stderr.log

+ deepspeed --master_port 19346 --module safe_rlhf.finetune --train_datasets inverse-json::/home/hansirui_1st/jiayi/resist/setting3/safety_data/training/safe/safe_30k.json --model_name_or_path /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T --max_length 2048 --trust_remote_code True --epochs 1 --per_device_train_batch_size 4 --per_device_eval_batch_size 4 --gradient_accumulation_steps 8 --gradient_checkpointing --learning_rate 1e-5 --lr_warmup_ratio 0 --weight_decay 0.0 --lr_scheduler_type constant --weight_decay 0.0 --seed 42 --output_dir /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-2T/tinyllama-2T-s3-Q1-30k --log_type wandb --log_run_name tinyllama-2T-s3-Q1-30k --log_project Inverse_Alignment --zero_stage 3 --offload none --bf16 True --tf32 True --save_16bit
[rank4]:[W529 17:26:48.462090986 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 4]  using GPU 4 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank3]:[W529 17:26:49.301168376 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 3]  using GPU 3 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank1]:[W529 17:26:49.611287717 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 1]  using GPU 1 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank5]:[W529 17:26:49.722383317 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 5]  using GPU 5 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank2]:[W529 17:26:49.736489114 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 2]  using GPU 2 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank6]:[W529 17:26:49.744524280 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 6]  using GPU 6 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank0]:[W529 17:26:49.745616340 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 0]  using GPU 0 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank7]:[W529 17:26:49.764576913 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 7]  using GPU 7 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/config.json
Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/model.safetensors
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

All model checkpoint weights were used when initializing LlamaForCausalLM.

All model checkpoint weights were used when initializing LlamaForCausalLM.

All model checkpoint weights were used when initializing LlamaForCausalLM.

All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

loading file tokenizer.model
loading file tokenizer.model
loading file tokenizer.model
loading file tokenizer.json
loading file tokenizer.json
loading file added_tokens.json
loading file tokenizer.json
loading file special_tokens_map.json
loading file added_tokens.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file special_tokens_map.json
loading file chat_template.jinja
loading file tokenizer_config.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file chat_template.jinja
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...

Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Detected CUDA files, patching ldflags
Emitting ninja build file /home/hansirui_1st/.cache/torch_extensions/py311_cu124/fused_adam/build.ninja...
/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/torch/utils/cpp_extension.py:2059: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation. 
If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST'].
  warnings.warn(
Building extension module fused_adam...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
wandb: Using wandb-core as the SDK backend.  Please refer to https://wandb.me/wandb-core for more information.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
wandb: Currently logged in as: xtom to https://api.wandb.ai. Use `wandb login --relogin` to force relogin
wandb: Tracking run with wandb version 0.19.8
wandb: Run data is saved locally in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-2T/tinyllama-2T-s3-Q1-30k/wandb/run-20250529_172723-8imc1sym
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run tinyllama-2T-s3-Q1-30k
wandb: ⭐️ View project at https://wandb.ai/xtom/Inverse_Alignment
wandb: 🚀 View run at https://wandb.ai/xtom/Inverse_Alignment/runs/8imc1sym

Training 1/1 epoch:   0%|          | 0/938 [00:00<?, ?it/s]`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.

Training 1/1 epoch (loss 2.0886):   0%|          | 0/938 [00:05<?, ?it/s]
Training 1/1 epoch (loss 2.0886):   0%|          | 1/938 [00:05<1:23:40,  5.36s/it]
Training 1/1 epoch (loss 2.0124):   0%|          | 1/938 [00:06<1:23:40,  5.36s/it]
Training 1/1 epoch (loss 2.0124):   0%|          | 2/938 [00:06<47:47,  3.06s/it]  
Training 1/1 epoch (loss 2.0405):   0%|          | 2/938 [00:07<47:47,  3.06s/it]
Training 1/1 epoch (loss 2.0405):   0%|          | 3/938 [00:07<28:19,  1.82s/it]
Training 1/1 epoch (loss 2.0534):   0%|          | 3/938 [00:07<28:19,  1.82s/it]
Training 1/1 epoch (loss 2.0534):   0%|          | 4/938 [00:07<19:01,  1.22s/it]
Training 1/1 epoch (loss 1.9724):   0%|          | 4/938 [00:07<19:01,  1.22s/it]
Training 1/1 epoch (loss 1.9724):   1%|          | 5/938 [00:07<13:51,  1.12it/s]
Training 1/1 epoch (loss 1.9856):   1%|          | 5/938 [00:08<13:51,  1.12it/s]
Training 1/1 epoch (loss 1.9856):   1%|          | 6/938 [00:08<10:54,  1.42it/s]
Training 1/1 epoch (loss 2.0163):   1%|          | 6/938 [00:08<10:54,  1.42it/s]
Training 1/1 epoch (loss 2.0163):   1%|          | 7/938 [00:08<08:50,  1.76it/s]
Training 1/1 epoch (loss 2.0723):   1%|          | 7/938 [00:08<08:50,  1.76it/s]
Training 1/1 epoch (loss 2.0723):   1%|          | 8/938 [00:08<07:59,  1.94it/s]
Training 1/1 epoch (loss 1.9441):   1%|          | 8/938 [00:09<07:59,  1.94it/s]
Training 1/1 epoch (loss 1.9441):   1%|          | 9/938 [00:09<06:56,  2.23it/s]
Training 1/1 epoch (loss 1.9509):   1%|          | 9/938 [00:09<06:56,  2.23it/s]
Training 1/1 epoch (loss 1.9509):   1%|          | 10/938 [00:09<06:17,  2.46it/s]
Training 1/1 epoch (loss 2.0250):   1%|          | 10/938 [00:09<06:17,  2.46it/s]
Training 1/1 epoch (loss 2.0250):   1%|          | 11/938 [00:09<06:08,  2.52it/s]
Training 1/1 epoch (loss 2.0769):   1%|          | 11/938 [00:10<06:08,  2.52it/s]
Training 1/1 epoch (loss 2.0769):   1%|▏         | 12/938 [00:10<05:50,  2.64it/s]
Training 1/1 epoch (loss 2.1021):   1%|▏         | 12/938 [00:10<05:50,  2.64it/s]
Training 1/1 epoch (loss 2.1021):   1%|▏         | 13/938 [00:10<05:46,  2.67it/s]
Training 1/1 epoch (loss 1.8803):   1%|▏         | 13/938 [00:10<05:46,  2.67it/s]
Training 1/1 epoch (loss 1.8803):   1%|▏         | 14/938 [00:10<05:27,  2.82it/s]
Training 1/1 epoch (loss 2.1142):   1%|▏         | 14/938 [00:11<05:27,  2.82it/s]
Training 1/1 epoch (loss 2.1142):   2%|▏         | 15/938 [00:11<05:11,  2.96it/s]
Training 1/1 epoch (loss 2.0380):   2%|▏         | 15/938 [00:11<05:11,  2.96it/s]
Training 1/1 epoch (loss 2.0380):   2%|▏         | 16/938 [00:11<05:06,  3.01it/s]
Training 1/1 epoch (loss 1.8542):   2%|▏         | 16/938 [00:11<05:06,  3.01it/s]
Training 1/1 epoch (loss 1.8542):   2%|▏         | 17/938 [00:11<05:03,  3.03it/s]
Training 1/1 epoch (loss 2.0572):   2%|▏         | 17/938 [00:12<05:03,  3.03it/s]
Training 1/1 epoch (loss 2.0572):   2%|▏         | 18/938 [00:12<05:12,  2.94it/s]
Training 1/1 epoch (loss 1.8676):   2%|▏         | 18/938 [00:12<05:12,  2.94it/s]
Training 1/1 epoch (loss 1.8676):   2%|▏         | 19/938 [00:12<05:06,  3.00it/s]
Training 1/1 epoch (loss 1.9604):   2%|▏         | 19/938 [00:12<05:06,  3.00it/s]
Training 1/1 epoch (loss 1.9604):   2%|▏         | 20/938 [00:12<04:58,  3.07it/s]
Training 1/1 epoch (loss 1.9735):   2%|▏         | 20/938 [00:13<04:58,  3.07it/s]
Training 1/1 epoch (loss 1.9735):   2%|▏         | 21/938 [00:13<04:58,  3.07it/s]
Training 1/1 epoch (loss 2.1057):   2%|▏         | 21/938 [00:13<04:58,  3.07it/s]
Training 1/1 epoch (loss 2.1057):   2%|▏         | 22/938 [00:13<04:54,  3.12it/s]
Training 1/1 epoch (loss 1.9158):   2%|▏         | 22/938 [00:13<04:54,  3.12it/s]
Training 1/1 epoch (loss 1.9158):   2%|▏         | 23/938 [00:13<04:52,  3.13it/s]
Training 1/1 epoch (loss 1.9458):   2%|▏         | 23/938 [00:14<04:52,  3.13it/s]
Training 1/1 epoch (loss 1.9458):   3%|▎         | 24/938 [00:14<05:05,  2.99it/s]
Training 1/1 epoch (loss 2.0418):   3%|▎         | 24/938 [00:14<05:05,  2.99it/s]
Training 1/1 epoch (loss 2.0418):   3%|▎         | 25/938 [00:14<04:58,  3.06it/s]
Training 1/1 epoch (loss 2.0315):   3%|▎         | 25/938 [00:14<04:58,  3.06it/s]
Training 1/1 epoch (loss 2.0315):   3%|▎         | 26/938 [00:14<04:54,  3.10it/s]
Training 1/1 epoch (loss 1.9305):   3%|▎         | 26/938 [00:15<04:54,  3.10it/s]
Training 1/1 epoch (loss 1.9305):   3%|▎         | 27/938 [00:15<04:58,  3.06it/s]
Training 1/1 epoch (loss 1.8989):   3%|▎         | 27/938 [00:15<04:58,  3.06it/s]
Training 1/1 epoch (loss 1.8989):   3%|▎         | 28/938 [00:15<04:51,  3.12it/s]
Training 1/1 epoch (loss 2.0149):   3%|▎         | 28/938 [00:15<04:51,  3.12it/s]
Training 1/1 epoch (loss 2.0149):   3%|▎         | 29/938 [00:15<04:54,  3.09it/s]
Training 1/1 epoch (loss 2.0278):   3%|▎         | 29/938 [00:16<04:54,  3.09it/s]
Training 1/1 epoch (loss 2.0278):   3%|▎         | 30/938 [00:16<05:03,  2.99it/s]
Training 1/1 epoch (loss 1.9639):   3%|▎         | 30/938 [00:16<05:03,  2.99it/s]
Training 1/1 epoch (loss 1.9639):   3%|▎         | 31/938 [00:16<04:54,  3.08it/s]
Training 1/1 epoch (loss 1.9956):   3%|▎         | 31/938 [00:16<04:54,  3.08it/s]
Training 1/1 epoch (loss 1.9956):   3%|▎         | 32/938 [00:16<04:55,  3.07it/s]
Training 1/1 epoch (loss 1.8724):   3%|▎         | 32/938 [00:16<04:55,  3.07it/s]
Training 1/1 epoch (loss 1.8724):   4%|▎         | 33/938 [00:16<04:50,  3.11it/s]
Training 1/1 epoch (loss 1.9103):   4%|▎         | 33/938 [00:17<04:50,  3.11it/s]
Training 1/1 epoch (loss 1.9103):   4%|▎         | 34/938 [00:17<04:46,  3.15it/s]
Training 1/1 epoch (loss 1.8523):   4%|▎         | 34/938 [00:17<04:46,  3.15it/s]
Training 1/1 epoch (loss 1.8523):   4%|▎         | 35/938 [00:17<04:43,  3.18it/s]
Training 1/1 epoch (loss 1.8735):   4%|▎         | 35/938 [00:17<04:43,  3.18it/s]
Training 1/1 epoch (loss 1.8735):   4%|▍         | 36/938 [00:17<04:59,  3.01it/s]
Training 1/1 epoch (loss 1.9303):   4%|▍         | 36/938 [00:18<04:59,  3.01it/s]
Training 1/1 epoch (loss 1.9303):   4%|▍         | 37/938 [00:18<04:49,  3.11it/s]
Training 1/1 epoch (loss 1.8985):   4%|▍         | 37/938 [00:18<04:49,  3.11it/s]
Training 1/1 epoch (loss 1.8985):   4%|▍         | 38/938 [00:18<04:44,  3.16it/s]
Training 1/1 epoch (loss 1.8596):   4%|▍         | 38/938 [00:18<04:44,  3.16it/s]
Training 1/1 epoch (loss 1.8596):   4%|▍         | 39/938 [00:18<04:41,  3.19it/s]
Training 1/1 epoch (loss 1.8876):   4%|▍         | 39/938 [00:19<04:41,  3.19it/s]
Training 1/1 epoch (loss 1.8876):   4%|▍         | 40/938 [00:19<04:48,  3.12it/s]
Training 1/1 epoch (loss 1.7993):   4%|▍         | 40/938 [00:19<04:48,  3.12it/s]
Training 1/1 epoch (loss 1.7993):   4%|▍         | 41/938 [00:19<04:43,  3.16it/s]
Training 1/1 epoch (loss 1.9353):   4%|▍         | 41/938 [00:19<04:43,  3.16it/s]
Training 1/1 epoch (loss 1.9353):   4%|▍         | 42/938 [00:19<04:47,  3.12it/s]
Training 1/1 epoch (loss 1.8642):   4%|▍         | 42/938 [00:20<04:47,  3.12it/s]
Training 1/1 epoch (loss 1.8642):   5%|▍         | 43/938 [00:20<05:03,  2.95it/s]
Training 1/1 epoch (loss 1.8517):   5%|▍         | 43/938 [00:20<05:03,  2.95it/s]
Training 1/1 epoch (loss 1.8517):   5%|▍         | 44/938 [00:20<04:54,  3.04it/s]
Training 1/1 epoch (loss 1.8349):   5%|▍         | 44/938 [00:20<04:54,  3.04it/s]
Training 1/1 epoch (loss 1.8349):   5%|▍         | 45/938 [00:20<05:00,  2.97it/s]
Training 1/1 epoch (loss 1.8698):   5%|▍         | 45/938 [00:21<05:00,  2.97it/s]
Training 1/1 epoch (loss 1.8698):   5%|▍         | 46/938 [00:21<04:49,  3.09it/s]
Training 1/1 epoch (loss 1.9042):   5%|▍         | 46/938 [00:21<04:49,  3.09it/s]
Training 1/1 epoch (loss 1.9042):   5%|▌         | 47/938 [00:21<04:44,  3.13it/s]
Training 1/1 epoch (loss 1.8854):   5%|▌         | 47/938 [00:21<04:44,  3.13it/s]
Training 1/1 epoch (loss 1.8854):   5%|▌         | 48/938 [00:21<04:51,  3.05it/s]
Training 1/1 epoch (loss 1.9489):   5%|▌         | 48/938 [00:22<04:51,  3.05it/s]
Training 1/1 epoch (loss 1.9489):   5%|▌         | 49/938 [00:22<04:54,  3.02it/s]
Training 1/1 epoch (loss 1.7554):   5%|▌         | 49/938 [00:22<04:54,  3.02it/s]
Training 1/1 epoch (loss 1.7554):   5%|▌         | 50/938 [00:22<04:47,  3.09it/s]
Training 1/1 epoch (loss 1.8064):   5%|▌         | 50/938 [00:22<04:47,  3.09it/s]
Training 1/1 epoch (loss 1.8064):   5%|▌         | 51/938 [00:22<04:42,  3.14it/s]
Training 1/1 epoch (loss 1.8808):   5%|▌         | 51/938 [00:23<04:42,  3.14it/s]
Training 1/1 epoch (loss 1.8808):   6%|▌         | 52/938 [00:23<04:39,  3.16it/s]
Training 1/1 epoch (loss 1.7632):   6%|▌         | 52/938 [00:23<04:39,  3.16it/s]
Training 1/1 epoch (loss 1.7632):   6%|▌         | 53/938 [00:23<04:39,  3.17it/s]
Training 1/1 epoch (loss 1.8720):   6%|▌         | 53/938 [00:23<04:39,  3.17it/s]
Training 1/1 epoch (loss 1.8720):   6%|▌         | 54/938 [00:23<04:30,  3.27it/s]
Training 1/1 epoch (loss 1.8994):   6%|▌         | 54/938 [00:24<04:30,  3.27it/s]
Training 1/1 epoch (loss 1.8994):   6%|▌         | 55/938 [00:24<04:39,  3.16it/s]
Training 1/1 epoch (loss 1.7690):   6%|▌         | 55/938 [00:24<04:39,  3.16it/s]
Training 1/1 epoch (loss 1.7690):   6%|▌         | 56/938 [00:24<04:38,  3.17it/s]
Training 1/1 epoch (loss 1.7478):   6%|▌         | 56/938 [00:24<04:38,  3.17it/s]
Training 1/1 epoch (loss 1.7478):   6%|▌         | 57/938 [00:24<04:36,  3.19it/s]
Training 1/1 epoch (loss 1.7764):   6%|▌         | 57/938 [00:24<04:36,  3.19it/s]
Training 1/1 epoch (loss 1.7764):   6%|▌         | 58/938 [00:24<04:36,  3.18it/s]
Training 1/1 epoch (loss 1.7674):   6%|▌         | 58/938 [00:25<04:36,  3.18it/s]
Training 1/1 epoch (loss 1.7674):   6%|▋         | 59/938 [00:25<04:32,  3.23it/s]
Training 1/1 epoch (loss 1.8116):   6%|▋         | 59/938 [00:25<04:32,  3.23it/s]
Training 1/1 epoch (loss 1.8116):   6%|▋         | 60/938 [00:25<04:37,  3.17it/s]
Training 1/1 epoch (loss 1.8339):   6%|▋         | 60/938 [00:25<04:37,  3.17it/s]
Training 1/1 epoch (loss 1.8339):   7%|▋         | 61/938 [00:25<04:43,  3.09it/s]
Training 1/1 epoch (loss 1.8119):   7%|▋         | 61/938 [00:26<04:43,  3.09it/s]
Training 1/1 epoch (loss 1.8119):   7%|▋         | 62/938 [00:26<04:45,  3.07it/s]
Training 1/1 epoch (loss 1.8281):   7%|▋         | 62/938 [00:26<04:45,  3.07it/s]
Training 1/1 epoch (loss 1.8281):   7%|▋         | 63/938 [00:26<04:38,  3.14it/s]
Training 1/1 epoch (loss 1.6736):   7%|▋         | 63/938 [00:26<04:38,  3.14it/s]
Training 1/1 epoch (loss 1.6736):   7%|▋         | 64/938 [00:26<04:41,  3.10it/s]
Training 1/1 epoch (loss 1.7686):   7%|▋         | 64/938 [00:27<04:41,  3.10it/s]
Training 1/1 epoch (loss 1.7686):   7%|▋         | 65/938 [00:27<04:38,  3.13it/s]
Training 1/1 epoch (loss 1.8431):   7%|▋         | 65/938 [00:27<04:38,  3.13it/s]
Training 1/1 epoch (loss 1.8431):   7%|▋         | 66/938 [00:27<04:35,  3.16it/s]
Training 1/1 epoch (loss 1.7686):   7%|▋         | 66/938 [00:27<04:35,  3.16it/s]
Training 1/1 epoch (loss 1.7686):   7%|▋         | 67/938 [00:27<04:43,  3.07it/s]
Training 1/1 epoch (loss 1.7111):   7%|▋         | 67/938 [00:28<04:43,  3.07it/s]
Training 1/1 epoch (loss 1.7111):   7%|▋         | 68/938 [00:28<04:56,  2.94it/s]
Training 1/1 epoch (loss 1.7841):   7%|▋         | 68/938 [00:28<04:56,  2.94it/s]
Training 1/1 epoch (loss 1.7841):   7%|▋         | 69/938 [00:28<04:49,  3.00it/s]
Training 1/1 epoch (loss 1.8431):   7%|▋         | 69/938 [00:28<04:49,  3.00it/s]
Training 1/1 epoch (loss 1.8431):   7%|▋         | 70/938 [00:28<04:41,  3.08it/s]
Training 1/1 epoch (loss 1.6584):   7%|▋         | 70/938 [00:29<04:41,  3.08it/s]
Training 1/1 epoch (loss 1.6584):   8%|▊         | 71/938 [00:29<04:33,  3.17it/s]
Training 1/1 epoch (loss 1.7463):   8%|▊         | 71/938 [00:29<04:33,  3.17it/s]
Training 1/1 epoch (loss 1.7463):   8%|▊         | 72/938 [00:29<04:34,  3.15it/s]
Training 1/1 epoch (loss 1.6684):   8%|▊         | 72/938 [00:29<04:34,  3.15it/s]
Training 1/1 epoch (loss 1.6684):   8%|▊         | 73/938 [00:29<04:34,  3.15it/s]
Training 1/1 epoch (loss 1.6903):   8%|▊         | 73/938 [00:30<04:34,  3.15it/s]
Training 1/1 epoch (loss 1.6903):   8%|▊         | 74/938 [00:30<04:53,  2.94it/s]
Training 1/1 epoch (loss 1.6414):   8%|▊         | 74/938 [00:30<04:53,  2.94it/s]
Training 1/1 epoch (loss 1.6414):   8%|▊         | 75/938 [00:30<04:47,  3.00it/s]
Training 1/1 epoch (loss 1.7005):   8%|▊         | 75/938 [00:30<04:47,  3.00it/s]
Training 1/1 epoch (loss 1.7005):   8%|▊         | 76/938 [00:30<04:46,  3.01it/s]
Training 1/1 epoch (loss 1.6195):   8%|▊         | 76/938 [00:31<04:46,  3.01it/s]
Training 1/1 epoch (loss 1.6195):   8%|▊         | 77/938 [00:31<04:40,  3.07it/s]
Training 1/1 epoch (loss 1.8112):   8%|▊         | 77/938 [00:31<04:40,  3.07it/s]
Training 1/1 epoch (loss 1.8112):   8%|▊         | 78/938 [00:31<04:34,  3.13it/s]
Training 1/1 epoch (loss 1.7361):   8%|▊         | 78/938 [00:31<04:34,  3.13it/s]
Training 1/1 epoch (loss 1.7361):   8%|▊         | 79/938 [00:31<04:35,  3.12it/s]
Training 1/1 epoch (loss 1.8144):   8%|▊         | 79/938 [00:32<04:35,  3.12it/s]
Training 1/1 epoch (loss 1.8144):   9%|▊         | 80/938 [00:32<04:48,  2.97it/s]
Training 1/1 epoch (loss 1.7534):   9%|▊         | 80/938 [00:32<04:48,  2.97it/s]
Training 1/1 epoch (loss 1.7534):   9%|▊         | 81/938 [00:32<04:56,  2.89it/s]
Training 1/1 epoch (loss 1.6256):   9%|▊         | 81/938 [00:32<04:56,  2.89it/s]
Training 1/1 epoch (loss 1.6256):   9%|▊         | 82/938 [00:32<04:58,  2.86it/s]
Training 1/1 epoch (loss 1.7068):   9%|▊         | 82/938 [00:33<04:58,  2.86it/s]
Training 1/1 epoch (loss 1.7068):   9%|▉         | 83/938 [00:33<04:46,  2.99it/s]
Training 1/1 epoch (loss 1.6722):   9%|▉         | 83/938 [00:33<04:46,  2.99it/s]
Training 1/1 epoch (loss 1.6722):   9%|▉         | 84/938 [00:33<04:38,  3.06it/s]
Training 1/1 epoch (loss 1.6786):   9%|▉         | 84/938 [00:33<04:38,  3.06it/s]
Training 1/1 epoch (loss 1.6786):   9%|▉         | 85/938 [00:33<04:38,  3.06it/s]
Training 1/1 epoch (loss 1.7165):   9%|▉         | 85/938 [00:34<04:38,  3.06it/s]
Training 1/1 epoch (loss 1.7165):   9%|▉         | 86/938 [00:34<04:41,  3.03it/s]
Training 1/1 epoch (loss 1.6949):   9%|▉         | 86/938 [00:34<04:41,  3.03it/s]
Training 1/1 epoch (loss 1.6949):   9%|▉         | 87/938 [00:34<04:38,  3.05it/s]
Training 1/1 epoch (loss 1.6626):   9%|▉         | 87/938 [00:34<04:38,  3.05it/s]
Training 1/1 epoch (loss 1.6626):   9%|▉         | 88/938 [00:34<04:34,  3.10it/s]
Training 1/1 epoch (loss 1.6971):   9%|▉         | 88/938 [00:35<04:34,  3.10it/s]
Training 1/1 epoch (loss 1.6971):   9%|▉         | 89/938 [00:35<04:29,  3.14it/s]
Training 1/1 epoch (loss 1.7156):   9%|▉         | 89/938 [00:35<04:29,  3.14it/s]
Training 1/1 epoch (loss 1.7156):  10%|▉         | 90/938 [00:35<04:28,  3.15it/s]
Training 1/1 epoch (loss 1.7658):  10%|▉         | 90/938 [00:35<04:28,  3.15it/s]
Training 1/1 epoch (loss 1.7658):  10%|▉         | 91/938 [00:35<04:35,  3.07it/s]
Training 1/1 epoch (loss 1.5823):  10%|▉         | 91/938 [00:36<04:35,  3.07it/s]
Training 1/1 epoch (loss 1.5823):  10%|▉         | 92/938 [00:36<04:36,  3.06it/s]
Training 1/1 epoch (loss 1.8280):  10%|▉         | 92/938 [00:36<04:36,  3.06it/s]
Training 1/1 epoch (loss 1.8280):  10%|▉         | 93/938 [00:36<04:37,  3.05it/s]
Training 1/1 epoch (loss 1.7043):  10%|▉         | 93/938 [00:36<04:37,  3.05it/s]
Training 1/1 epoch (loss 1.7043):  10%|█         | 94/938 [00:36<04:34,  3.07it/s]
Training 1/1 epoch (loss 1.6915):  10%|█         | 94/938 [00:37<04:34,  3.07it/s]
Training 1/1 epoch (loss 1.6915):  10%|█         | 95/938 [00:37<04:31,  3.11it/s]
Training 1/1 epoch (loss 1.7185):  10%|█         | 95/938 [00:37<04:31,  3.11it/s]
Training 1/1 epoch (loss 1.7185):  10%|█         | 96/938 [00:37<04:27,  3.15it/s]
Training 1/1 epoch (loss 1.7642):  10%|█         | 96/938 [00:37<04:27,  3.15it/s]
Training 1/1 epoch (loss 1.7642):  10%|█         | 97/938 [00:37<04:26,  3.16it/s]
Training 1/1 epoch (loss 1.6670):  10%|█         | 97/938 [00:37<04:26,  3.16it/s]
Training 1/1 epoch (loss 1.6670):  10%|█         | 98/938 [00:37<04:34,  3.06it/s]
Training 1/1 epoch (loss 1.7587):  10%|█         | 98/938 [00:38<04:34,  3.06it/s]
Training 1/1 epoch (loss 1.7587):  11%|█         | 99/938 [00:38<04:36,  3.04it/s]
Training 1/1 epoch (loss 1.5194):  11%|█         | 99/938 [00:38<04:36,  3.04it/s]
Training 1/1 epoch (loss 1.5194):  11%|█         | 100/938 [00:38<04:28,  3.12it/s]
Training 1/1 epoch (loss 1.6758):  11%|█         | 100/938 [00:38<04:28,  3.12it/s]
Training 1/1 epoch (loss 1.6758):  11%|█         | 101/938 [00:38<04:24,  3.16it/s]
Training 1/1 epoch (loss 1.6739):  11%|█         | 101/938 [00:39<04:24,  3.16it/s]
Training 1/1 epoch (loss 1.6739):  11%|█         | 102/938 [00:39<04:25,  3.15it/s]
Training 1/1 epoch (loss 1.6454):  11%|█         | 102/938 [00:39<04:25,  3.15it/s]
Training 1/1 epoch (loss 1.6454):  11%|█         | 103/938 [00:39<04:21,  3.19it/s]
Training 1/1 epoch (loss 1.7066):  11%|█         | 103/938 [00:39<04:21,  3.19it/s]
Training 1/1 epoch (loss 1.7066):  11%|█         | 104/938 [00:39<04:30,  3.09it/s]
Training 1/1 epoch (loss 1.7418):  11%|█         | 104/938 [00:40<04:30,  3.09it/s]
Training 1/1 epoch (loss 1.7418):  11%|█         | 105/938 [00:40<04:32,  3.06it/s]
Training 1/1 epoch (loss 1.7281):  11%|█         | 105/938 [00:40<04:32,  3.06it/s]
Training 1/1 epoch (loss 1.7281):  11%|█▏        | 106/938 [00:40<04:43,  2.94it/s]
Training 1/1 epoch (loss 1.7611):  11%|█▏        | 106/938 [00:40<04:43,  2.94it/s]
Training 1/1 epoch (loss 1.7611):  11%|█▏        | 107/938 [00:40<04:34,  3.03it/s]
Training 1/1 epoch (loss 1.8436):  11%|█▏        | 107/938 [00:41<04:34,  3.03it/s]
Training 1/1 epoch (loss 1.8436):  12%|█▏        | 108/938 [00:41<04:25,  3.12it/s]
Training 1/1 epoch (loss 1.7631):  12%|█▏        | 108/938 [00:41<04:25,  3.12it/s]
Training 1/1 epoch (loss 1.7631):  12%|█▏        | 109/938 [00:41<04:23,  3.14it/s]
Training 1/1 epoch (loss 1.8213):  12%|█▏        | 109/938 [00:41<04:23,  3.14it/s]
Training 1/1 epoch (loss 1.8213):  12%|█▏        | 110/938 [00:41<04:44,  2.91it/s]
Training 1/1 epoch (loss 1.5215):  12%|█▏        | 110/938 [00:42<04:44,  2.91it/s]
Training 1/1 epoch (loss 1.5215):  12%|█▏        | 111/938 [00:42<04:46,  2.88it/s]
Training 1/1 epoch (loss 1.7422):  12%|█▏        | 111/938 [00:42<04:46,  2.88it/s]
Training 1/1 epoch (loss 1.7422):  12%|█▏        | 112/938 [00:42<04:43,  2.92it/s]
Training 1/1 epoch (loss 1.6342):  12%|█▏        | 112/938 [00:42<04:43,  2.92it/s]
Training 1/1 epoch (loss 1.6342):  12%|█▏        | 113/938 [00:42<04:32,  3.03it/s]
Training 1/1 epoch (loss 1.6527):  12%|█▏        | 113/938 [00:43<04:32,  3.03it/s]
Training 1/1 epoch (loss 1.6527):  12%|█▏        | 114/938 [00:43<04:22,  3.14it/s]
Training 1/1 epoch (loss 1.6351):  12%|█▏        | 114/938 [00:43<04:22,  3.14it/s]
Training 1/1 epoch (loss 1.6351):  12%|█▏        | 115/938 [00:43<04:21,  3.15it/s]
Training 1/1 epoch (loss 1.7526):  12%|█▏        | 115/938 [00:43<04:21,  3.15it/s]
Training 1/1 epoch (loss 1.7526):  12%|█▏        | 116/938 [00:43<04:18,  3.18it/s]
Training 1/1 epoch (loss 1.6462):  12%|█▏        | 116/938 [00:44<04:18,  3.18it/s]
Training 1/1 epoch (loss 1.6462):  12%|█▏        | 117/938 [00:44<04:24,  3.10it/s]
Training 1/1 epoch (loss 1.6912):  12%|█▏        | 117/938 [00:44<04:24,  3.10it/s]
Training 1/1 epoch (loss 1.6912):  13%|█▎        | 118/938 [00:44<04:19,  3.17it/s]
Training 1/1 epoch (loss 1.7530):  13%|█▎        | 118/938 [00:44<04:19,  3.17it/s]
Training 1/1 epoch (loss 1.7530):  13%|█▎        | 119/938 [00:44<04:18,  3.17it/s]
Training 1/1 epoch (loss 1.5902):  13%|█▎        | 119/938 [00:45<04:18,  3.17it/s]
Training 1/1 epoch (loss 1.5902):  13%|█▎        | 120/938 [00:45<04:19,  3.15it/s]
Training 1/1 epoch (loss 1.6880):  13%|█▎        | 120/938 [00:45<04:19,  3.15it/s]
Training 1/1 epoch (loss 1.6880):  13%|█▎        | 121/938 [00:45<04:18,  3.17it/s]
Training 1/1 epoch (loss 1.6769):  13%|█▎        | 121/938 [00:45<04:18,  3.17it/s]
Training 1/1 epoch (loss 1.6769):  13%|█▎        | 122/938 [00:45<04:17,  3.17it/s]
Training 1/1 epoch (loss 1.6504):  13%|█▎        | 122/938 [00:46<04:17,  3.17it/s]
Training 1/1 epoch (loss 1.6504):  13%|█▎        | 123/938 [00:46<04:27,  3.05it/s]
Training 1/1 epoch (loss 1.6799):  13%|█▎        | 123/938 [00:46<04:27,  3.05it/s]
Training 1/1 epoch (loss 1.6799):  13%|█▎        | 124/938 [00:46<04:20,  3.12it/s]
Training 1/1 epoch (loss 1.7216):  13%|█▎        | 124/938 [00:46<04:20,  3.12it/s]
Training 1/1 epoch (loss 1.7216):  13%|█▎        | 125/938 [00:46<04:12,  3.22it/s]
Training 1/1 epoch (loss 1.7810):  13%|█▎        | 125/938 [00:47<04:12,  3.22it/s]
Training 1/1 epoch (loss 1.7810):  13%|█▎        | 126/938 [00:47<04:25,  3.06it/s]
Training 1/1 epoch (loss 1.5516):  13%|█▎        | 126/938 [00:47<04:25,  3.06it/s]
Training 1/1 epoch (loss 1.5516):  14%|█▎        | 127/938 [00:47<04:23,  3.07it/s]
Training 1/1 epoch (loss 1.6115):  14%|█▎        | 127/938 [00:47<04:23,  3.07it/s]
Training 1/1 epoch (loss 1.6115):  14%|█▎        | 128/938 [00:47<04:25,  3.06it/s]
Training 1/1 epoch (loss 1.6810):  14%|█▎        | 128/938 [00:48<04:25,  3.06it/s]
Training 1/1 epoch (loss 1.6810):  14%|█▍        | 129/938 [00:48<04:26,  3.04it/s]
Training 1/1 epoch (loss 1.5843):  14%|█▍        | 129/938 [00:48<04:26,  3.04it/s]
Training 1/1 epoch (loss 1.5843):  14%|█▍        | 130/938 [00:48<04:23,  3.07it/s]
Training 1/1 epoch (loss 1.6530):  14%|█▍        | 130/938 [00:48<04:23,  3.07it/s]
Training 1/1 epoch (loss 1.6530):  14%|█▍        | 131/938 [00:48<04:18,  3.12it/s]
Training 1/1 epoch (loss 1.5635):  14%|█▍        | 131/938 [00:48<04:18,  3.12it/s]
Training 1/1 epoch (loss 1.5635):  14%|█▍        | 132/938 [00:48<04:12,  3.20it/s]
Training 1/1 epoch (loss 1.6971):  14%|█▍        | 132/938 [00:49<04:12,  3.20it/s]
Training 1/1 epoch (loss 1.6971):  14%|█▍        | 133/938 [00:49<04:12,  3.19it/s]
Training 1/1 epoch (loss 1.7782):  14%|█▍        | 133/938 [00:49<04:12,  3.19it/s]
Training 1/1 epoch (loss 1.7782):  14%|█▍        | 134/938 [00:49<04:16,  3.14it/s]
Training 1/1 epoch (loss 1.5750):  14%|█▍        | 134/938 [00:49<04:16,  3.14it/s]
Training 1/1 epoch (loss 1.5750):  14%|█▍        | 135/938 [00:49<04:27,  3.00it/s]
Training 1/1 epoch (loss 1.7044):  14%|█▍        | 135/938 [00:50<04:27,  3.00it/s]
Training 1/1 epoch (loss 1.7044):  14%|█▍        | 136/938 [00:50<04:33,  2.93it/s]
Training 1/1 epoch (loss 1.6881):  14%|█▍        | 136/938 [00:50<04:33,  2.93it/s]
Training 1/1 epoch (loss 1.6881):  15%|█▍        | 137/938 [00:50<04:30,  2.96it/s]
Training 1/1 epoch (loss 1.6578):  15%|█▍        | 137/938 [00:50<04:30,  2.96it/s]
Training 1/1 epoch (loss 1.6578):  15%|█▍        | 138/938 [00:50<04:22,  3.05it/s]
Training 1/1 epoch (loss 1.6749):  15%|█▍        | 138/938 [00:51<04:22,  3.05it/s]
Training 1/1 epoch (loss 1.6749):  15%|█▍        | 139/938 [00:51<04:19,  3.08it/s]
Training 1/1 epoch (loss 1.6477):  15%|█▍        | 139/938 [00:51<04:19,  3.08it/s]
Training 1/1 epoch (loss 1.6477):  15%|█▍        | 140/938 [00:51<04:12,  3.16it/s]
Training 1/1 epoch (loss 1.7120):  15%|█▍        | 140/938 [00:51<04:12,  3.16it/s]
Training 1/1 epoch (loss 1.7120):  15%|█▌        | 141/938 [00:51<04:14,  3.14it/s]
Training 1/1 epoch (loss 1.6788):  15%|█▌        | 141/938 [00:52<04:14,  3.14it/s]
Training 1/1 epoch (loss 1.6788):  15%|█▌        | 142/938 [00:52<04:38,  2.85it/s]
Training 1/1 epoch (loss 1.6142):  15%|█▌        | 142/938 [00:52<04:38,  2.85it/s]
Training 1/1 epoch (loss 1.6142):  15%|█▌        | 143/938 [00:52<04:25,  2.99it/s]
Training 1/1 epoch (loss 1.7386):  15%|█▌        | 143/938 [00:52<04:25,  2.99it/s]
Training 1/1 epoch (loss 1.7386):  15%|█▌        | 144/938 [00:52<04:22,  3.03it/s]
Training 1/1 epoch (loss 1.6449):  15%|█▌        | 144/938 [00:53<04:22,  3.03it/s]
Training 1/1 epoch (loss 1.6449):  15%|█▌        | 145/938 [00:53<04:15,  3.10it/s]
Training 1/1 epoch (loss 1.6387):  15%|█▌        | 145/938 [00:53<04:15,  3.10it/s]
Training 1/1 epoch (loss 1.6387):  16%|█▌        | 146/938 [00:53<04:10,  3.16it/s]
Training 1/1 epoch (loss 1.4747):  16%|█▌        | 146/938 [00:53<04:10,  3.16it/s]
Training 1/1 epoch (loss 1.4747):  16%|█▌        | 147/938 [00:53<04:09,  3.17it/s]
Training 1/1 epoch (loss 1.7604):  16%|█▌        | 147/938 [00:54<04:09,  3.17it/s]
Training 1/1 epoch (loss 1.7604):  16%|█▌        | 148/938 [00:54<04:12,  3.12it/s]
Training 1/1 epoch (loss 1.7886):  16%|█▌        | 148/938 [00:54<04:12,  3.12it/s]
Training 1/1 epoch (loss 1.7886):  16%|█▌        | 149/938 [00:54<04:07,  3.18it/s]
Training 1/1 epoch (loss 1.5595):  16%|█▌        | 149/938 [00:54<04:07,  3.18it/s]
Training 1/1 epoch (loss 1.5595):  16%|█▌        | 150/938 [00:54<04:05,  3.21it/s]
Training 1/1 epoch (loss 1.6608):  16%|█▌        | 150/938 [00:55<04:05,  3.21it/s]
Training 1/1 epoch (loss 1.6608):  16%|█▌        | 151/938 [00:55<04:01,  3.26it/s]
Training 1/1 epoch (loss 1.6243):  16%|█▌        | 151/938 [00:55<04:01,  3.26it/s]
Training 1/1 epoch (loss 1.6243):  16%|█▌        | 152/938 [00:55<04:11,  3.13it/s]
Training 1/1 epoch (loss 1.6585):  16%|█▌        | 152/938 [00:55<04:11,  3.13it/s]
Training 1/1 epoch (loss 1.6585):  16%|█▋        | 153/938 [00:55<04:13,  3.09it/s]
Training 1/1 epoch (loss 1.5482):  16%|█▋        | 153/938 [00:56<04:13,  3.09it/s]
Training 1/1 epoch (loss 1.5482):  16%|█▋        | 154/938 [00:56<04:09,  3.14it/s]
Training 1/1 epoch (loss 1.7548):  16%|█▋        | 154/938 [00:56<04:09,  3.14it/s]
Training 1/1 epoch (loss 1.7548):  17%|█▋        | 155/938 [00:56<04:11,  3.11it/s]
Training 1/1 epoch (loss 1.6700):  17%|█▋        | 155/938 [00:56<04:11,  3.11it/s]
Training 1/1 epoch (loss 1.6700):  17%|█▋        | 156/938 [00:56<04:06,  3.17it/s]
Training 1/1 epoch (loss 1.6748):  17%|█▋        | 156/938 [00:57<04:06,  3.17it/s]
Training 1/1 epoch (loss 1.6748):  17%|█▋        | 157/938 [00:57<04:04,  3.19it/s]
Training 1/1 epoch (loss 1.5869):  17%|█▋        | 157/938 [00:57<04:04,  3.19it/s]
Training 1/1 epoch (loss 1.5869):  17%|█▋        | 158/938 [00:57<04:04,  3.19it/s]
Training 1/1 epoch (loss 1.7521):  17%|█▋        | 158/938 [00:57<04:04,  3.19it/s]
Training 1/1 epoch (loss 1.7521):  17%|█▋        | 159/938 [00:57<04:06,  3.16it/s]
Training 1/1 epoch (loss 1.6897):  17%|█▋        | 159/938 [00:58<04:06,  3.16it/s]
Training 1/1 epoch (loss 1.6897):  17%|█▋        | 160/938 [00:58<04:15,  3.04it/s]
Training 1/1 epoch (loss 1.6043):  17%|█▋        | 160/938 [00:58<04:15,  3.04it/s]
Training 1/1 epoch (loss 1.6043):  17%|█▋        | 161/938 [00:58<04:16,  3.02it/s]
Training 1/1 epoch (loss 1.6774):  17%|█▋        | 161/938 [00:58<04:16,  3.02it/s]
Training 1/1 epoch (loss 1.6774):  17%|█▋        | 162/938 [00:58<04:11,  3.09it/s]
Training 1/1 epoch (loss 1.6200):  17%|█▋        | 162/938 [00:58<04:11,  3.09it/s]
Training 1/1 epoch (loss 1.6200):  17%|█▋        | 163/938 [00:58<04:03,  3.18it/s]
Training 1/1 epoch (loss 1.6467):  17%|█▋        | 163/938 [00:59<04:03,  3.18it/s]
Training 1/1 epoch (loss 1.6467):  17%|█▋        | 164/938 [00:59<04:02,  3.19it/s]
Training 1/1 epoch (loss 1.6675):  17%|█▋        | 164/938 [00:59<04:02,  3.19it/s]
Training 1/1 epoch (loss 1.6675):  18%|█▊        | 165/938 [00:59<03:59,  3.23it/s]
Training 1/1 epoch (loss 1.6160):  18%|█▊        | 165/938 [00:59<03:59,  3.23it/s]
Training 1/1 epoch (loss 1.6160):  18%|█▊        | 166/938 [00:59<04:06,  3.14it/s]
Training 1/1 epoch (loss 1.6708):  18%|█▊        | 166/938 [01:00<04:06,  3.14it/s]
Training 1/1 epoch (loss 1.6708):  18%|█▊        | 167/938 [01:00<04:21,  2.95it/s]
Training 1/1 epoch (loss 1.5310):  18%|█▊        | 167/938 [01:00<04:21,  2.95it/s]
Training 1/1 epoch (loss 1.5310):  18%|█▊        | 168/938 [01:00<04:15,  3.01it/s]
Training 1/1 epoch (loss 1.7714):  18%|█▊        | 168/938 [01:00<04:15,  3.01it/s]
Training 1/1 epoch (loss 1.7714):  18%|█▊        | 169/938 [01:00<04:18,  2.97it/s]
Training 1/1 epoch (loss 1.6491):  18%|█▊        | 169/938 [01:01<04:18,  2.97it/s]
Training 1/1 epoch (loss 1.6491):  18%|█▊        | 170/938 [01:01<04:08,  3.08it/s]
Training 1/1 epoch (loss 1.6202):  18%|█▊        | 170/938 [01:01<04:08,  3.08it/s]
Training 1/1 epoch (loss 1.6202):  18%|█▊        | 171/938 [01:01<04:09,  3.08it/s]
Training 1/1 epoch (loss 1.5773):  18%|█▊        | 171/938 [01:01<04:09,  3.08it/s]
Training 1/1 epoch (loss 1.5773):  18%|█▊        | 172/938 [01:01<04:08,  3.08it/s]
Training 1/1 epoch (loss 1.6230):  18%|█▊        | 172/938 [01:02<04:08,  3.08it/s]
Training 1/1 epoch (loss 1.6230):  18%|█▊        | 173/938 [01:02<04:18,  2.96it/s]
Training 1/1 epoch (loss 1.7478):  18%|█▊        | 173/938 [01:02<04:18,  2.96it/s]
Training 1/1 epoch (loss 1.7478):  19%|█▊        | 174/938 [01:02<04:08,  3.08it/s]
Training 1/1 epoch (loss 1.7059):  19%|█▊        | 174/938 [01:02<04:08,  3.08it/s]
Training 1/1 epoch (loss 1.7059):  19%|█▊        | 175/938 [01:02<04:02,  3.14it/s]
Training 1/1 epoch (loss 1.4940):  19%|█▊        | 175/938 [01:03<04:02,  3.14it/s]
Training 1/1 epoch (loss 1.4940):  19%|█▉        | 176/938 [01:03<04:04,  3.12it/s]
Training 1/1 epoch (loss 1.6093):  19%|█▉        | 176/938 [01:03<04:04,  3.12it/s]
Training 1/1 epoch (loss 1.6093):  19%|█▉        | 177/938 [01:03<03:58,  3.19it/s]
Training 1/1 epoch (loss 1.6904):  19%|█▉        | 177/938 [01:03<03:58,  3.19it/s]
Training 1/1 epoch (loss 1.6904):  19%|█▉        | 178/938 [01:03<04:04,  3.11it/s]
Training 1/1 epoch (loss 1.7279):  19%|█▉        | 178/938 [01:04<04:04,  3.11it/s]
Training 1/1 epoch (loss 1.7279):  19%|█▉        | 179/938 [01:04<04:08,  3.05it/s]
Training 1/1 epoch (loss 1.5911):  19%|█▉        | 179/938 [01:04<04:08,  3.05it/s]
Training 1/1 epoch (loss 1.5911):  19%|█▉        | 180/938 [01:04<04:05,  3.08it/s]
Training 1/1 epoch (loss 1.6488):  19%|█▉        | 180/938 [01:04<04:05,  3.08it/s]
Training 1/1 epoch (loss 1.6488):  19%|█▉        | 181/938 [01:04<04:05,  3.08it/s]
Training 1/1 epoch (loss 1.4930):  19%|█▉        | 181/938 [01:05<04:05,  3.08it/s]
Training 1/1 epoch (loss 1.4930):  19%|█▉        | 182/938 [01:05<04:03,  3.10it/s]
Training 1/1 epoch (loss 1.7389):  19%|█▉        | 182/938 [01:05<04:03,  3.10it/s]
Training 1/1 epoch (loss 1.7389):  20%|█▉        | 183/938 [01:05<03:59,  3.15it/s]
Training 1/1 epoch (loss 1.6204):  20%|█▉        | 183/938 [01:05<03:59,  3.15it/s]
Training 1/1 epoch (loss 1.6204):  20%|█▉        | 184/938 [01:05<03:58,  3.16it/s]
Training 1/1 epoch (loss 1.6640):  20%|█▉        | 184/938 [01:06<03:58,  3.16it/s]
Training 1/1 epoch (loss 1.6640):  20%|█▉        | 185/938 [01:06<04:05,  3.06it/s]
Training 1/1 epoch (loss 1.6388):  20%|█▉        | 185/938 [01:06<04:05,  3.06it/s]
Training 1/1 epoch (loss 1.6388):  20%|█▉        | 186/938 [01:06<04:03,  3.09it/s]
Training 1/1 epoch (loss 1.5416):  20%|█▉        | 186/938 [01:06<04:03,  3.09it/s]
Training 1/1 epoch (loss 1.5416):  20%|█▉        | 187/938 [01:06<03:59,  3.13it/s]
Training 1/1 epoch (loss 1.5998):  20%|█▉        | 187/938 [01:07<03:59,  3.13it/s]
Training 1/1 epoch (loss 1.5998):  20%|██        | 188/938 [01:07<03:57,  3.16it/s]
Training 1/1 epoch (loss 1.6855):  20%|██        | 188/938 [01:07<03:57,  3.16it/s]
Training 1/1 epoch (loss 1.6855):  20%|██        | 189/938 [01:07<03:50,  3.25it/s]
Training 1/1 epoch (loss 1.8642):  20%|██        | 189/938 [01:07<03:50,  3.25it/s]
Training 1/1 epoch (loss 1.8642):  20%|██        | 190/938 [01:07<03:52,  3.21it/s]
Training 1/1 epoch (loss 1.5854):  20%|██        | 190/938 [01:08<03:52,  3.21it/s]
Training 1/1 epoch (loss 1.5854):  20%|██        | 191/938 [01:08<04:10,  2.99it/s]
Training 1/1 epoch (loss 1.6415):  20%|██        | 191/938 [01:08<04:10,  2.99it/s]
Training 1/1 epoch (loss 1.6415):  20%|██        | 192/938 [01:08<04:12,  2.96it/s]
Training 1/1 epoch (loss 1.6344):  20%|██        | 192/938 [01:08<04:12,  2.96it/s]
Training 1/1 epoch (loss 1.6344):  21%|██        | 193/938 [01:08<04:03,  3.06it/s]
Training 1/1 epoch (loss 1.6393):  21%|██        | 193/938 [01:08<04:03,  3.06it/s]
Training 1/1 epoch (loss 1.6393):  21%|██        | 194/938 [01:08<03:57,  3.13it/s]
Training 1/1 epoch (loss 1.6402):  21%|██        | 194/938 [01:09<03:57,  3.13it/s]
Training 1/1 epoch (loss 1.6402):  21%|██        | 195/938 [01:09<03:53,  3.19it/s]
Training 1/1 epoch (loss 1.6466):  21%|██        | 195/938 [01:09<03:53,  3.19it/s]
Training 1/1 epoch (loss 1.6466):  21%|██        | 196/938 [01:09<03:51,  3.20it/s]
Training 1/1 epoch (loss 1.6812):  21%|██        | 196/938 [01:09<03:51,  3.20it/s]
Training 1/1 epoch (loss 1.6812):  21%|██        | 197/938 [01:09<03:55,  3.15it/s]
Training 1/1 epoch (loss 1.5231):  21%|██        | 197/938 [01:10<03:55,  3.15it/s]
Training 1/1 epoch (loss 1.5231):  21%|██        | 198/938 [01:10<03:55,  3.15it/s]
Training 1/1 epoch (loss 1.6337):  21%|██        | 198/938 [01:10<03:55,  3.15it/s]
Training 1/1 epoch (loss 1.6337):  21%|██        | 199/938 [01:10<03:56,  3.13it/s]
Training 1/1 epoch (loss 1.5626):  21%|██        | 199/938 [01:10<03:56,  3.13it/s]
Training 1/1 epoch (loss 1.5626):  21%|██▏       | 200/938 [01:10<03:59,  3.08it/s]
Training 1/1 epoch (loss 1.6209):  21%|██▏       | 200/938 [01:11<03:59,  3.08it/s]
Training 1/1 epoch (loss 1.6209):  21%|██▏       | 201/938 [01:11<04:05,  3.01it/s]
Training 1/1 epoch (loss 1.6431):  21%|██▏       | 201/938 [01:11<04:05,  3.01it/s]
Training 1/1 epoch (loss 1.6431):  22%|██▏       | 202/938 [01:11<03:56,  3.11it/s]
Training 1/1 epoch (loss 1.6604):  22%|██▏       | 202/938 [01:11<03:56,  3.11it/s]
Training 1/1 epoch (loss 1.6604):  22%|██▏       | 203/938 [01:11<03:57,  3.09it/s]
Training 1/1 epoch (loss 1.6077):  22%|██▏       | 203/938 [01:12<03:57,  3.09it/s]
Training 1/1 epoch (loss 1.6077):  22%|██▏       | 204/938 [01:12<03:59,  3.07it/s]
Training 1/1 epoch (loss 1.6324):  22%|██▏       | 204/938 [01:12<03:59,  3.07it/s]
Training 1/1 epoch (loss 1.6324):  22%|██▏       | 205/938 [01:12<03:51,  3.16it/s]
Training 1/1 epoch (loss 1.7329):  22%|██▏       | 205/938 [01:12<03:51,  3.16it/s]
Training 1/1 epoch (loss 1.7329):  22%|██▏       | 206/938 [01:12<03:55,  3.11it/s]
Training 1/1 epoch (loss 1.6274):  22%|██▏       | 206/938 [01:13<03:55,  3.11it/s]
Training 1/1 epoch (loss 1.6274):  22%|██▏       | 207/938 [01:13<03:56,  3.09it/s]
Training 1/1 epoch (loss 1.6928):  22%|██▏       | 207/938 [01:13<03:56,  3.09it/s]
Training 1/1 epoch (loss 1.6928):  22%|██▏       | 208/938 [01:13<03:54,  3.12it/s]
Training 1/1 epoch (loss 1.6178):  22%|██▏       | 208/938 [01:13<03:54,  3.12it/s]
Training 1/1 epoch (loss 1.6178):  22%|██▏       | 209/938 [01:13<03:55,  3.10it/s]
Training 1/1 epoch (loss 1.5929):  22%|██▏       | 209/938 [01:14<03:55,  3.10it/s]
Training 1/1 epoch (loss 1.5929):  22%|██▏       | 210/938 [01:14<03:58,  3.05it/s]
Training 1/1 epoch (loss 1.6575):  22%|██▏       | 210/938 [01:14<03:58,  3.05it/s]
Training 1/1 epoch (loss 1.6575):  22%|██▏       | 211/938 [01:14<03:57,  3.06it/s]
Training 1/1 epoch (loss 1.5425):  22%|██▏       | 211/938 [01:14<03:57,  3.06it/s]
Training 1/1 epoch (loss 1.5425):  23%|██▎       | 212/938 [01:14<03:50,  3.15it/s]
Training 1/1 epoch (loss 1.5030):  23%|██▎       | 212/938 [01:15<03:50,  3.15it/s]
Training 1/1 epoch (loss 1.5030):  23%|██▎       | 213/938 [01:15<04:07,  2.93it/s]
Training 1/1 epoch (loss 1.5982):  23%|██▎       | 213/938 [01:15<04:07,  2.93it/s]
Training 1/1 epoch (loss 1.5982):  23%|██▎       | 214/938 [01:15<03:57,  3.05it/s]
Training 1/1 epoch (loss 1.6612):  23%|██▎       | 214/938 [01:15<03:57,  3.05it/s]
Training 1/1 epoch (loss 1.6612):  23%|██▎       | 215/938 [01:15<03:51,  3.13it/s]
Training 1/1 epoch (loss 1.5478):  23%|██▎       | 215/938 [01:16<03:51,  3.13it/s]
Training 1/1 epoch (loss 1.5478):  23%|██▎       | 216/938 [01:16<04:02,  2.98it/s]
Training 1/1 epoch (loss 1.7504):  23%|██▎       | 216/938 [01:16<04:02,  2.98it/s]
Training 1/1 epoch (loss 1.7504):  23%|██▎       | 217/938 [01:16<03:56,  3.04it/s]
Training 1/1 epoch (loss 1.6669):  23%|██▎       | 217/938 [01:16<03:56,  3.04it/s]
Training 1/1 epoch (loss 1.6669):  23%|██▎       | 218/938 [01:16<03:50,  3.12it/s]
Training 1/1 epoch (loss 1.6092):  23%|██▎       | 218/938 [01:17<03:50,  3.12it/s]
Training 1/1 epoch (loss 1.6092):  23%|██▎       | 219/938 [01:17<03:45,  3.18it/s]
Training 1/1 epoch (loss 1.5222):  23%|██▎       | 219/938 [01:17<03:45,  3.18it/s]
Training 1/1 epoch (loss 1.5222):  23%|██▎       | 220/938 [01:17<03:43,  3.21it/s]
Training 1/1 epoch (loss 1.6733):  23%|██▎       | 220/938 [01:17<03:43,  3.21it/s]
Training 1/1 epoch (loss 1.6733):  24%|██▎       | 221/938 [01:17<03:42,  3.22it/s]
Training 1/1 epoch (loss 1.6237):  24%|██▎       | 221/938 [01:18<03:42,  3.22it/s]
Training 1/1 epoch (loss 1.6237):  24%|██▎       | 222/938 [01:18<03:50,  3.11it/s]
Training 1/1 epoch (loss 1.7217):  24%|██▎       | 222/938 [01:18<03:50,  3.11it/s]
Training 1/1 epoch (loss 1.7217):  24%|██▍       | 223/938 [01:18<04:02,  2.95it/s]
Training 1/1 epoch (loss 1.6716):  24%|██▍       | 223/938 [01:18<04:02,  2.95it/s]
Training 1/1 epoch (loss 1.6716):  24%|██▍       | 224/938 [01:18<04:01,  2.96it/s]
Training 1/1 epoch (loss 1.6477):  24%|██▍       | 224/938 [01:19<04:01,  2.96it/s]
Training 1/1 epoch (loss 1.6477):  24%|██▍       | 225/938 [01:19<03:56,  3.01it/s]
Training 1/1 epoch (loss 1.7280):  24%|██▍       | 225/938 [01:19<03:56,  3.01it/s]
Training 1/1 epoch (loss 1.7280):  24%|██▍       | 226/938 [01:19<03:50,  3.09it/s]
Training 1/1 epoch (loss 1.6436):  24%|██▍       | 226/938 [01:19<03:50,  3.09it/s]
Training 1/1 epoch (loss 1.6436):  24%|██▍       | 227/938 [01:19<03:46,  3.14it/s]
Training 1/1 epoch (loss 1.6275):  24%|██▍       | 227/938 [01:19<03:46,  3.14it/s]
Training 1/1 epoch (loss 1.6275):  24%|██▍       | 228/938 [01:19<03:43,  3.17it/s]
Training 1/1 epoch (loss 1.6674):  24%|██▍       | 228/938 [01:20<03:43,  3.17it/s]
Training 1/1 epoch (loss 1.6674):  24%|██▍       | 229/938 [01:20<04:11,  2.82it/s]
Training 1/1 epoch (loss 1.6616):  24%|██▍       | 229/938 [01:20<04:11,  2.82it/s]
Training 1/1 epoch (loss 1.6616):  25%|██▍       | 230/938 [01:20<04:00,  2.95it/s]
Training 1/1 epoch (loss 1.5981):  25%|██▍       | 230/938 [01:21<04:00,  2.95it/s]
Training 1/1 epoch (loss 1.5981):  25%|██▍       | 231/938 [01:21<03:55,  3.00it/s]
Training 1/1 epoch (loss 1.6509):  25%|██▍       | 231/938 [01:21<03:55,  3.00it/s]
Training 1/1 epoch (loss 1.6509):  25%|██▍       | 232/938 [01:21<04:00,  2.93it/s]
Training 1/1 epoch (loss 1.6298):  25%|██▍       | 232/938 [01:21<04:00,  2.93it/s]
Training 1/1 epoch (loss 1.6298):  25%|██▍       | 233/938 [01:21<03:53,  3.02it/s]
Training 1/1 epoch (loss 1.5997):  25%|██▍       | 233/938 [01:22<03:53,  3.02it/s]
Training 1/1 epoch (loss 1.5997):  25%|██▍       | 234/938 [01:22<03:56,  2.98it/s]
Training 1/1 epoch (loss 1.5950):  25%|██▍       | 234/938 [01:22<03:56,  2.98it/s]
Training 1/1 epoch (loss 1.5950):  25%|██▌       | 235/938 [01:22<03:58,  2.94it/s]
Training 1/1 epoch (loss 1.5372):  25%|██▌       | 235/938 [01:22<03:58,  2.94it/s]
Training 1/1 epoch (loss 1.5372):  25%|██▌       | 236/938 [01:22<03:49,  3.06it/s]
Training 1/1 epoch (loss 1.7157):  25%|██▌       | 236/938 [01:22<03:49,  3.06it/s]
Training 1/1 epoch (loss 1.7157):  25%|██▌       | 237/938 [01:22<03:42,  3.14it/s]
Training 1/1 epoch (loss 1.5143):  25%|██▌       | 237/938 [01:23<03:42,  3.14it/s]
Training 1/1 epoch (loss 1.5143):  25%|██▌       | 238/938 [01:23<03:43,  3.14it/s]
Training 1/1 epoch (loss 1.5730):  25%|██▌       | 238/938 [01:23<03:43,  3.14it/s]
Training 1/1 epoch (loss 1.5730):  25%|██▌       | 239/938 [01:23<03:48,  3.06it/s]
Training 1/1 epoch (loss 1.6160):  25%|██▌       | 239/938 [01:24<03:48,  3.06it/s]
Training 1/1 epoch (loss 1.6160):  26%|██▌       | 240/938 [01:24<03:55,  2.96it/s]
Training 1/1 epoch (loss 1.7365):  26%|██▌       | 240/938 [01:24<03:55,  2.96it/s]
Training 1/1 epoch (loss 1.7365):  26%|██▌       | 241/938 [01:24<03:55,  2.96it/s]
Training 1/1 epoch (loss 1.6512):  26%|██▌       | 241/938 [01:24<03:55,  2.96it/s]
Training 1/1 epoch (loss 1.6512):  26%|██▌       | 242/938 [01:24<03:49,  3.03it/s]
Training 1/1 epoch (loss 1.6544):  26%|██▌       | 242/938 [01:24<03:49,  3.03it/s]
Training 1/1 epoch (loss 1.6544):  26%|██▌       | 243/938 [01:24<03:42,  3.13it/s]
Training 1/1 epoch (loss 1.5156):  26%|██▌       | 243/938 [01:25<03:42,  3.13it/s]
Training 1/1 epoch (loss 1.5156):  26%|██▌       | 244/938 [01:25<03:37,  3.19it/s]
Training 1/1 epoch (loss 1.5657):  26%|██▌       | 244/938 [01:25<03:37,  3.19it/s]
Training 1/1 epoch (loss 1.5657):  26%|██▌       | 245/938 [01:25<03:39,  3.15it/s]
Training 1/1 epoch (loss 1.6442):  26%|██▌       | 245/938 [01:25<03:39,  3.15it/s]
Training 1/1 epoch (loss 1.6442):  26%|██▌       | 246/938 [01:25<03:40,  3.14it/s]
Training 1/1 epoch (loss 1.5974):  26%|██▌       | 246/938 [01:26<03:40,  3.14it/s]
Training 1/1 epoch (loss 1.5974):  26%|██▋       | 247/938 [01:26<03:41,  3.12it/s]
Training 1/1 epoch (loss 1.5832):  26%|██▋       | 247/938 [01:26<03:41,  3.12it/s]
Training 1/1 epoch (loss 1.5832):  26%|██▋       | 248/938 [01:26<04:00,  2.87it/s]
Training 1/1 epoch (loss 1.6202):  26%|██▋       | 248/938 [01:26<04:00,  2.87it/s]
Training 1/1 epoch (loss 1.6202):  27%|██▋       | 249/938 [01:26<03:51,  2.97it/s]
Training 1/1 epoch (loss 1.4820):  27%|██▋       | 249/938 [01:27<03:51,  2.97it/s]
Training 1/1 epoch (loss 1.4820):  27%|██▋       | 250/938 [01:27<03:44,  3.07it/s]
Training 1/1 epoch (loss 1.7170):  27%|██▋       | 250/938 [01:27<03:44,  3.07it/s]
Training 1/1 epoch (loss 1.7170):  27%|██▋       | 251/938 [01:27<03:38,  3.14it/s]
Training 1/1 epoch (loss 1.6215):  27%|██▋       | 251/938 [01:27<03:38,  3.14it/s]
Training 1/1 epoch (loss 1.6215):  27%|██▋       | 252/938 [01:27<03:46,  3.03it/s]
Training 1/1 epoch (loss 1.6206):  27%|██▋       | 252/938 [01:28<03:46,  3.03it/s]
Training 1/1 epoch (loss 1.6206):  27%|██▋       | 253/938 [01:28<03:59,  2.86it/s]
Training 1/1 epoch (loss 1.6177):  27%|██▋       | 253/938 [01:28<03:59,  2.86it/s]
Training 1/1 epoch (loss 1.6177):  27%|██▋       | 254/938 [01:28<04:00,  2.84it/s]
Training 1/1 epoch (loss 1.6140):  27%|██▋       | 254/938 [01:29<04:00,  2.84it/s]
Training 1/1 epoch (loss 1.6140):  27%|██▋       | 255/938 [01:29<04:00,  2.85it/s]
Training 1/1 epoch (loss 1.5305):  27%|██▋       | 255/938 [01:29<04:00,  2.85it/s]
Training 1/1 epoch (loss 1.5305):  27%|██▋       | 256/938 [01:29<04:03,  2.80it/s]
Training 1/1 epoch (loss 1.6040):  27%|██▋       | 256/938 [01:29<04:03,  2.80it/s]
Training 1/1 epoch (loss 1.6040):  27%|██▋       | 257/938 [01:29<04:02,  2.80it/s]
Training 1/1 epoch (loss 1.6376):  27%|██▋       | 257/938 [01:30<04:02,  2.80it/s]
Training 1/1 epoch (loss 1.6376):  28%|██▊       | 258/938 [01:30<04:11,  2.70it/s]
Training 1/1 epoch (loss 1.6486):  28%|██▊       | 258/938 [01:30<04:11,  2.70it/s]
Training 1/1 epoch (loss 1.6486):  28%|██▊       | 259/938 [01:30<04:01,  2.81it/s]
Training 1/1 epoch (loss 1.5686):  28%|██▊       | 259/938 [01:30<04:01,  2.81it/s]
Training 1/1 epoch (loss 1.5686):  28%|██▊       | 260/938 [01:30<03:49,  2.95it/s]
Training 1/1 epoch (loss 1.5860):  28%|██▊       | 260/938 [01:31<03:49,  2.95it/s]
Training 1/1 epoch (loss 1.5860):  28%|██▊       | 261/938 [01:31<03:41,  3.05it/s]
Training 1/1 epoch (loss 1.5855):  28%|██▊       | 261/938 [01:31<03:41,  3.05it/s]
Training 1/1 epoch (loss 1.5855):  28%|██▊       | 262/938 [01:31<03:36,  3.12it/s]
Training 1/1 epoch (loss 1.4769):  28%|██▊       | 262/938 [01:31<03:36,  3.12it/s]
Training 1/1 epoch (loss 1.4769):  28%|██▊       | 263/938 [01:31<03:31,  3.19it/s]
Training 1/1 epoch (loss 1.7211):  28%|██▊       | 263/938 [01:32<03:31,  3.19it/s]
Training 1/1 epoch (loss 1.7211):  28%|██▊       | 264/938 [01:32<03:37,  3.10it/s]
Training 1/1 epoch (loss 1.5971):  28%|██▊       | 264/938 [01:32<03:37,  3.10it/s]
Training 1/1 epoch (loss 1.5971):  28%|██▊       | 265/938 [01:32<03:42,  3.03it/s]
Training 1/1 epoch (loss 1.6377):  28%|██▊       | 265/938 [01:32<03:42,  3.03it/s]
Training 1/1 epoch (loss 1.6377):  28%|██▊       | 266/938 [01:32<03:35,  3.12it/s]
Training 1/1 epoch (loss 1.5361):  28%|██▊       | 266/938 [01:32<03:35,  3.12it/s]
Training 1/1 epoch (loss 1.5361):  28%|██▊       | 267/938 [01:32<03:31,  3.18it/s]
Training 1/1 epoch (loss 1.5395):  28%|██▊       | 267/938 [01:33<03:31,  3.18it/s]
Training 1/1 epoch (loss 1.5395):  29%|██▊       | 268/938 [01:33<03:26,  3.25it/s]
Training 1/1 epoch (loss 1.5421):  29%|██▊       | 268/938 [01:33<03:26,  3.25it/s]
Training 1/1 epoch (loss 1.5421):  29%|██▊       | 269/938 [01:33<03:27,  3.22it/s]
Training 1/1 epoch (loss 1.6312):  29%|██▊       | 269/938 [01:33<03:27,  3.22it/s]
Training 1/1 epoch (loss 1.6312):  29%|██▉       | 270/938 [01:33<03:32,  3.14it/s]
Training 1/1 epoch (loss 1.7046):  29%|██▉       | 270/938 [01:34<03:32,  3.14it/s]
Training 1/1 epoch (loss 1.7046):  29%|██▉       | 271/938 [01:34<03:47,  2.93it/s]
Training 1/1 epoch (loss 1.5291):  29%|██▉       | 271/938 [01:34<03:47,  2.93it/s]
Training 1/1 epoch (loss 1.5291):  29%|██▉       | 272/938 [01:34<03:42,  2.99it/s]
Training 1/1 epoch (loss 1.6319):  29%|██▉       | 272/938 [01:34<03:42,  2.99it/s]
Training 1/1 epoch (loss 1.6319):  29%|██▉       | 273/938 [01:34<03:38,  3.05it/s]
Training 1/1 epoch (loss 1.6745):  29%|██▉       | 273/938 [01:35<03:38,  3.05it/s]
Training 1/1 epoch (loss 1.6745):  29%|██▉       | 274/938 [01:35<03:33,  3.11it/s]
Training 1/1 epoch (loss 1.7573):  29%|██▉       | 274/938 [01:35<03:33,  3.11it/s]
Training 1/1 epoch (loss 1.7573):  29%|██▉       | 275/938 [01:35<03:33,  3.10it/s]
Training 1/1 epoch (loss 1.6184):  29%|██▉       | 275/938 [01:35<03:33,  3.10it/s]
Training 1/1 epoch (loss 1.6184):  29%|██▉       | 276/938 [01:35<03:29,  3.17it/s]
Training 1/1 epoch (loss 1.6704):  29%|██▉       | 276/938 [01:36<03:29,  3.17it/s]
Training 1/1 epoch (loss 1.6704):  30%|██▉       | 277/938 [01:36<03:35,  3.06it/s]
Training 1/1 epoch (loss 1.6226):  30%|██▉       | 277/938 [01:36<03:35,  3.06it/s]
Training 1/1 epoch (loss 1.6226):  30%|██▉       | 278/938 [01:36<03:35,  3.07it/s]
Training 1/1 epoch (loss 1.6661):  30%|██▉       | 278/938 [01:36<03:35,  3.07it/s]
Training 1/1 epoch (loss 1.6661):  30%|██▉       | 279/938 [01:36<03:27,  3.17it/s]
Training 1/1 epoch (loss 1.4928):  30%|██▉       | 279/938 [01:37<03:27,  3.17it/s]
Training 1/1 epoch (loss 1.4928):  30%|██▉       | 280/938 [01:37<03:29,  3.15it/s]
Training 1/1 epoch (loss 1.5825):  30%|██▉       | 280/938 [01:37<03:29,  3.15it/s]
Training 1/1 epoch (loss 1.5825):  30%|██▉       | 281/938 [01:37<03:26,  3.19it/s]
Training 1/1 epoch (loss 1.5580):  30%|██▉       | 281/938 [01:37<03:26,  3.19it/s]
Training 1/1 epoch (loss 1.5580):  30%|███       | 282/938 [01:37<03:22,  3.23it/s]
Training 1/1 epoch (loss 1.7229):  30%|███       | 282/938 [01:38<03:22,  3.23it/s]
Training 1/1 epoch (loss 1.7229):  30%|███       | 283/938 [01:38<03:26,  3.17it/s]
Training 1/1 epoch (loss 1.5564):  30%|███       | 283/938 [01:38<03:26,  3.17it/s]
Training 1/1 epoch (loss 1.5564):  30%|███       | 284/938 [01:38<03:24,  3.20it/s]
Training 1/1 epoch (loss 1.5304):  30%|███       | 284/938 [01:38<03:24,  3.20it/s]
Training 1/1 epoch (loss 1.5304):  30%|███       | 285/938 [01:38<03:21,  3.25it/s]
Training 1/1 epoch (loss 1.6668):  30%|███       | 285/938 [01:38<03:21,  3.25it/s]
Training 1/1 epoch (loss 1.6668):  30%|███       | 286/938 [01:38<03:21,  3.24it/s]
Training 1/1 epoch (loss 1.5310):  30%|███       | 286/938 [01:39<03:21,  3.24it/s]
Training 1/1 epoch (loss 1.5310):  31%|███       | 287/938 [01:39<03:18,  3.28it/s]
Training 1/1 epoch (loss 1.7303):  31%|███       | 287/938 [01:39<03:18,  3.28it/s]
Training 1/1 epoch (loss 1.7303):  31%|███       | 288/938 [01:39<03:20,  3.24it/s]
Training 1/1 epoch (loss 1.7207):  31%|███       | 288/938 [01:39<03:20,  3.24it/s]
Training 1/1 epoch (loss 1.7207):  31%|███       | 289/938 [01:39<03:32,  3.05it/s]
Training 1/1 epoch (loss 1.5484):  31%|███       | 289/938 [01:40<03:32,  3.05it/s]
Training 1/1 epoch (loss 1.5484):  31%|███       | 290/938 [01:40<03:29,  3.09it/s]
Training 1/1 epoch (loss 1.6929):  31%|███       | 290/938 [01:40<03:29,  3.09it/s]
Training 1/1 epoch (loss 1.6929):  31%|███       | 291/938 [01:40<03:31,  3.06it/s]
Training 1/1 epoch (loss 1.5666):  31%|███       | 291/938 [01:40<03:31,  3.06it/s]
Training 1/1 epoch (loss 1.5666):  31%|███       | 292/938 [01:40<03:28,  3.10it/s]
Training 1/1 epoch (loss 1.5789):  31%|███       | 292/938 [01:41<03:28,  3.10it/s]
Training 1/1 epoch (loss 1.5789):  31%|███       | 293/938 [01:41<03:22,  3.19it/s]
Training 1/1 epoch (loss 1.5815):  31%|███       | 293/938 [01:41<03:22,  3.19it/s]
Training 1/1 epoch (loss 1.5815):  31%|███▏      | 294/938 [01:41<03:18,  3.24it/s]
Training 1/1 epoch (loss 1.4567):  31%|███▏      | 294/938 [01:41<03:18,  3.24it/s]
Training 1/1 epoch (loss 1.4567):  31%|███▏      | 295/938 [01:41<03:30,  3.05it/s]
Training 1/1 epoch (loss 1.5445):  31%|███▏      | 295/938 [01:42<03:30,  3.05it/s]
Training 1/1 epoch (loss 1.5445):  32%|███▏      | 296/938 [01:42<03:31,  3.03it/s]
Training 1/1 epoch (loss 1.6087):  32%|███▏      | 296/938 [01:42<03:31,  3.03it/s]
Training 1/1 epoch (loss 1.6087):  32%|███▏      | 297/938 [01:42<03:30,  3.04it/s]
Training 1/1 epoch (loss 1.5853):  32%|███▏      | 297/938 [01:42<03:30,  3.04it/s]
Training 1/1 epoch (loss 1.5853):  32%|███▏      | 298/938 [01:42<03:26,  3.10it/s]
Training 1/1 epoch (loss 1.6333):  32%|███▏      | 298/938 [01:43<03:26,  3.10it/s]
Training 1/1 epoch (loss 1.6333):  32%|███▏      | 299/938 [01:43<03:18,  3.21it/s]
Training 1/1 epoch (loss 1.5953):  32%|███▏      | 299/938 [01:43<03:18,  3.21it/s]
Training 1/1 epoch (loss 1.5953):  32%|███▏      | 300/938 [01:43<03:20,  3.18it/s]
Training 1/1 epoch (loss 1.6243):  32%|███▏      | 300/938 [01:43<03:20,  3.18it/s]
Training 1/1 epoch (loss 1.6243):  32%|███▏      | 301/938 [01:43<03:16,  3.25it/s]
Training 1/1 epoch (loss 1.6566):  32%|███▏      | 301/938 [01:44<03:16,  3.25it/s]
Training 1/1 epoch (loss 1.6566):  32%|███▏      | 302/938 [01:44<03:20,  3.17it/s]
Training 1/1 epoch (loss 1.5399):  32%|███▏      | 302/938 [01:44<03:20,  3.17it/s]
Training 1/1 epoch (loss 1.5399):  32%|███▏      | 303/938 [01:44<03:25,  3.10it/s]
Training 1/1 epoch (loss 1.7533):  32%|███▏      | 303/938 [01:44<03:25,  3.10it/s]
Training 1/1 epoch (loss 1.7533):  32%|███▏      | 304/938 [01:44<03:28,  3.03it/s]
Training 1/1 epoch (loss 1.5649):  32%|███▏      | 304/938 [01:45<03:28,  3.03it/s]
Training 1/1 epoch (loss 1.5649):  33%|███▎      | 305/938 [01:45<03:26,  3.07it/s]
Training 1/1 epoch (loss 1.6416):  33%|███▎      | 305/938 [01:45<03:26,  3.07it/s]
Training 1/1 epoch (loss 1.6416):  33%|███▎      | 306/938 [01:45<03:22,  3.12it/s]
Training 1/1 epoch (loss 1.6210):  33%|███▎      | 306/938 [01:45<03:22,  3.12it/s]
Training 1/1 epoch (loss 1.6210):  33%|███▎      | 307/938 [01:45<03:16,  3.20it/s]
Training 1/1 epoch (loss 1.5713):  33%|███▎      | 307/938 [01:46<03:16,  3.20it/s]
Training 1/1 epoch (loss 1.5713):  33%|███▎      | 308/938 [01:46<03:20,  3.14it/s]
Training 1/1 epoch (loss 1.4771):  33%|███▎      | 308/938 [01:46<03:20,  3.14it/s]
Training 1/1 epoch (loss 1.4771):  33%|███▎      | 309/938 [01:46<03:24,  3.08it/s]
Training 1/1 epoch (loss 1.6953):  33%|███▎      | 309/938 [01:46<03:24,  3.08it/s]
Training 1/1 epoch (loss 1.6953):  33%|███▎      | 310/938 [01:46<03:21,  3.11it/s]
Training 1/1 epoch (loss 1.4491):  33%|███▎      | 310/938 [01:46<03:21,  3.11it/s]
Training 1/1 epoch (loss 1.4491):  33%|███▎      | 311/938 [01:46<03:15,  3.21it/s]
Training 1/1 epoch (loss 1.5550):  33%|███▎      | 311/938 [01:47<03:15,  3.21it/s]
Training 1/1 epoch (loss 1.5550):  33%|███▎      | 312/938 [01:47<03:18,  3.15it/s]
Training 1/1 epoch (loss 1.6548):  33%|███▎      | 312/938 [01:47<03:18,  3.15it/s]
Training 1/1 epoch (loss 1.6548):  33%|███▎      | 313/938 [01:47<03:16,  3.18it/s]
Training 1/1 epoch (loss 1.6909):  33%|███▎      | 313/938 [01:47<03:16,  3.18it/s]
Training 1/1 epoch (loss 1.6909):  33%|███▎      | 314/938 [01:47<03:22,  3.09it/s]
Training 1/1 epoch (loss 1.5104):  33%|███▎      | 314/938 [01:48<03:22,  3.09it/s]
Training 1/1 epoch (loss 1.5104):  34%|███▎      | 315/938 [01:48<03:17,  3.15it/s]
Training 1/1 epoch (loss 1.5979):  34%|███▎      | 315/938 [01:48<03:17,  3.15it/s]
Training 1/1 epoch (loss 1.5979):  34%|███▎      | 316/938 [01:48<03:18,  3.14it/s]
Training 1/1 epoch (loss 1.5877):  34%|███▎      | 316/938 [01:48<03:18,  3.14it/s]
Training 1/1 epoch (loss 1.5877):  34%|███▍      | 317/938 [01:48<03:14,  3.18it/s]
Training 1/1 epoch (loss 1.6459):  34%|███▍      | 317/938 [01:49<03:14,  3.18it/s]
Training 1/1 epoch (loss 1.6459):  34%|███▍      | 318/938 [01:49<03:10,  3.25it/s]
Training 1/1 epoch (loss 1.5306):  34%|███▍      | 318/938 [01:49<03:10,  3.25it/s]
Training 1/1 epoch (loss 1.5306):  34%|███▍      | 319/938 [01:49<03:09,  3.27it/s]
Training 1/1 epoch (loss 1.4058):  34%|███▍      | 319/938 [01:49<03:09,  3.27it/s]
Training 1/1 epoch (loss 1.4058):  34%|███▍      | 320/938 [01:49<03:21,  3.07it/s]
Training 1/1 epoch (loss 1.5406):  34%|███▍      | 320/938 [01:50<03:21,  3.07it/s]
Training 1/1 epoch (loss 1.5406):  34%|███▍      | 321/938 [01:50<03:24,  3.01it/s]
Training 1/1 epoch (loss 1.6437):  34%|███▍      | 321/938 [01:50<03:24,  3.01it/s]
Training 1/1 epoch (loss 1.6437):  34%|███▍      | 322/938 [01:50<03:21,  3.05it/s]
Training 1/1 epoch (loss 1.7029):  34%|███▍      | 322/938 [01:50<03:21,  3.05it/s]
Training 1/1 epoch (loss 1.7029):  34%|███▍      | 323/938 [01:50<03:17,  3.11it/s]
Training 1/1 epoch (loss 1.6351):  34%|███▍      | 323/938 [01:51<03:17,  3.11it/s]
Training 1/1 epoch (loss 1.6351):  35%|███▍      | 324/938 [01:51<03:10,  3.22it/s]
Training 1/1 epoch (loss 1.5802):  35%|███▍      | 324/938 [01:51<03:10,  3.22it/s]
Training 1/1 epoch (loss 1.5802):  35%|███▍      | 325/938 [01:51<03:11,  3.21it/s]
Training 1/1 epoch (loss 1.6808):  35%|███▍      | 325/938 [01:51<03:11,  3.21it/s]
Training 1/1 epoch (loss 1.6808):  35%|███▍      | 326/938 [01:51<03:11,  3.20it/s]
Training 1/1 epoch (loss 1.5454):  35%|███▍      | 326/938 [01:52<03:11,  3.20it/s]
Training 1/1 epoch (loss 1.5454):  35%|███▍      | 327/938 [01:52<03:21,  3.03it/s]
Training 1/1 epoch (loss 1.5775):  35%|███▍      | 327/938 [01:52<03:21,  3.03it/s]
Training 1/1 epoch (loss 1.5775):  35%|███▍      | 328/938 [01:52<03:29,  2.91it/s]
Training 1/1 epoch (loss 1.4769):  35%|███▍      | 328/938 [01:52<03:29,  2.91it/s]
Training 1/1 epoch (loss 1.4769):  35%|███▌      | 329/938 [01:52<03:23,  2.99it/s]
Training 1/1 epoch (loss 1.6485):  35%|███▌      | 329/938 [01:53<03:23,  2.99it/s]
Training 1/1 epoch (loss 1.6485):  35%|███▌      | 330/938 [01:53<03:14,  3.12it/s]
Training 1/1 epoch (loss 1.5652):  35%|███▌      | 330/938 [01:53<03:14,  3.12it/s]
Training 1/1 epoch (loss 1.5652):  35%|███▌      | 331/938 [01:53<03:10,  3.18it/s]
Training 1/1 epoch (loss 1.6691):  35%|███▌      | 331/938 [01:53<03:10,  3.18it/s]
Training 1/1 epoch (loss 1.6691):  35%|███▌      | 332/938 [01:53<03:13,  3.14it/s]
Training 1/1 epoch (loss 1.5979):  35%|███▌      | 332/938 [01:54<03:13,  3.14it/s]
Training 1/1 epoch (loss 1.5979):  36%|███▌      | 333/938 [01:54<03:11,  3.16it/s]
Training 1/1 epoch (loss 1.5351):  36%|███▌      | 333/938 [01:54<03:11,  3.16it/s]
Training 1/1 epoch (loss 1.5351):  36%|███▌      | 334/938 [01:54<03:16,  3.07it/s]
Training 1/1 epoch (loss 1.6264):  36%|███▌      | 334/938 [01:54<03:16,  3.07it/s]
Training 1/1 epoch (loss 1.6264):  36%|███▌      | 335/938 [01:54<03:15,  3.09it/s]
Training 1/1 epoch (loss 1.5125):  36%|███▌      | 335/938 [01:55<03:15,  3.09it/s]
Training 1/1 epoch (loss 1.5125):  36%|███▌      | 336/938 [01:55<03:19,  3.01it/s]
Training 1/1 epoch (loss 1.6347):  36%|███▌      | 336/938 [01:55<03:19,  3.01it/s]
Training 1/1 epoch (loss 1.6347):  36%|███▌      | 337/938 [01:55<03:14,  3.08it/s]
Training 1/1 epoch (loss 1.5915):  36%|███▌      | 337/938 [01:56<03:14,  3.08it/s]
Training 1/1 epoch (loss 1.5915):  36%|███▌      | 338/938 [01:56<04:25,  2.26it/s]
Training 1/1 epoch (loss 1.6214):  36%|███▌      | 338/938 [01:56<04:25,  2.26it/s]
Training 1/1 epoch (loss 1.6214):  36%|███▌      | 339/938 [01:56<04:09,  2.40it/s]
Training 1/1 epoch (loss 1.5453):  36%|███▌      | 339/938 [01:56<04:09,  2.40it/s]
Training 1/1 epoch (loss 1.5453):  36%|███▌      | 340/938 [01:56<03:47,  2.63it/s]
Training 1/1 epoch (loss 1.5984):  36%|███▌      | 340/938 [01:57<03:47,  2.63it/s]
Training 1/1 epoch (loss 1.5984):  36%|███▋      | 341/938 [01:57<03:33,  2.80it/s]
Training 1/1 epoch (loss 1.5432):  36%|███▋      | 341/938 [01:57<03:33,  2.80it/s]
Training 1/1 epoch (loss 1.5432):  36%|███▋      | 342/938 [01:57<03:23,  2.94it/s]
Training 1/1 epoch (loss 1.6890):  36%|███▋      | 342/938 [01:57<03:23,  2.94it/s]
Training 1/1 epoch (loss 1.6890):  37%|███▋      | 343/938 [01:57<03:21,  2.96it/s]
Training 1/1 epoch (loss 1.5958):  37%|███▋      | 343/938 [01:58<03:21,  2.96it/s]
Training 1/1 epoch (loss 1.5958):  37%|███▋      | 344/938 [01:58<03:26,  2.87it/s]
Training 1/1 epoch (loss 1.5319):  37%|███▋      | 344/938 [01:58<03:26,  2.87it/s]
Training 1/1 epoch (loss 1.5319):  37%|███▋      | 345/938 [01:58<03:28,  2.84it/s]
Training 1/1 epoch (loss 1.5166):  37%|███▋      | 345/938 [01:58<03:28,  2.84it/s]
Training 1/1 epoch (loss 1.5166):  37%|███▋      | 346/938 [01:58<03:19,  2.97it/s]
Training 1/1 epoch (loss 1.5898):  37%|███▋      | 346/938 [01:59<03:19,  2.97it/s]
Training 1/1 epoch (loss 1.5898):  37%|███▋      | 347/938 [01:59<03:13,  3.05it/s]
Training 1/1 epoch (loss 1.4523):  37%|███▋      | 347/938 [01:59<03:13,  3.05it/s]
Training 1/1 epoch (loss 1.4523):  37%|███▋      | 348/938 [01:59<03:08,  3.13it/s]
Training 1/1 epoch (loss 1.6601):  37%|███▋      | 348/938 [01:59<03:08,  3.13it/s]
Training 1/1 epoch (loss 1.6601):  37%|███▋      | 349/938 [01:59<03:10,  3.10it/s]
Training 1/1 epoch (loss 1.6030):  37%|███▋      | 349/938 [01:59<03:10,  3.10it/s]
Training 1/1 epoch (loss 1.6030):  37%|███▋      | 350/938 [01:59<03:07,  3.13it/s]
Training 1/1 epoch (loss 1.6246):  37%|███▋      | 350/938 [02:00<03:07,  3.13it/s]
Training 1/1 epoch (loss 1.6246):  37%|███▋      | 351/938 [02:00<03:23,  2.89it/s]
Training 1/1 epoch (loss 1.5418):  37%|███▋      | 351/938 [02:00<03:23,  2.89it/s]
Training 1/1 epoch (loss 1.5418):  38%|███▊      | 352/938 [02:00<03:17,  2.96it/s]
Training 1/1 epoch (loss 1.7047):  38%|███▊      | 352/938 [02:00<03:17,  2.96it/s]
Training 1/1 epoch (loss 1.7047):  38%|███▊      | 353/938 [02:00<03:11,  3.05it/s]
Training 1/1 epoch (loss 1.6274):  38%|███▊      | 353/938 [02:01<03:11,  3.05it/s]
Training 1/1 epoch (loss 1.6274):  38%|███▊      | 354/938 [02:01<03:08,  3.10it/s]
Training 1/1 epoch (loss 1.6178):  38%|███▊      | 354/938 [02:01<03:08,  3.10it/s]
Training 1/1 epoch (loss 1.6178):  38%|███▊      | 355/938 [02:01<03:02,  3.19it/s]
Training 1/1 epoch (loss 1.5290):  38%|███▊      | 355/938 [02:01<03:02,  3.19it/s]
Training 1/1 epoch (loss 1.5290):  38%|███▊      | 356/938 [02:01<03:03,  3.17it/s]
Training 1/1 epoch (loss 1.5729):  38%|███▊      | 356/938 [02:02<03:03,  3.17it/s]
Training 1/1 epoch (loss 1.5729):  38%|███▊      | 357/938 [02:02<03:11,  3.03it/s]
Training 1/1 epoch (loss 1.5523):  38%|███▊      | 357/938 [02:02<03:11,  3.03it/s]
Training 1/1 epoch (loss 1.5523):  38%|███▊      | 358/938 [02:02<03:10,  3.05it/s]
Training 1/1 epoch (loss 1.5563):  38%|███▊      | 358/938 [02:02<03:10,  3.05it/s]
Training 1/1 epoch (loss 1.5563):  38%|███▊      | 359/938 [02:02<03:08,  3.07it/s]
Training 1/1 epoch (loss 1.6700):  38%|███▊      | 359/938 [02:03<03:08,  3.07it/s]
Training 1/1 epoch (loss 1.6700):  38%|███▊      | 360/938 [02:03<03:07,  3.07it/s]
Training 1/1 epoch (loss 1.7260):  38%|███▊      | 360/938 [02:03<03:07,  3.07it/s]
Training 1/1 epoch (loss 1.7260):  38%|███▊      | 361/938 [02:03<03:05,  3.12it/s]
Training 1/1 epoch (loss 1.5854):  38%|███▊      | 361/938 [02:03<03:05,  3.12it/s]
Training 1/1 epoch (loss 1.5854):  39%|███▊      | 362/938 [02:03<03:03,  3.14it/s]
Training 1/1 epoch (loss 1.6269):  39%|███▊      | 362/938 [02:04<03:03,  3.14it/s]
Training 1/1 epoch (loss 1.6269):  39%|███▊      | 363/938 [02:04<03:03,  3.14it/s]
Training 1/1 epoch (loss 1.6599):  39%|███▊      | 363/938 [02:04<03:03,  3.14it/s]
Training 1/1 epoch (loss 1.6599):  39%|███▉      | 364/938 [02:04<03:06,  3.07it/s]
Training 1/1 epoch (loss 1.5544):  39%|███▉      | 364/938 [02:04<03:06,  3.07it/s]
Training 1/1 epoch (loss 1.5544):  39%|███▉      | 365/938 [02:04<03:03,  3.12it/s]
Training 1/1 epoch (loss 1.5731):  39%|███▉      | 365/938 [02:05<03:03,  3.12it/s]
Training 1/1 epoch (loss 1.5731):  39%|███▉      | 366/938 [02:05<02:59,  3.18it/s]
Training 1/1 epoch (loss 1.6048):  39%|███▉      | 366/938 [02:05<02:59,  3.18it/s]
Training 1/1 epoch (loss 1.6048):  39%|███▉      | 367/938 [02:05<02:58,  3.19it/s]
Training 1/1 epoch (loss 1.6955):  39%|███▉      | 367/938 [02:05<02:58,  3.19it/s]
Training 1/1 epoch (loss 1.6955):  39%|███▉      | 368/938 [02:05<03:04,  3.10it/s]
Training 1/1 epoch (loss 1.5247):  39%|███▉      | 368/938 [02:06<03:04,  3.10it/s]
Training 1/1 epoch (loss 1.5247):  39%|███▉      | 369/938 [02:06<03:17,  2.89it/s]
Training 1/1 epoch (loss 1.6465):  39%|███▉      | 369/938 [02:06<03:17,  2.89it/s]
Training 1/1 epoch (loss 1.6465):  39%|███▉      | 370/938 [02:06<03:20,  2.83it/s]
Training 1/1 epoch (loss 1.5579):  39%|███▉      | 370/938 [02:06<03:20,  2.83it/s]
Training 1/1 epoch (loss 1.5579):  40%|███▉      | 371/938 [02:06<03:08,  3.01it/s]
Training 1/1 epoch (loss 1.6024):  40%|███▉      | 371/938 [02:07<03:08,  3.01it/s]
Training 1/1 epoch (loss 1.6024):  40%|███▉      | 372/938 [02:07<03:06,  3.04it/s]
Training 1/1 epoch (loss 1.5064):  40%|███▉      | 372/938 [02:07<03:06,  3.04it/s]
Training 1/1 epoch (loss 1.5064):  40%|███▉      | 373/938 [02:07<02:59,  3.14it/s]
Training 1/1 epoch (loss 1.4443):  40%|███▉      | 373/938 [02:07<02:59,  3.14it/s]
Training 1/1 epoch (loss 1.4443):  40%|███▉      | 374/938 [02:07<02:57,  3.17it/s]
Training 1/1 epoch (loss 1.6487):  40%|███▉      | 374/938 [02:08<02:57,  3.17it/s]
Training 1/1 epoch (loss 1.6487):  40%|███▉      | 375/938 [02:08<02:58,  3.16it/s]
Training 1/1 epoch (loss 1.6285):  40%|███▉      | 375/938 [02:08<02:58,  3.16it/s]
Training 1/1 epoch (loss 1.6285):  40%|████      | 376/938 [02:08<02:59,  3.13it/s]
Training 1/1 epoch (loss 1.7372):  40%|████      | 376/938 [02:08<02:59,  3.13it/s]
Training 1/1 epoch (loss 1.7372):  40%|████      | 377/938 [02:08<02:58,  3.15it/s]
Training 1/1 epoch (loss 1.4899):  40%|████      | 377/938 [02:09<02:58,  3.15it/s]
Training 1/1 epoch (loss 1.4899):  40%|████      | 378/938 [02:09<02:56,  3.18it/s]
Training 1/1 epoch (loss 1.5597):  40%|████      | 378/938 [02:09<02:56,  3.18it/s]
Training 1/1 epoch (loss 1.5597):  40%|████      | 379/938 [02:09<02:52,  3.24it/s]
Training 1/1 epoch (loss 1.5139):  40%|████      | 379/938 [02:09<02:52,  3.24it/s]
Training 1/1 epoch (loss 1.5139):  41%|████      | 380/938 [02:09<02:52,  3.24it/s]
Training 1/1 epoch (loss 1.5972):  41%|████      | 380/938 [02:09<02:52,  3.24it/s]
Training 1/1 epoch (loss 1.5972):  41%|████      | 381/938 [02:09<02:52,  3.22it/s]
Training 1/1 epoch (loss 1.4537):  41%|████      | 381/938 [02:10<02:52,  3.22it/s]
Training 1/1 epoch (loss 1.4537):  41%|████      | 382/938 [02:10<02:52,  3.22it/s]
Training 1/1 epoch (loss 1.6371):  41%|████      | 382/938 [02:10<02:52,  3.22it/s]
Training 1/1 epoch (loss 1.6371):  41%|████      | 383/938 [02:10<03:02,  3.04it/s]
Training 1/1 epoch (loss 1.6368):  41%|████      | 383/938 [02:10<03:02,  3.04it/s]
Training 1/1 epoch (loss 1.6368):  41%|████      | 384/938 [02:10<02:59,  3.09it/s]
Training 1/1 epoch (loss 1.5469):  41%|████      | 384/938 [02:11<02:59,  3.09it/s]
Training 1/1 epoch (loss 1.5469):  41%|████      | 385/938 [02:11<02:58,  3.11it/s]
Training 1/1 epoch (loss 1.5926):  41%|████      | 385/938 [02:11<02:58,  3.11it/s]
Training 1/1 epoch (loss 1.5926):  41%|████      | 386/938 [02:11<02:53,  3.18it/s]
Training 1/1 epoch (loss 1.6435):  41%|████      | 386/938 [02:11<02:53,  3.18it/s]
Training 1/1 epoch (loss 1.6435):  41%|████▏     | 387/938 [02:11<02:49,  3.25it/s]
Training 1/1 epoch (loss 1.5837):  41%|████▏     | 387/938 [02:12<02:49,  3.25it/s]
Training 1/1 epoch (loss 1.5837):  41%|████▏     | 388/938 [02:12<02:49,  3.25it/s]
Training 1/1 epoch (loss 1.7125):  41%|████▏     | 388/938 [02:12<02:49,  3.25it/s]
Training 1/1 epoch (loss 1.7125):  41%|████▏     | 389/938 [02:12<03:05,  2.95it/s]
Training 1/1 epoch (loss 1.5776):  41%|████▏     | 389/938 [02:12<03:05,  2.95it/s]
Training 1/1 epoch (loss 1.5776):  42%|████▏     | 390/938 [02:12<02:59,  3.06it/s]
Training 1/1 epoch (loss 1.5869):  42%|████▏     | 390/938 [02:13<02:59,  3.06it/s]
Training 1/1 epoch (loss 1.5869):  42%|████▏     | 391/938 [02:13<02:53,  3.15it/s]
Training 1/1 epoch (loss 1.6618):  42%|████▏     | 391/938 [02:13<02:53,  3.15it/s]
Training 1/1 epoch (loss 1.6618):  42%|████▏     | 392/938 [02:13<02:57,  3.07it/s]
Training 1/1 epoch (loss 1.5847):  42%|████▏     | 392/938 [02:13<02:57,  3.07it/s]
Training 1/1 epoch (loss 1.5847):  42%|████▏     | 393/938 [02:13<02:54,  3.12it/s]
Training 1/1 epoch (loss 1.7094):  42%|████▏     | 393/938 [02:14<02:54,  3.12it/s]
Training 1/1 epoch (loss 1.7094):  42%|████▏     | 394/938 [02:14<02:51,  3.16it/s]
Training 1/1 epoch (loss 1.5726):  42%|████▏     | 394/938 [02:14<02:51,  3.16it/s]
Training 1/1 epoch (loss 1.5726):  42%|████▏     | 395/938 [02:14<02:56,  3.07it/s]
Training 1/1 epoch (loss 1.5782):  42%|████▏     | 395/938 [02:14<02:56,  3.07it/s]
Training 1/1 epoch (loss 1.5782):  42%|████▏     | 396/938 [02:14<02:52,  3.14it/s]
Training 1/1 epoch (loss 1.5406):  42%|████▏     | 396/938 [02:15<02:52,  3.14it/s]
Training 1/1 epoch (loss 1.5406):  42%|████▏     | 397/938 [02:15<02:54,  3.10it/s]
Training 1/1 epoch (loss 1.5744):  42%|████▏     | 397/938 [02:15<02:54,  3.10it/s]
Training 1/1 epoch (loss 1.5744):  42%|████▏     | 398/938 [02:15<02:50,  3.17it/s]
Training 1/1 epoch (loss 1.5748):  42%|████▏     | 398/938 [02:15<02:50,  3.17it/s]
Training 1/1 epoch (loss 1.5748):  43%|████▎     | 399/938 [02:15<02:52,  3.12it/s]
Training 1/1 epoch (loss 1.6720):  43%|████▎     | 399/938 [02:16<02:52,  3.12it/s]
Training 1/1 epoch (loss 1.6720):  43%|████▎     | 400/938 [02:16<03:07,  2.87it/s]
Training 1/1 epoch (loss 1.6497):  43%|████▎     | 400/938 [02:16<03:07,  2.87it/s]
Training 1/1 epoch (loss 1.6497):  43%|████▎     | 401/938 [02:16<03:14,  2.76it/s]
Training 1/1 epoch (loss 1.6547):  43%|████▎     | 401/938 [02:16<03:14,  2.76it/s]
Training 1/1 epoch (loss 1.6547):  43%|████▎     | 402/938 [02:16<03:02,  2.94it/s]
Training 1/1 epoch (loss 1.6072):  43%|████▎     | 402/938 [02:17<03:02,  2.94it/s]
Training 1/1 epoch (loss 1.6072):  43%|████▎     | 403/938 [02:17<02:56,  3.03it/s]
Training 1/1 epoch (loss 1.6052):  43%|████▎     | 403/938 [02:17<02:56,  3.03it/s]
Training 1/1 epoch (loss 1.6052):  43%|████▎     | 404/938 [02:17<02:54,  3.07it/s]
Training 1/1 epoch (loss 1.5037):  43%|████▎     | 404/938 [02:17<02:54,  3.07it/s]
Training 1/1 epoch (loss 1.5037):  43%|████▎     | 405/938 [02:17<02:51,  3.11it/s]
Training 1/1 epoch (loss 1.7081):  43%|████▎     | 405/938 [02:18<02:51,  3.11it/s]
Training 1/1 epoch (loss 1.7081):  43%|████▎     | 406/938 [02:18<02:49,  3.15it/s]
Training 1/1 epoch (loss 1.6542):  43%|████▎     | 406/938 [02:18<02:49,  3.15it/s]
Training 1/1 epoch (loss 1.6542):  43%|████▎     | 407/938 [02:18<02:49,  3.12it/s]
Training 1/1 epoch (loss 1.5937):  43%|████▎     | 407/938 [02:18<02:49,  3.12it/s]
Training 1/1 epoch (loss 1.5937):  43%|████▎     | 408/938 [02:18<02:51,  3.08it/s]
Training 1/1 epoch (loss 1.6034):  43%|████▎     | 408/938 [02:19<02:51,  3.08it/s]
Training 1/1 epoch (loss 1.6034):  44%|████▎     | 409/938 [02:19<02:48,  3.14it/s]
Training 1/1 epoch (loss 1.6069):  44%|████▎     | 409/938 [02:19<02:48,  3.14it/s]
Training 1/1 epoch (loss 1.6069):  44%|████▎     | 410/938 [02:19<02:45,  3.20it/s]
Training 1/1 epoch (loss 1.4521):  44%|████▎     | 410/938 [02:19<02:45,  3.20it/s]
Training 1/1 epoch (loss 1.4521):  44%|████▍     | 411/938 [02:19<02:42,  3.24it/s]
Training 1/1 epoch (loss 1.5936):  44%|████▍     | 411/938 [02:19<02:42,  3.24it/s]
Training 1/1 epoch (loss 1.5936):  44%|████▍     | 412/938 [02:19<02:43,  3.21it/s]
Training 1/1 epoch (loss 1.5057):  44%|████▍     | 412/938 [02:20<02:43,  3.21it/s]
Training 1/1 epoch (loss 1.5057):  44%|████▍     | 413/938 [02:20<02:43,  3.22it/s]
Training 1/1 epoch (loss 1.6066):  44%|████▍     | 413/938 [02:20<02:43,  3.22it/s]
Training 1/1 epoch (loss 1.6066):  44%|████▍     | 414/938 [02:20<02:44,  3.18it/s]
Training 1/1 epoch (loss 1.5208):  44%|████▍     | 414/938 [02:20<02:44,  3.18it/s]
Training 1/1 epoch (loss 1.5208):  44%|████▍     | 415/938 [02:20<02:43,  3.21it/s]
Training 1/1 epoch (loss 1.4734):  44%|████▍     | 415/938 [02:21<02:43,  3.21it/s]
Training 1/1 epoch (loss 1.4734):  44%|████▍     | 416/938 [02:21<02:51,  3.04it/s]
Training 1/1 epoch (loss 1.4801):  44%|████▍     | 416/938 [02:21<02:51,  3.04it/s]
Training 1/1 epoch (loss 1.4801):  44%|████▍     | 417/938 [02:21<02:50,  3.05it/s]
Training 1/1 epoch (loss 1.5382):  44%|████▍     | 417/938 [02:21<02:50,  3.05it/s]
Training 1/1 epoch (loss 1.5382):  45%|████▍     | 418/938 [02:21<02:46,  3.12it/s]
Training 1/1 epoch (loss 1.6077):  45%|████▍     | 418/938 [02:22<02:46,  3.12it/s]
Training 1/1 epoch (loss 1.6077):  45%|████▍     | 419/938 [02:22<02:48,  3.09it/s]
Training 1/1 epoch (loss 1.6380):  45%|████▍     | 419/938 [02:22<02:48,  3.09it/s]
Training 1/1 epoch (loss 1.6380):  45%|████▍     | 420/938 [02:22<02:49,  3.06it/s]
Training 1/1 epoch (loss 1.5027):  45%|████▍     | 420/938 [02:22<02:49,  3.06it/s]
Training 1/1 epoch (loss 1.5027):  45%|████▍     | 421/938 [02:22<02:47,  3.09it/s]
Training 1/1 epoch (loss 1.5051):  45%|████▍     | 421/938 [02:23<02:47,  3.09it/s]
Training 1/1 epoch (loss 1.5051):  45%|████▍     | 422/938 [02:23<02:44,  3.13it/s]
Training 1/1 epoch (loss 1.5146):  45%|████▍     | 422/938 [02:23<02:44,  3.13it/s]
Training 1/1 epoch (loss 1.5146):  45%|████▌     | 423/938 [02:23<02:40,  3.20it/s]
Training 1/1 epoch (loss 1.4337):  45%|████▌     | 423/938 [02:23<02:40,  3.20it/s]
Training 1/1 epoch (loss 1.4337):  45%|████▌     | 424/938 [02:23<02:43,  3.14it/s]
Training 1/1 epoch (loss 1.5847):  45%|████▌     | 424/938 [02:24<02:43,  3.14it/s]
Training 1/1 epoch (loss 1.5847):  45%|████▌     | 425/938 [02:24<02:45,  3.09it/s]
Training 1/1 epoch (loss 1.5756):  45%|████▌     | 425/938 [02:24<02:45,  3.09it/s]
Training 1/1 epoch (loss 1.5756):  45%|████▌     | 426/938 [02:24<02:46,  3.08it/s]
Training 1/1 epoch (loss 1.6179):  45%|████▌     | 426/938 [02:24<02:46,  3.08it/s]
Training 1/1 epoch (loss 1.6179):  46%|████▌     | 427/938 [02:24<02:41,  3.16it/s]
Training 1/1 epoch (loss 1.5129):  46%|████▌     | 427/938 [02:25<02:41,  3.16it/s]
Training 1/1 epoch (loss 1.5129):  46%|████▌     | 428/938 [02:25<02:39,  3.20it/s]
Training 1/1 epoch (loss 1.5487):  46%|████▌     | 428/938 [02:25<02:39,  3.20it/s]
Training 1/1 epoch (loss 1.5487):  46%|████▌     | 429/938 [02:25<02:36,  3.25it/s]
Training 1/1 epoch (loss 1.5964):  46%|████▌     | 429/938 [02:25<02:36,  3.25it/s]
Training 1/1 epoch (loss 1.5964):  46%|████▌     | 430/938 [02:25<02:38,  3.20it/s]
Training 1/1 epoch (loss 1.5463):  46%|████▌     | 430/938 [02:26<02:38,  3.20it/s]
Training 1/1 epoch (loss 1.5463):  46%|████▌     | 431/938 [02:26<02:43,  3.11it/s]
Training 1/1 epoch (loss 1.6136):  46%|████▌     | 431/938 [02:26<02:43,  3.11it/s]
Training 1/1 epoch (loss 1.6136):  46%|████▌     | 432/938 [02:26<02:50,  2.97it/s]
Training 1/1 epoch (loss 1.5624):  46%|████▌     | 432/938 [02:26<02:50,  2.97it/s]
Training 1/1 epoch (loss 1.5624):  46%|████▌     | 433/938 [02:26<02:45,  3.05it/s]
Training 1/1 epoch (loss 1.6630):  46%|████▌     | 433/938 [02:27<02:45,  3.05it/s]
Training 1/1 epoch (loss 1.6630):  46%|████▋     | 434/938 [02:27<02:44,  3.06it/s]
Training 1/1 epoch (loss 1.5841):  46%|████▋     | 434/938 [02:27<02:44,  3.06it/s]
Training 1/1 epoch (loss 1.5841):  46%|████▋     | 435/938 [02:27<02:38,  3.17it/s]
Training 1/1 epoch (loss 1.6449):  46%|████▋     | 435/938 [02:27<02:38,  3.17it/s]
Training 1/1 epoch (loss 1.6449):  46%|████▋     | 436/938 [02:27<02:37,  3.19it/s]
Training 1/1 epoch (loss 1.6017):  46%|████▋     | 436/938 [02:27<02:37,  3.19it/s]
Training 1/1 epoch (loss 1.6017):  47%|████▋     | 437/938 [02:27<02:40,  3.12it/s]
Training 1/1 epoch (loss 1.4830):  47%|████▋     | 437/938 [02:28<02:40,  3.12it/s]
Training 1/1 epoch (loss 1.4830):  47%|████▋     | 438/938 [02:28<02:37,  3.18it/s]
Training 1/1 epoch (loss 1.6160):  47%|████▋     | 438/938 [02:28<02:37,  3.18it/s]
Training 1/1 epoch (loss 1.6160):  47%|████▋     | 439/938 [02:28<02:39,  3.13it/s]
Training 1/1 epoch (loss 1.5659):  47%|████▋     | 439/938 [02:28<02:39,  3.13it/s]
Training 1/1 epoch (loss 1.5659):  47%|████▋     | 440/938 [02:28<02:36,  3.17it/s]
Training 1/1 epoch (loss 1.5767):  47%|████▋     | 440/938 [02:29<02:36,  3.17it/s]
Training 1/1 epoch (loss 1.5767):  47%|████▋     | 441/938 [02:29<02:35,  3.19it/s]
Training 1/1 epoch (loss 1.5149):  47%|████▋     | 441/938 [02:29<02:35,  3.19it/s]
Training 1/1 epoch (loss 1.5149):  47%|████▋     | 442/938 [02:29<02:34,  3.21it/s]
Training 1/1 epoch (loss 1.5733):  47%|████▋     | 442/938 [02:29<02:34,  3.21it/s]
Training 1/1 epoch (loss 1.5733):  47%|████▋     | 443/938 [02:29<02:33,  3.22it/s]
Training 1/1 epoch (loss 1.6274):  47%|████▋     | 443/938 [02:30<02:33,  3.22it/s]
Training 1/1 epoch (loss 1.6274):  47%|████▋     | 444/938 [02:30<02:36,  3.15it/s]
Training 1/1 epoch (loss 1.4172):  47%|████▋     | 444/938 [02:30<02:36,  3.15it/s]
Training 1/1 epoch (loss 1.4172):  47%|████▋     | 445/938 [02:30<02:37,  3.13it/s]
Training 1/1 epoch (loss 1.4824):  47%|████▋     | 445/938 [02:30<02:37,  3.13it/s]
Training 1/1 epoch (loss 1.4824):  48%|████▊     | 446/938 [02:30<02:35,  3.17it/s]
Training 1/1 epoch (loss 1.6436):  48%|████▊     | 446/938 [02:31<02:35,  3.17it/s]
Training 1/1 epoch (loss 1.6436):  48%|████▊     | 447/938 [02:31<02:31,  3.24it/s]
Training 1/1 epoch (loss 1.5732):  48%|████▊     | 447/938 [02:31<02:31,  3.24it/s]
Training 1/1 epoch (loss 1.5732):  48%|████▊     | 448/938 [02:31<02:31,  3.22it/s]
Training 1/1 epoch (loss 1.6792):  48%|████▊     | 448/938 [02:31<02:31,  3.22it/s]
Training 1/1 epoch (loss 1.6792):  48%|████▊     | 449/938 [02:31<02:40,  3.05it/s]
Training 1/1 epoch (loss 1.4278):  48%|████▊     | 449/938 [02:32<02:40,  3.05it/s]
Training 1/1 epoch (loss 1.4278):  48%|████▊     | 450/938 [02:32<02:40,  3.04it/s]
Training 1/1 epoch (loss 1.5390):  48%|████▊     | 450/938 [02:32<02:40,  3.04it/s]
Training 1/1 epoch (loss 1.5390):  48%|████▊     | 451/938 [02:32<02:41,  3.02it/s]
Training 1/1 epoch (loss 1.5586):  48%|████▊     | 451/938 [02:32<02:41,  3.02it/s]
Training 1/1 epoch (loss 1.5586):  48%|████▊     | 452/938 [02:32<02:38,  3.07it/s]
Training 1/1 epoch (loss 1.5522):  48%|████▊     | 452/938 [02:33<02:38,  3.07it/s]
Training 1/1 epoch (loss 1.5522):  48%|████▊     | 453/938 [02:33<02:34,  3.15it/s]
Training 1/1 epoch (loss 1.4834):  48%|████▊     | 453/938 [02:33<02:34,  3.15it/s]
Training 1/1 epoch (loss 1.4834):  48%|████▊     | 454/938 [02:33<02:31,  3.19it/s]
Training 1/1 epoch (loss 1.6138):  48%|████▊     | 454/938 [02:33<02:31,  3.19it/s]
Training 1/1 epoch (loss 1.6138):  49%|████▊     | 455/938 [02:33<02:30,  3.22it/s]
Training 1/1 epoch (loss 1.5585):  49%|████▊     | 455/938 [02:34<02:30,  3.22it/s]
Training 1/1 epoch (loss 1.5585):  49%|████▊     | 456/938 [02:34<02:33,  3.14it/s]
Training 1/1 epoch (loss 1.6769):  49%|████▊     | 456/938 [02:34<02:33,  3.14it/s]
Training 1/1 epoch (loss 1.6769):  49%|████▊     | 457/938 [02:34<02:40,  3.00it/s]
Training 1/1 epoch (loss 1.6343):  49%|████▊     | 457/938 [02:34<02:40,  3.00it/s]
Training 1/1 epoch (loss 1.6343):  49%|████▉     | 458/938 [02:34<02:36,  3.07it/s]
Training 1/1 epoch (loss 1.6533):  49%|████▉     | 458/938 [02:34<02:36,  3.07it/s]
Training 1/1 epoch (loss 1.6533):  49%|████▉     | 459/938 [02:34<02:33,  3.11it/s]
Training 1/1 epoch (loss 1.6076):  49%|████▉     | 459/938 [02:35<02:33,  3.11it/s]
Training 1/1 epoch (loss 1.6076):  49%|████▉     | 460/938 [02:35<02:32,  3.12it/s]
Training 1/1 epoch (loss 1.5475):  49%|████▉     | 460/938 [02:35<02:32,  3.12it/s]
Training 1/1 epoch (loss 1.5475):  49%|████▉     | 461/938 [02:35<02:33,  3.10it/s]
Training 1/1 epoch (loss 1.5164):  49%|████▉     | 461/938 [02:35<02:33,  3.10it/s]
Training 1/1 epoch (loss 1.5164):  49%|████▉     | 462/938 [02:35<02:32,  3.12it/s]
Training 1/1 epoch (loss 1.5070):  49%|████▉     | 462/938 [02:36<02:32,  3.12it/s]
Training 1/1 epoch (loss 1.5070):  49%|████▉     | 463/938 [02:36<02:37,  3.02it/s]
Training 1/1 epoch (loss 1.5728):  49%|████▉     | 463/938 [02:36<02:37,  3.02it/s]
Training 1/1 epoch (loss 1.5728):  49%|████▉     | 464/938 [02:36<02:46,  2.84it/s]
Training 1/1 epoch (loss 1.5448):  49%|████▉     | 464/938 [02:37<02:46,  2.84it/s]
Training 1/1 epoch (loss 1.5448):  50%|████▉     | 465/938 [02:37<02:44,  2.87it/s]
Training 1/1 epoch (loss 1.6206):  50%|████▉     | 465/938 [02:37<02:44,  2.87it/s]
Training 1/1 epoch (loss 1.6206):  50%|████▉     | 466/938 [02:37<02:36,  3.02it/s]
Training 1/1 epoch (loss 1.6557):  50%|████▉     | 466/938 [02:37<02:36,  3.02it/s]
Training 1/1 epoch (loss 1.6557):  50%|████▉     | 467/938 [02:37<02:35,  3.02it/s]
Training 1/1 epoch (loss 1.6861):  50%|████▉     | 467/938 [02:38<02:35,  3.02it/s]
Training 1/1 epoch (loss 1.6861):  50%|████▉     | 468/938 [02:38<03:04,  2.54it/s]
Training 1/1 epoch (loss 1.6396):  50%|████▉     | 468/938 [02:38<03:04,  2.54it/s]
Training 1/1 epoch (loss 1.6396):  50%|█████     | 469/938 [02:38<02:58,  2.63it/s]
Training 1/1 epoch (loss 1.4508):  50%|█████     | 469/938 [02:38<02:58,  2.63it/s]
Training 1/1 epoch (loss 1.4508):  50%|█████     | 470/938 [02:38<02:51,  2.73it/s]
Training 1/1 epoch (loss 1.5791):  50%|█████     | 470/938 [02:39<02:51,  2.73it/s]
Training 1/1 epoch (loss 1.5791):  50%|█████     | 471/938 [02:39<02:41,  2.88it/s]
Training 1/1 epoch (loss 1.5884):  50%|█████     | 471/938 [02:39<02:41,  2.88it/s]
Training 1/1 epoch (loss 1.5884):  50%|█████     | 472/938 [02:39<02:40,  2.91it/s]
Training 1/1 epoch (loss 1.4805):  50%|█████     | 472/938 [02:39<02:40,  2.91it/s]
Training 1/1 epoch (loss 1.4805):  50%|█████     | 473/938 [02:39<02:38,  2.93it/s]
Training 1/1 epoch (loss 1.6152):  50%|█████     | 473/938 [02:40<02:38,  2.93it/s]
Training 1/1 epoch (loss 1.6152):  51%|█████     | 474/938 [02:40<02:42,  2.86it/s]
Training 1/1 epoch (loss 1.5073):  51%|█████     | 474/938 [02:40<02:42,  2.86it/s]
Training 1/1 epoch (loss 1.5073):  51%|█████     | 475/938 [02:40<02:45,  2.79it/s]
Training 1/1 epoch (loss 1.5516):  51%|█████     | 475/938 [02:40<02:45,  2.79it/s]
Training 1/1 epoch (loss 1.5516):  51%|█████     | 476/938 [02:40<02:36,  2.95it/s]
Training 1/1 epoch (loss 1.5174):  51%|█████     | 476/938 [02:41<02:36,  2.95it/s]
Training 1/1 epoch (loss 1.5174):  51%|█████     | 477/938 [02:41<02:30,  3.05it/s]
Training 1/1 epoch (loss 1.4812):  51%|█████     | 477/938 [02:41<02:30,  3.05it/s]
Training 1/1 epoch (loss 1.4812):  51%|█████     | 478/938 [02:41<02:27,  3.11it/s]
Training 1/1 epoch (loss 1.6080):  51%|█████     | 478/938 [02:41<02:27,  3.11it/s]
Training 1/1 epoch (loss 1.6080):  51%|█████     | 479/938 [02:41<02:25,  3.16it/s]
Training 1/1 epoch (loss 1.5620):  51%|█████     | 479/938 [02:42<02:25,  3.16it/s]
Training 1/1 epoch (loss 1.5620):  51%|█████     | 480/938 [02:42<02:34,  2.96it/s]
Training 1/1 epoch (loss 1.5897):  51%|█████     | 480/938 [02:42<02:34,  2.96it/s]
Training 1/1 epoch (loss 1.5897):  51%|█████▏    | 481/938 [02:42<02:34,  2.96it/s]
Training 1/1 epoch (loss 1.5537):  51%|█████▏    | 481/938 [02:42<02:34,  2.96it/s]
Training 1/1 epoch (loss 1.5537):  51%|█████▏    | 482/938 [02:42<02:29,  3.04it/s]
Training 1/1 epoch (loss 1.5076):  51%|█████▏    | 482/938 [02:43<02:29,  3.04it/s]
Training 1/1 epoch (loss 1.5076):  51%|█████▏    | 483/938 [02:43<02:28,  3.06it/s]
Training 1/1 epoch (loss 1.6870):  51%|█████▏    | 483/938 [02:43<02:28,  3.06it/s]
Training 1/1 epoch (loss 1.6870):  52%|█████▏    | 484/938 [02:43<02:24,  3.14it/s]
Training 1/1 epoch (loss 1.6892):  52%|█████▏    | 484/938 [02:43<02:24,  3.14it/s]
Training 1/1 epoch (loss 1.6892):  52%|█████▏    | 485/938 [02:43<02:24,  3.13it/s]
Training 1/1 epoch (loss 1.5762):  52%|█████▏    | 485/938 [02:44<02:24,  3.13it/s]
Training 1/1 epoch (loss 1.5762):  52%|█████▏    | 486/938 [02:44<02:26,  3.09it/s]
Training 1/1 epoch (loss 1.5624):  52%|█████▏    | 486/938 [02:44<02:26,  3.09it/s]
Training 1/1 epoch (loss 1.5624):  52%|█████▏    | 487/938 [02:44<02:29,  3.02it/s]
Training 1/1 epoch (loss 1.5573):  52%|█████▏    | 487/938 [02:44<02:29,  3.02it/s]
Training 1/1 epoch (loss 1.5573):  52%|█████▏    | 488/938 [02:44<02:28,  3.03it/s]
Training 1/1 epoch (loss 1.5748):  52%|█████▏    | 488/938 [02:45<02:28,  3.03it/s]
Training 1/1 epoch (loss 1.5748):  52%|█████▏    | 489/938 [02:45<02:31,  2.96it/s]
Training 1/1 epoch (loss 1.6076):  52%|█████▏    | 489/938 [02:45<02:31,  2.96it/s]
Training 1/1 epoch (loss 1.6076):  52%|█████▏    | 490/938 [02:45<02:26,  3.06it/s]
Training 1/1 epoch (loss 1.6670):  52%|█████▏    | 490/938 [02:45<02:26,  3.06it/s]
Training 1/1 epoch (loss 1.6670):  52%|█████▏    | 491/938 [02:45<02:21,  3.16it/s]
Training 1/1 epoch (loss 1.6545):  52%|█████▏    | 491/938 [02:46<02:21,  3.16it/s]
Training 1/1 epoch (loss 1.6545):  52%|█████▏    | 492/938 [02:46<02:22,  3.12it/s]
Training 1/1 epoch (loss 1.5241):  52%|█████▏    | 492/938 [02:46<02:22,  3.12it/s]
Training 1/1 epoch (loss 1.5241):  53%|█████▎    | 493/938 [02:46<02:20,  3.16it/s]
Training 1/1 epoch (loss 1.5698):  53%|█████▎    | 493/938 [02:46<02:20,  3.16it/s]
Training 1/1 epoch (loss 1.5698):  53%|█████▎    | 494/938 [02:46<02:20,  3.16it/s]
Training 1/1 epoch (loss 1.5969):  53%|█████▎    | 494/938 [02:47<02:20,  3.16it/s]
Training 1/1 epoch (loss 1.5969):  53%|█████▎    | 495/938 [02:47<02:18,  3.19it/s]
Training 1/1 epoch (loss 1.4482):  53%|█████▎    | 495/938 [02:47<02:18,  3.19it/s]
Training 1/1 epoch (loss 1.4482):  53%|█████▎    | 496/938 [02:47<02:24,  3.05it/s]
Training 1/1 epoch (loss 1.6345):  53%|█████▎    | 496/938 [02:47<02:24,  3.05it/s]
Training 1/1 epoch (loss 1.6345):  53%|█████▎    | 497/938 [02:47<02:23,  3.08it/s]
Training 1/1 epoch (loss 1.5356):  53%|█████▎    | 497/938 [02:48<02:23,  3.08it/s]
Training 1/1 epoch (loss 1.5356):  53%|█████▎    | 498/938 [02:48<02:22,  3.10it/s]
Training 1/1 epoch (loss 1.5320):  53%|█████▎    | 498/938 [02:48<02:22,  3.10it/s]
Training 1/1 epoch (loss 1.5320):  53%|█████▎    | 499/938 [02:48<02:20,  3.14it/s]
Training 1/1 epoch (loss 1.4697):  53%|█████▎    | 499/938 [02:48<02:20,  3.14it/s]
Training 1/1 epoch (loss 1.4697):  53%|█████▎    | 500/938 [02:48<02:22,  3.06it/s]
Training 1/1 epoch (loss 1.6312):  53%|█████▎    | 500/938 [02:48<02:22,  3.06it/s]
Training 1/1 epoch (loss 1.6312):  53%|█████▎    | 501/938 [02:48<02:17,  3.17it/s]
Training 1/1 epoch (loss 1.6236):  53%|█████▎    | 501/938 [02:49<02:17,  3.17it/s]
Training 1/1 epoch (loss 1.6236):  54%|█████▎    | 502/938 [02:49<02:17,  3.17it/s]
Training 1/1 epoch (loss 1.5805):  54%|█████▎    | 502/938 [02:49<02:17,  3.17it/s]
Training 1/1 epoch (loss 1.5805):  54%|█████▎    | 503/938 [02:49<02:14,  3.24it/s]
Training 1/1 epoch (loss 1.4941):  54%|█████▎    | 503/938 [02:49<02:14,  3.24it/s]
Training 1/1 epoch (loss 1.4941):  54%|█████▎    | 504/938 [02:49<02:15,  3.20it/s]
Training 1/1 epoch (loss 1.5493):  54%|█████▎    | 504/938 [02:50<02:15,  3.20it/s]
Training 1/1 epoch (loss 1.5493):  54%|█████▍    | 505/938 [02:50<02:17,  3.15it/s]
Training 1/1 epoch (loss 1.5651):  54%|█████▍    | 505/938 [02:50<02:17,  3.15it/s]
Training 1/1 epoch (loss 1.5651):  54%|█████▍    | 506/938 [02:50<02:30,  2.88it/s]
Training 1/1 epoch (loss 1.6176):  54%|█████▍    | 506/938 [02:51<02:30,  2.88it/s]
Training 1/1 epoch (loss 1.6176):  54%|█████▍    | 507/938 [02:51<02:35,  2.78it/s]
Training 1/1 epoch (loss 1.5751):  54%|█████▍    | 507/938 [02:51<02:35,  2.78it/s]
Training 1/1 epoch (loss 1.5751):  54%|█████▍    | 508/938 [02:51<02:40,  2.68it/s]
Training 1/1 epoch (loss 1.5599):  54%|█████▍    | 508/938 [02:51<02:40,  2.68it/s]
Training 1/1 epoch (loss 1.5599):  54%|█████▍    | 509/938 [02:51<02:42,  2.64it/s]
Training 1/1 epoch (loss 1.6340):  54%|█████▍    | 509/938 [02:52<02:42,  2.64it/s]
Training 1/1 epoch (loss 1.6340):  54%|█████▍    | 510/938 [02:52<02:50,  2.52it/s]
Training 1/1 epoch (loss 1.7228):  54%|█████▍    | 510/938 [02:52<02:50,  2.52it/s]
Training 1/1 epoch (loss 1.7228):  54%|█████▍    | 511/938 [02:52<03:03,  2.32it/s]
Training 1/1 epoch (loss 1.5460):  54%|█████▍    | 511/938 [02:53<03:03,  2.32it/s]
Training 1/1 epoch (loss 1.5460):  55%|█████▍    | 512/938 [02:53<03:00,  2.37it/s]
Training 1/1 epoch (loss 1.5685):  55%|█████▍    | 512/938 [02:53<03:00,  2.37it/s]
Training 1/1 epoch (loss 1.5685):  55%|█████▍    | 513/938 [02:53<02:58,  2.38it/s]
Training 1/1 epoch (loss 1.6172):  55%|█████▍    | 513/938 [02:53<02:58,  2.38it/s]
Training 1/1 epoch (loss 1.6172):  55%|█████▍    | 514/938 [02:53<02:57,  2.39it/s]
Training 1/1 epoch (loss 1.5346):  55%|█████▍    | 514/938 [02:54<02:57,  2.39it/s]
Training 1/1 epoch (loss 1.5346):  55%|█████▍    | 515/938 [02:54<02:54,  2.42it/s]
Training 1/1 epoch (loss 1.5898):  55%|█████▍    | 515/938 [02:54<02:54,  2.42it/s]
Training 1/1 epoch (loss 1.5898):  55%|█████▌    | 516/938 [02:54<02:54,  2.42it/s]
Training 1/1 epoch (loss 1.6138):  55%|█████▌    | 516/938 [02:55<02:54,  2.42it/s]
Training 1/1 epoch (loss 1.6138):  55%|█████▌    | 517/938 [02:55<02:49,  2.48it/s]
Training 1/1 epoch (loss 1.5388):  55%|█████▌    | 517/938 [02:55<02:49,  2.48it/s]
Training 1/1 epoch (loss 1.5388):  55%|█████▌    | 518/938 [02:55<02:41,  2.60it/s]
Training 1/1 epoch (loss 1.6172):  55%|█████▌    | 518/938 [02:55<02:41,  2.60it/s]
Training 1/1 epoch (loss 1.6172):  55%|█████▌    | 519/938 [02:55<02:32,  2.75it/s]
Training 1/1 epoch (loss 1.6425):  55%|█████▌    | 519/938 [02:56<02:32,  2.75it/s]
Training 1/1 epoch (loss 1.6425):  55%|█████▌    | 520/938 [02:56<02:29,  2.79it/s]
Training 1/1 epoch (loss 1.4823):  55%|█████▌    | 520/938 [02:56<02:29,  2.79it/s]
Training 1/1 epoch (loss 1.4823):  56%|█████▌    | 521/938 [02:56<02:24,  2.89it/s]
Training 1/1 epoch (loss 1.5558):  56%|█████▌    | 521/938 [02:56<02:24,  2.89it/s]
Training 1/1 epoch (loss 1.5558):  56%|█████▌    | 522/938 [02:56<02:24,  2.87it/s]
Training 1/1 epoch (loss 1.5334):  56%|█████▌    | 522/938 [02:57<02:24,  2.87it/s]
Training 1/1 epoch (loss 1.5334):  56%|█████▌    | 523/938 [02:57<02:19,  2.97it/s]
Training 1/1 epoch (loss 1.4454):  56%|█████▌    | 523/938 [02:57<02:19,  2.97it/s]
Training 1/1 epoch (loss 1.4454):  56%|█████▌    | 524/938 [02:57<02:15,  3.06it/s]
Training 1/1 epoch (loss 1.5476):  56%|█████▌    | 524/938 [02:57<02:15,  3.06it/s]
Training 1/1 epoch (loss 1.5476):  56%|█████▌    | 525/938 [02:57<02:18,  2.99it/s]
Training 1/1 epoch (loss 1.6010):  56%|█████▌    | 525/938 [02:58<02:18,  2.99it/s]
Training 1/1 epoch (loss 1.6010):  56%|█████▌    | 526/938 [02:58<02:17,  2.99it/s]
Training 1/1 epoch (loss 1.5598):  56%|█████▌    | 526/938 [02:58<02:17,  2.99it/s]
Training 1/1 epoch (loss 1.5598):  56%|█████▌    | 527/938 [02:58<02:17,  2.99it/s]
Training 1/1 epoch (loss 1.5859):  56%|█████▌    | 527/938 [02:58<02:17,  2.99it/s]
Training 1/1 epoch (loss 1.5859):  56%|█████▋    | 528/938 [02:58<02:18,  2.95it/s]
Training 1/1 epoch (loss 1.5165):  56%|█████▋    | 528/938 [02:59<02:18,  2.95it/s]
Training 1/1 epoch (loss 1.5165):  56%|█████▋    | 529/938 [02:59<02:15,  3.03it/s]
Training 1/1 epoch (loss 1.5363):  56%|█████▋    | 529/938 [02:59<02:15,  3.03it/s]
Training 1/1 epoch (loss 1.5363):  57%|█████▋    | 530/938 [02:59<02:11,  3.11it/s]
Training 1/1 epoch (loss 1.5001):  57%|█████▋    | 530/938 [02:59<02:11,  3.11it/s]
Training 1/1 epoch (loss 1.5001):  57%|█████▋    | 531/938 [02:59<02:09,  3.13it/s]
Training 1/1 epoch (loss 1.6523):  57%|█████▋    | 531/938 [03:00<02:09,  3.13it/s]
Training 1/1 epoch (loss 1.6523):  57%|█████▋    | 532/938 [03:00<02:14,  3.02it/s]
Training 1/1 epoch (loss 1.5929):  57%|█████▋    | 532/938 [03:00<02:14,  3.02it/s]
Training 1/1 epoch (loss 1.5929):  57%|█████▋    | 533/938 [03:00<02:13,  3.03it/s]
Training 1/1 epoch (loss 1.5308):  57%|█████▋    | 533/938 [03:00<02:13,  3.03it/s]
Training 1/1 epoch (loss 1.5308):  57%|█████▋    | 534/938 [03:00<02:11,  3.07it/s]
Training 1/1 epoch (loss 1.5366):  57%|█████▋    | 534/938 [03:01<02:11,  3.07it/s]
Training 1/1 epoch (loss 1.5366):  57%|█████▋    | 535/938 [03:01<02:08,  3.14it/s]
Training 1/1 epoch (loss 1.4742):  57%|█████▋    | 535/938 [03:01<02:08,  3.14it/s]
Training 1/1 epoch (loss 1.4742):  57%|█████▋    | 536/938 [03:01<02:11,  3.06it/s]
Training 1/1 epoch (loss 1.5881):  57%|█████▋    | 536/938 [03:01<02:11,  3.06it/s]
Training 1/1 epoch (loss 1.5881):  57%|█████▋    | 537/938 [03:01<02:09,  3.09it/s]
Training 1/1 epoch (loss 1.5053):  57%|█████▋    | 537/938 [03:02<02:09,  3.09it/s]
Training 1/1 epoch (loss 1.5053):  57%|█████▋    | 538/938 [03:02<02:06,  3.17it/s]
Training 1/1 epoch (loss 1.5543):  57%|█████▋    | 538/938 [03:02<02:06,  3.17it/s]
Training 1/1 epoch (loss 1.5543):  57%|█████▋    | 539/938 [03:02<02:07,  3.13it/s]
Training 1/1 epoch (loss 1.5671):  57%|█████▋    | 539/938 [03:02<02:07,  3.13it/s]
Training 1/1 epoch (loss 1.5671):  58%|█████▊    | 540/938 [03:02<02:07,  3.13it/s]
Training 1/1 epoch (loss 1.4403):  58%|█████▊    | 540/938 [03:03<02:07,  3.13it/s]
Training 1/1 epoch (loss 1.4403):  58%|█████▊    | 541/938 [03:03<02:12,  3.01it/s]
Training 1/1 epoch (loss 1.5155):  58%|█████▊    | 541/938 [03:03<02:12,  3.01it/s]
Training 1/1 epoch (loss 1.5155):  58%|█████▊    | 542/938 [03:03<02:06,  3.12it/s]
Training 1/1 epoch (loss 1.5976):  58%|█████▊    | 542/938 [03:03<02:06,  3.12it/s]
Training 1/1 epoch (loss 1.5976):  58%|█████▊    | 543/938 [03:03<02:06,  3.12it/s]
Training 1/1 epoch (loss 1.5115):  58%|█████▊    | 543/938 [03:03<02:06,  3.12it/s]
Training 1/1 epoch (loss 1.5115):  58%|█████▊    | 544/938 [03:03<02:08,  3.07it/s]
Training 1/1 epoch (loss 1.4575):  58%|█████▊    | 544/938 [03:04<02:08,  3.07it/s]
Training 1/1 epoch (loss 1.4575):  58%|█████▊    | 545/938 [03:04<02:08,  3.06it/s]
Training 1/1 epoch (loss 1.4824):  58%|█████▊    | 545/938 [03:04<02:08,  3.06it/s]
Training 1/1 epoch (loss 1.4824):  58%|█████▊    | 546/938 [03:04<02:06,  3.09it/s]
Training 1/1 epoch (loss 1.4837):  58%|█████▊    | 546/938 [03:04<02:06,  3.09it/s]
Training 1/1 epoch (loss 1.4837):  58%|█████▊    | 547/938 [03:04<02:04,  3.13it/s]
Training 1/1 epoch (loss 1.6152):  58%|█████▊    | 547/938 [03:05<02:04,  3.13it/s]
Training 1/1 epoch (loss 1.6152):  58%|█████▊    | 548/938 [03:05<02:03,  3.17it/s]
Training 1/1 epoch (loss 1.4821):  58%|█████▊    | 548/938 [03:05<02:03,  3.17it/s]
Training 1/1 epoch (loss 1.4821):  59%|█████▊    | 549/938 [03:05<02:00,  3.23it/s]
Training 1/1 epoch (loss 1.6039):  59%|█████▊    | 549/938 [03:05<02:00,  3.23it/s]
Training 1/1 epoch (loss 1.6039):  59%|█████▊    | 550/938 [03:05<01:59,  3.24it/s]
Training 1/1 epoch (loss 1.5533):  59%|█████▊    | 550/938 [03:06<01:59,  3.24it/s]
Training 1/1 epoch (loss 1.5533):  59%|█████▊    | 551/938 [03:06<02:01,  3.19it/s]
Training 1/1 epoch (loss 1.5321):  59%|█████▊    | 551/938 [03:06<02:01,  3.19it/s]
Training 1/1 epoch (loss 1.5321):  59%|█████▉    | 552/938 [03:06<02:01,  3.17it/s]
Training 1/1 epoch (loss 1.6364):  59%|█████▉    | 552/938 [03:06<02:01,  3.17it/s]
Training 1/1 epoch (loss 1.6364):  59%|█████▉    | 553/938 [03:06<02:07,  3.01it/s]
Training 1/1 epoch (loss 1.5900):  59%|█████▉    | 553/938 [03:07<02:07,  3.01it/s]
Training 1/1 epoch (loss 1.5900):  59%|█████▉    | 554/938 [03:07<02:05,  3.07it/s]
Training 1/1 epoch (loss 1.4983):  59%|█████▉    | 554/938 [03:07<02:05,  3.07it/s]
Training 1/1 epoch (loss 1.4983):  59%|█████▉    | 555/938 [03:07<02:02,  3.13it/s]
Training 1/1 epoch (loss 1.6046):  59%|█████▉    | 555/938 [03:07<02:02,  3.13it/s]
Training 1/1 epoch (loss 1.6046):  59%|█████▉    | 556/938 [03:07<02:00,  3.17it/s]
Training 1/1 epoch (loss 1.5894):  59%|█████▉    | 556/938 [03:08<02:00,  3.17it/s]
Training 1/1 epoch (loss 1.5894):  59%|█████▉    | 557/938 [03:08<02:04,  3.07it/s]
Training 1/1 epoch (loss 1.5874):  59%|█████▉    | 557/938 [03:08<02:04,  3.07it/s]
Training 1/1 epoch (loss 1.5874):  59%|█████▉    | 558/938 [03:08<02:03,  3.08it/s]
Training 1/1 epoch (loss 1.5958):  59%|█████▉    | 558/938 [03:08<02:03,  3.08it/s]
Training 1/1 epoch (loss 1.5958):  60%|█████▉    | 559/938 [03:08<02:04,  3.05it/s]
Training 1/1 epoch (loss 1.4838):  60%|█████▉    | 559/938 [03:09<02:04,  3.05it/s]
Training 1/1 epoch (loss 1.4838):  60%|█████▉    | 560/938 [03:09<02:02,  3.08it/s]
Training 1/1 epoch (loss 1.5990):  60%|█████▉    | 560/938 [03:09<02:02,  3.08it/s]
Training 1/1 epoch (loss 1.5990):  60%|█████▉    | 561/938 [03:09<02:01,  3.11it/s]
Training 1/1 epoch (loss 1.6684):  60%|█████▉    | 561/938 [03:09<02:01,  3.11it/s]
Training 1/1 epoch (loss 1.6684):  60%|█████▉    | 562/938 [03:09<01:59,  3.15it/s]
Training 1/1 epoch (loss 1.4778):  60%|█████▉    | 562/938 [03:10<01:59,  3.15it/s]
Training 1/1 epoch (loss 1.4778):  60%|██████    | 563/938 [03:10<01:56,  3.21it/s]
Training 1/1 epoch (loss 1.5886):  60%|██████    | 563/938 [03:10<01:56,  3.21it/s]
Training 1/1 epoch (loss 1.5886):  60%|██████    | 564/938 [03:10<01:58,  3.15it/s]
Training 1/1 epoch (loss 1.5525):  60%|██████    | 564/938 [03:10<01:58,  3.15it/s]
Training 1/1 epoch (loss 1.5525):  60%|██████    | 565/938 [03:10<02:01,  3.07it/s]
Training 1/1 epoch (loss 1.5454):  60%|██████    | 565/938 [03:11<02:01,  3.07it/s]
Training 1/1 epoch (loss 1.5454):  60%|██████    | 566/938 [03:11<01:58,  3.15it/s]
Training 1/1 epoch (loss 1.5654):  60%|██████    | 566/938 [03:11<01:58,  3.15it/s]
Training 1/1 epoch (loss 1.5654):  60%|██████    | 567/938 [03:11<01:56,  3.20it/s]
Training 1/1 epoch (loss 1.6219):  60%|██████    | 567/938 [03:11<01:56,  3.20it/s]
Training 1/1 epoch (loss 1.6219):  61%|██████    | 568/938 [03:11<01:57,  3.16it/s]
Training 1/1 epoch (loss 1.6935):  61%|██████    | 568/938 [03:12<01:57,  3.16it/s]
Training 1/1 epoch (loss 1.6935):  61%|██████    | 569/938 [03:12<02:05,  2.95it/s]
Training 1/1 epoch (loss 1.6091):  61%|██████    | 569/938 [03:12<02:05,  2.95it/s]
Training 1/1 epoch (loss 1.6091):  61%|██████    | 570/938 [03:12<01:59,  3.07it/s]
Training 1/1 epoch (loss 1.6279):  61%|██████    | 570/938 [03:12<01:59,  3.07it/s]
Training 1/1 epoch (loss 1.6279):  61%|██████    | 571/938 [03:12<02:02,  2.99it/s]
Training 1/1 epoch (loss 1.6155):  61%|██████    | 571/938 [03:12<02:02,  2.99it/s]
Training 1/1 epoch (loss 1.6155):  61%|██████    | 572/938 [03:12<01:56,  3.15it/s]
Training 1/1 epoch (loss 1.6044):  61%|██████    | 572/938 [03:13<01:56,  3.15it/s]
Training 1/1 epoch (loss 1.6044):  61%|██████    | 573/938 [03:13<01:58,  3.08it/s]
Training 1/1 epoch (loss 1.5567):  61%|██████    | 573/938 [03:13<01:58,  3.08it/s]
Training 1/1 epoch (loss 1.5567):  61%|██████    | 574/938 [03:13<02:01,  3.01it/s]
Training 1/1 epoch (loss 1.5404):  61%|██████    | 574/938 [03:13<02:01,  3.01it/s]
Training 1/1 epoch (loss 1.5404):  61%|██████▏   | 575/938 [03:13<01:59,  3.03it/s]
Training 1/1 epoch (loss 1.5661):  61%|██████▏   | 575/938 [03:14<01:59,  3.03it/s]
Training 1/1 epoch (loss 1.5661):  61%|██████▏   | 576/938 [03:14<02:00,  3.00it/s]
Training 1/1 epoch (loss 1.5116):  61%|██████▏   | 576/938 [03:14<02:00,  3.00it/s]
Training 1/1 epoch (loss 1.5116):  62%|██████▏   | 577/938 [03:14<01:59,  3.02it/s]
Training 1/1 epoch (loss 1.6470):  62%|██████▏   | 577/938 [03:14<01:59,  3.02it/s]
Training 1/1 epoch (loss 1.6470):  62%|██████▏   | 578/938 [03:14<01:58,  3.04it/s]
Training 1/1 epoch (loss 1.6173):  62%|██████▏   | 578/938 [03:15<01:58,  3.04it/s]
Training 1/1 epoch (loss 1.6173):  62%|██████▏   | 579/938 [03:15<01:54,  3.13it/s]
Training 1/1 epoch (loss 1.5724):  62%|██████▏   | 579/938 [03:15<01:54,  3.13it/s]
Training 1/1 epoch (loss 1.5724):  62%|██████▏   | 580/938 [03:15<01:52,  3.18it/s]
Training 1/1 epoch (loss 1.5508):  62%|██████▏   | 580/938 [03:15<01:52,  3.18it/s]
Training 1/1 epoch (loss 1.5508):  62%|██████▏   | 581/938 [03:15<01:50,  3.24it/s]
Training 1/1 epoch (loss 1.4746):  62%|██████▏   | 581/938 [03:16<01:50,  3.24it/s]
Training 1/1 epoch (loss 1.4746):  62%|██████▏   | 582/938 [03:16<01:50,  3.23it/s]
Training 1/1 epoch (loss 1.6138):  62%|██████▏   | 582/938 [03:16<01:50,  3.23it/s]
Training 1/1 epoch (loss 1.6138):  62%|██████▏   | 583/938 [03:16<01:52,  3.17it/s]
Training 1/1 epoch (loss 1.4368):  62%|██████▏   | 583/938 [03:16<01:52,  3.17it/s]
Training 1/1 epoch (loss 1.4368):  62%|██████▏   | 584/938 [03:16<01:54,  3.09it/s]
Training 1/1 epoch (loss 1.4547):  62%|██████▏   | 584/938 [03:17<01:54,  3.09it/s]
Training 1/1 epoch (loss 1.4547):  62%|██████▏   | 585/938 [03:17<01:53,  3.12it/s]
Training 1/1 epoch (loss 1.6004):  62%|██████▏   | 585/938 [03:17<01:53,  3.12it/s]
Training 1/1 epoch (loss 1.6004):  62%|██████▏   | 586/938 [03:17<01:51,  3.17it/s]
Training 1/1 epoch (loss 1.5961):  62%|██████▏   | 586/938 [03:17<01:51,  3.17it/s]
Training 1/1 epoch (loss 1.5961):  63%|██████▎   | 587/938 [03:17<01:49,  3.21it/s]
Training 1/1 epoch (loss 1.5942):  63%|██████▎   | 587/938 [03:18<01:49,  3.21it/s]
Training 1/1 epoch (loss 1.5942):  63%|██████▎   | 588/938 [03:18<01:47,  3.26it/s]
Training 1/1 epoch (loss 1.6173):  63%|██████▎   | 588/938 [03:18<01:47,  3.26it/s]
Training 1/1 epoch (loss 1.6173):  63%|██████▎   | 589/938 [03:18<01:53,  3.07it/s]
Training 1/1 epoch (loss 1.6200):  63%|██████▎   | 589/938 [03:18<01:53,  3.07it/s]
Training 1/1 epoch (loss 1.6200):  63%|██████▎   | 590/938 [03:18<01:56,  2.97it/s]
Training 1/1 epoch (loss 1.5579):  63%|██████▎   | 590/938 [03:19<01:56,  2.97it/s]
Training 1/1 epoch (loss 1.5579):  63%|██████▎   | 591/938 [03:19<01:52,  3.08it/s]
Training 1/1 epoch (loss 1.4550):  63%|██████▎   | 591/938 [03:19<01:52,  3.08it/s]
Training 1/1 epoch (loss 1.4550):  63%|██████▎   | 592/938 [03:19<01:52,  3.08it/s]
Training 1/1 epoch (loss 1.6276):  63%|██████▎   | 592/938 [03:19<01:52,  3.08it/s]
Training 1/1 epoch (loss 1.6276):  63%|██████▎   | 593/938 [03:19<01:50,  3.12it/s]
Training 1/1 epoch (loss 1.5924):  63%|██████▎   | 593/938 [03:20<01:50,  3.12it/s]
Training 1/1 epoch (loss 1.5924):  63%|██████▎   | 594/938 [03:20<01:50,  3.11it/s]
Training 1/1 epoch (loss 1.5105):  63%|██████▎   | 594/938 [03:20<01:50,  3.11it/s]
Training 1/1 epoch (loss 1.5105):  63%|██████▎   | 595/938 [03:20<01:49,  3.13it/s]
Training 1/1 epoch (loss 1.6402):  63%|██████▎   | 595/938 [03:20<01:49,  3.13it/s]
Training 1/1 epoch (loss 1.6402):  64%|██████▎   | 596/938 [03:20<01:52,  3.05it/s]
Training 1/1 epoch (loss 1.5250):  64%|██████▎   | 596/938 [03:21<01:52,  3.05it/s]
Training 1/1 epoch (loss 1.5250):  64%|██████▎   | 597/938 [03:21<01:50,  3.10it/s]
Training 1/1 epoch (loss 1.6509):  64%|██████▎   | 597/938 [03:21<01:50,  3.10it/s]
Training 1/1 epoch (loss 1.6509):  64%|██████▍   | 598/938 [03:21<01:47,  3.16it/s]
Training 1/1 epoch (loss 1.6092):  64%|██████▍   | 598/938 [03:21<01:47,  3.16it/s]
Training 1/1 epoch (loss 1.6092):  64%|██████▍   | 599/938 [03:21<01:46,  3.19it/s]
Training 1/1 epoch (loss 1.5281):  64%|██████▍   | 599/938 [03:21<01:46,  3.19it/s]
Training 1/1 epoch (loss 1.5281):  64%|██████▍   | 600/938 [03:21<01:46,  3.18it/s]
Training 1/1 epoch (loss 1.6105):  64%|██████▍   | 600/938 [03:22<01:46,  3.18it/s]
Training 1/1 epoch (loss 1.6105):  64%|██████▍   | 601/938 [03:22<01:48,  3.09it/s]
Training 1/1 epoch (loss 1.4614):  64%|██████▍   | 601/938 [03:22<01:48,  3.09it/s]
Training 1/1 epoch (loss 1.4614):  64%|██████▍   | 602/938 [03:22<01:49,  3.08it/s]
Training 1/1 epoch (loss 1.5772):  64%|██████▍   | 602/938 [03:22<01:49,  3.08it/s]
Training 1/1 epoch (loss 1.5772):  64%|██████▍   | 603/938 [03:22<01:47,  3.11it/s]
Training 1/1 epoch (loss 1.6313):  64%|██████▍   | 603/938 [03:23<01:47,  3.11it/s]
Training 1/1 epoch (loss 1.6313):  64%|██████▍   | 604/938 [03:23<01:44,  3.19it/s]
Training 1/1 epoch (loss 1.5476):  64%|██████▍   | 604/938 [03:23<01:44,  3.19it/s]
Training 1/1 epoch (loss 1.5476):  64%|██████▍   | 605/938 [03:23<01:43,  3.21it/s]
Training 1/1 epoch (loss 1.6078):  64%|██████▍   | 605/938 [03:23<01:43,  3.21it/s]
Training 1/1 epoch (loss 1.6078):  65%|██████▍   | 606/938 [03:23<01:44,  3.18it/s]
Training 1/1 epoch (loss 1.5457):  65%|██████▍   | 606/938 [03:24<01:44,  3.18it/s]
Training 1/1 epoch (loss 1.5457):  65%|██████▍   | 607/938 [03:24<01:49,  3.02it/s]
Training 1/1 epoch (loss 1.6533):  65%|██████▍   | 607/938 [03:24<01:49,  3.02it/s]
Training 1/1 epoch (loss 1.6533):  65%|██████▍   | 608/938 [03:24<01:49,  3.01it/s]
Training 1/1 epoch (loss 1.5588):  65%|██████▍   | 608/938 [03:24<01:49,  3.01it/s]
Training 1/1 epoch (loss 1.5588):  65%|██████▍   | 609/938 [03:24<01:49,  3.00it/s]
Training 1/1 epoch (loss 1.5401):  65%|██████▍   | 609/938 [03:25<01:49,  3.00it/s]
Training 1/1 epoch (loss 1.5401):  65%|██████▌   | 610/938 [03:25<01:46,  3.09it/s]
Training 1/1 epoch (loss 1.6145):  65%|██████▌   | 610/938 [03:25<01:46,  3.09it/s]
Training 1/1 epoch (loss 1.6145):  65%|██████▌   | 611/938 [03:25<01:44,  3.13it/s]
Training 1/1 epoch (loss 1.5095):  65%|██████▌   | 611/938 [03:25<01:44,  3.13it/s]
Training 1/1 epoch (loss 1.5095):  65%|██████▌   | 612/938 [03:25<01:42,  3.18it/s]
Training 1/1 epoch (loss 1.3935):  65%|██████▌   | 612/938 [03:26<01:42,  3.18it/s]
Training 1/1 epoch (loss 1.3935):  65%|██████▌   | 613/938 [03:26<01:45,  3.09it/s]
Training 1/1 epoch (loss 1.6739):  65%|██████▌   | 613/938 [03:26<01:45,  3.09it/s]
Training 1/1 epoch (loss 1.6739):  65%|██████▌   | 614/938 [03:26<01:44,  3.10it/s]
Training 1/1 epoch (loss 1.4317):  65%|██████▌   | 614/938 [03:26<01:44,  3.10it/s]
Training 1/1 epoch (loss 1.4317):  66%|██████▌   | 615/938 [03:26<01:43,  3.14it/s]
Training 1/1 epoch (loss 1.5907):  66%|██████▌   | 615/938 [03:27<01:43,  3.14it/s]
Training 1/1 epoch (loss 1.5907):  66%|██████▌   | 616/938 [03:27<01:43,  3.12it/s]
Training 1/1 epoch (loss 1.4520):  66%|██████▌   | 616/938 [03:27<01:43,  3.12it/s]
Training 1/1 epoch (loss 1.4520):  66%|██████▌   | 617/938 [03:27<01:41,  3.15it/s]
Training 1/1 epoch (loss 1.5291):  66%|██████▌   | 617/938 [03:27<01:41,  3.15it/s]
Training 1/1 epoch (loss 1.5291):  66%|██████▌   | 618/938 [03:27<01:39,  3.22it/s]
Training 1/1 epoch (loss 1.5793):  66%|██████▌   | 618/938 [03:28<01:39,  3.22it/s]
Training 1/1 epoch (loss 1.5793):  66%|██████▌   | 619/938 [03:28<01:41,  3.14it/s]
Training 1/1 epoch (loss 1.4663):  66%|██████▌   | 619/938 [03:28<01:41,  3.14it/s]
Training 1/1 epoch (loss 1.4663):  66%|██████▌   | 620/938 [03:28<01:39,  3.20it/s]
Training 1/1 epoch (loss 1.4455):  66%|██████▌   | 620/938 [03:28<01:39,  3.20it/s]
Training 1/1 epoch (loss 1.4455):  66%|██████▌   | 621/938 [03:28<01:40,  3.15it/s]
Training 1/1 epoch (loss 1.6780):  66%|██████▌   | 621/938 [03:28<01:40,  3.15it/s]
Training 1/1 epoch (loss 1.6780):  66%|██████▋   | 622/938 [03:28<01:39,  3.18it/s]
Training 1/1 epoch (loss 1.4908):  66%|██████▋   | 622/938 [03:29<01:39,  3.18it/s]
Training 1/1 epoch (loss 1.4908):  66%|██████▋   | 623/938 [03:29<01:41,  3.11it/s]
Training 1/1 epoch (loss 1.4927):  66%|██████▋   | 623/938 [03:29<01:41,  3.11it/s]
Training 1/1 epoch (loss 1.4927):  67%|██████▋   | 624/938 [03:29<01:39,  3.14it/s]
Training 1/1 epoch (loss 1.4076):  67%|██████▋   | 624/938 [03:29<01:39,  3.14it/s]
Training 1/1 epoch (loss 1.4076):  67%|██████▋   | 625/938 [03:29<01:40,  3.10it/s]
Training 1/1 epoch (loss 1.5538):  67%|██████▋   | 625/938 [03:30<01:40,  3.10it/s]
Training 1/1 epoch (loss 1.5538):  67%|██████▋   | 626/938 [03:30<01:37,  3.20it/s]
Training 1/1 epoch (loss 1.5698):  67%|██████▋   | 626/938 [03:30<01:37,  3.20it/s]
Training 1/1 epoch (loss 1.5698):  67%|██████▋   | 627/938 [03:30<01:38,  3.15it/s]
Training 1/1 epoch (loss 1.5335):  67%|██████▋   | 627/938 [03:30<01:38,  3.15it/s]
Training 1/1 epoch (loss 1.5335):  67%|██████▋   | 628/938 [03:30<01:37,  3.17it/s]
Training 1/1 epoch (loss 1.5108):  67%|██████▋   | 628/938 [03:31<01:37,  3.17it/s]
Training 1/1 epoch (loss 1.5108):  67%|██████▋   | 629/938 [03:31<01:35,  3.24it/s]
Training 1/1 epoch (loss 1.5615):  67%|██████▋   | 629/938 [03:31<01:35,  3.24it/s]
Training 1/1 epoch (loss 1.5615):  67%|██████▋   | 630/938 [03:31<01:35,  3.23it/s]
Training 1/1 epoch (loss 1.4929):  67%|██████▋   | 630/938 [03:31<01:35,  3.23it/s]
Training 1/1 epoch (loss 1.4929):  67%|██████▋   | 631/938 [03:31<01:33,  3.28it/s]
Training 1/1 epoch (loss 1.6252):  67%|██████▋   | 631/938 [03:32<01:33,  3.28it/s]
Training 1/1 epoch (loss 1.6252):  67%|██████▋   | 632/938 [03:32<01:37,  3.14it/s]
Training 1/1 epoch (loss 1.4273):  67%|██████▋   | 632/938 [03:32<01:37,  3.14it/s]
Training 1/1 epoch (loss 1.4273):  67%|██████▋   | 633/938 [03:32<01:36,  3.15it/s]
Training 1/1 epoch (loss 1.5811):  67%|██████▋   | 633/938 [03:32<01:36,  3.15it/s]
Training 1/1 epoch (loss 1.5811):  68%|██████▊   | 634/938 [03:32<01:38,  3.10it/s]
Training 1/1 epoch (loss 1.5471):  68%|██████▊   | 634/938 [03:33<01:38,  3.10it/s]
Training 1/1 epoch (loss 1.5471):  68%|██████▊   | 635/938 [03:33<01:35,  3.17it/s]
Training 1/1 epoch (loss 1.4875):  68%|██████▊   | 635/938 [03:33<01:35,  3.17it/s]
Training 1/1 epoch (loss 1.4875):  68%|██████▊   | 636/938 [03:33<01:33,  3.24it/s]
Training 1/1 epoch (loss 1.6209):  68%|██████▊   | 636/938 [03:33<01:33,  3.24it/s]
Training 1/1 epoch (loss 1.6209):  68%|██████▊   | 637/938 [03:33<01:33,  3.22it/s]
Training 1/1 epoch (loss 1.5611):  68%|██████▊   | 637/938 [03:34<01:33,  3.22it/s]
Training 1/1 epoch (loss 1.5611):  68%|██████▊   | 638/938 [03:34<01:33,  3.22it/s]
Training 1/1 epoch (loss 1.4530):  68%|██████▊   | 638/938 [03:34<01:33,  3.22it/s]
Training 1/1 epoch (loss 1.4530):  68%|██████▊   | 639/938 [03:34<01:45,  2.85it/s]
Training 1/1 epoch (loss 1.5948):  68%|██████▊   | 639/938 [03:34<01:45,  2.85it/s]
Training 1/1 epoch (loss 1.5948):  68%|██████▊   | 640/938 [03:34<01:45,  2.83it/s]
Training 1/1 epoch (loss 1.4304):  68%|██████▊   | 640/938 [03:35<01:45,  2.83it/s]
Training 1/1 epoch (loss 1.4304):  68%|██████▊   | 641/938 [03:35<01:40,  2.95it/s]
Training 1/1 epoch (loss 1.4964):  68%|██████▊   | 641/938 [03:35<01:40,  2.95it/s]
Training 1/1 epoch (loss 1.4964):  68%|██████▊   | 642/938 [03:35<01:38,  3.01it/s]
Training 1/1 epoch (loss 1.6106):  68%|██████▊   | 642/938 [03:35<01:38,  3.01it/s]
Training 1/1 epoch (loss 1.6106):  69%|██████▊   | 643/938 [03:35<01:34,  3.11it/s]
Training 1/1 epoch (loss 1.4385):  69%|██████▊   | 643/938 [03:36<01:34,  3.11it/s]
Training 1/1 epoch (loss 1.4385):  69%|██████▊   | 644/938 [03:36<01:34,  3.11it/s]
Training 1/1 epoch (loss 1.5954):  69%|██████▊   | 644/938 [03:36<01:34,  3.11it/s]
Training 1/1 epoch (loss 1.5954):  69%|██████▉   | 645/938 [03:36<01:33,  3.12it/s]
Training 1/1 epoch (loss 1.6235):  69%|██████▉   | 645/938 [03:36<01:33,  3.12it/s]
Training 1/1 epoch (loss 1.6235):  69%|██████▉   | 646/938 [03:36<01:33,  3.11it/s]
Training 1/1 epoch (loss 1.6103):  69%|██████▉   | 646/938 [03:36<01:33,  3.11it/s]
Training 1/1 epoch (loss 1.6103):  69%|██████▉   | 647/938 [03:36<01:31,  3.17it/s]
Training 1/1 epoch (loss 1.6081):  69%|██████▉   | 647/938 [03:37<01:31,  3.17it/s]
Training 1/1 epoch (loss 1.6081):  69%|██████▉   | 648/938 [03:37<01:33,  3.11it/s]
Training 1/1 epoch (loss 1.5501):  69%|██████▉   | 648/938 [03:37<01:33,  3.11it/s]
Training 1/1 epoch (loss 1.5501):  69%|██████▉   | 649/938 [03:37<01:31,  3.16it/s]
Training 1/1 epoch (loss 1.5011):  69%|██████▉   | 649/938 [03:37<01:31,  3.16it/s]
Training 1/1 epoch (loss 1.5011):  69%|██████▉   | 650/938 [03:37<01:31,  3.16it/s]
Training 1/1 epoch (loss 1.5149):  69%|██████▉   | 650/938 [03:38<01:31,  3.16it/s]
Training 1/1 epoch (loss 1.5149):  69%|██████▉   | 651/938 [03:38<01:30,  3.17it/s]
Training 1/1 epoch (loss 1.5174):  69%|██████▉   | 651/938 [03:38<01:30,  3.17it/s]
Training 1/1 epoch (loss 1.5174):  70%|██████▉   | 652/938 [03:38<01:30,  3.18it/s]
Training 1/1 epoch (loss 1.6725):  70%|██████▉   | 652/938 [03:38<01:30,  3.18it/s]
Training 1/1 epoch (loss 1.6725):  70%|██████▉   | 653/938 [03:38<01:30,  3.13it/s]
Training 1/1 epoch (loss 1.6095):  70%|██████▉   | 653/938 [03:39<01:30,  3.13it/s]
Training 1/1 epoch (loss 1.6095):  70%|██████▉   | 654/938 [03:39<01:28,  3.20it/s]
Training 1/1 epoch (loss 1.5075):  70%|██████▉   | 654/938 [03:39<01:28,  3.20it/s]
Training 1/1 epoch (loss 1.5075):  70%|██████▉   | 655/938 [03:39<01:27,  3.25it/s]
Training 1/1 epoch (loss 1.5683):  70%|██████▉   | 655/938 [03:39<01:27,  3.25it/s]
Training 1/1 epoch (loss 1.5683):  70%|██████▉   | 656/938 [03:39<01:29,  3.15it/s]
Training 1/1 epoch (loss 1.6426):  70%|██████▉   | 656/938 [03:40<01:29,  3.15it/s]
Training 1/1 epoch (loss 1.6426):  70%|███████   | 657/938 [03:40<01:30,  3.12it/s]
Training 1/1 epoch (loss 1.5477):  70%|███████   | 657/938 [03:40<01:30,  3.12it/s]
Training 1/1 epoch (loss 1.5477):  70%|███████   | 658/938 [03:40<01:33,  2.99it/s]
Training 1/1 epoch (loss 1.6125):  70%|███████   | 658/938 [03:40<01:33,  2.99it/s]
Training 1/1 epoch (loss 1.6125):  70%|███████   | 659/938 [03:40<01:31,  3.04it/s]
Training 1/1 epoch (loss 1.5697):  70%|███████   | 659/938 [03:41<01:31,  3.04it/s]
Training 1/1 epoch (loss 1.5697):  70%|███████   | 660/938 [03:41<01:29,  3.10it/s]
Training 1/1 epoch (loss 1.6092):  70%|███████   | 660/938 [03:41<01:29,  3.10it/s]
Training 1/1 epoch (loss 1.6092):  70%|███████   | 661/938 [03:41<01:27,  3.15it/s]
Training 1/1 epoch (loss 1.5572):  70%|███████   | 661/938 [03:41<01:27,  3.15it/s]
Training 1/1 epoch (loss 1.5572):  71%|███████   | 662/938 [03:41<01:24,  3.27it/s]
Training 1/1 epoch (loss 1.4584):  71%|███████   | 662/938 [03:42<01:24,  3.27it/s]
Training 1/1 epoch (loss 1.4584):  71%|███████   | 663/938 [03:42<01:31,  3.01it/s]
Training 1/1 epoch (loss 1.5627):  71%|███████   | 663/938 [03:42<01:31,  3.01it/s]
Training 1/1 epoch (loss 1.5627):  71%|███████   | 664/938 [03:42<01:30,  3.01it/s]
Training 1/1 epoch (loss 1.5734):  71%|███████   | 664/938 [03:42<01:30,  3.01it/s]
Training 1/1 epoch (loss 1.5734):  71%|███████   | 665/938 [03:42<01:33,  2.93it/s]
Training 1/1 epoch (loss 1.5088):  71%|███████   | 665/938 [03:43<01:33,  2.93it/s]
Training 1/1 epoch (loss 1.5088):  71%|███████   | 666/938 [03:43<01:32,  2.96it/s]
Training 1/1 epoch (loss 1.5149):  71%|███████   | 666/938 [03:43<01:32,  2.96it/s]
Training 1/1 epoch (loss 1.5149):  71%|███████   | 667/938 [03:43<01:28,  3.06it/s]
Training 1/1 epoch (loss 1.4213):  71%|███████   | 667/938 [03:43<01:28,  3.06it/s]
Training 1/1 epoch (loss 1.4213):  71%|███████   | 668/938 [03:43<01:26,  3.10it/s]
Training 1/1 epoch (loss 1.4748):  71%|███████   | 668/938 [03:44<01:26,  3.10it/s]
Training 1/1 epoch (loss 1.4748):  71%|███████▏  | 669/938 [03:44<01:27,  3.07it/s]
Training 1/1 epoch (loss 1.4353):  71%|███████▏  | 669/938 [03:44<01:27,  3.07it/s]
Training 1/1 epoch (loss 1.4353):  71%|███████▏  | 670/938 [03:44<01:29,  2.98it/s]
Training 1/1 epoch (loss 1.6206):  71%|███████▏  | 670/938 [03:44<01:29,  2.98it/s]
Training 1/1 epoch (loss 1.6206):  72%|███████▏  | 671/938 [03:44<01:32,  2.89it/s]
Training 1/1 epoch (loss 1.7323):  72%|███████▏  | 671/938 [03:45<01:32,  2.89it/s]
Training 1/1 epoch (loss 1.7323):  72%|███████▏  | 672/938 [03:45<01:35,  2.78it/s]
Training 1/1 epoch (loss 1.5270):  72%|███████▏  | 672/938 [03:45<01:35,  2.78it/s]
Training 1/1 epoch (loss 1.5270):  72%|███████▏  | 673/938 [03:45<01:30,  2.93it/s]
Training 1/1 epoch (loss 1.5611):  72%|███████▏  | 673/938 [03:45<01:30,  2.93it/s]
Training 1/1 epoch (loss 1.5611):  72%|███████▏  | 674/938 [03:45<01:26,  3.05it/s]
Training 1/1 epoch (loss 1.5257):  72%|███████▏  | 674/938 [03:46<01:26,  3.05it/s]
Training 1/1 epoch (loss 1.5257):  72%|███████▏  | 675/938 [03:46<01:24,  3.12it/s]
Training 1/1 epoch (loss 1.5568):  72%|███████▏  | 675/938 [03:46<01:24,  3.12it/s]
Training 1/1 epoch (loss 1.5568):  72%|███████▏  | 676/938 [03:46<01:24,  3.09it/s]
Training 1/1 epoch (loss 1.6458):  72%|███████▏  | 676/938 [03:46<01:24,  3.09it/s]
Training 1/1 epoch (loss 1.6458):  72%|███████▏  | 677/938 [03:46<01:24,  3.09it/s]
Training 1/1 epoch (loss 1.4791):  72%|███████▏  | 677/938 [03:47<01:24,  3.09it/s]
Training 1/1 epoch (loss 1.4791):  72%|███████▏  | 678/938 [03:47<01:23,  3.11it/s]
Training 1/1 epoch (loss 1.5308):  72%|███████▏  | 678/938 [03:47<01:23,  3.11it/s]
Training 1/1 epoch (loss 1.5308):  72%|███████▏  | 679/938 [03:47<01:22,  3.15it/s]
Training 1/1 epoch (loss 1.5858):  72%|███████▏  | 679/938 [03:47<01:22,  3.15it/s]
Training 1/1 epoch (loss 1.5858):  72%|███████▏  | 680/938 [03:47<01:23,  3.09it/s]
Training 1/1 epoch (loss 1.5595):  72%|███████▏  | 680/938 [03:48<01:23,  3.09it/s]
Training 1/1 epoch (loss 1.5595):  73%|███████▎  | 681/938 [03:48<01:24,  3.05it/s]
Training 1/1 epoch (loss 1.5776):  73%|███████▎  | 681/938 [03:48<01:24,  3.05it/s]
Training 1/1 epoch (loss 1.5776):  73%|███████▎  | 682/938 [03:48<01:24,  3.04it/s]
Training 1/1 epoch (loss 1.6080):  73%|███████▎  | 682/938 [03:48<01:24,  3.04it/s]
Training 1/1 epoch (loss 1.6080):  73%|███████▎  | 683/938 [03:48<01:25,  2.97it/s]
Training 1/1 epoch (loss 1.5678):  73%|███████▎  | 683/938 [03:49<01:25,  2.97it/s]
Training 1/1 epoch (loss 1.5678):  73%|███████▎  | 684/938 [03:49<01:23,  3.03it/s]
Training 1/1 epoch (loss 1.5794):  73%|███████▎  | 684/938 [03:49<01:23,  3.03it/s]
Training 1/1 epoch (loss 1.5794):  73%|███████▎  | 685/938 [03:49<01:21,  3.11it/s]
Training 1/1 epoch (loss 1.6389):  73%|███████▎  | 685/938 [03:49<01:21,  3.11it/s]
Training 1/1 epoch (loss 1.6389):  73%|███████▎  | 686/938 [03:49<01:19,  3.15it/s]
Training 1/1 epoch (loss 1.5603):  73%|███████▎  | 686/938 [03:50<01:19,  3.15it/s]
Training 1/1 epoch (loss 1.5603):  73%|███████▎  | 687/938 [03:50<01:25,  2.95it/s]
Training 1/1 epoch (loss 1.5706):  73%|███████▎  | 687/938 [03:50<01:25,  2.95it/s]
Training 1/1 epoch (loss 1.5706):  73%|███████▎  | 688/938 [03:50<01:32,  2.70it/s]
Training 1/1 epoch (loss 1.5967):  73%|███████▎  | 688/938 [03:50<01:32,  2.70it/s]
Training 1/1 epoch (loss 1.5967):  73%|███████▎  | 689/938 [03:50<01:29,  2.77it/s]
Training 1/1 epoch (loss 1.6062):  73%|███████▎  | 689/938 [03:51<01:29,  2.77it/s]
Training 1/1 epoch (loss 1.6062):  74%|███████▎  | 690/938 [03:51<01:25,  2.90it/s]
Training 1/1 epoch (loss 1.6382):  74%|███████▎  | 690/938 [03:51<01:25,  2.90it/s]
Training 1/1 epoch (loss 1.6382):  74%|███████▎  | 691/938 [03:51<01:23,  2.96it/s]
Training 1/1 epoch (loss 1.5799):  74%|███████▎  | 691/938 [03:51<01:23,  2.96it/s]
Training 1/1 epoch (loss 1.5799):  74%|███████▍  | 692/938 [03:51<01:21,  3.00it/s]
Training 1/1 epoch (loss 1.5219):  74%|███████▍  | 692/938 [03:52<01:21,  3.00it/s]
Training 1/1 epoch (loss 1.5219):  74%|███████▍  | 693/938 [03:52<01:21,  2.99it/s]
Training 1/1 epoch (loss 1.4773):  74%|███████▍  | 693/938 [03:52<01:21,  2.99it/s]
Training 1/1 epoch (loss 1.4773):  74%|███████▍  | 694/938 [03:52<01:22,  2.95it/s]
Training 1/1 epoch (loss 1.5627):  74%|███████▍  | 694/938 [03:52<01:22,  2.95it/s]
Training 1/1 epoch (loss 1.5627):  74%|███████▍  | 695/938 [03:52<01:22,  2.96it/s]
Training 1/1 epoch (loss 1.3962):  74%|███████▍  | 695/938 [03:53<01:22,  2.96it/s]
Training 1/1 epoch (loss 1.3962):  74%|███████▍  | 696/938 [03:53<01:19,  3.04it/s]
Training 1/1 epoch (loss 1.6194):  74%|███████▍  | 696/938 [03:53<01:19,  3.04it/s]
Training 1/1 epoch (loss 1.6194):  74%|███████▍  | 697/938 [03:53<01:18,  3.07it/s]
Training 1/1 epoch (loss 1.5220):  74%|███████▍  | 697/938 [03:53<01:18,  3.07it/s]
Training 1/1 epoch (loss 1.5220):  74%|███████▍  | 698/938 [03:53<01:16,  3.15it/s]
Training 1/1 epoch (loss 1.4612):  74%|███████▍  | 698/938 [03:54<01:16,  3.15it/s]
Training 1/1 epoch (loss 1.4612):  75%|███████▍  | 699/938 [03:54<01:17,  3.07it/s]
Training 1/1 epoch (loss 1.5786):  75%|███████▍  | 699/938 [03:54<01:17,  3.07it/s]
Training 1/1 epoch (loss 1.5786):  75%|███████▍  | 700/938 [03:54<01:20,  2.95it/s]
Training 1/1 epoch (loss 1.4870):  75%|███████▍  | 700/938 [03:54<01:20,  2.95it/s]
Training 1/1 epoch (loss 1.4870):  75%|███████▍  | 701/938 [03:54<01:24,  2.81it/s]
Training 1/1 epoch (loss 1.4649):  75%|███████▍  | 701/938 [03:55<01:24,  2.81it/s]
Training 1/1 epoch (loss 1.4649):  75%|███████▍  | 702/938 [03:55<01:20,  2.92it/s]
Training 1/1 epoch (loss 1.5970):  75%|███████▍  | 702/938 [03:55<01:20,  2.92it/s]
Training 1/1 epoch (loss 1.5970):  75%|███████▍  | 703/938 [03:55<01:20,  2.93it/s]
Training 1/1 epoch (loss 1.4781):  75%|███████▍  | 703/938 [03:55<01:20,  2.93it/s]
Training 1/1 epoch (loss 1.4781):  75%|███████▌  | 704/938 [03:55<01:18,  2.96it/s]
Training 1/1 epoch (loss 1.5448):  75%|███████▌  | 704/938 [03:56<01:18,  2.96it/s]
Training 1/1 epoch (loss 1.5448):  75%|███████▌  | 705/938 [03:56<01:19,  2.94it/s]
Training 1/1 epoch (loss 1.4522):  75%|███████▌  | 705/938 [03:56<01:19,  2.94it/s]
Training 1/1 epoch (loss 1.4522):  75%|███████▌  | 706/938 [03:56<01:17,  3.01it/s]
Training 1/1 epoch (loss 1.5497):  75%|███████▌  | 706/938 [03:56<01:17,  3.01it/s]
Training 1/1 epoch (loss 1.5497):  75%|███████▌  | 707/938 [03:56<01:16,  3.00it/s]
Training 1/1 epoch (loss 1.5887):  75%|███████▌  | 707/938 [03:57<01:16,  3.00it/s]
Training 1/1 epoch (loss 1.5887):  75%|███████▌  | 708/938 [03:57<01:17,  2.98it/s]
Training 1/1 epoch (loss 1.6578):  75%|███████▌  | 708/938 [03:57<01:17,  2.98it/s]
Training 1/1 epoch (loss 1.6578):  76%|███████▌  | 709/938 [03:57<01:14,  3.06it/s]
Training 1/1 epoch (loss 1.5301):  76%|███████▌  | 709/938 [03:57<01:14,  3.06it/s]
Training 1/1 epoch (loss 1.5301):  76%|███████▌  | 710/938 [03:57<01:12,  3.15it/s]
Training 1/1 epoch (loss 1.5908):  76%|███████▌  | 710/938 [03:58<01:12,  3.15it/s]
Training 1/1 epoch (loss 1.5908):  76%|███████▌  | 711/938 [03:58<01:12,  3.14it/s]
Training 1/1 epoch (loss 1.5124):  76%|███████▌  | 711/938 [03:58<01:12,  3.14it/s]
Training 1/1 epoch (loss 1.5124):  76%|███████▌  | 712/938 [03:58<01:14,  3.01it/s]
Training 1/1 epoch (loss 1.5890):  76%|███████▌  | 712/938 [03:58<01:14,  3.01it/s]
Training 1/1 epoch (loss 1.5890):  76%|███████▌  | 713/938 [03:58<01:13,  3.07it/s]
Training 1/1 epoch (loss 1.5139):  76%|███████▌  | 713/938 [03:59<01:13,  3.07it/s]
Training 1/1 epoch (loss 1.5139):  76%|███████▌  | 714/938 [03:59<01:11,  3.13it/s]
Training 1/1 epoch (loss 1.5305):  76%|███████▌  | 714/938 [03:59<01:11,  3.13it/s]
Training 1/1 epoch (loss 1.5305):  76%|███████▌  | 715/938 [03:59<01:10,  3.15it/s]
Training 1/1 epoch (loss 1.5059):  76%|███████▌  | 715/938 [03:59<01:10,  3.15it/s]
Training 1/1 epoch (loss 1.5059):  76%|███████▋  | 716/938 [03:59<01:09,  3.21it/s]
Training 1/1 epoch (loss 1.5183):  76%|███████▋  | 716/938 [04:00<01:09,  3.21it/s]
Training 1/1 epoch (loss 1.5183):  76%|███████▋  | 717/938 [04:00<01:14,  2.97it/s]
Training 1/1 epoch (loss 1.5360):  76%|███████▋  | 717/938 [04:00<01:14,  2.97it/s]
Training 1/1 epoch (loss 1.5360):  77%|███████▋  | 718/938 [04:00<01:11,  3.07it/s]
Training 1/1 epoch (loss 1.5554):  77%|███████▋  | 718/938 [04:00<01:11,  3.07it/s]
Training 1/1 epoch (loss 1.5554):  77%|███████▋  | 719/938 [04:00<01:12,  3.03it/s]
Training 1/1 epoch (loss 1.5172):  77%|███████▋  | 719/938 [04:01<01:12,  3.03it/s]
Training 1/1 epoch (loss 1.5172):  77%|███████▋  | 720/938 [04:01<01:13,  2.95it/s]
Training 1/1 epoch (loss 1.4905):  77%|███████▋  | 720/938 [04:01<01:13,  2.95it/s]
Training 1/1 epoch (loss 1.4905):  77%|███████▋  | 721/938 [04:01<01:11,  3.03it/s]
Training 1/1 epoch (loss 1.4959):  77%|███████▋  | 721/938 [04:01<01:11,  3.03it/s]
Training 1/1 epoch (loss 1.4959):  77%|███████▋  | 722/938 [04:01<01:08,  3.15it/s]
Training 1/1 epoch (loss 1.4015):  77%|███████▋  | 722/938 [04:01<01:08,  3.15it/s]
Training 1/1 epoch (loss 1.4015):  77%|███████▋  | 723/938 [04:01<01:07,  3.19it/s]
Training 1/1 epoch (loss 1.4422):  77%|███████▋  | 723/938 [04:02<01:07,  3.19it/s]
Training 1/1 epoch (loss 1.4422):  77%|███████▋  | 724/938 [04:02<01:07,  3.18it/s]
Training 1/1 epoch (loss 1.4943):  77%|███████▋  | 724/938 [04:02<01:07,  3.18it/s]
Training 1/1 epoch (loss 1.4943):  77%|███████▋  | 725/938 [04:02<01:07,  3.16it/s]
Training 1/1 epoch (loss 1.4814):  77%|███████▋  | 725/938 [04:02<01:07,  3.16it/s]
Training 1/1 epoch (loss 1.4814):  77%|███████▋  | 726/938 [04:02<01:06,  3.18it/s]
Training 1/1 epoch (loss 1.5570):  77%|███████▋  | 726/938 [04:03<01:06,  3.18it/s]
Training 1/1 epoch (loss 1.5570):  78%|███████▊  | 727/938 [04:03<01:06,  3.20it/s]
Training 1/1 epoch (loss 1.4316):  78%|███████▊  | 727/938 [04:03<01:06,  3.20it/s]
Training 1/1 epoch (loss 1.4316):  78%|███████▊  | 728/938 [04:03<01:05,  3.18it/s]
Training 1/1 epoch (loss 1.5077):  78%|███████▊  | 728/938 [04:03<01:05,  3.18it/s]
Training 1/1 epoch (loss 1.5077):  78%|███████▊  | 729/938 [04:03<01:05,  3.19it/s]
Training 1/1 epoch (loss 1.5535):  78%|███████▊  | 729/938 [04:04<01:05,  3.19it/s]
Training 1/1 epoch (loss 1.5535):  78%|███████▊  | 730/938 [04:04<01:05,  3.20it/s]
Training 1/1 epoch (loss 1.5956):  78%|███████▊  | 730/938 [04:04<01:05,  3.20it/s]
Training 1/1 epoch (loss 1.5956):  78%|███████▊  | 731/938 [04:04<01:04,  3.20it/s]
Training 1/1 epoch (loss 1.5431):  78%|███████▊  | 731/938 [04:04<01:04,  3.20it/s]
Training 1/1 epoch (loss 1.5431):  78%|███████▊  | 732/938 [04:04<01:06,  3.11it/s]
Training 1/1 epoch (loss 1.6304):  78%|███████▊  | 732/938 [04:05<01:06,  3.11it/s]
Training 1/1 epoch (loss 1.6304):  78%|███████▊  | 733/938 [04:05<01:05,  3.15it/s]
Training 1/1 epoch (loss 1.5368):  78%|███████▊  | 733/938 [04:05<01:05,  3.15it/s]
Training 1/1 epoch (loss 1.5368):  78%|███████▊  | 734/938 [04:05<01:04,  3.16it/s]
Training 1/1 epoch (loss 1.5698):  78%|███████▊  | 734/938 [04:05<01:04,  3.16it/s]
Training 1/1 epoch (loss 1.5698):  78%|███████▊  | 735/938 [04:05<01:02,  3.22it/s]
Training 1/1 epoch (loss 1.5325):  78%|███████▊  | 735/938 [04:06<01:02,  3.22it/s]
Training 1/1 epoch (loss 1.5325):  78%|███████▊  | 736/938 [04:06<01:06,  3.03it/s]
Training 1/1 epoch (loss 1.5415):  78%|███████▊  | 736/938 [04:06<01:06,  3.03it/s]
Training 1/1 epoch (loss 1.5415):  79%|███████▊  | 737/938 [04:06<01:07,  2.97it/s]
Training 1/1 epoch (loss 1.5268):  79%|███████▊  | 737/938 [04:06<01:07,  2.97it/s]
Training 1/1 epoch (loss 1.5268):  79%|███████▊  | 738/938 [04:06<01:05,  3.04it/s]
Training 1/1 epoch (loss 1.4781):  79%|███████▊  | 738/938 [04:07<01:05,  3.04it/s]
Training 1/1 epoch (loss 1.4781):  79%|███████▉  | 739/938 [04:07<01:03,  3.14it/s]
Training 1/1 epoch (loss 1.5546):  79%|███████▉  | 739/938 [04:07<01:03,  3.14it/s]
Training 1/1 epoch (loss 1.5546):  79%|███████▉  | 740/938 [04:07<01:02,  3.15it/s]
Training 1/1 epoch (loss 1.5922):  79%|███████▉  | 740/938 [04:07<01:02,  3.15it/s]
Training 1/1 epoch (loss 1.5922):  79%|███████▉  | 741/938 [04:07<01:01,  3.20it/s]
Training 1/1 epoch (loss 1.5520):  79%|███████▉  | 741/938 [04:08<01:01,  3.20it/s]
Training 1/1 epoch (loss 1.5520):  79%|███████▉  | 742/938 [04:08<01:02,  3.15it/s]
Training 1/1 epoch (loss 1.5593):  79%|███████▉  | 742/938 [04:08<01:02,  3.15it/s]
Training 1/1 epoch (loss 1.5593):  79%|███████▉  | 743/938 [04:08<01:03,  3.05it/s]
Training 1/1 epoch (loss 1.5317):  79%|███████▉  | 743/938 [04:08<01:03,  3.05it/s]
Training 1/1 epoch (loss 1.5317):  79%|███████▉  | 744/938 [04:08<01:04,  3.03it/s]
Training 1/1 epoch (loss 1.6153):  79%|███████▉  | 744/938 [04:09<01:04,  3.03it/s]
Training 1/1 epoch (loss 1.6153):  79%|███████▉  | 745/938 [04:09<01:02,  3.09it/s]
Training 1/1 epoch (loss 1.4852):  79%|███████▉  | 745/938 [04:09<01:02,  3.09it/s]
Training 1/1 epoch (loss 1.4852):  80%|███████▉  | 746/938 [04:09<01:02,  3.10it/s]
Training 1/1 epoch (loss 1.6571):  80%|███████▉  | 746/938 [04:09<01:02,  3.10it/s]
Training 1/1 epoch (loss 1.6571):  80%|███████▉  | 747/938 [04:09<01:00,  3.17it/s]
Training 1/1 epoch (loss 1.5035):  80%|███████▉  | 747/938 [04:09<01:00,  3.17it/s]
Training 1/1 epoch (loss 1.5035):  80%|███████▉  | 748/938 [04:09<01:00,  3.15it/s]
Training 1/1 epoch (loss 1.4767):  80%|███████▉  | 748/938 [04:10<01:00,  3.15it/s]
Training 1/1 epoch (loss 1.4767):  80%|███████▉  | 749/938 [04:10<00:59,  3.18it/s]
Training 1/1 epoch (loss 1.4858):  80%|███████▉  | 749/938 [04:10<00:59,  3.18it/s]
Training 1/1 epoch (loss 1.4858):  80%|███████▉  | 750/938 [04:10<01:00,  3.11it/s]
Training 1/1 epoch (loss 1.5336):  80%|███████▉  | 750/938 [04:10<01:00,  3.11it/s]
Training 1/1 epoch (loss 1.5336):  80%|████████  | 751/938 [04:10<00:59,  3.13it/s]
Training 1/1 epoch (loss 1.5274):  80%|████████  | 751/938 [04:11<00:59,  3.13it/s]
Training 1/1 epoch (loss 1.5274):  80%|████████  | 752/938 [04:11<01:01,  3.00it/s]
Training 1/1 epoch (loss 1.5871):  80%|████████  | 752/938 [04:11<01:01,  3.00it/s]
Training 1/1 epoch (loss 1.5871):  80%|████████  | 753/938 [04:11<01:00,  3.07it/s]
Training 1/1 epoch (loss 1.5743):  80%|████████  | 753/938 [04:11<01:00,  3.07it/s]
Training 1/1 epoch (loss 1.5743):  80%|████████  | 754/938 [04:11<00:59,  3.09it/s]
Training 1/1 epoch (loss 1.5395):  80%|████████  | 754/938 [04:12<00:59,  3.09it/s]
Training 1/1 epoch (loss 1.5395):  80%|████████  | 755/938 [04:12<00:59,  3.08it/s]
Training 1/1 epoch (loss 1.5602):  80%|████████  | 755/938 [04:12<00:59,  3.08it/s]
Training 1/1 epoch (loss 1.5602):  81%|████████  | 756/938 [04:12<01:00,  3.00it/s]
Training 1/1 epoch (loss 1.5234):  81%|████████  | 756/938 [04:12<01:00,  3.00it/s]
Training 1/1 epoch (loss 1.5234):  81%|████████  | 757/938 [04:12<00:58,  3.12it/s]
Training 1/1 epoch (loss 1.6487):  81%|████████  | 757/938 [04:13<00:58,  3.12it/s]
Training 1/1 epoch (loss 1.6487):  81%|████████  | 758/938 [04:13<00:56,  3.16it/s]
Training 1/1 epoch (loss 1.5287):  81%|████████  | 758/938 [04:13<00:56,  3.16it/s]
Training 1/1 epoch (loss 1.5287):  81%|████████  | 759/938 [04:13<00:56,  3.18it/s]
Training 1/1 epoch (loss 1.3763):  81%|████████  | 759/938 [04:13<00:56,  3.18it/s]
Training 1/1 epoch (loss 1.3763):  81%|████████  | 760/938 [04:13<00:55,  3.21it/s]
Training 1/1 epoch (loss 1.6733):  81%|████████  | 760/938 [04:14<00:55,  3.21it/s]
Training 1/1 epoch (loss 1.6733):  81%|████████  | 761/938 [04:14<00:57,  3.10it/s]
Training 1/1 epoch (loss 1.5245):  81%|████████  | 761/938 [04:14<00:57,  3.10it/s]
Training 1/1 epoch (loss 1.5245):  81%|████████  | 762/938 [04:14<00:57,  3.08it/s]
Training 1/1 epoch (loss 1.6487):  81%|████████  | 762/938 [04:14<00:57,  3.08it/s]
Training 1/1 epoch (loss 1.6487):  81%|████████▏ | 763/938 [04:14<00:57,  3.06it/s]
Training 1/1 epoch (loss 1.3808):  81%|████████▏ | 763/938 [04:15<00:57,  3.06it/s]
Training 1/1 epoch (loss 1.3808):  81%|████████▏ | 764/938 [04:15<00:58,  2.99it/s]
Training 1/1 epoch (loss 1.6058):  81%|████████▏ | 764/938 [04:15<00:58,  2.99it/s]
Training 1/1 epoch (loss 1.6058):  82%|████████▏ | 765/938 [04:15<00:55,  3.13it/s]
Training 1/1 epoch (loss 1.5348):  82%|████████▏ | 765/938 [04:15<00:55,  3.13it/s]
Training 1/1 epoch (loss 1.5348):  82%|████████▏ | 766/938 [04:15<00:54,  3.13it/s]
Training 1/1 epoch (loss 1.5916):  82%|████████▏ | 766/938 [04:16<00:54,  3.13it/s]
Training 1/1 epoch (loss 1.5916):  82%|████████▏ | 767/938 [04:16<00:53,  3.18it/s]
Training 1/1 epoch (loss 1.4585):  82%|████████▏ | 767/938 [04:16<00:53,  3.18it/s]
Training 1/1 epoch (loss 1.4585):  82%|████████▏ | 768/938 [04:16<00:55,  3.04it/s]
Training 1/1 epoch (loss 1.7167):  82%|████████▏ | 768/938 [04:16<00:55,  3.04it/s]
Training 1/1 epoch (loss 1.7167):  82%|████████▏ | 769/938 [04:16<00:55,  3.02it/s]
Training 1/1 epoch (loss 1.5531):  82%|████████▏ | 769/938 [04:17<00:55,  3.02it/s]
Training 1/1 epoch (loss 1.5531):  82%|████████▏ | 770/938 [04:17<00:54,  3.08it/s]
Training 1/1 epoch (loss 1.4547):  82%|████████▏ | 770/938 [04:17<00:54,  3.08it/s]
Training 1/1 epoch (loss 1.4547):  82%|████████▏ | 771/938 [04:17<00:54,  3.09it/s]
Training 1/1 epoch (loss 1.4792):  82%|████████▏ | 771/938 [04:17<00:54,  3.09it/s]
Training 1/1 epoch (loss 1.4792):  82%|████████▏ | 772/938 [04:17<00:52,  3.15it/s]
Training 1/1 epoch (loss 1.4779):  82%|████████▏ | 772/938 [04:18<00:52,  3.15it/s]
Training 1/1 epoch (loss 1.4779):  82%|████████▏ | 773/938 [04:18<00:52,  3.16it/s]
Training 1/1 epoch (loss 1.4138):  82%|████████▏ | 773/938 [04:18<00:52,  3.16it/s]
Training 1/1 epoch (loss 1.4138):  83%|████████▎ | 774/938 [04:18<00:50,  3.23it/s]
Training 1/1 epoch (loss 1.5065):  83%|████████▎ | 774/938 [04:18<00:50,  3.23it/s]
Training 1/1 epoch (loss 1.5065):  83%|████████▎ | 775/938 [04:18<00:51,  3.15it/s]
Training 1/1 epoch (loss 1.6044):  83%|████████▎ | 775/938 [04:18<00:51,  3.15it/s]
Training 1/1 epoch (loss 1.6044):  83%|████████▎ | 776/938 [04:18<00:51,  3.13it/s]
Training 1/1 epoch (loss 1.5497):  83%|████████▎ | 776/938 [04:19<00:51,  3.13it/s]
Training 1/1 epoch (loss 1.5497):  83%|████████▎ | 777/938 [04:19<00:51,  3.16it/s]
Training 1/1 epoch (loss 1.4901):  83%|████████▎ | 777/938 [04:19<00:51,  3.16it/s]
Training 1/1 epoch (loss 1.4901):  83%|████████▎ | 778/938 [04:19<00:50,  3.14it/s]
Training 1/1 epoch (loss 1.4382):  83%|████████▎ | 778/938 [04:19<00:50,  3.14it/s]
Training 1/1 epoch (loss 1.4382):  83%|████████▎ | 779/938 [04:19<00:49,  3.20it/s]
Training 1/1 epoch (loss 1.6184):  83%|████████▎ | 779/938 [04:20<00:49,  3.20it/s]
Training 1/1 epoch (loss 1.6184):  83%|████████▎ | 780/938 [04:20<00:49,  3.17it/s]
Training 1/1 epoch (loss 1.6159):  83%|████████▎ | 780/938 [04:20<00:49,  3.17it/s]
Training 1/1 epoch (loss 1.6159):  83%|████████▎ | 781/938 [04:20<00:50,  3.09it/s]
Training 1/1 epoch (loss 1.5002):  83%|████████▎ | 781/938 [04:20<00:50,  3.09it/s]
Training 1/1 epoch (loss 1.5002):  83%|████████▎ | 782/938 [04:20<00:50,  3.08it/s]
Training 1/1 epoch (loss 1.5458):  83%|████████▎ | 782/938 [04:21<00:50,  3.08it/s]
Training 1/1 epoch (loss 1.5458):  83%|████████▎ | 783/938 [04:21<00:49,  3.13it/s]
Training 1/1 epoch (loss 1.5048):  83%|████████▎ | 783/938 [04:21<00:49,  3.13it/s]
Training 1/1 epoch (loss 1.5048):  84%|████████▎ | 784/938 [04:21<00:51,  3.01it/s]
Training 1/1 epoch (loss 1.5504):  84%|████████▎ | 784/938 [04:21<00:51,  3.01it/s]
Training 1/1 epoch (loss 1.5504):  84%|████████▎ | 785/938 [04:21<00:51,  2.98it/s]
Training 1/1 epoch (loss 1.5692):  84%|████████▎ | 785/938 [04:22<00:51,  2.98it/s]
Training 1/1 epoch (loss 1.5692):  84%|████████▍ | 786/938 [04:22<00:49,  3.05it/s]
Training 1/1 epoch (loss 1.4996):  84%|████████▍ | 786/938 [04:22<00:49,  3.05it/s]
Training 1/1 epoch (loss 1.4996):  84%|████████▍ | 787/938 [04:22<00:49,  3.08it/s]
Training 1/1 epoch (loss 1.6763):  84%|████████▍ | 787/938 [04:22<00:49,  3.08it/s]
Training 1/1 epoch (loss 1.6763):  84%|████████▍ | 788/938 [04:22<00:48,  3.10it/s]
Training 1/1 epoch (loss 1.4414):  84%|████████▍ | 788/938 [04:23<00:48,  3.10it/s]
Training 1/1 epoch (loss 1.4414):  84%|████████▍ | 789/938 [04:23<00:48,  3.09it/s]
Training 1/1 epoch (loss 1.4197):  84%|████████▍ | 789/938 [04:23<00:48,  3.09it/s]
Training 1/1 epoch (loss 1.4197):  84%|████████▍ | 790/938 [04:23<00:46,  3.15it/s]
Training 1/1 epoch (loss 1.5575):  84%|████████▍ | 790/938 [04:23<00:46,  3.15it/s]
Training 1/1 epoch (loss 1.5575):  84%|████████▍ | 791/938 [04:23<00:45,  3.24it/s]
Training 1/1 epoch (loss 1.4733):  84%|████████▍ | 791/938 [04:24<00:45,  3.24it/s]
Training 1/1 epoch (loss 1.4733):  84%|████████▍ | 792/938 [04:24<00:46,  3.14it/s]
Training 1/1 epoch (loss 1.6012):  84%|████████▍ | 792/938 [04:24<00:46,  3.14it/s]
Training 1/1 epoch (loss 1.6012):  85%|████████▍ | 793/938 [04:24<00:46,  3.12it/s]
Training 1/1 epoch (loss 1.6115):  85%|████████▍ | 793/938 [04:24<00:46,  3.12it/s]
Training 1/1 epoch (loss 1.6115):  85%|████████▍ | 794/938 [04:24<00:46,  3.11it/s]
Training 1/1 epoch (loss 1.5284):  85%|████████▍ | 794/938 [04:25<00:46,  3.11it/s]
Training 1/1 epoch (loss 1.5284):  85%|████████▍ | 795/938 [04:25<00:44,  3.22it/s]
Training 1/1 epoch (loss 1.5278):  85%|████████▍ | 795/938 [04:25<00:44,  3.22it/s]
Training 1/1 epoch (loss 1.5278):  85%|████████▍ | 796/938 [04:25<00:46,  3.05it/s]
Training 1/1 epoch (loss 1.4962):  85%|████████▍ | 796/938 [04:25<00:46,  3.05it/s]
Training 1/1 epoch (loss 1.4962):  85%|████████▍ | 797/938 [04:25<00:45,  3.08it/s]
Training 1/1 epoch (loss 1.5980):  85%|████████▍ | 797/938 [04:26<00:45,  3.08it/s]
Training 1/1 epoch (loss 1.5980):  85%|████████▌ | 798/938 [04:26<00:45,  3.09it/s]
Training 1/1 epoch (loss 1.5752):  85%|████████▌ | 798/938 [04:26<00:45,  3.09it/s]
Training 1/1 epoch (loss 1.5752):  85%|████████▌ | 799/938 [04:26<00:44,  3.11it/s]
Training 1/1 epoch (loss 1.5379):  85%|████████▌ | 799/938 [04:26<00:44,  3.11it/s]
Training 1/1 epoch (loss 1.5379):  85%|████████▌ | 800/938 [04:26<00:46,  2.98it/s]
Training 1/1 epoch (loss 1.6042):  85%|████████▌ | 800/938 [04:27<00:46,  2.98it/s]
Training 1/1 epoch (loss 1.6042):  85%|████████▌ | 801/938 [04:27<00:47,  2.87it/s]
Training 1/1 epoch (loss 1.6044):  85%|████████▌ | 801/938 [04:27<00:47,  2.87it/s]
Training 1/1 epoch (loss 1.6044):  86%|████████▌ | 802/938 [04:27<00:45,  2.98it/s]
Training 1/1 epoch (loss 1.5680):  86%|████████▌ | 802/938 [04:27<00:45,  2.98it/s]
Training 1/1 epoch (loss 1.5680):  86%|████████▌ | 803/938 [04:27<00:43,  3.08it/s]
Training 1/1 epoch (loss 1.4906):  86%|████████▌ | 803/938 [04:28<00:43,  3.08it/s]
Training 1/1 epoch (loss 1.4906):  86%|████████▌ | 804/938 [04:28<00:42,  3.12it/s]
Training 1/1 epoch (loss 1.5307):  86%|████████▌ | 804/938 [04:28<00:42,  3.12it/s]
Training 1/1 epoch (loss 1.5307):  86%|████████▌ | 805/938 [04:28<00:42,  3.13it/s]
Training 1/1 epoch (loss 1.5009):  86%|████████▌ | 805/938 [04:28<00:42,  3.13it/s]
Training 1/1 epoch (loss 1.5009):  86%|████████▌ | 806/938 [04:28<00:42,  3.12it/s]
Training 1/1 epoch (loss 1.5395):  86%|████████▌ | 806/938 [04:29<00:42,  3.12it/s]
Training 1/1 epoch (loss 1.5395):  86%|████████▌ | 807/938 [04:29<00:41,  3.12it/s]
Training 1/1 epoch (loss 1.5626):  86%|████████▌ | 807/938 [04:29<00:41,  3.12it/s]
Training 1/1 epoch (loss 1.5626):  86%|████████▌ | 808/938 [04:29<00:41,  3.11it/s]
Training 1/1 epoch (loss 1.4485):  86%|████████▌ | 808/938 [04:29<00:41,  3.11it/s]
Training 1/1 epoch (loss 1.4485):  86%|████████▌ | 809/938 [04:29<00:40,  3.18it/s]
Training 1/1 epoch (loss 1.5608):  86%|████████▌ | 809/938 [04:29<00:40,  3.18it/s]
Training 1/1 epoch (loss 1.5608):  86%|████████▋ | 810/938 [04:29<00:40,  3.16it/s]
Training 1/1 epoch (loss 1.4432):  86%|████████▋ | 810/938 [04:30<00:40,  3.16it/s]
Training 1/1 epoch (loss 1.4432):  86%|████████▋ | 811/938 [04:30<00:41,  3.03it/s]
Training 1/1 epoch (loss 1.5502):  86%|████████▋ | 811/938 [04:30<00:41,  3.03it/s]
Training 1/1 epoch (loss 1.5502):  87%|████████▋ | 812/938 [04:30<00:41,  3.05it/s]
Training 1/1 epoch (loss 1.5991):  87%|████████▋ | 812/938 [04:30<00:41,  3.05it/s]
Training 1/1 epoch (loss 1.5991):  87%|████████▋ | 813/938 [04:30<00:41,  3.04it/s]
Training 1/1 epoch (loss 1.5209):  87%|████████▋ | 813/938 [04:31<00:41,  3.04it/s]
Training 1/1 epoch (loss 1.5209):  87%|████████▋ | 814/938 [04:31<00:39,  3.10it/s]
Training 1/1 epoch (loss 1.5304):  87%|████████▋ | 814/938 [04:31<00:39,  3.10it/s]
Training 1/1 epoch (loss 1.5304):  87%|████████▋ | 815/938 [04:31<00:39,  3.15it/s]
Training 1/1 epoch (loss 1.5446):  87%|████████▋ | 815/938 [04:31<00:39,  3.15it/s]
Training 1/1 epoch (loss 1.5446):  87%|████████▋ | 816/938 [04:31<00:39,  3.12it/s]
Training 1/1 epoch (loss 1.5223):  87%|████████▋ | 816/938 [04:32<00:39,  3.12it/s]
Training 1/1 epoch (loss 1.5223):  87%|████████▋ | 817/938 [04:32<00:41,  2.95it/s]
Training 1/1 epoch (loss 1.4721):  87%|████████▋ | 817/938 [04:32<00:41,  2.95it/s]
Training 1/1 epoch (loss 1.4721):  87%|████████▋ | 818/938 [04:32<00:40,  2.94it/s]
Training 1/1 epoch (loss 1.6526):  87%|████████▋ | 818/938 [04:32<00:40,  2.94it/s]
Training 1/1 epoch (loss 1.6526):  87%|████████▋ | 819/938 [04:32<00:39,  2.98it/s]
Training 1/1 epoch (loss 1.4840):  87%|████████▋ | 819/938 [04:33<00:39,  2.98it/s]
Training 1/1 epoch (loss 1.4840):  87%|████████▋ | 820/938 [04:33<00:37,  3.11it/s]
Training 1/1 epoch (loss 1.5689):  87%|████████▋ | 820/938 [04:33<00:37,  3.11it/s]
Training 1/1 epoch (loss 1.5689):  88%|████████▊ | 821/938 [04:33<00:37,  3.15it/s]
Training 1/1 epoch (loss 1.6443):  88%|████████▊ | 821/938 [04:33<00:37,  3.15it/s]
Training 1/1 epoch (loss 1.6443):  88%|████████▊ | 822/938 [04:33<00:36,  3.21it/s]
Training 1/1 epoch (loss 1.5241):  88%|████████▊ | 822/938 [04:34<00:36,  3.21it/s]
Training 1/1 epoch (loss 1.5241):  88%|████████▊ | 823/938 [04:34<00:36,  3.19it/s]
Training 1/1 epoch (loss 1.4930):  88%|████████▊ | 823/938 [04:34<00:36,  3.19it/s]
Training 1/1 epoch (loss 1.4930):  88%|████████▊ | 824/938 [04:34<00:36,  3.09it/s]
Training 1/1 epoch (loss 1.4653):  88%|████████▊ | 824/938 [04:34<00:36,  3.09it/s]
Training 1/1 epoch (loss 1.4653):  88%|████████▊ | 825/938 [04:34<00:37,  3.02it/s]
Training 1/1 epoch (loss 1.5567):  88%|████████▊ | 825/938 [04:35<00:37,  3.02it/s]
Training 1/1 epoch (loss 1.5567):  88%|████████▊ | 826/938 [04:35<00:36,  3.06it/s]
Training 1/1 epoch (loss 1.5262):  88%|████████▊ | 826/938 [04:35<00:36,  3.06it/s]
Training 1/1 epoch (loss 1.5262):  88%|████████▊ | 827/938 [04:35<00:36,  3.05it/s]
Training 1/1 epoch (loss 1.6177):  88%|████████▊ | 827/938 [04:35<00:36,  3.05it/s]
Training 1/1 epoch (loss 1.6177):  88%|████████▊ | 828/938 [04:35<00:36,  3.03it/s]
Training 1/1 epoch (loss 1.5426):  88%|████████▊ | 828/938 [04:36<00:36,  3.03it/s]
Training 1/1 epoch (loss 1.5426):  88%|████████▊ | 829/938 [04:36<00:35,  3.07it/s]
Training 1/1 epoch (loss 1.5885):  88%|████████▊ | 829/938 [04:36<00:35,  3.07it/s]
Training 1/1 epoch (loss 1.5885):  88%|████████▊ | 830/938 [04:36<00:34,  3.12it/s]
Training 1/1 epoch (loss 1.6284):  88%|████████▊ | 830/938 [04:36<00:34,  3.12it/s]
Training 1/1 epoch (loss 1.6284):  89%|████████▊ | 831/938 [04:36<00:35,  3.05it/s]
Training 1/1 epoch (loss 1.5572):  89%|████████▊ | 831/938 [04:37<00:35,  3.05it/s]
Training 1/1 epoch (loss 1.5572):  89%|████████▊ | 832/938 [04:37<00:35,  3.03it/s]
Training 1/1 epoch (loss 1.5540):  89%|████████▊ | 832/938 [04:37<00:35,  3.03it/s]
Training 1/1 epoch (loss 1.5540):  89%|████████▉ | 833/938 [04:37<00:35,  2.94it/s]
Training 1/1 epoch (loss 1.5392):  89%|████████▉ | 833/938 [04:37<00:35,  2.94it/s]
Training 1/1 epoch (loss 1.5392):  89%|████████▉ | 834/938 [04:37<00:34,  3.04it/s]
Training 1/1 epoch (loss 1.4836):  89%|████████▉ | 834/938 [04:38<00:34,  3.04it/s]
Training 1/1 epoch (loss 1.4836):  89%|████████▉ | 835/938 [04:38<00:33,  3.05it/s]
Training 1/1 epoch (loss 1.5001):  89%|████████▉ | 835/938 [04:38<00:33,  3.05it/s]
Training 1/1 epoch (loss 1.5001):  89%|████████▉ | 836/938 [04:38<00:32,  3.10it/s]
Training 1/1 epoch (loss 1.6516):  89%|████████▉ | 836/938 [04:38<00:32,  3.10it/s]
Training 1/1 epoch (loss 1.6516):  89%|████████▉ | 837/938 [04:38<00:33,  3.03it/s]
Training 1/1 epoch (loss 1.5345):  89%|████████▉ | 837/938 [04:39<00:33,  3.03it/s]
Training 1/1 epoch (loss 1.5345):  89%|████████▉ | 838/938 [04:39<00:32,  3.09it/s]
Training 1/1 epoch (loss 1.5013):  89%|████████▉ | 838/938 [04:39<00:32,  3.09it/s]
Training 1/1 epoch (loss 1.5013):  89%|████████▉ | 839/938 [04:39<00:31,  3.18it/s]
Training 1/1 epoch (loss 1.4530):  89%|████████▉ | 839/938 [04:39<00:31,  3.18it/s]
Training 1/1 epoch (loss 1.4530):  90%|████████▉ | 840/938 [04:39<00:31,  3.14it/s]
Training 1/1 epoch (loss 1.6245):  90%|████████▉ | 840/938 [04:40<00:31,  3.14it/s]
Training 1/1 epoch (loss 1.6245):  90%|████████▉ | 841/938 [04:40<00:30,  3.18it/s]
Training 1/1 epoch (loss 1.4918):  90%|████████▉ | 841/938 [04:40<00:30,  3.18it/s]
Training 1/1 epoch (loss 1.4918):  90%|████████▉ | 842/938 [04:40<00:30,  3.17it/s]
Training 1/1 epoch (loss 1.6224):  90%|████████▉ | 842/938 [04:40<00:30,  3.17it/s]
Training 1/1 epoch (loss 1.6224):  90%|████████▉ | 843/938 [04:40<00:30,  3.08it/s]
Training 1/1 epoch (loss 1.4425):  90%|████████▉ | 843/938 [04:41<00:30,  3.08it/s]
Training 1/1 epoch (loss 1.4425):  90%|████████▉ | 844/938 [04:41<00:30,  3.08it/s]
Training 1/1 epoch (loss 1.5912):  90%|████████▉ | 844/938 [04:41<00:30,  3.08it/s]
Training 1/1 epoch (loss 1.5912):  90%|█████████ | 845/938 [04:41<00:29,  3.12it/s]
Training 1/1 epoch (loss 1.6644):  90%|█████████ | 845/938 [04:41<00:29,  3.12it/s]
Training 1/1 epoch (loss 1.6644):  90%|█████████ | 846/938 [04:41<00:28,  3.17it/s]
Training 1/1 epoch (loss 1.5229):  90%|█████████ | 846/938 [04:41<00:28,  3.17it/s]
Training 1/1 epoch (loss 1.5229):  90%|█████████ | 847/938 [04:41<00:29,  3.14it/s]
Training 1/1 epoch (loss 1.5059):  90%|█████████ | 847/938 [04:42<00:29,  3.14it/s]
Training 1/1 epoch (loss 1.5059):  90%|█████████ | 848/938 [04:42<00:28,  3.12it/s]
Training 1/1 epoch (loss 1.5519):  90%|█████████ | 848/938 [04:42<00:28,  3.12it/s]
Training 1/1 epoch (loss 1.5519):  91%|█████████ | 849/938 [04:42<00:31,  2.87it/s]
Training 1/1 epoch (loss 1.4814):  91%|█████████ | 849/938 [04:43<00:31,  2.87it/s]
Training 1/1 epoch (loss 1.4814):  91%|█████████ | 850/938 [04:43<00:29,  2.97it/s]
Training 1/1 epoch (loss 1.6227):  91%|█████████ | 850/938 [04:43<00:29,  2.97it/s]
Training 1/1 epoch (loss 1.6227):  91%|█████████ | 851/938 [04:43<00:28,  3.07it/s]
Training 1/1 epoch (loss 1.6425):  91%|█████████ | 851/938 [04:43<00:28,  3.07it/s]
Training 1/1 epoch (loss 1.6425):  91%|█████████ | 852/938 [04:43<00:27,  3.10it/s]
Training 1/1 epoch (loss 1.5761):  91%|█████████ | 852/938 [04:43<00:27,  3.10it/s]
Training 1/1 epoch (loss 1.5761):  91%|█████████ | 853/938 [04:43<00:27,  3.12it/s]
Training 1/1 epoch (loss 1.5306):  91%|█████████ | 853/938 [04:44<00:27,  3.12it/s]
Training 1/1 epoch (loss 1.5306):  91%|█████████ | 854/938 [04:44<00:26,  3.23it/s]
Training 1/1 epoch (loss 1.6081):  91%|█████████ | 854/938 [04:44<00:26,  3.23it/s]
Training 1/1 epoch (loss 1.6081):  91%|█████████ | 855/938 [04:44<00:26,  3.19it/s]
Training 1/1 epoch (loss 1.5339):  91%|█████████ | 855/938 [04:44<00:26,  3.19it/s]
Training 1/1 epoch (loss 1.5339):  91%|█████████▏| 856/938 [04:44<00:26,  3.14it/s]
Training 1/1 epoch (loss 1.5924):  91%|█████████▏| 856/938 [04:45<00:26,  3.14it/s]
Training 1/1 epoch (loss 1.5924):  91%|█████████▏| 857/938 [04:45<00:25,  3.13it/s]
Training 1/1 epoch (loss 1.5376):  91%|█████████▏| 857/938 [04:45<00:25,  3.13it/s]
Training 1/1 epoch (loss 1.5376):  91%|█████████▏| 858/938 [04:45<00:25,  3.15it/s]
Training 1/1 epoch (loss 1.5534):  91%|█████████▏| 858/938 [04:45<00:25,  3.15it/s]
Training 1/1 epoch (loss 1.5534):  92%|█████████▏| 859/938 [04:45<00:25,  3.07it/s]
Training 1/1 epoch (loss 1.4396):  92%|█████████▏| 859/938 [04:46<00:25,  3.07it/s]
Training 1/1 epoch (loss 1.4396):  92%|█████████▏| 860/938 [04:46<00:25,  3.04it/s]
Training 1/1 epoch (loss 1.5445):  92%|█████████▏| 860/938 [04:46<00:25,  3.04it/s]
Training 1/1 epoch (loss 1.5445):  92%|█████████▏| 861/938 [04:46<00:25,  3.05it/s]
Training 1/1 epoch (loss 1.6179):  92%|█████████▏| 861/938 [04:46<00:25,  3.05it/s]
Training 1/1 epoch (loss 1.6179):  92%|█████████▏| 862/938 [04:46<00:24,  3.06it/s]
Training 1/1 epoch (loss 1.6457):  92%|█████████▏| 862/938 [04:47<00:24,  3.06it/s]
Training 1/1 epoch (loss 1.6457):  92%|█████████▏| 863/938 [04:47<00:23,  3.13it/s]
Training 1/1 epoch (loss 1.5193):  92%|█████████▏| 863/938 [04:47<00:23,  3.13it/s]
Training 1/1 epoch (loss 1.5193):  92%|█████████▏| 864/938 [04:47<00:23,  3.17it/s]
Training 1/1 epoch (loss 1.5789):  92%|█████████▏| 864/938 [04:47<00:23,  3.17it/s]
Training 1/1 epoch (loss 1.5789):  92%|█████████▏| 865/938 [04:47<00:22,  3.22it/s]
Training 1/1 epoch (loss 1.4011):  92%|█████████▏| 865/938 [04:48<00:22,  3.22it/s]
Training 1/1 epoch (loss 1.4011):  92%|█████████▏| 866/938 [04:48<00:23,  3.01it/s]
Training 1/1 epoch (loss 1.3718):  92%|█████████▏| 866/938 [04:48<00:23,  3.01it/s]
Training 1/1 epoch (loss 1.3718):  92%|█████████▏| 867/938 [04:48<00:23,  3.04it/s]
Training 1/1 epoch (loss 1.4525):  92%|█████████▏| 867/938 [04:48<00:23,  3.04it/s]
Training 1/1 epoch (loss 1.4525):  93%|█████████▎| 868/938 [04:48<00:22,  3.06it/s]
Training 1/1 epoch (loss 1.4947):  93%|█████████▎| 868/938 [04:49<00:22,  3.06it/s]
Training 1/1 epoch (loss 1.4947):  93%|█████████▎| 869/938 [04:49<00:22,  3.11it/s]
Training 1/1 epoch (loss 1.6176):  93%|█████████▎| 869/938 [04:49<00:22,  3.11it/s]
Training 1/1 epoch (loss 1.6176):  93%|█████████▎| 870/938 [04:49<00:21,  3.21it/s]
Training 1/1 epoch (loss 1.4931):  93%|█████████▎| 870/938 [04:49<00:21,  3.21it/s]
Training 1/1 epoch (loss 1.4931):  93%|█████████▎| 871/938 [04:49<00:20,  3.20it/s]
Training 1/1 epoch (loss 1.5443):  93%|█████████▎| 871/938 [04:50<00:20,  3.20it/s]
Training 1/1 epoch (loss 1.5443):  93%|█████████▎| 872/938 [04:50<00:20,  3.19it/s]
Training 1/1 epoch (loss 1.6516):  93%|█████████▎| 872/938 [04:50<00:20,  3.19it/s]
Training 1/1 epoch (loss 1.6516):  93%|█████████▎| 873/938 [04:50<00:20,  3.13it/s]
Training 1/1 epoch (loss 1.5384):  93%|█████████▎| 873/938 [04:50<00:20,  3.13it/s]
Training 1/1 epoch (loss 1.5384):  93%|█████████▎| 874/938 [04:50<00:20,  3.15it/s]
Training 1/1 epoch (loss 1.5957):  93%|█████████▎| 874/938 [04:50<00:20,  3.15it/s]
Training 1/1 epoch (loss 1.5957):  93%|█████████▎| 875/938 [04:50<00:19,  3.16it/s]
Training 1/1 epoch (loss 1.5189):  93%|█████████▎| 875/938 [04:51<00:19,  3.16it/s]
Training 1/1 epoch (loss 1.5189):  93%|█████████▎| 876/938 [04:51<00:19,  3.20it/s]
Training 1/1 epoch (loss 1.5224):  93%|█████████▎| 876/938 [04:51<00:19,  3.20it/s]
Training 1/1 epoch (loss 1.5224):  93%|█████████▎| 877/938 [04:51<00:19,  3.20it/s]
Training 1/1 epoch (loss 1.4918):  93%|█████████▎| 877/938 [04:51<00:19,  3.20it/s]
Training 1/1 epoch (loss 1.4918):  94%|█████████▎| 878/938 [04:51<00:18,  3.25it/s]
Training 1/1 epoch (loss 1.4729):  94%|█████████▎| 878/938 [04:52<00:18,  3.25it/s]
Training 1/1 epoch (loss 1.4729):  94%|█████████▎| 879/938 [04:52<00:18,  3.24it/s]
Training 1/1 epoch (loss 1.6702):  94%|█████████▎| 879/938 [04:52<00:18,  3.24it/s]
Training 1/1 epoch (loss 1.6702):  94%|█████████▍| 880/938 [04:52<00:18,  3.15it/s]
Training 1/1 epoch (loss 1.5397):  94%|█████████▍| 880/938 [04:52<00:18,  3.15it/s]
Training 1/1 epoch (loss 1.5397):  94%|█████████▍| 881/938 [04:52<00:18,  3.10it/s]
Training 1/1 epoch (loss 1.5487):  94%|█████████▍| 881/938 [04:53<00:18,  3.10it/s]
Training 1/1 epoch (loss 1.5487):  94%|█████████▍| 882/938 [04:53<00:17,  3.16it/s]
Training 1/1 epoch (loss 1.6189):  94%|█████████▍| 882/938 [04:53<00:17,  3.16it/s]
Training 1/1 epoch (loss 1.6189):  94%|█████████▍| 883/938 [04:53<00:17,  3.23it/s]
Training 1/1 epoch (loss 1.5976):  94%|█████████▍| 883/938 [04:53<00:17,  3.23it/s]
Training 1/1 epoch (loss 1.5976):  94%|█████████▍| 884/938 [04:53<00:16,  3.22it/s]
Training 1/1 epoch (loss 1.5391):  94%|█████████▍| 884/938 [04:54<00:16,  3.22it/s]
Training 1/1 epoch (loss 1.5391):  94%|█████████▍| 885/938 [04:54<00:16,  3.25it/s]
Training 1/1 epoch (loss 1.4369):  94%|█████████▍| 885/938 [04:54<00:16,  3.25it/s]
Training 1/1 epoch (loss 1.4369):  94%|█████████▍| 886/938 [04:54<00:16,  3.24it/s]
Training 1/1 epoch (loss 1.5440):  94%|█████████▍| 886/938 [04:54<00:16,  3.24it/s]
Training 1/1 epoch (loss 1.5440):  95%|█████████▍| 887/938 [04:54<00:15,  3.21it/s]
Training 1/1 epoch (loss 1.4927):  95%|█████████▍| 887/938 [04:55<00:15,  3.21it/s]
Training 1/1 epoch (loss 1.4927):  95%|█████████▍| 888/938 [04:55<00:15,  3.17it/s]
Training 1/1 epoch (loss 1.5418):  95%|█████████▍| 888/938 [04:55<00:15,  3.17it/s]
Training 1/1 epoch (loss 1.5418):  95%|█████████▍| 889/938 [04:55<00:15,  3.13it/s]
Training 1/1 epoch (loss 1.4197):  95%|█████████▍| 889/938 [04:55<00:15,  3.13it/s]
Training 1/1 epoch (loss 1.4197):  95%|█████████▍| 890/938 [04:55<00:15,  3.10it/s]
Training 1/1 epoch (loss 1.5274):  95%|█████████▍| 890/938 [04:56<00:15,  3.10it/s]
Training 1/1 epoch (loss 1.5274):  95%|█████████▍| 891/938 [04:56<00:15,  3.04it/s]
Training 1/1 epoch (loss 1.5815):  95%|█████████▍| 891/938 [04:56<00:15,  3.04it/s]
Training 1/1 epoch (loss 1.5815):  95%|█████████▌| 892/938 [04:56<00:14,  3.09it/s]
Training 1/1 epoch (loss 1.6290):  95%|█████████▌| 892/938 [04:56<00:14,  3.09it/s]
Training 1/1 epoch (loss 1.6290):  95%|█████████▌| 893/938 [04:56<00:14,  3.07it/s]
Training 1/1 epoch (loss 1.4577):  95%|█████████▌| 893/938 [04:56<00:14,  3.07it/s]
Training 1/1 epoch (loss 1.4577):  95%|█████████▌| 894/938 [04:56<00:14,  3.11it/s]
Training 1/1 epoch (loss 1.5334):  95%|█████████▌| 894/938 [04:57<00:14,  3.11it/s]
Training 1/1 epoch (loss 1.5334):  95%|█████████▌| 895/938 [04:57<00:13,  3.17it/s]
Training 1/1 epoch (loss 1.5004):  95%|█████████▌| 895/938 [04:57<00:13,  3.17it/s]
Training 1/1 epoch (loss 1.5004):  96%|█████████▌| 896/938 [04:57<00:13,  3.18it/s]
Training 1/1 epoch (loss 1.5021):  96%|█████████▌| 896/938 [04:57<00:13,  3.18it/s]
Training 1/1 epoch (loss 1.5021):  96%|█████████▌| 897/938 [04:57<00:12,  3.20it/s]
Training 1/1 epoch (loss 1.5338):  96%|█████████▌| 897/938 [04:58<00:12,  3.20it/s]
Training 1/1 epoch (loss 1.5338):  96%|█████████▌| 898/938 [04:58<00:12,  3.20it/s]
Training 1/1 epoch (loss 1.5456):  96%|█████████▌| 898/938 [04:58<00:12,  3.20it/s]
Training 1/1 epoch (loss 1.5456):  96%|█████████▌| 899/938 [04:58<00:12,  3.07it/s]
Training 1/1 epoch (loss 1.5286):  96%|█████████▌| 899/938 [04:58<00:12,  3.07it/s]
Training 1/1 epoch (loss 1.5286):  96%|█████████▌| 900/938 [04:58<00:12,  3.01it/s]
Training 1/1 epoch (loss 1.4384):  96%|█████████▌| 900/938 [04:59<00:12,  3.01it/s]
Training 1/1 epoch (loss 1.4384):  96%|█████████▌| 901/938 [04:59<00:12,  3.01it/s]
Training 1/1 epoch (loss 1.4648):  96%|█████████▌| 901/938 [04:59<00:12,  3.01it/s]
Training 1/1 epoch (loss 1.4648):  96%|█████████▌| 902/938 [04:59<00:11,  3.14it/s]
Training 1/1 epoch (loss 1.5321):  96%|█████████▌| 902/938 [04:59<00:11,  3.14it/s]
Training 1/1 epoch (loss 1.5321):  96%|█████████▋| 903/938 [04:59<00:11,  3.13it/s]
Training 1/1 epoch (loss 1.6033):  96%|█████████▋| 903/938 [05:00<00:11,  3.13it/s]
Training 1/1 epoch (loss 1.6033):  96%|█████████▋| 904/938 [05:00<00:11,  3.03it/s]
Training 1/1 epoch (loss 1.5703):  96%|█████████▋| 904/938 [05:00<00:11,  3.03it/s]
Training 1/1 epoch (loss 1.5703):  96%|█████████▋| 905/938 [05:00<00:10,  3.12it/s]
Training 1/1 epoch (loss 1.5849):  96%|█████████▋| 905/938 [05:00<00:10,  3.12it/s]
Training 1/1 epoch (loss 1.5849):  97%|█████████▋| 906/938 [05:00<00:10,  3.09it/s]
Training 1/1 epoch (loss 1.4479):  97%|█████████▋| 906/938 [05:01<00:10,  3.09it/s]
Training 1/1 epoch (loss 1.4479):  97%|█████████▋| 907/938 [05:01<00:09,  3.14it/s]
Training 1/1 epoch (loss 1.5448):  97%|█████████▋| 907/938 [05:01<00:09,  3.14it/s]
Training 1/1 epoch (loss 1.5448):  97%|█████████▋| 908/938 [05:01<00:09,  3.20it/s]
Training 1/1 epoch (loss 1.4818):  97%|█████████▋| 908/938 [05:01<00:09,  3.20it/s]
Training 1/1 epoch (loss 1.4818):  97%|█████████▋| 909/938 [05:01<00:08,  3.24it/s]
Training 1/1 epoch (loss 1.5678):  97%|█████████▋| 909/938 [05:02<00:08,  3.24it/s]
Training 1/1 epoch (loss 1.5678):  97%|█████████▋| 910/938 [05:02<00:08,  3.29it/s]
Training 1/1 epoch (loss 1.5252):  97%|█████████▋| 910/938 [05:02<00:08,  3.29it/s]
Training 1/1 epoch (loss 1.5252):  97%|█████████▋| 911/938 [05:02<00:08,  3.19it/s]
Training 1/1 epoch (loss 1.5444):  97%|█████████▋| 911/938 [05:02<00:08,  3.19it/s]
Training 1/1 epoch (loss 1.5444):  97%|█████████▋| 912/938 [05:02<00:08,  3.11it/s]
Training 1/1 epoch (loss 1.5509):  97%|█████████▋| 912/938 [05:03<00:08,  3.11it/s]
Training 1/1 epoch (loss 1.5509):  97%|█████████▋| 913/938 [05:03<00:07,  3.13it/s]
Training 1/1 epoch (loss 1.5627):  97%|█████████▋| 913/938 [05:03<00:07,  3.13it/s]
Training 1/1 epoch (loss 1.5627):  97%|█████████▋| 914/938 [05:03<00:07,  3.16it/s]
Training 1/1 epoch (loss 1.5139):  97%|█████████▋| 914/938 [05:03<00:07,  3.16it/s]
Training 1/1 epoch (loss 1.5139):  98%|█████████▊| 915/938 [05:03<00:07,  3.18it/s]
Training 1/1 epoch (loss 1.5852):  98%|█████████▊| 915/938 [05:03<00:07,  3.18it/s]
Training 1/1 epoch (loss 1.5852):  98%|█████████▊| 916/938 [05:03<00:06,  3.24it/s]
Training 1/1 epoch (loss 1.5107):  98%|█████████▊| 916/938 [05:04<00:06,  3.24it/s]
Training 1/1 epoch (loss 1.5107):  98%|█████████▊| 917/938 [05:04<00:06,  3.15it/s]
Training 1/1 epoch (loss 1.4975):  98%|█████████▊| 917/938 [05:04<00:06,  3.15it/s]
Training 1/1 epoch (loss 1.4975):  98%|█████████▊| 918/938 [05:04<00:06,  3.13it/s]
Training 1/1 epoch (loss 1.5054):  98%|█████████▊| 918/938 [05:04<00:06,  3.13it/s]
Training 1/1 epoch (loss 1.5054):  98%|█████████▊| 919/938 [05:04<00:06,  3.06it/s]
Training 1/1 epoch (loss 1.5309):  98%|█████████▊| 919/938 [05:05<00:06,  3.06it/s]
Training 1/1 epoch (loss 1.5309):  98%|█████████▊| 920/938 [05:05<00:05,  3.04it/s]
Training 1/1 epoch (loss 1.5070):  98%|█████████▊| 920/938 [05:05<00:05,  3.04it/s]
Training 1/1 epoch (loss 1.5070):  98%|█████████▊| 921/938 [05:05<00:05,  3.11it/s]
Training 1/1 epoch (loss 1.4739):  98%|█████████▊| 921/938 [05:05<00:05,  3.11it/s]
Training 1/1 epoch (loss 1.4739):  98%|█████████▊| 922/938 [05:05<00:05,  3.15it/s]
Training 1/1 epoch (loss 1.4511):  98%|█████████▊| 922/938 [05:06<00:05,  3.15it/s]
Training 1/1 epoch (loss 1.4511):  98%|█████████▊| 923/938 [05:06<00:05,  3.00it/s]
Training 1/1 epoch (loss 1.5123):  98%|█████████▊| 923/938 [05:06<00:05,  3.00it/s]
Training 1/1 epoch (loss 1.5123):  99%|█████████▊| 924/938 [05:06<00:04,  3.01it/s]
Training 1/1 epoch (loss 1.5278):  99%|█████████▊| 924/938 [05:06<00:04,  3.01it/s]
Training 1/1 epoch (loss 1.5278):  99%|█████████▊| 925/938 [05:06<00:04,  3.08it/s]
Training 1/1 epoch (loss 1.4407):  99%|█████████▊| 925/938 [05:07<00:04,  3.08it/s]
Training 1/1 epoch (loss 1.4407):  99%|█████████▊| 926/938 [05:07<00:03,  3.15it/s]
Training 1/1 epoch (loss 1.5205):  99%|█████████▊| 926/938 [05:07<00:03,  3.15it/s]
Training 1/1 epoch (loss 1.5205):  99%|█████████▉| 927/938 [05:07<00:03,  3.18it/s]
Training 1/1 epoch (loss 1.5022):  99%|█████████▉| 927/938 [05:07<00:03,  3.18it/s]
Training 1/1 epoch (loss 1.5022):  99%|█████████▉| 928/938 [05:07<00:03,  3.17it/s]
Training 1/1 epoch (loss 1.4787):  99%|█████████▉| 928/938 [05:08<00:03,  3.17it/s]
Training 1/1 epoch (loss 1.4787):  99%|█████████▉| 929/938 [05:08<00:02,  3.18it/s]
Training 1/1 epoch (loss 1.5318):  99%|█████████▉| 929/938 [05:08<00:02,  3.18it/s]
Training 1/1 epoch (loss 1.5318):  99%|█████████▉| 930/938 [05:08<00:02,  3.16it/s]
Training 1/1 epoch (loss 1.4681):  99%|█████████▉| 930/938 [05:08<00:02,  3.16it/s]
Training 1/1 epoch (loss 1.4681):  99%|█████████▉| 931/938 [05:08<00:02,  3.18it/s]
Training 1/1 epoch (loss 1.5486):  99%|█████████▉| 931/938 [05:09<00:02,  3.18it/s]
Training 1/1 epoch (loss 1.5486):  99%|█████████▉| 932/938 [05:09<00:01,  3.07it/s]
Training 1/1 epoch (loss 1.5389):  99%|█████████▉| 932/938 [05:09<00:01,  3.07it/s]
Training 1/1 epoch (loss 1.5389):  99%|█████████▉| 933/938 [05:09<00:01,  3.10it/s]
Training 1/1 epoch (loss 1.5312):  99%|█████████▉| 933/938 [05:09<00:01,  3.10it/s]
Training 1/1 epoch (loss 1.5312): 100%|█████████▉| 934/938 [05:09<00:01,  3.16it/s]
Training 1/1 epoch (loss 1.6599): 100%|█████████▉| 934/938 [05:10<00:01,  3.16it/s]
Training 1/1 epoch (loss 1.6599): 100%|█████████▉| 935/938 [05:10<00:00,  3.18it/s]
Training 1/1 epoch (loss 1.5348): 100%|█████████▉| 935/938 [05:10<00:00,  3.18it/s]
Training 1/1 epoch (loss 1.5348): 100%|█████████▉| 936/938 [05:10<00:00,  3.07it/s]
Training 1/1 epoch (loss 1.5127): 100%|█████████▉| 936/938 [05:10<00:00,  3.07it/s]
Training 1/1 epoch (loss 1.5127): 100%|█████████▉| 937/938 [05:10<00:00,  2.99it/s]
Training 1/1 epoch (loss 1.5660): 100%|█████████▉| 937/938 [05:11<00:00,  2.99it/s]
Training 1/1 epoch (loss 1.5660): 100%|██████████| 938/938 [05:11<00:00,  3.02it/s]
Training 1/1 epoch (loss 1.5660): 100%|██████████| 938/938 [05:11<00:00,  3.02it/s]
tokenizer config file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-2T/tinyllama-2T-s3-Q1-30k/tokenizer_config.json
Special tokens file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-2T/tinyllama-2T-s3-Q1-30k/special_tokens_map.json
wandb: ERROR Problem finishing run
Exception ignored in atexit callback: <bound method rank_zero_only.<locals>.wrapper of <safe_rlhf.logger.Logger object at 0x15505f9df950>>
Traceback (most recent call last):
  File "/home/hansirui_1st/jiayi/resist/setting3/safe_rlhf/utils.py", line 212, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/hansirui_1st/jiayi/resist/setting3/safe_rlhf/logger.py", line 183, in close
    self.wandb.finish()
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 449, in wrapper
    return func(self, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 391, in wrapper
    return func(self, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2106, in finish
    return self._finish(exit_code)
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2127, in _finish
    self._atexit_cleanup(exit_code=exit_code)
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2352, in _atexit_cleanup
    self._on_finish()
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2609, in _on_finish
    wait_with_progress(
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/mailbox/wait_with_progress.py", line 24, in wait_with_progress
    return wait_all_with_progress(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/mailbox/wait_with_progress.py", line 87, in wait_all_with_progress
    return asyncio_compat.run(progress_loop_with_timeout)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/lib/asyncio_compat.py", line 27, in run
    future = executor.submit(runner.run, fn)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/concurrent/futures/thread.py", line 169, in submit
    raise RuntimeError('cannot schedule new futures after '
RuntimeError: cannot schedule new futures after interpreter shutdown