906136ms step_avg:178.90ms step:5076/5100 train_loss:3.2490 train_time:906311ms step_avg:178.90ms step:5077/5100 train_loss:3.2270 train_time:906487ms step_avg:178.90ms step:5078/5100 train_loss:3.3063 train_time:906663ms step_avg:178.90ms step:5079/5100 train_loss:3.4387 train_time:906839ms step_avg:178.90ms step:5080/5100 train_loss:3.4087 train_time:907016ms step_avg:178.90ms step:5081/5100 train_loss:3.2316 train_time:907193ms step_avg:178.90ms step:5082/5100 train_loss:3.3596 train_time:907369ms step_avg:178.90ms step:5083/5100 train_loss:3.2189 train_time:907545ms step_avg:178.90ms step:5084/5100 train_loss:3.3031 train_time:907721ms step_avg:178.90ms step:5085/5100 train_loss:3.1912 train_time:907898ms step_avg:178.90ms step:5086/5100 train_loss:4.0435 train_time:908075ms step_avg:178.90ms step:5087/5100 train_loss:3.3364 train_time:908251ms step_avg:178.90ms step:5088/5100 train_loss:3.2485 train_time:908427ms step_avg:178.89ms step:5089/5100 train_loss:3.2520 train_time:908603ms step_avg:178.89ms step:5090/5100 train_loss:3.3942 train_time:908780ms step_avg:178.89ms step:5091/5100 train_loss:3.3124 train_time:908957ms step_avg:178.89ms step:5092/5100 train_loss:3.2092 train_time:909134ms step_avg:178.89ms step:5093/5100 train_loss:3.2327 train_time:909311ms step_avg:178.89ms step:5094/5100 train_loss:3.2298 train_time:909487ms step_avg:178.89ms step:5095/5100 train_loss:3.1553 train_time:909664ms step_avg:178.89ms step:5096/5100 train_loss:3.2771 train_time:909840ms step_avg:178.89ms step:5097/5100 train_loss:3.0641 train_time:910017ms step_avg:178.89ms step:5098/5100 train_loss:3.3601 train_time:910193ms step_avg:178.89ms step:5099/5100 train_loss:3.2174 train_time:910369ms step_avg:178.89ms step:5100/5100 train_loss:3.2686 train_time:910545ms step_avg:178.89ms step:5100/5100 val_loss:3.2741 train_time:910546ms step_avg:178.89ms