stderr.log

+ deepspeed --master_port 23492 --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-1431k-3T --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-3T/tinyllama-3T-s3-Q1-30k --log_type wandb --log_run_name tinyllama-3T-s3-Q1-30k --log_project Inverse_Alignment --zero_stage 3 --offload none --bf16 True --tf32 True --save_16bit
[rank1]:[W529 02:14:35.331153081 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.
[rank7]:[W529 02:14:35.375353911 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.
[rank4]:[W529 02:14:35.400698596 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.
[rank6]:[W529 02:14:35.402550858 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.
[rank2]:[W529 02:14:35.498670538 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.
[rank5]:[W529 02:14:35.583207408 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.
[rank0]:[W529 02:14:35.727667235 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.
[rank3]:[W529 02:14:35.785046247 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.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/config.json
Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T",
  "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-1431k-3T",
  "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-1431k-3T",
  "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-1431k-3T",
  "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-1431k-3T",
  "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-1431k-3T",
  "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-1431k-3T",
  "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-1431k-3T",
  "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-1431k-3T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/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.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/model.safetensors
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
}

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
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-1431k-3T.
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-1431k-3T.
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-1431k-3T.
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-1431k-3T.
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-1431k-3T.
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-1431k-3T.
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-1431k-3T.
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-1431k-3T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/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
}

loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/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
}

loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T/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.model
loading file tokenizer.model
loading file tokenizer.json
loading file tokenizer.json
loading file tokenizer.model
loading file added_tokens.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer.model
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file tokenizer.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file added_tokens.json
loading file chat_template.jinja
loading file tokenizer.json
loading file special_tokens_map.json
loading file added_tokens.json
loading file tokenizer_config.json
loading file special_tokens_map.json
loading file chat_template.jinja
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer.model
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.json
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file added_tokens.json
loading file tokenizer_config.json
loading file special_tokens_map.json
loading file chat_template.jinja
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-1431k-3T.
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-1431k-3T/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-3T/tinyllama-3T-s3-Q1-30k/wandb/run-20250529_021446-1oeru2an
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run tinyllama-3T-s3-Q1-30k
wandb: ⭐️ View project at https://wandb.ai/xtom/Inverse_Alignment
wandb: 🚀 View run at https://wandb.ai/xtom/Inverse_Alignment/runs/1oeru2an

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 2.0020):   0%|          | 0/938 [00:06<?, ?it/s]
Training 1/1 epoch (loss 2.0020):   0%|          | 1/938 [00:06<1:46:48,  6.84s/it]
Training 1/1 epoch (loss 1.9623):   0%|          | 1/938 [00:08<1:46:48,  6.84s/it]
Training 1/1 epoch (loss 1.9623):   0%|          | 2/938 [00:08<1:01:05,  3.92s/it]
Training 1/1 epoch (loss 1.9595):   0%|          | 2/938 [00:09<1:01:05,  3.92s/it]
Training 1/1 epoch (loss 1.9595):   0%|          | 3/938 [00:09<35:56,  2.31s/it]  
Training 1/1 epoch (loss 1.9877):   0%|          | 3/938 [00:09<35:56,  2.31s/it]
Training 1/1 epoch (loss 1.9877):   0%|          | 4/938 [00:09<24:03,  1.55s/it]
Training 1/1 epoch (loss 1.9219):   0%|          | 4/938 [00:09<24:03,  1.55s/it]
Training 1/1 epoch (loss 1.9219):   1%|          | 5/938 [00:09<17:30,  1.13s/it]
Training 1/1 epoch (loss 1.9014):   1%|          | 5/938 [00:10<17:30,  1.13s/it]
Training 1/1 epoch (loss 1.9014):   1%|          | 6/938 [00:10<13:36,  1.14it/s]
Training 1/1 epoch (loss 1.9475):   1%|          | 6/938 [00:10<13:36,  1.14it/s]
Training 1/1 epoch (loss 1.9475):   1%|          | 7/938 [00:10<11:01,  1.41it/s]
Training 1/1 epoch (loss 2.0038):   1%|          | 7/938 [00:11<11:01,  1.41it/s]
Training 1/1 epoch (loss 2.0038):   1%|          | 8/938 [00:11<10:06,  1.53it/s]
Training 1/1 epoch (loss 1.7789):   1%|          | 8/938 [00:11<10:06,  1.53it/s]
Training 1/1 epoch (loss 1.7789):   1%|          | 9/938 [00:11<09:00,  1.72it/s]
Training 1/1 epoch (loss 1.7788):   1%|          | 9/938 [00:11<09:00,  1.72it/s]
Training 1/1 epoch (loss 1.7788):   1%|          | 10/938 [00:11<07:59,  1.94it/s]
Training 1/1 epoch (loss 1.8612):   1%|          | 10/938 [00:12<07:59,  1.94it/s]
Training 1/1 epoch (loss 1.8612):   1%|          | 11/938 [00:12<07:23,  2.09it/s]
Training 1/1 epoch (loss 1.8916):   1%|          | 11/938 [00:12<07:23,  2.09it/s]
Training 1/1 epoch (loss 1.8916):   1%|▏         | 12/938 [00:12<06:47,  2.27it/s]
Training 1/1 epoch (loss 1.9388):   1%|▏         | 12/938 [00:13<06:47,  2.27it/s]
Training 1/1 epoch (loss 1.9388):   1%|▏         | 13/938 [00:13<06:31,  2.36it/s]
Training 1/1 epoch (loss 1.6995):   1%|▏         | 13/938 [00:13<06:31,  2.36it/s]
Training 1/1 epoch (loss 1.6995):   1%|▏         | 14/938 [00:13<06:31,  2.36it/s]
Training 1/1 epoch (loss 1.9198):   1%|▏         | 14/938 [00:13<06:31,  2.36it/s]
Training 1/1 epoch (loss 1.9198):   2%|▏         | 15/938 [00:13<06:18,  2.44it/s]
Training 1/1 epoch (loss 1.8617):   2%|▏         | 15/938 [00:14<06:18,  2.44it/s]
Training 1/1 epoch (loss 1.8617):   2%|▏         | 16/938 [00:14<06:16,  2.45it/s]
Training 1/1 epoch (loss 1.6163):   2%|▏         | 16/938 [00:14<06:16,  2.45it/s]
Training 1/1 epoch (loss 1.6163):   2%|▏         | 17/938 [00:14<06:42,  2.29it/s]
Training 1/1 epoch (loss 1.7746):   2%|▏         | 17/938 [00:15<06:42,  2.29it/s]
Training 1/1 epoch (loss 1.7746):   2%|▏         | 18/938 [00:15<06:38,  2.31it/s]
Training 1/1 epoch (loss 1.5777):   2%|▏         | 18/938 [00:15<06:38,  2.31it/s]
Training 1/1 epoch (loss 1.5777):   2%|▏         | 19/938 [00:15<06:31,  2.35it/s]
Training 1/1 epoch (loss 1.6825):   2%|▏         | 19/938 [00:15<06:31,  2.35it/s]
Training 1/1 epoch (loss 1.6825):   2%|▏         | 20/938 [00:15<06:10,  2.48it/s]
Training 1/1 epoch (loss 1.7394):   2%|▏         | 20/938 [00:16<06:10,  2.48it/s]
Training 1/1 epoch (loss 1.7394):   2%|▏         | 21/938 [00:16<06:06,  2.50it/s]
Training 1/1 epoch (loss 1.7963):   2%|▏         | 21/938 [00:16<06:06,  2.50it/s]
Training 1/1 epoch (loss 1.7963):   2%|▏         | 22/938 [00:16<06:03,  2.52it/s]
Training 1/1 epoch (loss 1.6556):   2%|▏         | 22/938 [00:17<06:03,  2.52it/s]
Training 1/1 epoch (loss 1.6556):   2%|▏         | 23/938 [00:17<05:59,  2.55it/s]
Training 1/1 epoch (loss 1.7121):   2%|▏         | 23/938 [00:17<05:59,  2.55it/s]
Training 1/1 epoch (loss 1.7121):   3%|▎         | 24/938 [00:17<05:57,  2.56it/s]
Training 1/1 epoch (loss 1.7337):   3%|▎         | 24/938 [00:17<05:57,  2.56it/s]
Training 1/1 epoch (loss 1.7337):   3%|▎         | 25/938 [00:17<06:04,  2.50it/s]
Training 1/1 epoch (loss 1.7382):   3%|▎         | 25/938 [00:18<06:04,  2.50it/s]
Training 1/1 epoch (loss 1.7382):   3%|▎         | 26/938 [00:18<05:59,  2.54it/s]
Training 1/1 epoch (loss 1.6627):   3%|▎         | 26/938 [00:18<05:59,  2.54it/s]
Training 1/1 epoch (loss 1.6627):   3%|▎         | 27/938 [00:18<05:54,  2.57it/s]
Training 1/1 epoch (loss 1.5895):   3%|▎         | 27/938 [00:19<05:54,  2.57it/s]
Training 1/1 epoch (loss 1.5895):   3%|▎         | 28/938 [00:19<05:49,  2.60it/s]
Training 1/1 epoch (loss 1.7182):   3%|▎         | 28/938 [00:19<05:49,  2.60it/s]
Training 1/1 epoch (loss 1.7182):   3%|▎         | 29/938 [00:19<05:49,  2.60it/s]
Training 1/1 epoch (loss 1.7012):   3%|▎         | 29/938 [00:19<05:49,  2.60it/s]
Training 1/1 epoch (loss 1.7012):   3%|▎         | 30/938 [00:19<05:46,  2.62it/s]
Training 1/1 epoch (loss 1.7042):   3%|▎         | 30/938 [00:20<05:46,  2.62it/s]
Training 1/1 epoch (loss 1.7042):   3%|▎         | 31/938 [00:20<05:42,  2.65it/s]
Training 1/1 epoch (loss 1.7146):   3%|▎         | 31/938 [00:20<05:42,  2.65it/s]
Training 1/1 epoch (loss 1.7146):   3%|▎         | 32/938 [00:20<05:38,  2.67it/s]
Training 1/1 epoch (loss 1.6340):   3%|▎         | 32/938 [00:20<05:38,  2.67it/s]
Training 1/1 epoch (loss 1.6340):   4%|▎         | 33/938 [00:20<05:45,  2.62it/s]
Training 1/1 epoch (loss 1.6364):   4%|▎         | 33/938 [00:21<05:45,  2.62it/s]
Training 1/1 epoch (loss 1.6364):   4%|▎         | 34/938 [00:21<05:43,  2.63it/s]
Training 1/1 epoch (loss 1.6306):   4%|▎         | 34/938 [00:21<05:43,  2.63it/s]
Training 1/1 epoch (loss 1.6306):   4%|▎         | 35/938 [00:21<06:00,  2.50it/s]
Training 1/1 epoch (loss 1.6232):   4%|▎         | 35/938 [00:22<06:00,  2.50it/s]
Training 1/1 epoch (loss 1.6232):   4%|▍         | 36/938 [00:22<05:46,  2.60it/s]
Training 1/1 epoch (loss 1.7123):   4%|▍         | 36/938 [00:22<05:46,  2.60it/s]
Training 1/1 epoch (loss 1.7123):   4%|▍         | 37/938 [00:22<05:46,  2.60it/s]
Training 1/1 epoch (loss 1.6155):   4%|▍         | 37/938 [00:22<05:46,  2.60it/s]
Training 1/1 epoch (loss 1.6155):   4%|▍         | 38/938 [00:22<05:48,  2.59it/s]
Training 1/1 epoch (loss 1.6060):   4%|▍         | 38/938 [00:23<05:48,  2.59it/s]
Training 1/1 epoch (loss 1.6060):   4%|▍         | 39/938 [00:23<05:46,  2.59it/s]
Training 1/1 epoch (loss 1.6385):   4%|▍         | 39/938 [00:23<05:46,  2.59it/s]
Training 1/1 epoch (loss 1.6385):   4%|▍         | 40/938 [00:23<05:43,  2.61it/s]
Training 1/1 epoch (loss 1.5081):   4%|▍         | 40/938 [00:24<05:43,  2.61it/s]
Training 1/1 epoch (loss 1.5081):   4%|▍         | 41/938 [00:24<05:53,  2.53it/s]
Training 1/1 epoch (loss 1.6386):   4%|▍         | 41/938 [00:24<05:53,  2.53it/s]
Training 1/1 epoch (loss 1.6386):   4%|▍         | 42/938 [00:24<05:46,  2.59it/s]
Training 1/1 epoch (loss 1.6192):   4%|▍         | 42/938 [00:24<05:46,  2.59it/s]
Training 1/1 epoch (loss 1.6192):   5%|▍         | 43/938 [00:24<05:51,  2.55it/s]
Training 1/1 epoch (loss 1.5711):   5%|▍         | 43/938 [00:25<05:51,  2.55it/s]
Training 1/1 epoch (loss 1.5711):   5%|▍         | 44/938 [00:25<06:08,  2.43it/s]
Training 1/1 epoch (loss 1.5964):   5%|▍         | 44/938 [00:25<06:08,  2.43it/s]
Training 1/1 epoch (loss 1.5964):   5%|▍         | 45/938 [00:25<05:48,  2.56it/s]
Training 1/1 epoch (loss 1.5633):   5%|▍         | 45/938 [00:26<05:48,  2.56it/s]
Training 1/1 epoch (loss 1.5633):   5%|▍         | 46/938 [00:26<05:49,  2.56it/s]
Training 1/1 epoch (loss 1.6139):   5%|▍         | 46/938 [00:26<05:49,  2.56it/s]
Training 1/1 epoch (loss 1.6139):   5%|▌         | 47/938 [00:26<05:48,  2.55it/s]
Training 1/1 epoch (loss 1.6456):   5%|▌         | 47/938 [00:26<05:48,  2.55it/s]
Training 1/1 epoch (loss 1.6456):   5%|▌         | 48/938 [00:26<05:50,  2.54it/s]
Training 1/1 epoch (loss 1.6700):   5%|▌         | 48/938 [00:27<05:50,  2.54it/s]
Training 1/1 epoch (loss 1.6700):   5%|▌         | 49/938 [00:27<06:08,  2.41it/s]
Training 1/1 epoch (loss 1.4809):   5%|▌         | 49/938 [00:27<06:08,  2.41it/s]
Training 1/1 epoch (loss 1.4809):   5%|▌         | 50/938 [00:27<05:59,  2.47it/s]
Training 1/1 epoch (loss 1.5375):   5%|▌         | 50/938 [00:28<05:59,  2.47it/s]
Training 1/1 epoch (loss 1.5375):   5%|▌         | 51/938 [00:28<05:55,  2.50it/s]
Training 1/1 epoch (loss 1.5325):   5%|▌         | 51/938 [00:28<05:55,  2.50it/s]
Training 1/1 epoch (loss 1.5325):   6%|▌         | 52/938 [00:28<05:49,  2.54it/s]
Training 1/1 epoch (loss 1.4956):   6%|▌         | 52/938 [00:28<05:49,  2.54it/s]
Training 1/1 epoch (loss 1.4956):   6%|▌         | 53/938 [00:28<05:42,  2.58it/s]
Training 1/1 epoch (loss 1.6047):   6%|▌         | 53/938 [00:29<05:42,  2.58it/s]
Training 1/1 epoch (loss 1.6047):   6%|▌         | 54/938 [00:29<05:35,  2.63it/s]
Training 1/1 epoch (loss 1.6696):   6%|▌         | 54/938 [00:29<05:35,  2.63it/s]
Training 1/1 epoch (loss 1.6696):   6%|▌         | 55/938 [00:29<05:55,  2.49it/s]
Training 1/1 epoch (loss 1.5259):   6%|▌         | 55/938 [00:30<05:55,  2.49it/s]
Training 1/1 epoch (loss 1.5259):   6%|▌         | 56/938 [00:30<06:01,  2.44it/s]
Training 1/1 epoch (loss 1.5290):   6%|▌         | 56/938 [00:30<06:01,  2.44it/s]
Training 1/1 epoch (loss 1.5290):   6%|▌         | 57/938 [00:30<05:56,  2.47it/s]
Training 1/1 epoch (loss 1.5763):   6%|▌         | 57/938 [00:30<05:56,  2.47it/s]
Training 1/1 epoch (loss 1.5763):   6%|▌         | 58/938 [00:30<05:47,  2.53it/s]
Training 1/1 epoch (loss 1.5916):   6%|▌         | 58/938 [00:31<05:47,  2.53it/s]
Training 1/1 epoch (loss 1.5916):   6%|▋         | 59/938 [00:31<05:38,  2.60it/s]
Training 1/1 epoch (loss 1.6278):   6%|▋         | 59/938 [00:31<05:38,  2.60it/s]
Training 1/1 epoch (loss 1.6278):   6%|▋         | 60/938 [00:31<05:44,  2.55it/s]
Training 1/1 epoch (loss 1.6524):   6%|▋         | 60/938 [00:31<05:44,  2.55it/s]
Training 1/1 epoch (loss 1.6524):   7%|▋         | 61/938 [00:31<05:36,  2.60it/s]
Training 1/1 epoch (loss 1.6300):   7%|▋         | 61/938 [00:32<05:36,  2.60it/s]
Training 1/1 epoch (loss 1.6300):   7%|▋         | 62/938 [00:32<05:49,  2.51it/s]
Training 1/1 epoch (loss 1.6445):   7%|▋         | 62/938 [00:32<05:49,  2.51it/s]
Training 1/1 epoch (loss 1.6445):   7%|▋         | 63/938 [00:32<05:41,  2.57it/s]
Training 1/1 epoch (loss 1.5066):   7%|▋         | 63/938 [00:33<05:41,  2.57it/s]
Training 1/1 epoch (loss 1.5066):   7%|▋         | 64/938 [00:33<05:55,  2.46it/s]
Training 1/1 epoch (loss 1.6075):   7%|▋         | 64/938 [00:33<05:55,  2.46it/s]
Training 1/1 epoch (loss 1.6075):   7%|▋         | 65/938 [00:33<05:55,  2.46it/s]
Training 1/1 epoch (loss 1.6110):   7%|▋         | 65/938 [00:34<05:55,  2.46it/s]
Training 1/1 epoch (loss 1.6110):   7%|▋         | 66/938 [00:34<05:43,  2.54it/s]
Training 1/1 epoch (loss 1.6042):   7%|▋         | 66/938 [00:34<05:43,  2.54it/s]
Training 1/1 epoch (loss 1.6042):   7%|▋         | 67/938 [00:34<05:51,  2.48it/s]
Training 1/1 epoch (loss 1.5220):   7%|▋         | 67/938 [00:34<05:51,  2.48it/s]
Training 1/1 epoch (loss 1.5220):   7%|▋         | 68/938 [00:34<05:51,  2.48it/s]
Training 1/1 epoch (loss 1.5888):   7%|▋         | 68/938 [00:35<05:51,  2.48it/s]
Training 1/1 epoch (loss 1.5888):   7%|▋         | 69/938 [00:35<07:04,  2.05it/s]
Training 1/1 epoch (loss 1.6713):   7%|▋         | 69/938 [00:36<07:04,  2.05it/s]
Training 1/1 epoch (loss 1.6713):   7%|▋         | 70/938 [00:36<07:29,  1.93it/s]
Training 1/1 epoch (loss 1.4924):   7%|▋         | 70/938 [00:36<07:29,  1.93it/s]
Training 1/1 epoch (loss 1.4924):   8%|▊         | 71/938 [00:36<06:50,  2.11it/s]
Training 1/1 epoch (loss 1.5830):   8%|▊         | 71/938 [00:36<06:50,  2.11it/s]
Training 1/1 epoch (loss 1.5830):   8%|▊         | 72/938 [00:36<06:28,  2.23it/s]
Training 1/1 epoch (loss 1.5051):   8%|▊         | 72/938 [00:37<06:28,  2.23it/s]
Training 1/1 epoch (loss 1.5051):   8%|▊         | 73/938 [00:37<06:08,  2.35it/s]
Training 1/1 epoch (loss 1.5194):   8%|▊         | 73/938 [00:37<06:08,  2.35it/s]
Training 1/1 epoch (loss 1.5194):   8%|▊         | 74/938 [00:37<06:05,  2.37it/s]
Training 1/1 epoch (loss 1.4526):   8%|▊         | 74/938 [00:38<06:05,  2.37it/s]
Training 1/1 epoch (loss 1.4526):   8%|▊         | 75/938 [00:38<05:52,  2.45it/s]
Training 1/1 epoch (loss 1.5372):   8%|▊         | 75/938 [00:38<05:52,  2.45it/s]
Training 1/1 epoch (loss 1.5372):   8%|▊         | 76/938 [00:38<05:51,  2.45it/s]
Training 1/1 epoch (loss 1.4428):   8%|▊         | 76/938 [00:38<05:51,  2.45it/s]
Training 1/1 epoch (loss 1.4428):   8%|▊         | 77/938 [00:38<05:40,  2.53it/s]
Training 1/1 epoch (loss 1.6149):   8%|▊         | 77/938 [00:39<05:40,  2.53it/s]
Training 1/1 epoch (loss 1.6149):   8%|▊         | 78/938 [00:39<05:37,  2.54it/s]
Training 1/1 epoch (loss 1.5406):   8%|▊         | 78/938 [00:39<05:37,  2.54it/s]
Training 1/1 epoch (loss 1.5406):   8%|▊         | 79/938 [00:39<05:35,  2.56it/s]
Training 1/1 epoch (loss 1.6405):   8%|▊         | 79/938 [00:40<05:35,  2.56it/s]
Training 1/1 epoch (loss 1.6405):   9%|▊         | 80/938 [00:40<05:43,  2.50it/s]
Training 1/1 epoch (loss 1.5750):   9%|▊         | 80/938 [00:40<05:43,  2.50it/s]
Training 1/1 epoch (loss 1.5750):   9%|▊         | 81/938 [00:40<05:46,  2.47it/s]
Training 1/1 epoch (loss 1.4918):   9%|▊         | 81/938 [00:40<05:46,  2.47it/s]
Training 1/1 epoch (loss 1.4918):   9%|▊         | 82/938 [00:40<05:47,  2.47it/s]
Training 1/1 epoch (loss 1.5317):   9%|▊         | 82/938 [00:41<05:47,  2.47it/s]
Training 1/1 epoch (loss 1.5317):   9%|▉         | 83/938 [00:41<05:52,  2.42it/s]
Training 1/1 epoch (loss 1.5125):   9%|▉         | 83/938 [00:41<05:52,  2.42it/s]
Training 1/1 epoch (loss 1.5125):   9%|▉         | 84/938 [00:41<05:41,  2.50it/s]
Training 1/1 epoch (loss 1.5397):   9%|▉         | 84/938 [00:41<05:41,  2.50it/s]
Training 1/1 epoch (loss 1.5397):   9%|▉         | 85/938 [00:41<05:35,  2.55it/s]
Training 1/1 epoch (loss 1.5386):   9%|▉         | 85/938 [00:42<05:35,  2.55it/s]
Training 1/1 epoch (loss 1.5386):   9%|▉         | 86/938 [00:42<05:42,  2.49it/s]
Training 1/1 epoch (loss 1.5002):   9%|▉         | 86/938 [00:42<05:42,  2.49it/s]
Training 1/1 epoch (loss 1.5002):   9%|▉         | 87/938 [00:42<05:40,  2.50it/s]
Training 1/1 epoch (loss 1.5102):   9%|▉         | 87/938 [00:43<05:40,  2.50it/s]
Training 1/1 epoch (loss 1.5102):   9%|▉         | 88/938 [00:43<05:41,  2.49it/s]
Training 1/1 epoch (loss 1.5528):   9%|▉         | 88/938 [00:43<05:41,  2.49it/s]
Training 1/1 epoch (loss 1.5528):   9%|▉         | 89/938 [00:43<05:37,  2.52it/s]
Training 1/1 epoch (loss 1.5526):   9%|▉         | 89/938 [00:44<05:37,  2.52it/s]
Training 1/1 epoch (loss 1.5526):  10%|▉         | 90/938 [00:44<05:39,  2.50it/s]
Training 1/1 epoch (loss 1.5946):  10%|▉         | 90/938 [00:44<05:39,  2.50it/s]
Training 1/1 epoch (loss 1.5946):  10%|▉         | 91/938 [00:44<05:43,  2.46it/s]
Training 1/1 epoch (loss 1.4170):  10%|▉         | 91/938 [00:44<05:43,  2.46it/s]
Training 1/1 epoch (loss 1.4170):  10%|▉         | 92/938 [00:44<05:47,  2.44it/s]
Training 1/1 epoch (loss 1.6410):  10%|▉         | 92/938 [00:45<05:47,  2.44it/s]
Training 1/1 epoch (loss 1.6410):  10%|▉         | 93/938 [00:45<05:50,  2.41it/s]
Training 1/1 epoch (loss 1.5597):  10%|▉         | 93/938 [00:45<05:50,  2.41it/s]
Training 1/1 epoch (loss 1.5597):  10%|█         | 94/938 [00:45<05:49,  2.41it/s]
Training 1/1 epoch (loss 1.5092):  10%|█         | 94/938 [00:46<05:49,  2.41it/s]
Training 1/1 epoch (loss 1.5092):  10%|█         | 95/938 [00:46<05:44,  2.45it/s]
Training 1/1 epoch (loss 1.5375):  10%|█         | 95/938 [00:46<05:44,  2.45it/s]
Training 1/1 epoch (loss 1.5375):  10%|█         | 96/938 [00:46<05:44,  2.44it/s]
Training 1/1 epoch (loss 1.6066):  10%|█         | 96/938 [00:46<05:44,  2.44it/s]
Training 1/1 epoch (loss 1.6066):  10%|█         | 97/938 [00:46<05:35,  2.51it/s]
Training 1/1 epoch (loss 1.5110):  10%|█         | 97/938 [00:47<05:35,  2.51it/s]
Training 1/1 epoch (loss 1.5110):  10%|█         | 98/938 [00:47<05:28,  2.56it/s]
Training 1/1 epoch (loss 1.6181):  10%|█         | 98/938 [00:47<05:28,  2.56it/s]
Training 1/1 epoch (loss 1.6181):  11%|█         | 99/938 [00:47<05:27,  2.57it/s]
Training 1/1 epoch (loss 1.3632):  11%|█         | 99/938 [00:48<05:27,  2.57it/s]
Training 1/1 epoch (loss 1.3632):  11%|█         | 100/938 [00:48<05:37,  2.48it/s]
Training 1/1 epoch (loss 1.5292):  11%|█         | 100/938 [00:48<05:37,  2.48it/s]
Training 1/1 epoch (loss 1.5292):  11%|█         | 101/938 [00:48<05:33,  2.51it/s]
Training 1/1 epoch (loss 1.5342):  11%|█         | 101/938 [00:48<05:33,  2.51it/s]
Training 1/1 epoch (loss 1.5342):  11%|█         | 102/938 [00:48<05:29,  2.54it/s]
Training 1/1 epoch (loss 1.4986):  11%|█         | 102/938 [00:49<05:29,  2.54it/s]
Training 1/1 epoch (loss 1.4986):  11%|█         | 103/938 [00:49<05:26,  2.56it/s]
Training 1/1 epoch (loss 1.5745):  11%|█         | 103/938 [00:49<05:26,  2.56it/s]
Training 1/1 epoch (loss 1.5745):  11%|█         | 104/938 [00:49<05:27,  2.55it/s]
Training 1/1 epoch (loss 1.5777):  11%|█         | 104/938 [00:50<05:27,  2.55it/s]
Training 1/1 epoch (loss 1.5777):  11%|█         | 105/938 [00:50<05:24,  2.57it/s]
Training 1/1 epoch (loss 1.5610):  11%|█         | 105/938 [00:50<05:24,  2.57it/s]
Training 1/1 epoch (loss 1.5610):  11%|█▏        | 106/938 [00:50<05:26,  2.55it/s]
Training 1/1 epoch (loss 1.5748):  11%|█▏        | 106/938 [00:50<05:26,  2.55it/s]
Training 1/1 epoch (loss 1.5748):  11%|█▏        | 107/938 [00:50<05:33,  2.49it/s]
Training 1/1 epoch (loss 1.6864):  11%|█▏        | 107/938 [00:51<05:33,  2.49it/s]
Training 1/1 epoch (loss 1.6864):  12%|█▏        | 108/938 [00:51<05:26,  2.54it/s]
Training 1/1 epoch (loss 1.6250):  12%|█▏        | 108/938 [00:51<05:26,  2.54it/s]
Training 1/1 epoch (loss 1.6250):  12%|█▏        | 109/938 [00:51<05:20,  2.58it/s]
Training 1/1 epoch (loss 1.6531):  12%|█▏        | 109/938 [00:51<05:20,  2.58it/s]
Training 1/1 epoch (loss 1.6531):  12%|█▏        | 110/938 [00:51<05:25,  2.54it/s]
Training 1/1 epoch (loss 1.3879):  12%|█▏        | 110/938 [00:52<05:25,  2.54it/s]
Training 1/1 epoch (loss 1.3879):  12%|█▏        | 111/938 [00:52<05:26,  2.53it/s]
Training 1/1 epoch (loss 1.6206):  12%|█▏        | 111/938 [00:52<05:26,  2.53it/s]
Training 1/1 epoch (loss 1.6206):  12%|█▏        | 112/938 [00:52<05:17,  2.60it/s]
Training 1/1 epoch (loss 1.5157):  12%|█▏        | 112/938 [00:53<05:17,  2.60it/s]
Training 1/1 epoch (loss 1.5157):  12%|█▏        | 113/938 [00:53<05:30,  2.50it/s]
Training 1/1 epoch (loss 1.4934):  12%|█▏        | 113/938 [00:53<05:30,  2.50it/s]
Training 1/1 epoch (loss 1.4934):  12%|█▏        | 114/938 [00:53<05:29,  2.50it/s]
Training 1/1 epoch (loss 1.4553):  12%|█▏        | 114/938 [00:53<05:29,  2.50it/s]
Training 1/1 epoch (loss 1.4553):  12%|█▏        | 115/938 [00:53<05:23,  2.54it/s]
Training 1/1 epoch (loss 1.5928):  12%|█▏        | 115/938 [00:54<05:23,  2.54it/s]
Training 1/1 epoch (loss 1.5928):  12%|█▏        | 116/938 [00:54<05:22,  2.55it/s]
Training 1/1 epoch (loss 1.5098):  12%|█▏        | 116/938 [00:54<05:22,  2.55it/s]
Training 1/1 epoch (loss 1.5098):  12%|█▏        | 117/938 [00:54<05:19,  2.57it/s]
Training 1/1 epoch (loss 1.5226):  12%|█▏        | 117/938 [00:55<05:19,  2.57it/s]
Training 1/1 epoch (loss 1.5226):  13%|█▎        | 118/938 [00:55<05:29,  2.49it/s]
Training 1/1 epoch (loss 1.5912):  13%|█▎        | 118/938 [00:55<05:29,  2.49it/s]
Training 1/1 epoch (loss 1.5912):  13%|█▎        | 119/938 [00:55<05:25,  2.51it/s]
Training 1/1 epoch (loss 1.4741):  13%|█▎        | 119/938 [00:55<05:25,  2.51it/s]
Training 1/1 epoch (loss 1.4741):  13%|█▎        | 120/938 [00:55<05:24,  2.52it/s]
Training 1/1 epoch (loss 1.5596):  13%|█▎        | 120/938 [00:56<05:24,  2.52it/s]
Training 1/1 epoch (loss 1.5596):  13%|█▎        | 121/938 [00:56<06:01,  2.26it/s]
Training 1/1 epoch (loss 1.5409):  13%|█▎        | 121/938 [00:56<06:01,  2.26it/s]
Training 1/1 epoch (loss 1.5409):  13%|█▎        | 122/938 [00:56<05:46,  2.35it/s]
Training 1/1 epoch (loss 1.4897):  13%|█▎        | 122/938 [00:57<05:46,  2.35it/s]
Training 1/1 epoch (loss 1.4897):  13%|█▎        | 123/938 [00:57<05:43,  2.37it/s]
Training 1/1 epoch (loss 1.5358):  13%|█▎        | 123/938 [00:57<05:43,  2.37it/s]
Training 1/1 epoch (loss 1.5358):  13%|█▎        | 124/938 [00:57<05:28,  2.48it/s]
Training 1/1 epoch (loss 1.5794):  13%|█▎        | 124/938 [00:58<05:28,  2.48it/s]
Training 1/1 epoch (loss 1.5794):  13%|█▎        | 125/938 [00:58<05:17,  2.56it/s]
Training 1/1 epoch (loss 1.6394):  13%|█▎        | 125/938 [00:58<05:17,  2.56it/s]
Training 1/1 epoch (loss 1.6394):  13%|█▎        | 126/938 [00:58<05:34,  2.43it/s]
Training 1/1 epoch (loss 1.4156):  13%|█▎        | 126/938 [00:58<05:34,  2.43it/s]
Training 1/1 epoch (loss 1.4156):  14%|█▎        | 127/938 [00:58<05:25,  2.49it/s]
Training 1/1 epoch (loss 1.4734):  14%|█▎        | 127/938 [00:59<05:25,  2.49it/s]
Training 1/1 epoch (loss 1.4734):  14%|█▎        | 128/938 [00:59<05:19,  2.53it/s]
Training 1/1 epoch (loss 1.5361):  14%|█▎        | 128/938 [00:59<05:19,  2.53it/s]
Training 1/1 epoch (loss 1.5361):  14%|█▍        | 129/938 [00:59<05:16,  2.56it/s]
Training 1/1 epoch (loss 1.4609):  14%|█▍        | 129/938 [01:00<05:16,  2.56it/s]
Training 1/1 epoch (loss 1.4609):  14%|█▍        | 130/938 [01:00<05:17,  2.55it/s]
Training 1/1 epoch (loss 1.5205):  14%|█▍        | 130/938 [01:00<05:17,  2.55it/s]
Training 1/1 epoch (loss 1.5205):  14%|█▍        | 131/938 [01:00<05:23,  2.49it/s]
Training 1/1 epoch (loss 1.4458):  14%|█▍        | 131/938 [01:00<05:23,  2.49it/s]
Training 1/1 epoch (loss 1.4458):  14%|█▍        | 132/938 [01:00<05:15,  2.56it/s]
Training 1/1 epoch (loss 1.5556):  14%|█▍        | 132/938 [01:01<05:15,  2.56it/s]
Training 1/1 epoch (loss 1.5556):  14%|█▍        | 133/938 [01:01<05:13,  2.57it/s]
Training 1/1 epoch (loss 1.6735):  14%|█▍        | 133/938 [01:01<05:13,  2.57it/s]
Training 1/1 epoch (loss 1.6735):  14%|█▍        | 134/938 [01:01<05:16,  2.54it/s]
Training 1/1 epoch (loss 1.4471):  14%|█▍        | 134/938 [01:01<05:16,  2.54it/s]
Training 1/1 epoch (loss 1.4471):  14%|█▍        | 135/938 [01:01<05:12,  2.57it/s]
Training 1/1 epoch (loss 1.5605):  14%|█▍        | 135/938 [01:02<05:12,  2.57it/s]
Training 1/1 epoch (loss 1.5605):  14%|█▍        | 136/938 [01:02<05:25,  2.47it/s]
Training 1/1 epoch (loss 1.5365):  14%|█▍        | 136/938 [01:02<05:25,  2.47it/s]
Training 1/1 epoch (loss 1.5365):  15%|█▍        | 137/938 [01:02<05:12,  2.56it/s]
Training 1/1 epoch (loss 1.5132):  15%|█▍        | 137/938 [01:03<05:12,  2.56it/s]
Training 1/1 epoch (loss 1.5132):  15%|█▍        | 138/938 [01:03<05:11,  2.57it/s]
Training 1/1 epoch (loss 1.5261):  15%|█▍        | 138/938 [01:03<05:11,  2.57it/s]
Training 1/1 epoch (loss 1.5261):  15%|█▍        | 139/938 [01:03<05:16,  2.52it/s]
Training 1/1 epoch (loss 1.5091):  15%|█▍        | 139/938 [01:03<05:16,  2.52it/s]
Training 1/1 epoch (loss 1.5091):  15%|█▍        | 140/938 [01:03<05:24,  2.46it/s]
Training 1/1 epoch (loss 1.5720):  15%|█▍        | 140/938 [01:04<05:24,  2.46it/s]
Training 1/1 epoch (loss 1.5720):  15%|█▌        | 141/938 [01:04<05:20,  2.49it/s]
Training 1/1 epoch (loss 1.5545):  15%|█▌        | 141/938 [01:04<05:20,  2.49it/s]
Training 1/1 epoch (loss 1.5545):  15%|█▌        | 142/938 [01:04<05:39,  2.34it/s]
Training 1/1 epoch (loss 1.4848):  15%|█▌        | 142/938 [01:05<05:39,  2.34it/s]
Training 1/1 epoch (loss 1.4848):  15%|█▌        | 143/938 [01:05<06:04,  2.18it/s]
Training 1/1 epoch (loss 1.5924):  15%|█▌        | 143/938 [01:06<06:04,  2.18it/s]
Training 1/1 epoch (loss 1.5924):  15%|█▌        | 144/938 [01:06<06:47,  1.95it/s]
Training 1/1 epoch (loss 1.5086):  15%|█▌        | 144/938 [01:06<06:47,  1.95it/s]
Training 1/1 epoch (loss 1.5086):  15%|█▌        | 145/938 [01:06<06:18,  2.10it/s]
Training 1/1 epoch (loss 1.4889):  15%|█▌        | 145/938 [01:06<06:18,  2.10it/s]
Training 1/1 epoch (loss 1.4889):  16%|█▌        | 146/938 [01:06<05:52,  2.25it/s]
Training 1/1 epoch (loss 1.3625):  16%|█▌        | 146/938 [01:07<05:52,  2.25it/s]
Training 1/1 epoch (loss 1.3625):  16%|█▌        | 147/938 [01:07<05:38,  2.34it/s]
Training 1/1 epoch (loss 1.6068):  16%|█▌        | 147/938 [01:07<05:38,  2.34it/s]
Training 1/1 epoch (loss 1.6068):  16%|█▌        | 148/938 [01:07<05:24,  2.44it/s]
Training 1/1 epoch (loss 1.6343):  16%|█▌        | 148/938 [01:07<05:24,  2.44it/s]
Training 1/1 epoch (loss 1.6343):  16%|█▌        | 149/938 [01:07<05:11,  2.53it/s]
Training 1/1 epoch (loss 1.4105):  16%|█▌        | 149/938 [01:08<05:11,  2.53it/s]
Training 1/1 epoch (loss 1.4105):  16%|█▌        | 150/938 [01:08<05:13,  2.51it/s]
Training 1/1 epoch (loss 1.5200):  16%|█▌        | 150/938 [01:08<05:13,  2.51it/s]
Training 1/1 epoch (loss 1.5200):  16%|█▌        | 151/938 [01:08<05:10,  2.53it/s]
Training 1/1 epoch (loss 1.4667):  16%|█▌        | 151/938 [01:09<05:10,  2.53it/s]
Training 1/1 epoch (loss 1.4667):  16%|█▌        | 152/938 [01:09<05:23,  2.43it/s]
Training 1/1 epoch (loss 1.5153):  16%|█▌        | 152/938 [01:09<05:23,  2.43it/s]
Training 1/1 epoch (loss 1.5153):  16%|█▋        | 153/938 [01:09<05:16,  2.48it/s]
Training 1/1 epoch (loss 1.4149):  16%|█▋        | 153/938 [01:09<05:16,  2.48it/s]
Training 1/1 epoch (loss 1.4149):  16%|█▋        | 154/938 [01:09<05:05,  2.57it/s]
Training 1/1 epoch (loss 1.6328):  16%|█▋        | 154/938 [01:10<05:05,  2.57it/s]
Training 1/1 epoch (loss 1.6328):  17%|█▋        | 155/938 [01:10<05:04,  2.57it/s]
Training 1/1 epoch (loss 1.5349):  17%|█▋        | 155/938 [01:10<05:04,  2.57it/s]
Training 1/1 epoch (loss 1.5349):  17%|█▋        | 156/938 [01:10<04:58,  2.62it/s]
Training 1/1 epoch (loss 1.5426):  17%|█▋        | 156/938 [01:11<04:58,  2.62it/s]
Training 1/1 epoch (loss 1.5426):  17%|█▋        | 157/938 [01:11<04:54,  2.65it/s]
Training 1/1 epoch (loss 1.4327):  17%|█▋        | 157/938 [01:11<04:54,  2.65it/s]
Training 1/1 epoch (loss 1.4327):  17%|█▋        | 158/938 [01:11<05:05,  2.56it/s]
Training 1/1 epoch (loss 1.6094):  17%|█▋        | 158/938 [01:11<05:05,  2.56it/s]
Training 1/1 epoch (loss 1.6094):  17%|█▋        | 159/938 [01:11<04:55,  2.64it/s]
Training 1/1 epoch (loss 1.5255):  17%|█▋        | 159/938 [01:12<04:55,  2.64it/s]
Training 1/1 epoch (loss 1.5255):  17%|█▋        | 160/938 [01:12<05:03,  2.56it/s]
Training 1/1 epoch (loss 1.4489):  17%|█▋        | 160/938 [01:12<05:03,  2.56it/s]
Training 1/1 epoch (loss 1.4489):  17%|█▋        | 161/938 [01:12<05:00,  2.58it/s]
Training 1/1 epoch (loss 1.5201):  17%|█▋        | 161/938 [01:13<05:00,  2.58it/s]
Training 1/1 epoch (loss 1.5201):  17%|█▋        | 162/938 [01:13<05:09,  2.50it/s]
Training 1/1 epoch (loss 1.5152):  17%|█▋        | 162/938 [01:13<05:09,  2.50it/s]
Training 1/1 epoch (loss 1.5152):  17%|█▋        | 163/938 [01:13<05:18,  2.44it/s]
Training 1/1 epoch (loss 1.5225):  17%|█▋        | 163/938 [01:13<05:18,  2.44it/s]
Training 1/1 epoch (loss 1.5225):  17%|█▋        | 164/938 [01:13<05:12,  2.48it/s]
Training 1/1 epoch (loss 1.5483):  17%|█▋        | 164/938 [01:14<05:12,  2.48it/s]
Training 1/1 epoch (loss 1.5483):  18%|█▊        | 165/938 [01:14<05:34,  2.31it/s]
Training 1/1 epoch (loss 1.4924):  18%|█▊        | 165/938 [01:14<05:34,  2.31it/s]
Training 1/1 epoch (loss 1.4924):  18%|█▊        | 166/938 [01:14<05:26,  2.36it/s]
Training 1/1 epoch (loss 1.5315):  18%|█▊        | 166/938 [01:15<05:26,  2.36it/s]
Training 1/1 epoch (loss 1.5315):  18%|█▊        | 167/938 [01:15<05:19,  2.41it/s]
Training 1/1 epoch (loss 1.4051):  18%|█▊        | 167/938 [01:15<05:19,  2.41it/s]
Training 1/1 epoch (loss 1.4051):  18%|█▊        | 168/938 [01:15<05:19,  2.41it/s]
Training 1/1 epoch (loss 1.6316):  18%|█▊        | 168/938 [01:15<05:19,  2.41it/s]
Training 1/1 epoch (loss 1.6316):  18%|█▊        | 169/938 [01:15<05:05,  2.52it/s]
Training 1/1 epoch (loss 1.5233):  18%|█▊        | 169/938 [01:16<05:05,  2.52it/s]
Training 1/1 epoch (loss 1.5233):  18%|█▊        | 170/938 [01:16<05:05,  2.51it/s]
Training 1/1 epoch (loss 1.4981):  18%|█▊        | 170/938 [01:16<05:05,  2.51it/s]
Training 1/1 epoch (loss 1.4981):  18%|█▊        | 171/938 [01:16<04:56,  2.59it/s]
Training 1/1 epoch (loss 1.4350):  18%|█▊        | 171/938 [01:17<04:56,  2.59it/s]
Training 1/1 epoch (loss 1.4350):  18%|█▊        | 172/938 [01:17<05:11,  2.46it/s]
Training 1/1 epoch (loss 1.5016):  18%|█▊        | 172/938 [01:17<05:11,  2.46it/s]
Training 1/1 epoch (loss 1.5016):  18%|█▊        | 173/938 [01:17<05:10,  2.47it/s]
Training 1/1 epoch (loss 1.6200):  18%|█▊        | 173/938 [01:17<05:10,  2.47it/s]
Training 1/1 epoch (loss 1.6200):  19%|█▊        | 174/938 [01:17<05:12,  2.45it/s]
Training 1/1 epoch (loss 1.5493):  19%|█▊        | 174/938 [01:18<05:12,  2.45it/s]
Training 1/1 epoch (loss 1.5493):  19%|█▊        | 175/938 [01:18<05:04,  2.50it/s]
Training 1/1 epoch (loss 1.3655):  19%|█▊        | 175/938 [01:18<05:04,  2.50it/s]
Training 1/1 epoch (loss 1.3655):  19%|█▉        | 176/938 [01:18<05:11,  2.45it/s]
Training 1/1 epoch (loss 1.4817):  19%|█▉        | 176/938 [01:19<05:11,  2.45it/s]
Training 1/1 epoch (loss 1.4817):  19%|█▉        | 177/938 [01:19<05:03,  2.51it/s]
Training 1/1 epoch (loss 1.5334):  19%|█▉        | 177/938 [01:19<05:03,  2.51it/s]
Training 1/1 epoch (loss 1.5334):  19%|█▉        | 178/938 [01:19<05:15,  2.41it/s]
Training 1/1 epoch (loss 1.5951):  19%|█▉        | 178/938 [01:20<05:15,  2.41it/s]
Training 1/1 epoch (loss 1.5951):  19%|█▉        | 179/938 [01:20<05:17,  2.39it/s]
Training 1/1 epoch (loss 1.4618):  19%|█▉        | 179/938 [01:20<05:17,  2.39it/s]
Training 1/1 epoch (loss 1.4618):  19%|█▉        | 180/938 [01:20<05:14,  2.41it/s]
Training 1/1 epoch (loss 1.4838):  19%|█▉        | 180/938 [01:20<05:14,  2.41it/s]
Training 1/1 epoch (loss 1.4838):  19%|█▉        | 181/938 [01:20<05:00,  2.52it/s]
Training 1/1 epoch (loss 1.3436):  19%|█▉        | 181/938 [01:21<05:00,  2.52it/s]
Training 1/1 epoch (loss 1.3436):  19%|█▉        | 182/938 [01:21<04:58,  2.53it/s]
Training 1/1 epoch (loss 1.6164):  19%|█▉        | 182/938 [01:21<04:58,  2.53it/s]
Training 1/1 epoch (loss 1.6164):  20%|█▉        | 183/938 [01:21<05:00,  2.51it/s]
Training 1/1 epoch (loss 1.4871):  20%|█▉        | 183/938 [01:21<05:00,  2.51it/s]
Training 1/1 epoch (loss 1.4871):  20%|█▉        | 184/938 [01:21<04:57,  2.54it/s]
Training 1/1 epoch (loss 1.5292):  20%|█▉        | 184/938 [01:22<04:57,  2.54it/s]
Training 1/1 epoch (loss 1.5292):  20%|█▉        | 185/938 [01:22<04:55,  2.55it/s]
Training 1/1 epoch (loss 1.5097):  20%|█▉        | 185/938 [01:22<04:55,  2.55it/s]
Training 1/1 epoch (loss 1.5097):  20%|█▉        | 186/938 [01:22<04:46,  2.63it/s]
Training 1/1 epoch (loss 1.4307):  20%|█▉        | 186/938 [01:23<04:46,  2.63it/s]
Training 1/1 epoch (loss 1.4307):  20%|█▉        | 187/938 [01:23<04:48,  2.60it/s]
Training 1/1 epoch (loss 1.4849):  20%|█▉        | 187/938 [01:23<04:48,  2.60it/s]
Training 1/1 epoch (loss 1.4849):  20%|██        | 188/938 [01:23<04:52,  2.57it/s]
Training 1/1 epoch (loss 1.5396):  20%|██        | 188/938 [01:23<04:52,  2.57it/s]
Training 1/1 epoch (loss 1.5396):  20%|██        | 189/938 [01:23<04:48,  2.59it/s]
Training 1/1 epoch (loss 1.7218):  20%|██        | 189/938 [01:24<04:48,  2.59it/s]
Training 1/1 epoch (loss 1.7218):  20%|██        | 190/938 [01:24<04:49,  2.59it/s]
Training 1/1 epoch (loss 1.4634):  20%|██        | 190/938 [01:24<04:49,  2.59it/s]
Training 1/1 epoch (loss 1.4634):  20%|██        | 191/938 [01:24<04:51,  2.57it/s]
Training 1/1 epoch (loss 1.4780):  20%|██        | 191/938 [01:25<04:51,  2.57it/s]
Training 1/1 epoch (loss 1.4780):  20%|██        | 192/938 [01:25<05:10,  2.40it/s]
Training 1/1 epoch (loss 1.5118):  20%|██        | 192/938 [01:25<05:10,  2.40it/s]
Training 1/1 epoch (loss 1.5118):  21%|██        | 193/938 [01:25<05:16,  2.35it/s]
Training 1/1 epoch (loss 1.4923):  21%|██        | 193/938 [01:25<05:16,  2.35it/s]
Training 1/1 epoch (loss 1.4923):  21%|██        | 194/938 [01:25<05:10,  2.40it/s]
Training 1/1 epoch (loss 1.4912):  21%|██        | 194/938 [01:26<05:10,  2.40it/s]
Training 1/1 epoch (loss 1.4912):  21%|██        | 195/938 [01:26<05:33,  2.23it/s]
Training 1/1 epoch (loss 1.5236):  21%|██        | 195/938 [01:26<05:33,  2.23it/s]
Training 1/1 epoch (loss 1.5236):  21%|██        | 196/938 [01:26<05:26,  2.27it/s]
Training 1/1 epoch (loss 1.5551):  21%|██        | 196/938 [01:27<05:26,  2.27it/s]
Training 1/1 epoch (loss 1.5551):  21%|██        | 197/938 [01:27<05:40,  2.18it/s]
Training 1/1 epoch (loss 1.4136):  21%|██        | 197/938 [01:27<05:40,  2.18it/s]
Training 1/1 epoch (loss 1.4136):  21%|██        | 198/938 [01:27<05:37,  2.19it/s]
Training 1/1 epoch (loss 1.4954):  21%|██        | 198/938 [01:28<05:37,  2.19it/s]
Training 1/1 epoch (loss 1.4954):  21%|██        | 199/938 [01:28<05:16,  2.34it/s]
Training 1/1 epoch (loss 1.4490):  21%|██        | 199/938 [01:28<05:16,  2.34it/s]
Training 1/1 epoch (loss 1.4490):  21%|██▏       | 200/938 [01:28<05:17,  2.33it/s]
Training 1/1 epoch (loss 1.4882):  21%|██▏       | 200/938 [01:29<05:17,  2.33it/s]
Training 1/1 epoch (loss 1.4882):  21%|██▏       | 201/938 [01:29<05:01,  2.44it/s]
Training 1/1 epoch (loss 1.5032):  21%|██▏       | 201/938 [01:29<05:01,  2.44it/s]
Training 1/1 epoch (loss 1.5032):  22%|██▏       | 202/938 [01:29<05:00,  2.45it/s]
Training 1/1 epoch (loss 1.5195):  22%|██▏       | 202/938 [01:29<05:00,  2.45it/s]
Training 1/1 epoch (loss 1.5195):  22%|██▏       | 203/938 [01:29<05:00,  2.44it/s]
Training 1/1 epoch (loss 1.4513):  22%|██▏       | 203/938 [01:30<05:00,  2.44it/s]
Training 1/1 epoch (loss 1.4513):  22%|██▏       | 204/938 [01:30<05:07,  2.39it/s]
Training 1/1 epoch (loss 1.5058):  22%|██▏       | 204/938 [01:30<05:07,  2.39it/s]
Training 1/1 epoch (loss 1.5058):  22%|██▏       | 205/938 [01:30<04:57,  2.46it/s]
Training 1/1 epoch (loss 1.5842):  22%|██▏       | 205/938 [01:31<04:57,  2.46it/s]
Training 1/1 epoch (loss 1.5842):  22%|██▏       | 206/938 [01:31<04:48,  2.54it/s]
Training 1/1 epoch (loss 1.4993):  22%|██▏       | 206/938 [01:31<04:48,  2.54it/s]
Training 1/1 epoch (loss 1.4993):  22%|██▏       | 207/938 [01:31<04:52,  2.50it/s]
Training 1/1 epoch (loss 1.5575):  22%|██▏       | 207/938 [01:31<04:52,  2.50it/s]
Training 1/1 epoch (loss 1.5575):  22%|██▏       | 208/938 [01:31<04:57,  2.46it/s]
Training 1/1 epoch (loss 1.4829):  22%|██▏       | 208/938 [01:32<04:57,  2.46it/s]
Training 1/1 epoch (loss 1.4829):  22%|██▏       | 209/938 [01:32<05:08,  2.37it/s]
Training 1/1 epoch (loss 1.4617):  22%|██▏       | 209/938 [01:32<05:08,  2.37it/s]
Training 1/1 epoch (loss 1.4617):  22%|██▏       | 210/938 [01:32<04:55,  2.47it/s]
Training 1/1 epoch (loss 1.4719):  22%|██▏       | 210/938 [01:33<04:55,  2.47it/s]
Training 1/1 epoch (loss 1.4719):  22%|██▏       | 211/938 [01:33<04:54,  2.47it/s]
Training 1/1 epoch (loss 1.3829):  22%|██▏       | 211/938 [01:33<04:54,  2.47it/s]
Training 1/1 epoch (loss 1.3829):  23%|██▎       | 212/938 [01:33<04:58,  2.43it/s]
Training 1/1 epoch (loss 1.4132):  23%|██▎       | 212/938 [01:33<04:58,  2.43it/s]
Training 1/1 epoch (loss 1.4132):  23%|██▎       | 213/938 [01:33<05:01,  2.40it/s]
Training 1/1 epoch (loss 1.4269):  23%|██▎       | 213/938 [01:34<05:01,  2.40it/s]
Training 1/1 epoch (loss 1.4269):  23%|██▎       | 214/938 [01:34<04:57,  2.43it/s]
Training 1/1 epoch (loss 1.5241):  23%|██▎       | 214/938 [01:34<04:57,  2.43it/s]
Training 1/1 epoch (loss 1.5241):  23%|██▎       | 215/938 [01:34<04:53,  2.46it/s]
Training 1/1 epoch (loss 1.4263):  23%|██▎       | 215/938 [01:35<04:53,  2.46it/s]
Training 1/1 epoch (loss 1.4263):  23%|██▎       | 216/938 [01:35<05:33,  2.16it/s]
Training 1/1 epoch (loss 1.6101):  23%|██▎       | 216/938 [01:36<05:33,  2.16it/s]
Training 1/1 epoch (loss 1.6101):  23%|██▎       | 217/938 [01:36<06:39,  1.81it/s]
Training 1/1 epoch (loss 1.5387):  23%|██▎       | 217/938 [01:36<06:39,  1.81it/s]
Training 1/1 epoch (loss 1.5387):  23%|██▎       | 218/938 [01:36<06:10,  1.94it/s]
Training 1/1 epoch (loss 1.4693):  23%|██▎       | 218/938 [01:36<06:10,  1.94it/s]
Training 1/1 epoch (loss 1.4693):  23%|██▎       | 219/938 [01:36<05:38,  2.12it/s]
Training 1/1 epoch (loss 1.4113):  23%|██▎       | 219/938 [01:37<05:38,  2.12it/s]
Training 1/1 epoch (loss 1.4113):  23%|██▎       | 220/938 [01:37<05:37,  2.13it/s]
Training 1/1 epoch (loss 1.5499):  23%|██▎       | 220/938 [01:37<05:37,  2.13it/s]
Training 1/1 epoch (loss 1.5499):  24%|██▎       | 221/938 [01:37<05:18,  2.25it/s]
Training 1/1 epoch (loss 1.4759):  24%|██▎       | 221/938 [01:38<05:18,  2.25it/s]
Training 1/1 epoch (loss 1.4759):  24%|██▎       | 222/938 [01:38<05:06,  2.34it/s]
Training 1/1 epoch (loss 1.6005):  24%|██▎       | 222/938 [01:38<05:06,  2.34it/s]
Training 1/1 epoch (loss 1.6005):  24%|██▍       | 223/938 [01:38<04:55,  2.42it/s]
Training 1/1 epoch (loss 1.5403):  24%|██▍       | 223/938 [01:38<04:55,  2.42it/s]
Training 1/1 epoch (loss 1.5403):  24%|██▍       | 224/938 [01:38<04:46,  2.49it/s]
Training 1/1 epoch (loss 1.5050):  24%|██▍       | 224/938 [01:39<04:46,  2.49it/s]
Training 1/1 epoch (loss 1.5050):  24%|██▍       | 225/938 [01:39<04:56,  2.41it/s]
Training 1/1 epoch (loss 1.5974):  24%|██▍       | 225/938 [01:39<04:56,  2.41it/s]
Training 1/1 epoch (loss 1.5974):  24%|██▍       | 226/938 [01:39<04:45,  2.49it/s]
Training 1/1 epoch (loss 1.4900):  24%|██▍       | 226/938 [01:40<04:45,  2.49it/s]
Training 1/1 epoch (loss 1.4900):  24%|██▍       | 227/938 [01:40<04:52,  2.43it/s]
Training 1/1 epoch (loss 1.5132):  24%|██▍       | 227/938 [01:40<04:52,  2.43it/s]
Training 1/1 epoch (loss 1.5132):  24%|██▍       | 228/938 [01:40<04:47,  2.47it/s]
Training 1/1 epoch (loss 1.5400):  24%|██▍       | 228/938 [01:40<04:47,  2.47it/s]
Training 1/1 epoch (loss 1.5400):  24%|██▍       | 229/938 [01:40<04:40,  2.53it/s]
Training 1/1 epoch (loss 1.5459):  24%|██▍       | 229/938 [01:41<04:40,  2.53it/s]
Training 1/1 epoch (loss 1.5459):  25%|██▍       | 230/938 [01:41<04:37,  2.55it/s]
Training 1/1 epoch (loss 1.4707):  25%|██▍       | 230/938 [01:41<04:37,  2.55it/s]
Training 1/1 epoch (loss 1.4707):  25%|██▍       | 231/938 [01:41<04:32,  2.59it/s]
Training 1/1 epoch (loss 1.5035):  25%|██▍       | 231/938 [01:42<04:32,  2.59it/s]
Training 1/1 epoch (loss 1.5035):  25%|██▍       | 232/938 [01:42<04:39,  2.53it/s]
Training 1/1 epoch (loss 1.5022):  25%|██▍       | 232/938 [01:42<04:39,  2.53it/s]
Training 1/1 epoch (loss 1.5022):  25%|██▍       | 233/938 [01:42<04:38,  2.53it/s]
Training 1/1 epoch (loss 1.4794):  25%|██▍       | 233/938 [01:42<04:38,  2.53it/s]
Training 1/1 epoch (loss 1.4794):  25%|██▍       | 234/938 [01:42<04:33,  2.58it/s]
Training 1/1 epoch (loss 1.4822):  25%|██▍       | 234/938 [01:43<04:33,  2.58it/s]
Training 1/1 epoch (loss 1.4822):  25%|██▌       | 235/938 [01:43<04:31,  2.59it/s]
Training 1/1 epoch (loss 1.4054):  25%|██▌       | 235/938 [01:43<04:31,  2.59it/s]
Training 1/1 epoch (loss 1.4054):  25%|██▌       | 236/938 [01:43<04:32,  2.58it/s]
Training 1/1 epoch (loss 1.5899):  25%|██▌       | 236/938 [01:43<04:32,  2.58it/s]
Training 1/1 epoch (loss 1.5899):  25%|██▌       | 237/938 [01:43<04:30,  2.60it/s]
Training 1/1 epoch (loss 1.3846):  25%|██▌       | 237/938 [01:44<04:30,  2.60it/s]
Training 1/1 epoch (loss 1.3846):  25%|██▌       | 238/938 [01:44<04:26,  2.63it/s]
Training 1/1 epoch (loss 1.4586):  25%|██▌       | 238/938 [01:44<04:26,  2.63it/s]
Training 1/1 epoch (loss 1.4586):  25%|██▌       | 239/938 [01:44<04:42,  2.48it/s]
Training 1/1 epoch (loss 1.4638):  25%|██▌       | 239/938 [01:45<04:42,  2.48it/s]
Training 1/1 epoch (loss 1.4638):  26%|██▌       | 240/938 [01:45<05:01,  2.31it/s]
Training 1/1 epoch (loss 1.6142):  26%|██▌       | 240/938 [01:45<05:01,  2.31it/s]
Training 1/1 epoch (loss 1.6142):  26%|██▌       | 241/938 [01:45<05:02,  2.30it/s]
Training 1/1 epoch (loss 1.5047):  26%|██▌       | 241/938 [01:46<05:02,  2.30it/s]
Training 1/1 epoch (loss 1.5047):  26%|██▌       | 242/938 [01:46<04:53,  2.37it/s]
Training 1/1 epoch (loss 1.5184):  26%|██▌       | 242/938 [01:46<04:53,  2.37it/s]
Training 1/1 epoch (loss 1.5184):  26%|██▌       | 243/938 [01:46<04:45,  2.43it/s]
Training 1/1 epoch (loss 1.3770):  26%|██▌       | 243/938 [01:46<04:45,  2.43it/s]
Training 1/1 epoch (loss 1.3770):  26%|██▌       | 244/938 [01:46<04:38,  2.49it/s]
Training 1/1 epoch (loss 1.4266):  26%|██▌       | 244/938 [01:47<04:38,  2.49it/s]
Training 1/1 epoch (loss 1.4266):  26%|██▌       | 245/938 [01:47<04:33,  2.53it/s]
Training 1/1 epoch (loss 1.4827):  26%|██▌       | 245/938 [01:47<04:33,  2.53it/s]
Training 1/1 epoch (loss 1.4827):  26%|██▌       | 246/938 [01:47<04:29,  2.57it/s]
Training 1/1 epoch (loss 1.4735):  26%|██▌       | 246/938 [01:48<04:29,  2.57it/s]
Training 1/1 epoch (loss 1.4735):  26%|██▋       | 247/938 [01:48<04:47,  2.40it/s]
Training 1/1 epoch (loss 1.4460):  26%|██▋       | 247/938 [01:48<04:47,  2.40it/s]
Training 1/1 epoch (loss 1.4460):  26%|██▋       | 248/938 [01:48<04:50,  2.38it/s]
Training 1/1 epoch (loss 1.5027):  26%|██▋       | 248/938 [01:49<04:50,  2.38it/s]
Training 1/1 epoch (loss 1.5027):  27%|██▋       | 249/938 [01:49<04:53,  2.35it/s]
Training 1/1 epoch (loss 1.3708):  27%|██▋       | 249/938 [01:49<04:53,  2.35it/s]
Training 1/1 epoch (loss 1.3708):  27%|██▋       | 250/938 [01:49<04:43,  2.43it/s]
Training 1/1 epoch (loss 1.6117):  27%|██▋       | 250/938 [01:49<04:43,  2.43it/s]
Training 1/1 epoch (loss 1.6117):  27%|██▋       | 251/938 [01:49<04:36,  2.49it/s]
Training 1/1 epoch (loss 1.5280):  27%|██▋       | 251/938 [01:50<04:36,  2.49it/s]
Training 1/1 epoch (loss 1.5280):  27%|██▋       | 252/938 [01:50<04:36,  2.48it/s]
Training 1/1 epoch (loss 1.4955):  27%|██▋       | 252/938 [01:50<04:36,  2.48it/s]
Training 1/1 epoch (loss 1.4955):  27%|██▋       | 253/938 [01:50<04:29,  2.54it/s]
Training 1/1 epoch (loss 1.5053):  27%|██▋       | 253/938 [01:50<04:29,  2.54it/s]
Training 1/1 epoch (loss 1.5053):  27%|██▋       | 254/938 [01:50<04:27,  2.55it/s]
Training 1/1 epoch (loss 1.5049):  27%|██▋       | 254/938 [01:51<04:27,  2.55it/s]
Training 1/1 epoch (loss 1.5049):  27%|██▋       | 255/938 [01:51<04:33,  2.50it/s]
Training 1/1 epoch (loss 1.3954):  27%|██▋       | 255/938 [01:51<04:33,  2.50it/s]
Training 1/1 epoch (loss 1.3954):  27%|██▋       | 256/938 [01:51<04:26,  2.56it/s]
Training 1/1 epoch (loss 1.4757):  27%|██▋       | 256/938 [01:52<04:26,  2.56it/s]
Training 1/1 epoch (loss 1.4757):  27%|██▋       | 257/938 [01:52<04:35,  2.48it/s]
Training 1/1 epoch (loss 1.4993):  27%|██▋       | 257/938 [01:52<04:35,  2.48it/s]
Training 1/1 epoch (loss 1.4993):  28%|██▊       | 258/938 [01:52<04:30,  2.51it/s]
Training 1/1 epoch (loss 1.5188):  28%|██▊       | 258/938 [01:52<04:30,  2.51it/s]
Training 1/1 epoch (loss 1.5188):  28%|██▊       | 259/938 [01:52<04:22,  2.59it/s]
Training 1/1 epoch (loss 1.4522):  28%|██▊       | 259/938 [01:53<04:22,  2.59it/s]
Training 1/1 epoch (loss 1.4522):  28%|██▊       | 260/938 [01:53<04:24,  2.56it/s]
Training 1/1 epoch (loss 1.4686):  28%|██▊       | 260/938 [01:53<04:24,  2.56it/s]
Training 1/1 epoch (loss 1.4686):  28%|██▊       | 261/938 [01:53<04:22,  2.58it/s]
Training 1/1 epoch (loss 1.4515):  28%|██▊       | 261/938 [01:54<04:22,  2.58it/s]
Training 1/1 epoch (loss 1.4515):  28%|██▊       | 262/938 [01:54<04:22,  2.58it/s]
Training 1/1 epoch (loss 1.3550):  28%|██▊       | 262/938 [01:54<04:22,  2.58it/s]
Training 1/1 epoch (loss 1.3550):  28%|██▊       | 263/938 [01:54<04:34,  2.46it/s]
Training 1/1 epoch (loss 1.5947):  28%|██▊       | 263/938 [01:55<04:34,  2.46it/s]
Training 1/1 epoch (loss 1.5947):  28%|██▊       | 264/938 [01:55<04:57,  2.27it/s]
Training 1/1 epoch (loss 1.4540):  28%|██▊       | 264/938 [01:55<04:57,  2.27it/s]
Training 1/1 epoch (loss 1.4540):  28%|██▊       | 265/938 [01:55<05:23,  2.08it/s]
Training 1/1 epoch (loss 1.4999):  28%|██▊       | 265/938 [01:56<05:23,  2.08it/s]
Training 1/1 epoch (loss 1.4999):  28%|██▊       | 266/938 [01:56<05:25,  2.07it/s]
Training 1/1 epoch (loss 1.4054):  28%|██▊       | 266/938 [01:56<05:25,  2.07it/s]
Training 1/1 epoch (loss 1.4054):  28%|██▊       | 267/938 [01:56<05:29,  2.04it/s]
Training 1/1 epoch (loss 1.4059):  28%|██▊       | 267/938 [01:57<05:29,  2.04it/s]
Training 1/1 epoch (loss 1.4059):  29%|██▊       | 268/938 [01:57<05:15,  2.12it/s]
Training 1/1 epoch (loss 1.4112):  29%|██▊       | 268/938 [01:57<05:15,  2.12it/s]
Training 1/1 epoch (loss 1.4112):  29%|██▊       | 269/938 [01:57<05:05,  2.19it/s]
Training 1/1 epoch (loss 1.5069):  29%|██▊       | 269/938 [01:57<05:05,  2.19it/s]
Training 1/1 epoch (loss 1.5069):  29%|██▉       | 270/938 [01:57<04:51,  2.29it/s]
Training 1/1 epoch (loss 1.5692):  29%|██▉       | 270/938 [01:58<04:51,  2.29it/s]
Training 1/1 epoch (loss 1.5692):  29%|██▉       | 271/938 [01:58<04:41,  2.37it/s]
Training 1/1 epoch (loss 1.3878):  29%|██▉       | 271/938 [01:58<04:41,  2.37it/s]
Training 1/1 epoch (loss 1.3878):  29%|██▉       | 272/938 [01:58<04:39,  2.39it/s]
Training 1/1 epoch (loss 1.5019):  29%|██▉       | 272/938 [01:59<04:39,  2.39it/s]
Training 1/1 epoch (loss 1.5019):  29%|██▉       | 273/938 [01:59<04:31,  2.44it/s]
Training 1/1 epoch (loss 1.5486):  29%|██▉       | 273/938 [01:59<04:31,  2.44it/s]
Training 1/1 epoch (loss 1.5486):  29%|██▉       | 274/938 [01:59<04:25,  2.50it/s]
Training 1/1 epoch (loss 1.6470):  29%|██▉       | 274/938 [01:59<04:25,  2.50it/s]
Training 1/1 epoch (loss 1.6470):  29%|██▉       | 275/938 [01:59<04:18,  2.56it/s]
Training 1/1 epoch (loss 1.4853):  29%|██▉       | 275/938 [02:00<04:18,  2.56it/s]
Training 1/1 epoch (loss 1.4853):  29%|██▉       | 276/938 [02:00<04:25,  2.49it/s]
Training 1/1 epoch (loss 1.5508):  29%|██▉       | 276/938 [02:00<04:25,  2.49it/s]
Training 1/1 epoch (loss 1.5508):  30%|██▉       | 277/938 [02:00<04:28,  2.46it/s]
Training 1/1 epoch (loss 1.5001):  30%|██▉       | 277/938 [02:01<04:28,  2.46it/s]
Training 1/1 epoch (loss 1.5001):  30%|██▉       | 278/938 [02:01<04:35,  2.40it/s]
Training 1/1 epoch (loss 1.5307):  30%|██▉       | 278/938 [02:01<04:35,  2.40it/s]
Training 1/1 epoch (loss 1.5307):  30%|██▉       | 279/938 [02:01<04:40,  2.35it/s]
Training 1/1 epoch (loss 1.3540):  30%|██▉       | 279/938 [02:01<04:40,  2.35it/s]
Training 1/1 epoch (loss 1.3540):  30%|██▉       | 280/938 [02:01<04:33,  2.40it/s]
Training 1/1 epoch (loss 1.4501):  30%|██▉       | 280/938 [02:02<04:33,  2.40it/s]
Training 1/1 epoch (loss 1.4501):  30%|██▉       | 281/938 [02:02<04:27,  2.46it/s]
Training 1/1 epoch (loss 1.4280):  30%|██▉       | 281/938 [02:02<04:27,  2.46it/s]
Training 1/1 epoch (loss 1.4280):  30%|███       | 282/938 [02:02<04:18,  2.54it/s]
Training 1/1 epoch (loss 1.5630):  30%|███       | 282/938 [02:03<04:18,  2.54it/s]
Training 1/1 epoch (loss 1.5630):  30%|███       | 283/938 [02:03<04:17,  2.54it/s]
Training 1/1 epoch (loss 1.4455):  30%|███       | 283/938 [02:03<04:17,  2.54it/s]
Training 1/1 epoch (loss 1.4455):  30%|███       | 284/938 [02:03<04:15,  2.56it/s]
Training 1/1 epoch (loss 1.4256):  30%|███       | 284/938 [02:03<04:15,  2.56it/s]
Training 1/1 epoch (loss 1.4256):  30%|███       | 285/938 [02:03<04:11,  2.60it/s]
Training 1/1 epoch (loss 1.5283):  30%|███       | 285/938 [02:04<04:11,  2.60it/s]
Training 1/1 epoch (loss 1.5283):  30%|███       | 286/938 [02:04<04:10,  2.60it/s]
Training 1/1 epoch (loss 1.3993):  30%|███       | 286/938 [02:04<04:10,  2.60it/s]
Training 1/1 epoch (loss 1.3993):  31%|███       | 287/938 [02:04<04:15,  2.55it/s]
Training 1/1 epoch (loss 1.5870):  31%|███       | 287/938 [02:05<04:15,  2.55it/s]
Training 1/1 epoch (loss 1.5870):  31%|███       | 288/938 [02:05<04:41,  2.31it/s]
Training 1/1 epoch (loss 1.5873):  31%|███       | 288/938 [02:05<04:41,  2.31it/s]
Training 1/1 epoch (loss 1.5873):  31%|███       | 289/938 [02:05<05:18,  2.04it/s]
Training 1/1 epoch (loss 1.4596):  31%|███       | 289/938 [02:06<05:18,  2.04it/s]
Training 1/1 epoch (loss 1.4596):  31%|███       | 290/938 [02:06<05:24,  2.00it/s]
Training 1/1 epoch (loss 1.5695):  31%|███       | 290/938 [02:06<05:24,  2.00it/s]
Training 1/1 epoch (loss 1.5695):  31%|███       | 291/938 [02:06<05:06,  2.11it/s]
Training 1/1 epoch (loss 1.4230):  31%|███       | 291/938 [02:07<05:06,  2.11it/s]
Training 1/1 epoch (loss 1.4230):  31%|███       | 292/938 [02:07<04:45,  2.26it/s]
Training 1/1 epoch (loss 1.4538):  31%|███       | 292/938 [02:07<04:45,  2.26it/s]
Training 1/1 epoch (loss 1.4538):  31%|███       | 293/938 [02:07<04:39,  2.31it/s]
Training 1/1 epoch (loss 1.4569):  31%|███       | 293/938 [02:07<04:39,  2.31it/s]
Training 1/1 epoch (loss 1.4569):  31%|███▏      | 294/938 [02:07<04:28,  2.40it/s]
Training 1/1 epoch (loss 1.3152):  31%|███▏      | 294/938 [02:08<04:28,  2.40it/s]
Training 1/1 epoch (loss 1.3152):  31%|███▏      | 295/938 [02:08<04:38,  2.31it/s]
Training 1/1 epoch (loss 1.4196):  31%|███▏      | 295/938 [02:08<04:38,  2.31it/s]
Training 1/1 epoch (loss 1.4196):  32%|███▏      | 296/938 [02:08<04:28,  2.39it/s]
Training 1/1 epoch (loss 1.4875):  32%|███▏      | 296/938 [02:09<04:28,  2.39it/s]
Training 1/1 epoch (loss 1.4875):  32%|███▏      | 297/938 [02:09<04:18,  2.48it/s]
Training 1/1 epoch (loss 1.4340):  32%|███▏      | 297/938 [02:09<04:18,  2.48it/s]
Training 1/1 epoch (loss 1.4340):  32%|███▏      | 298/938 [02:09<04:22,  2.44it/s]
Training 1/1 epoch (loss 1.5025):  32%|███▏      | 298/938 [02:09<04:22,  2.44it/s]
Training 1/1 epoch (loss 1.5025):  32%|███▏      | 299/938 [02:09<04:20,  2.45it/s]
Training 1/1 epoch (loss 1.4822):  32%|███▏      | 299/938 [02:10<04:20,  2.45it/s]
Training 1/1 epoch (loss 1.4822):  32%|███▏      | 300/938 [02:10<04:13,  2.52it/s]
Training 1/1 epoch (loss 1.4999):  32%|███▏      | 300/938 [02:10<04:13,  2.52it/s]
Training 1/1 epoch (loss 1.4999):  32%|███▏      | 301/938 [02:10<04:10,  2.54it/s]
Training 1/1 epoch (loss 1.5312):  32%|███▏      | 301/938 [02:11<04:10,  2.54it/s]
Training 1/1 epoch (loss 1.5312):  32%|███▏      | 302/938 [02:11<04:08,  2.55it/s]
Training 1/1 epoch (loss 1.4063):  32%|███▏      | 302/938 [02:11<04:08,  2.55it/s]
Training 1/1 epoch (loss 1.4063):  32%|███▏      | 303/938 [02:11<04:05,  2.58it/s]
Training 1/1 epoch (loss 1.6265):  32%|███▏      | 303/938 [02:11<04:05,  2.58it/s]
Training 1/1 epoch (loss 1.6265):  32%|███▏      | 304/938 [02:11<04:01,  2.62it/s]
Training 1/1 epoch (loss 1.4051):  32%|███▏      | 304/938 [02:12<04:01,  2.62it/s]
Training 1/1 epoch (loss 1.4051):  33%|███▎      | 305/938 [02:12<04:09,  2.54it/s]
Training 1/1 epoch (loss 1.4983):  33%|███▎      | 305/938 [02:12<04:09,  2.54it/s]
Training 1/1 epoch (loss 1.4983):  33%|███▎      | 306/938 [02:12<04:11,  2.51it/s]
Training 1/1 epoch (loss 1.5102):  33%|███▎      | 306/938 [02:13<04:11,  2.51it/s]
Training 1/1 epoch (loss 1.5102):  33%|███▎      | 307/938 [02:13<04:09,  2.53it/s]
Training 1/1 epoch (loss 1.4304):  33%|███▎      | 307/938 [02:13<04:09,  2.53it/s]
Training 1/1 epoch (loss 1.4304):  33%|███▎      | 308/938 [02:13<04:04,  2.58it/s]
Training 1/1 epoch (loss 1.3522):  33%|███▎      | 308/938 [02:13<04:04,  2.58it/s]
Training 1/1 epoch (loss 1.3522):  33%|███▎      | 309/938 [02:13<03:58,  2.63it/s]
Training 1/1 epoch (loss 1.5598):  33%|███▎      | 309/938 [02:14<03:58,  2.63it/s]
Training 1/1 epoch (loss 1.5598):  33%|███▎      | 310/938 [02:14<03:58,  2.63it/s]
Training 1/1 epoch (loss 1.2966):  33%|███▎      | 310/938 [02:14<03:58,  2.63it/s]
Training 1/1 epoch (loss 1.2966):  33%|███▎      | 311/938 [02:14<04:00,  2.61it/s]
Training 1/1 epoch (loss 1.4529):  33%|███▎      | 311/938 [02:15<04:00,  2.61it/s]
Training 1/1 epoch (loss 1.4529):  33%|███▎      | 312/938 [02:15<04:22,  2.39it/s]
Training 1/1 epoch (loss 1.5081):  33%|███▎      | 312/938 [02:15<04:22,  2.39it/s]
Training 1/1 epoch (loss 1.5081):  33%|███▎      | 313/938 [02:15<04:26,  2.34it/s]
Training 1/1 epoch (loss 1.5764):  33%|███▎      | 313/938 [02:15<04:26,  2.34it/s]
Training 1/1 epoch (loss 1.5764):  33%|███▎      | 314/938 [02:15<04:17,  2.43it/s]
Training 1/1 epoch (loss 1.3901):  33%|███▎      | 314/938 [02:16<04:17,  2.43it/s]
Training 1/1 epoch (loss 1.3901):  34%|███▎      | 315/938 [02:16<04:11,  2.47it/s]
Training 1/1 epoch (loss 1.5032):  34%|███▎      | 315/938 [02:16<04:11,  2.47it/s]
Training 1/1 epoch (loss 1.5032):  34%|███▎      | 316/938 [02:16<04:21,  2.38it/s]
Training 1/1 epoch (loss 1.4703):  34%|███▎      | 316/938 [02:17<04:21,  2.38it/s]
Training 1/1 epoch (loss 1.4703):  34%|███▍      | 317/938 [02:17<04:10,  2.48it/s]
Training 1/1 epoch (loss 1.5055):  34%|███▍      | 317/938 [02:17<04:10,  2.48it/s]
Training 1/1 epoch (loss 1.5055):  34%|███▍      | 318/938 [02:17<04:04,  2.53it/s]
Training 1/1 epoch (loss 1.3727):  34%|███▍      | 318/938 [02:17<04:04,  2.53it/s]
Training 1/1 epoch (loss 1.3727):  34%|███▍      | 319/938 [02:17<04:03,  2.54it/s]
Training 1/1 epoch (loss 1.3223):  34%|███▍      | 319/938 [02:18<04:03,  2.54it/s]
Training 1/1 epoch (loss 1.3223):  34%|███▍      | 320/938 [02:18<04:20,  2.38it/s]
Training 1/1 epoch (loss 1.4042):  34%|███▍      | 320/938 [02:18<04:20,  2.38it/s]
Training 1/1 epoch (loss 1.4042):  34%|███▍      | 321/938 [02:18<04:26,  2.31it/s]
Training 1/1 epoch (loss 1.4896):  34%|███▍      | 321/938 [02:19<04:26,  2.31it/s]
Training 1/1 epoch (loss 1.4896):  34%|███▍      | 322/938 [02:19<04:18,  2.38it/s]
Training 1/1 epoch (loss 1.5974):  34%|███▍      | 322/938 [02:19<04:18,  2.38it/s]
Training 1/1 epoch (loss 1.5974):  34%|███▍      | 323/938 [02:19<04:12,  2.43it/s]
Training 1/1 epoch (loss 1.5037):  34%|███▍      | 323/938 [02:19<04:12,  2.43it/s]
Training 1/1 epoch (loss 1.5037):  35%|███▍      | 324/938 [02:19<04:08,  2.48it/s]
Training 1/1 epoch (loss 1.4564):  35%|███▍      | 324/938 [02:20<04:08,  2.48it/s]
Training 1/1 epoch (loss 1.4564):  35%|███▍      | 325/938 [02:20<04:01,  2.54it/s]
Training 1/1 epoch (loss 1.5180):  35%|███▍      | 325/938 [02:20<04:01,  2.54it/s]
Training 1/1 epoch (loss 1.5180):  35%|███▍      | 326/938 [02:20<03:59,  2.55it/s]
Training 1/1 epoch (loss 1.4490):  35%|███▍      | 326/938 [02:21<03:59,  2.55it/s]
Training 1/1 epoch (loss 1.4490):  35%|███▍      | 327/938 [02:21<03:56,  2.59it/s]
Training 1/1 epoch (loss 1.4143):  35%|███▍      | 327/938 [02:21<03:56,  2.59it/s]
Training 1/1 epoch (loss 1.4143):  35%|███▍      | 328/938 [02:21<03:57,  2.57it/s]
Training 1/1 epoch (loss 1.3630):  35%|███▍      | 328/938 [02:21<03:57,  2.57it/s]
Training 1/1 epoch (loss 1.3630):  35%|███▌      | 329/938 [02:21<03:52,  2.62it/s]
Training 1/1 epoch (loss 1.5279):  35%|███▌      | 329/938 [02:22<03:52,  2.62it/s]
Training 1/1 epoch (loss 1.5279):  35%|███▌      | 330/938 [02:22<03:59,  2.54it/s]
Training 1/1 epoch (loss 1.4438):  35%|███▌      | 330/938 [02:22<03:59,  2.54it/s]
Training 1/1 epoch (loss 1.4438):  35%|███▌      | 331/938 [02:22<04:03,  2.49it/s]
Training 1/1 epoch (loss 1.5463):  35%|███▌      | 331/938 [02:23<04:03,  2.49it/s]
Training 1/1 epoch (loss 1.5463):  35%|███▌      | 332/938 [02:23<03:57,  2.55it/s]
Training 1/1 epoch (loss 1.4554):  35%|███▌      | 332/938 [02:23<03:57,  2.55it/s]
Training 1/1 epoch (loss 1.4554):  36%|███▌      | 333/938 [02:23<03:55,  2.57it/s]
Training 1/1 epoch (loss 1.4105):  36%|███▌      | 333/938 [02:23<03:55,  2.57it/s]
Training 1/1 epoch (loss 1.4105):  36%|███▌      | 334/938 [02:23<03:51,  2.61it/s]
Training 1/1 epoch (loss 1.4954):  36%|███▌      | 334/938 [02:24<03:51,  2.61it/s]
Training 1/1 epoch (loss 1.4954):  36%|███▌      | 335/938 [02:24<03:56,  2.55it/s]
Training 1/1 epoch (loss 1.3607):  36%|███▌      | 335/938 [02:24<03:56,  2.55it/s]
Training 1/1 epoch (loss 1.3607):  36%|███▌      | 336/938 [02:24<03:57,  2.54it/s]
Training 1/1 epoch (loss 1.4846):  36%|███▌      | 336/938 [02:25<03:57,  2.54it/s]
Training 1/1 epoch (loss 1.4846):  36%|███▌      | 337/938 [02:25<04:10,  2.40it/s]
Training 1/1 epoch (loss 1.4665):  36%|███▌      | 337/938 [02:25<04:10,  2.40it/s]
Training 1/1 epoch (loss 1.4665):  36%|███▌      | 338/938 [02:25<05:39,  1.77it/s]
Training 1/1 epoch (loss 1.4970):  36%|███▌      | 338/938 [02:26<05:39,  1.77it/s]
Training 1/1 epoch (loss 1.4970):  36%|███▌      | 339/938 [02:26<05:21,  1.87it/s]
Training 1/1 epoch (loss 1.3939):  36%|███▌      | 339/938 [02:26<05:21,  1.87it/s]
Training 1/1 epoch (loss 1.3939):  36%|███▌      | 340/938 [02:26<04:56,  2.02it/s]
Training 1/1 epoch (loss 1.4814):  36%|███▌      | 340/938 [02:27<04:56,  2.02it/s]
Training 1/1 epoch (loss 1.4814):  36%|███▋      | 341/938 [02:27<04:31,  2.20it/s]
Training 1/1 epoch (loss 1.4269):  36%|███▋      | 341/938 [02:27<04:31,  2.20it/s]
Training 1/1 epoch (loss 1.4269):  36%|███▋      | 342/938 [02:27<04:25,  2.25it/s]
Training 1/1 epoch (loss 1.5264):  36%|███▋      | 342/938 [02:28<04:25,  2.25it/s]
Training 1/1 epoch (loss 1.5264):  37%|███▋      | 343/938 [02:28<04:13,  2.34it/s]
Training 1/1 epoch (loss 1.4660):  37%|███▋      | 343/938 [02:28<04:13,  2.34it/s]
Training 1/1 epoch (loss 1.4660):  37%|███▋      | 344/938 [02:28<04:09,  2.38it/s]
Training 1/1 epoch (loss 1.4003):  37%|███▋      | 344/938 [02:28<04:09,  2.38it/s]
Training 1/1 epoch (loss 1.4003):  37%|███▋      | 345/938 [02:28<04:10,  2.36it/s]
Training 1/1 epoch (loss 1.4083):  37%|███▋      | 345/938 [02:29<04:10,  2.36it/s]
Training 1/1 epoch (loss 1.4083):  37%|███▋      | 346/938 [02:29<04:01,  2.45it/s]
Training 1/1 epoch (loss 1.4849):  37%|███▋      | 346/938 [02:29<04:01,  2.45it/s]
Training 1/1 epoch (loss 1.4849):  37%|███▋      | 347/938 [02:29<03:56,  2.50it/s]
Training 1/1 epoch (loss 1.3135):  37%|███▋      | 347/938 [02:30<03:56,  2.50it/s]
Training 1/1 epoch (loss 1.3135):  37%|███▋      | 348/938 [02:30<04:02,  2.44it/s]
Training 1/1 epoch (loss 1.5207):  37%|███▋      | 348/938 [02:30<04:02,  2.44it/s]
Training 1/1 epoch (loss 1.5207):  37%|███▋      | 349/938 [02:30<04:02,  2.43it/s]
Training 1/1 epoch (loss 1.4674):  37%|███▋      | 349/938 [02:30<04:02,  2.43it/s]
Training 1/1 epoch (loss 1.4674):  37%|███▋      | 350/938 [02:30<03:56,  2.49it/s]
Training 1/1 epoch (loss 1.4884):  37%|███▋      | 350/938 [02:31<03:56,  2.49it/s]
Training 1/1 epoch (loss 1.4884):  37%|███▋      | 351/938 [02:31<03:47,  2.58it/s]
Training 1/1 epoch (loss 1.4194):  37%|███▋      | 351/938 [02:31<03:47,  2.58it/s]
Training 1/1 epoch (loss 1.4194):  38%|███▊      | 352/938 [02:31<03:58,  2.45it/s]
Training 1/1 epoch (loss 1.5693):  38%|███▊      | 352/938 [02:32<03:58,  2.45it/s]
Training 1/1 epoch (loss 1.5693):  38%|███▊      | 353/938 [02:32<03:52,  2.52it/s]
Training 1/1 epoch (loss 1.4810):  38%|███▊      | 353/938 [02:32<03:52,  2.52it/s]
Training 1/1 epoch (loss 1.4810):  38%|███▊      | 354/938 [02:32<03:51,  2.52it/s]
Training 1/1 epoch (loss 1.4861):  38%|███▊      | 354/938 [02:32<03:51,  2.52it/s]
Training 1/1 epoch (loss 1.4861):  38%|███▊      | 355/938 [02:32<03:53,  2.49it/s]
Training 1/1 epoch (loss 1.3907):  38%|███▊      | 355/938 [02:33<03:53,  2.49it/s]
Training 1/1 epoch (loss 1.3907):  38%|███▊      | 356/938 [02:33<03:51,  2.52it/s]
Training 1/1 epoch (loss 1.4490):  38%|███▊      | 356/938 [02:33<03:51,  2.52it/s]
Training 1/1 epoch (loss 1.4490):  38%|███▊      | 357/938 [02:33<03:49,  2.53it/s]
Training 1/1 epoch (loss 1.4268):  38%|███▊      | 357/938 [02:33<03:49,  2.53it/s]
Training 1/1 epoch (loss 1.4268):  38%|███▊      | 358/938 [02:33<03:45,  2.57it/s]
Training 1/1 epoch (loss 1.4253):  38%|███▊      | 358/938 [02:34<03:45,  2.57it/s]
Training 1/1 epoch (loss 1.4253):  38%|███▊      | 359/938 [02:34<03:44,  2.58it/s]
Training 1/1 epoch (loss 1.5393):  38%|███▊      | 359/938 [02:34<03:44,  2.58it/s]
Training 1/1 epoch (loss 1.5393):  38%|███▊      | 360/938 [02:34<04:08,  2.33it/s]
Training 1/1 epoch (loss 1.6007):  38%|███▊      | 360/938 [02:35<04:08,  2.33it/s]
Training 1/1 epoch (loss 1.6007):  38%|███▊      | 361/938 [02:35<04:26,  2.17it/s]
Training 1/1 epoch (loss 1.4588):  38%|███▊      | 361/938 [02:36<04:26,  2.17it/s]
Training 1/1 epoch (loss 1.4588):  39%|███▊      | 362/938 [02:36<05:29,  1.75it/s]
Training 1/1 epoch (loss 1.4901):  39%|███▊      | 362/938 [02:36<05:29,  1.75it/s]
Training 1/1 epoch (loss 1.4901):  39%|███▊      | 363/938 [02:36<04:57,  1.93it/s]
Training 1/1 epoch (loss 1.5194):  39%|███▊      | 363/938 [02:37<04:57,  1.93it/s]
Training 1/1 epoch (loss 1.5194):  39%|███▉      | 364/938 [02:37<04:36,  2.08it/s]
Training 1/1 epoch (loss 1.4126):  39%|███▉      | 364/938 [02:37<04:36,  2.08it/s]
Training 1/1 epoch (loss 1.4126):  39%|███▉      | 365/938 [02:37<04:19,  2.21it/s]
Training 1/1 epoch (loss 1.4577):  39%|███▉      | 365/938 [02:37<04:19,  2.21it/s]
Training 1/1 epoch (loss 1.4577):  39%|███▉      | 366/938 [02:37<04:02,  2.36it/s]
Training 1/1 epoch (loss 1.4588):  39%|███▉      | 366/938 [02:38<04:02,  2.36it/s]
Training 1/1 epoch (loss 1.4588):  39%|███▉      | 367/938 [02:38<04:05,  2.33it/s]
Training 1/1 epoch (loss 1.5916):  39%|███▉      | 367/938 [02:38<04:05,  2.33it/s]
Training 1/1 epoch (loss 1.5916):  39%|███▉      | 368/938 [02:38<03:55,  2.42it/s]
Training 1/1 epoch (loss 1.3872):  39%|███▉      | 368/938 [02:39<03:55,  2.42it/s]
Training 1/1 epoch (loss 1.3872):  39%|███▉      | 369/938 [02:39<03:57,  2.40it/s]
Training 1/1 epoch (loss 1.5077):  39%|███▉      | 369/938 [02:39<03:57,  2.40it/s]
Training 1/1 epoch (loss 1.5077):  39%|███▉      | 370/938 [02:39<03:51,  2.45it/s]
Training 1/1 epoch (loss 1.4422):  39%|███▉      | 370/938 [02:39<03:51,  2.45it/s]
Training 1/1 epoch (loss 1.4422):  40%|███▉      | 371/938 [02:39<03:46,  2.51it/s]
Training 1/1 epoch (loss 1.4806):  40%|███▉      | 371/938 [02:40<03:46,  2.51it/s]
Training 1/1 epoch (loss 1.4806):  40%|███▉      | 372/938 [02:40<03:40,  2.56it/s]
Training 1/1 epoch (loss 1.3837):  40%|███▉      | 372/938 [02:40<03:40,  2.56it/s]
Training 1/1 epoch (loss 1.3837):  40%|███▉      | 373/938 [02:40<03:42,  2.54it/s]
Training 1/1 epoch (loss 1.3241):  40%|███▉      | 373/938 [02:40<03:42,  2.54it/s]
Training 1/1 epoch (loss 1.3241):  40%|███▉      | 374/938 [02:40<03:44,  2.52it/s]
Training 1/1 epoch (loss 1.5083):  40%|███▉      | 374/938 [02:41<03:44,  2.52it/s]
Training 1/1 epoch (loss 1.5083):  40%|███▉      | 375/938 [02:41<03:45,  2.50it/s]
Training 1/1 epoch (loss 1.5153):  40%|███▉      | 375/938 [02:41<03:45,  2.50it/s]
Training 1/1 epoch (loss 1.5153):  40%|████      | 376/938 [02:41<03:46,  2.48it/s]
Training 1/1 epoch (loss 1.6042):  40%|████      | 376/938 [02:42<03:46,  2.48it/s]
Training 1/1 epoch (loss 1.6042):  40%|████      | 377/938 [02:42<03:50,  2.44it/s]
Training 1/1 epoch (loss 1.3718):  40%|████      | 377/938 [02:42<03:50,  2.44it/s]
Training 1/1 epoch (loss 1.3718):  40%|████      | 378/938 [02:42<03:43,  2.50it/s]
Training 1/1 epoch (loss 1.4299):  40%|████      | 378/938 [02:42<03:43,  2.50it/s]
Training 1/1 epoch (loss 1.4299):  40%|████      | 379/938 [02:42<03:37,  2.57it/s]
Training 1/1 epoch (loss 1.3832):  40%|████      | 379/938 [02:43<03:37,  2.57it/s]
Training 1/1 epoch (loss 1.3832):  41%|████      | 380/938 [02:43<03:42,  2.51it/s]
Training 1/1 epoch (loss 1.4755):  41%|████      | 380/938 [02:43<03:42,  2.51it/s]
Training 1/1 epoch (loss 1.4755):  41%|████      | 381/938 [02:43<03:37,  2.56it/s]
Training 1/1 epoch (loss 1.3441):  41%|████      | 381/938 [02:44<03:37,  2.56it/s]
Training 1/1 epoch (loss 1.3441):  41%|████      | 382/938 [02:44<03:32,  2.61it/s]
Training 1/1 epoch (loss 1.5032):  41%|████      | 382/938 [02:44<03:32,  2.61it/s]
Training 1/1 epoch (loss 1.5032):  41%|████      | 383/938 [02:44<03:36,  2.57it/s]
Training 1/1 epoch (loss 1.5103):  41%|████      | 383/938 [02:44<03:36,  2.57it/s]
Training 1/1 epoch (loss 1.5103):  41%|████      | 384/938 [02:44<03:44,  2.47it/s]
Training 1/1 epoch (loss 1.4589):  41%|████      | 384/938 [02:45<03:44,  2.47it/s]
Training 1/1 epoch (loss 1.4589):  41%|████      | 385/938 [02:45<03:42,  2.48it/s]
Training 1/1 epoch (loss 1.4775):  41%|████      | 385/938 [02:45<03:42,  2.48it/s]
Training 1/1 epoch (loss 1.4775):  41%|████      | 386/938 [02:45<03:37,  2.54it/s]
Training 1/1 epoch (loss 1.4988):  41%|████      | 386/938 [02:46<03:37,  2.54it/s]
Training 1/1 epoch (loss 1.4988):  41%|████▏     | 387/938 [02:46<03:47,  2.42it/s]
Training 1/1 epoch (loss 1.4546):  41%|████▏     | 387/938 [02:46<03:47,  2.42it/s]
Training 1/1 epoch (loss 1.4546):  41%|████▏     | 388/938 [02:46<03:57,  2.32it/s]
Training 1/1 epoch (loss 1.5923):  41%|████▏     | 388/938 [02:47<03:57,  2.32it/s]
Training 1/1 epoch (loss 1.5923):  41%|████▏     | 389/938 [02:47<04:03,  2.25it/s]
Training 1/1 epoch (loss 1.4333):  41%|████▏     | 389/938 [02:47<04:03,  2.25it/s]
Training 1/1 epoch (loss 1.4333):  42%|████▏     | 390/938 [02:47<03:57,  2.30it/s]
Training 1/1 epoch (loss 1.4409):  42%|████▏     | 390/938 [02:47<03:57,  2.30it/s]
Training 1/1 epoch (loss 1.4409):  42%|████▏     | 391/938 [02:47<04:00,  2.27it/s]
Training 1/1 epoch (loss 1.5660):  42%|████▏     | 391/938 [02:48<04:00,  2.27it/s]
Training 1/1 epoch (loss 1.5660):  42%|████▏     | 392/938 [02:48<03:54,  2.33it/s]
Training 1/1 epoch (loss 1.4685):  42%|████▏     | 392/938 [02:48<03:54,  2.33it/s]
Training 1/1 epoch (loss 1.4685):  42%|████▏     | 393/938 [02:48<03:54,  2.32it/s]
Training 1/1 epoch (loss 1.5751):  42%|████▏     | 393/938 [02:49<03:54,  2.32it/s]
Training 1/1 epoch (loss 1.5751):  42%|████▏     | 394/938 [02:49<03:44,  2.42it/s]
Training 1/1 epoch (loss 1.4699):  42%|████▏     | 394/938 [02:49<03:44,  2.42it/s]
Training 1/1 epoch (loss 1.4699):  42%|████▏     | 395/938 [02:49<03:38,  2.48it/s]
Training 1/1 epoch (loss 1.4325):  42%|████▏     | 395/938 [02:49<03:38,  2.48it/s]
Training 1/1 epoch (loss 1.4325):  42%|████▏     | 396/938 [02:49<03:32,  2.55it/s]
Training 1/1 epoch (loss 1.4216):  42%|████▏     | 396/938 [02:50<03:32,  2.55it/s]
Training 1/1 epoch (loss 1.4216):  42%|████▏     | 397/938 [02:50<03:35,  2.51it/s]
Training 1/1 epoch (loss 1.4322):  42%|████▏     | 397/938 [02:50<03:35,  2.51it/s]
Training 1/1 epoch (loss 1.4322):  42%|████▏     | 398/938 [02:50<03:30,  2.57it/s]
Training 1/1 epoch (loss 1.4677):  42%|████▏     | 398/938 [02:51<03:30,  2.57it/s]
Training 1/1 epoch (loss 1.4677):  43%|████▎     | 399/938 [02:51<03:29,  2.58it/s]
Training 1/1 epoch (loss 1.5291):  43%|████▎     | 399/938 [02:51<03:29,  2.58it/s]
Training 1/1 epoch (loss 1.5291):  43%|████▎     | 400/938 [02:51<03:34,  2.51it/s]
Training 1/1 epoch (loss 1.4989):  43%|████▎     | 400/938 [02:51<03:34,  2.51it/s]
Training 1/1 epoch (loss 1.4989):  43%|████▎     | 401/938 [02:51<03:30,  2.56it/s]
Training 1/1 epoch (loss 1.5097):  43%|████▎     | 401/938 [02:52<03:30,  2.56it/s]
Training 1/1 epoch (loss 1.5097):  43%|████▎     | 402/938 [02:52<03:30,  2.55it/s]
Training 1/1 epoch (loss 1.4836):  43%|████▎     | 402/938 [02:52<03:30,  2.55it/s]
Training 1/1 epoch (loss 1.4836):  43%|████▎     | 403/938 [02:52<03:36,  2.47it/s]
Training 1/1 epoch (loss 1.4746):  43%|████▎     | 403/938 [02:53<03:36,  2.47it/s]
Training 1/1 epoch (loss 1.4746):  43%|████▎     | 404/938 [02:53<03:32,  2.52it/s]
Training 1/1 epoch (loss 1.3487):  43%|████▎     | 404/938 [02:53<03:32,  2.52it/s]
Training 1/1 epoch (loss 1.3487):  43%|████▎     | 405/938 [02:53<03:30,  2.53it/s]
Training 1/1 epoch (loss 1.5845):  43%|████▎     | 405/938 [02:53<03:30,  2.53it/s]
Training 1/1 epoch (loss 1.5845):  43%|████▎     | 406/938 [02:53<03:30,  2.53it/s]
Training 1/1 epoch (loss 1.5436):  43%|████▎     | 406/938 [02:54<03:30,  2.53it/s]
Training 1/1 epoch (loss 1.5436):  43%|████▎     | 407/938 [02:54<03:36,  2.45it/s]
Training 1/1 epoch (loss 1.4320):  43%|████▎     | 407/938 [02:54<03:36,  2.45it/s]
Training 1/1 epoch (loss 1.4320):  43%|████▎     | 408/938 [02:54<03:35,  2.46it/s]
Training 1/1 epoch (loss 1.4859):  43%|████▎     | 408/938 [02:55<03:35,  2.46it/s]
Training 1/1 epoch (loss 1.4859):  44%|████▎     | 409/938 [02:55<03:37,  2.43it/s]
Training 1/1 epoch (loss 1.4669):  44%|████▎     | 409/938 [02:55<03:37,  2.43it/s]
Training 1/1 epoch (loss 1.4669):  44%|████▎     | 410/938 [02:55<03:41,  2.39it/s]
Training 1/1 epoch (loss 1.3210):  44%|████▎     | 410/938 [02:55<03:41,  2.39it/s]
Training 1/1 epoch (loss 1.3210):  44%|████▍     | 411/938 [02:55<03:32,  2.48it/s]
Training 1/1 epoch (loss 1.4350):  44%|████▍     | 411/938 [02:56<03:32,  2.48it/s]
Training 1/1 epoch (loss 1.4350):  44%|████▍     | 412/938 [02:56<03:40,  2.38it/s]
Training 1/1 epoch (loss 1.3684):  44%|████▍     | 412/938 [02:56<03:40,  2.38it/s]
Training 1/1 epoch (loss 1.3684):  44%|████▍     | 413/938 [02:56<03:29,  2.50it/s]
Training 1/1 epoch (loss 1.4595):  44%|████▍     | 413/938 [02:57<03:29,  2.50it/s]
Training 1/1 epoch (loss 1.4595):  44%|████▍     | 414/938 [02:57<03:27,  2.53it/s]
Training 1/1 epoch (loss 1.3959):  44%|████▍     | 414/938 [02:57<03:27,  2.53it/s]
Training 1/1 epoch (loss 1.3959):  44%|████▍     | 415/938 [02:57<03:26,  2.54it/s]
Training 1/1 epoch (loss 1.3413):  44%|████▍     | 415/938 [02:57<03:26,  2.54it/s]
Training 1/1 epoch (loss 1.3413):  44%|████▍     | 416/938 [02:57<03:22,  2.57it/s]
Training 1/1 epoch (loss 1.3607):  44%|████▍     | 416/938 [02:58<03:22,  2.57it/s]
Training 1/1 epoch (loss 1.3607):  44%|████▍     | 417/938 [02:58<03:25,  2.54it/s]
Training 1/1 epoch (loss 1.4108):  44%|████▍     | 417/938 [02:58<03:25,  2.54it/s]
Training 1/1 epoch (loss 1.4108):  45%|████▍     | 418/938 [02:58<03:23,  2.56it/s]
Training 1/1 epoch (loss 1.5189):  45%|████▍     | 418/938 [02:59<03:23,  2.56it/s]
Training 1/1 epoch (loss 1.5189):  45%|████▍     | 419/938 [02:59<03:21,  2.57it/s]
Training 1/1 epoch (loss 1.5435):  45%|████▍     | 419/938 [02:59<03:21,  2.57it/s]
Training 1/1 epoch (loss 1.5435):  45%|████▍     | 420/938 [02:59<03:49,  2.26it/s]
Training 1/1 epoch (loss 1.3876):  45%|████▍     | 420/938 [03:00<03:49,  2.26it/s]
Training 1/1 epoch (loss 1.3876):  45%|████▍     | 421/938 [03:00<03:46,  2.29it/s]
Training 1/1 epoch (loss 1.3856):  45%|████▍     | 421/938 [03:00<03:46,  2.29it/s]
Training 1/1 epoch (loss 1.3856):  45%|████▍     | 422/938 [03:00<03:42,  2.32it/s]
Training 1/1 epoch (loss 1.3910):  45%|████▍     | 422/938 [03:00<03:42,  2.32it/s]
Training 1/1 epoch (loss 1.3910):  45%|████▌     | 423/938 [03:00<03:33,  2.41it/s]
Training 1/1 epoch (loss 1.3055):  45%|████▌     | 423/938 [03:01<03:33,  2.41it/s]
Training 1/1 epoch (loss 1.3055):  45%|████▌     | 424/938 [03:01<03:30,  2.44it/s]
Training 1/1 epoch (loss 1.4462):  45%|████▌     | 424/938 [03:01<03:30,  2.44it/s]
Training 1/1 epoch (loss 1.4462):  45%|████▌     | 425/938 [03:01<03:30,  2.44it/s]
Training 1/1 epoch (loss 1.4751):  45%|████▌     | 425/938 [03:02<03:30,  2.44it/s]
Training 1/1 epoch (loss 1.4751):  45%|████▌     | 426/938 [03:02<03:30,  2.43it/s]
Training 1/1 epoch (loss 1.4718):  45%|████▌     | 426/938 [03:02<03:30,  2.43it/s]
Training 1/1 epoch (loss 1.4718):  46%|████▌     | 427/938 [03:02<03:22,  2.52it/s]
Training 1/1 epoch (loss 1.3801):  46%|████▌     | 427/938 [03:02<03:22,  2.52it/s]
Training 1/1 epoch (loss 1.3801):  46%|████▌     | 428/938 [03:02<03:27,  2.45it/s]
Training 1/1 epoch (loss 1.4292):  46%|████▌     | 428/938 [03:03<03:27,  2.45it/s]
Training 1/1 epoch (loss 1.4292):  46%|████▌     | 429/938 [03:03<03:22,  2.51it/s]
Training 1/1 epoch (loss 1.4506):  46%|████▌     | 429/938 [03:03<03:22,  2.51it/s]
Training 1/1 epoch (loss 1.4506):  46%|████▌     | 430/938 [03:03<03:24,  2.48it/s]
Training 1/1 epoch (loss 1.3936):  46%|████▌     | 430/938 [03:04<03:24,  2.48it/s]
Training 1/1 epoch (loss 1.3936):  46%|████▌     | 431/938 [03:04<03:20,  2.53it/s]
Training 1/1 epoch (loss 1.4919):  46%|████▌     | 431/938 [03:04<03:20,  2.53it/s]
Training 1/1 epoch (loss 1.4919):  46%|████▌     | 432/938 [03:04<03:25,  2.47it/s]
Training 1/1 epoch (loss 1.4207):  46%|████▌     | 432/938 [03:05<03:25,  2.47it/s]
Training 1/1 epoch (loss 1.4207):  46%|████▌     | 433/938 [03:05<03:45,  2.24it/s]
Training 1/1 epoch (loss 1.5332):  46%|████▌     | 433/938 [03:05<03:45,  2.24it/s]
Training 1/1 epoch (loss 1.5332):  46%|████▋     | 434/938 [03:05<03:59,  2.11it/s]
Training 1/1 epoch (loss 1.4357):  46%|████▋     | 434/938 [03:06<03:59,  2.11it/s]
Training 1/1 epoch (loss 1.4357):  46%|████▋     | 435/938 [03:06<03:52,  2.17it/s]
Training 1/1 epoch (loss 1.5064):  46%|████▋     | 435/938 [03:06<03:52,  2.17it/s]
Training 1/1 epoch (loss 1.5064):  46%|████▋     | 436/938 [03:06<03:50,  2.18it/s]
Training 1/1 epoch (loss 1.5057):  46%|████▋     | 436/938 [03:06<03:50,  2.18it/s]
Training 1/1 epoch (loss 1.5057):  47%|████▋     | 437/938 [03:06<03:40,  2.27it/s]
Training 1/1 epoch (loss 1.3359):  47%|████▋     | 437/938 [03:07<03:40,  2.27it/s]
Training 1/1 epoch (loss 1.3359):  47%|████▋     | 438/938 [03:07<03:27,  2.41it/s]
Training 1/1 epoch (loss 1.4965):  47%|████▋     | 438/938 [03:07<03:27,  2.41it/s]
Training 1/1 epoch (loss 1.4965):  47%|████▋     | 439/938 [03:07<03:19,  2.50it/s]
Training 1/1 epoch (loss 1.4540):  47%|████▋     | 439/938 [03:07<03:19,  2.50it/s]
Training 1/1 epoch (loss 1.4540):  47%|████▋     | 440/938 [03:07<03:17,  2.52it/s]
Training 1/1 epoch (loss 1.4493):  47%|████▋     | 440/938 [03:08<03:17,  2.52it/s]
Training 1/1 epoch (loss 1.4493):  47%|████▋     | 441/938 [03:08<03:18,  2.50it/s]
Training 1/1 epoch (loss 1.3906):  47%|████▋     | 441/938 [03:08<03:18,  2.50it/s]
Training 1/1 epoch (loss 1.3906):  47%|████▋     | 442/938 [03:08<03:15,  2.54it/s]
Training 1/1 epoch (loss 1.4740):  47%|████▋     | 442/938 [03:09<03:15,  2.54it/s]
Training 1/1 epoch (loss 1.4740):  47%|████▋     | 443/938 [03:09<03:11,  2.58it/s]
Training 1/1 epoch (loss 1.5165):  47%|████▋     | 443/938 [03:09<03:11,  2.58it/s]
Training 1/1 epoch (loss 1.5165):  47%|████▋     | 444/938 [03:09<03:12,  2.56it/s]
Training 1/1 epoch (loss 1.2764):  47%|████▋     | 444/938 [03:09<03:12,  2.56it/s]
Training 1/1 epoch (loss 1.2764):  47%|████▋     | 445/938 [03:09<03:20,  2.46it/s]
Training 1/1 epoch (loss 1.3626):  47%|████▋     | 445/938 [03:10<03:20,  2.46it/s]
Training 1/1 epoch (loss 1.3626):  48%|████▊     | 446/938 [03:10<03:22,  2.43it/s]
Training 1/1 epoch (loss 1.5081):  48%|████▊     | 446/938 [03:10<03:22,  2.43it/s]
Training 1/1 epoch (loss 1.5081):  48%|████▊     | 447/938 [03:10<03:15,  2.51it/s]
Training 1/1 epoch (loss 1.4538):  48%|████▊     | 447/938 [03:11<03:15,  2.51it/s]
Training 1/1 epoch (loss 1.4538):  48%|████▊     | 448/938 [03:11<03:13,  2.53it/s]
Training 1/1 epoch (loss 1.5580):  48%|████▊     | 448/938 [03:11<03:13,  2.53it/s]
Training 1/1 epoch (loss 1.5580):  48%|████▊     | 449/938 [03:11<03:26,  2.37it/s]
Training 1/1 epoch (loss 1.2832):  48%|████▊     | 449/938 [03:12<03:26,  2.37it/s]
Training 1/1 epoch (loss 1.2832):  48%|████▊     | 450/938 [03:12<03:22,  2.41it/s]
Training 1/1 epoch (loss 1.3934):  48%|████▊     | 450/938 [03:12<03:22,  2.41it/s]
Training 1/1 epoch (loss 1.3934):  48%|████▊     | 451/938 [03:12<03:23,  2.39it/s]
Training 1/1 epoch (loss 1.4359):  48%|████▊     | 451/938 [03:12<03:23,  2.39it/s]
Training 1/1 epoch (loss 1.4359):  48%|████▊     | 452/938 [03:12<03:23,  2.38it/s]
Training 1/1 epoch (loss 1.4446):  48%|████▊     | 452/938 [03:13<03:23,  2.38it/s]
Training 1/1 epoch (loss 1.4446):  48%|████▊     | 453/938 [03:13<03:14,  2.50it/s]
Training 1/1 epoch (loss 1.3829):  48%|████▊     | 453/938 [03:13<03:14,  2.50it/s]
Training 1/1 epoch (loss 1.3829):  48%|████▊     | 454/938 [03:13<03:08,  2.56it/s]
Training 1/1 epoch (loss 1.4857):  48%|████▊     | 454/938 [03:14<03:08,  2.56it/s]
Training 1/1 epoch (loss 1.4857):  49%|████▊     | 455/938 [03:14<03:07,  2.57it/s]
Training 1/1 epoch (loss 1.4338):  49%|████▊     | 455/938 [03:14<03:07,  2.57it/s]
Training 1/1 epoch (loss 1.4338):  49%|████▊     | 456/938 [03:14<03:10,  2.53it/s]
Training 1/1 epoch (loss 1.5224):  49%|████▊     | 456/938 [03:14<03:10,  2.53it/s]
Training 1/1 epoch (loss 1.5224):  49%|████▊     | 457/938 [03:14<03:11,  2.51it/s]
Training 1/1 epoch (loss 1.5185):  49%|████▊     | 457/938 [03:15<03:11,  2.51it/s]
Training 1/1 epoch (loss 1.5185):  49%|████▉     | 458/938 [03:15<03:17,  2.43it/s]
Training 1/1 epoch (loss 1.5330):  49%|████▉     | 458/938 [03:15<03:17,  2.43it/s]
Training 1/1 epoch (loss 1.5330):  49%|████▉     | 459/938 [03:15<03:16,  2.44it/s]
Training 1/1 epoch (loss 1.4590):  49%|████▉     | 459/938 [03:16<03:16,  2.44it/s]
Training 1/1 epoch (loss 1.4590):  49%|████▉     | 460/938 [03:16<03:21,  2.37it/s]
Training 1/1 epoch (loss 1.4250):  49%|████▉     | 460/938 [03:16<03:21,  2.37it/s]
Training 1/1 epoch (loss 1.4250):  49%|████▉     | 461/938 [03:16<03:25,  2.32it/s]
Training 1/1 epoch (loss 1.3977):  49%|████▉     | 461/938 [03:16<03:25,  2.32it/s]
Training 1/1 epoch (loss 1.3977):  49%|████▉     | 462/938 [03:16<03:17,  2.41it/s]
Training 1/1 epoch (loss 1.3802):  49%|████▉     | 462/938 [03:17<03:17,  2.41it/s]
Training 1/1 epoch (loss 1.3802):  49%|████▉     | 463/938 [03:17<03:16,  2.42it/s]
Training 1/1 epoch (loss 1.4555):  49%|████▉     | 463/938 [03:17<03:16,  2.42it/s]
Training 1/1 epoch (loss 1.4555):  49%|████▉     | 464/938 [03:17<03:12,  2.46it/s]
Training 1/1 epoch (loss 1.4169):  49%|████▉     | 464/938 [03:18<03:12,  2.46it/s]
Training 1/1 epoch (loss 1.4169):  50%|████▉     | 465/938 [03:18<03:22,  2.34it/s]
Training 1/1 epoch (loss 1.5026):  50%|████▉     | 465/938 [03:18<03:22,  2.34it/s]
Training 1/1 epoch (loss 1.5026):  50%|████▉     | 466/938 [03:18<03:21,  2.34it/s]
Training 1/1 epoch (loss 1.5275):  50%|████▉     | 466/938 [03:19<03:21,  2.34it/s]
Training 1/1 epoch (loss 1.5275):  50%|████▉     | 467/938 [03:19<03:42,  2.11it/s]
Training 1/1 epoch (loss 1.5689):  50%|████▉     | 467/938 [03:19<03:42,  2.11it/s]
Training 1/1 epoch (loss 1.5689):  50%|████▉     | 468/938 [03:19<03:28,  2.25it/s]
Training 1/1 epoch (loss 1.5137):  50%|████▉     | 468/938 [03:19<03:28,  2.25it/s]
Training 1/1 epoch (loss 1.5137):  50%|█████     | 469/938 [03:19<03:16,  2.39it/s]
Training 1/1 epoch (loss 1.3367):  50%|█████     | 469/938 [03:20<03:16,  2.39it/s]
Training 1/1 epoch (loss 1.3367):  50%|█████     | 470/938 [03:20<03:27,  2.25it/s]
Training 1/1 epoch (loss 1.4369):  50%|█████     | 470/938 [03:20<03:27,  2.25it/s]
Training 1/1 epoch (loss 1.4369):  50%|█████     | 471/938 [03:20<03:17,  2.37it/s]
Training 1/1 epoch (loss 1.4782):  50%|█████     | 471/938 [03:21<03:17,  2.37it/s]
Training 1/1 epoch (loss 1.4782):  50%|█████     | 472/938 [03:21<03:13,  2.40it/s]
Training 1/1 epoch (loss 1.3356):  50%|█████     | 472/938 [03:21<03:13,  2.40it/s]
Training 1/1 epoch (loss 1.3356):  50%|█████     | 473/938 [03:21<03:07,  2.48it/s]
Training 1/1 epoch (loss 1.5026):  50%|█████     | 473/938 [03:21<03:07,  2.48it/s]
Training 1/1 epoch (loss 1.5026):  51%|█████     | 474/938 [03:21<03:02,  2.55it/s]
Training 1/1 epoch (loss 1.3640):  51%|█████     | 474/938 [03:22<03:02,  2.55it/s]
Training 1/1 epoch (loss 1.3640):  51%|█████     | 475/938 [03:22<02:56,  2.62it/s]
Training 1/1 epoch (loss 1.4038):  51%|█████     | 475/938 [03:22<02:56,  2.62it/s]
Training 1/1 epoch (loss 1.4038):  51%|█████     | 476/938 [03:22<03:01,  2.54it/s]
Training 1/1 epoch (loss 1.4088):  51%|█████     | 476/938 [03:23<03:01,  2.54it/s]
Training 1/1 epoch (loss 1.4088):  51%|█████     | 477/938 [03:23<03:07,  2.45it/s]
Training 1/1 epoch (loss 1.3746):  51%|█████     | 477/938 [03:23<03:07,  2.45it/s]
Training 1/1 epoch (loss 1.3746):  51%|█████     | 478/938 [03:23<03:05,  2.48it/s]
Training 1/1 epoch (loss 1.4837):  51%|█████     | 478/938 [03:23<03:05,  2.48it/s]
Training 1/1 epoch (loss 1.4837):  51%|█████     | 479/938 [03:23<02:59,  2.55it/s]
Training 1/1 epoch (loss 1.3984):  51%|█████     | 479/938 [03:24<02:59,  2.55it/s]
Training 1/1 epoch (loss 1.3984):  51%|█████     | 480/938 [03:24<03:07,  2.44it/s]
Training 1/1 epoch (loss 1.4526):  51%|█████     | 480/938 [03:24<03:07,  2.44it/s]
Training 1/1 epoch (loss 1.4526):  51%|█████▏    | 481/938 [03:24<03:08,  2.43it/s]
Training 1/1 epoch (loss 1.4546):  51%|█████▏    | 481/938 [03:25<03:08,  2.43it/s]
Training 1/1 epoch (loss 1.4546):  51%|█████▏    | 482/938 [03:25<03:03,  2.49it/s]
Training 1/1 epoch (loss 1.3856):  51%|█████▏    | 482/938 [03:25<03:03,  2.49it/s]
Training 1/1 epoch (loss 1.3856):  51%|█████▏    | 483/938 [03:25<03:04,  2.47it/s]
Training 1/1 epoch (loss 1.5815):  51%|█████▏    | 483/938 [03:26<03:04,  2.47it/s]
Training 1/1 epoch (loss 1.5815):  52%|█████▏    | 484/938 [03:26<03:00,  2.52it/s]
Training 1/1 epoch (loss 1.5574):  52%|█████▏    | 484/938 [03:26<03:00,  2.52it/s]
Training 1/1 epoch (loss 1.5574):  52%|█████▏    | 485/938 [03:26<02:55,  2.58it/s]
Training 1/1 epoch (loss 1.4542):  52%|█████▏    | 485/938 [03:26<02:55,  2.58it/s]
Training 1/1 epoch (loss 1.4542):  52%|█████▏    | 486/938 [03:26<02:53,  2.61it/s]
Training 1/1 epoch (loss 1.4495):  52%|█████▏    | 486/938 [03:27<02:53,  2.61it/s]
Training 1/1 epoch (loss 1.4495):  52%|█████▏    | 487/938 [03:27<02:54,  2.58it/s]
Training 1/1 epoch (loss 1.4466):  52%|█████▏    | 487/938 [03:27<02:54,  2.58it/s]
Training 1/1 epoch (loss 1.4466):  52%|█████▏    | 488/938 [03:27<02:58,  2.53it/s]
Training 1/1 epoch (loss 1.4632):  52%|█████▏    | 488/938 [03:27<02:58,  2.53it/s]
Training 1/1 epoch (loss 1.4632):  52%|█████▏    | 489/938 [03:27<02:52,  2.60it/s]
Training 1/1 epoch (loss 1.4731):  52%|█████▏    | 489/938 [03:28<02:52,  2.60it/s]
Training 1/1 epoch (loss 1.4731):  52%|█████▏    | 490/938 [03:28<02:53,  2.59it/s]
Training 1/1 epoch (loss 1.5514):  52%|█████▏    | 490/938 [03:28<02:53,  2.59it/s]
Training 1/1 epoch (loss 1.5514):  52%|█████▏    | 491/938 [03:28<02:57,  2.51it/s]
Training 1/1 epoch (loss 1.5310):  52%|█████▏    | 491/938 [03:29<02:57,  2.51it/s]
Training 1/1 epoch (loss 1.5310):  52%|█████▏    | 492/938 [03:29<02:56,  2.53it/s]
Training 1/1 epoch (loss 1.3902):  52%|█████▏    | 492/938 [03:29<02:56,  2.53it/s]
Training 1/1 epoch (loss 1.3902):  53%|█████▎    | 493/938 [03:29<02:51,  2.60it/s]
Training 1/1 epoch (loss 1.4367):  53%|█████▎    | 493/938 [03:29<02:51,  2.60it/s]
Training 1/1 epoch (loss 1.4367):  53%|█████▎    | 494/938 [03:29<02:54,  2.54it/s]
Training 1/1 epoch (loss 1.4665):  53%|█████▎    | 494/938 [03:30<02:54,  2.54it/s]
Training 1/1 epoch (loss 1.4665):  53%|█████▎    | 495/938 [03:30<02:53,  2.56it/s]
Training 1/1 epoch (loss 1.3390):  53%|█████▎    | 495/938 [03:30<02:53,  2.56it/s]
Training 1/1 epoch (loss 1.3390):  53%|█████▎    | 496/938 [03:30<03:01,  2.43it/s]
Training 1/1 epoch (loss 1.4904):  53%|█████▎    | 496/938 [03:31<03:01,  2.43it/s]
Training 1/1 epoch (loss 1.4904):  53%|█████▎    | 497/938 [03:31<03:05,  2.38it/s]
Training 1/1 epoch (loss 1.4114):  53%|█████▎    | 497/938 [03:31<03:05,  2.38it/s]
Training 1/1 epoch (loss 1.4114):  53%|█████▎    | 498/938 [03:31<02:59,  2.45it/s]
Training 1/1 epoch (loss 1.3625):  53%|█████▎    | 498/938 [03:31<02:59,  2.45it/s]
Training 1/1 epoch (loss 1.3625):  53%|█████▎    | 499/938 [03:31<02:52,  2.55it/s]
Training 1/1 epoch (loss 1.3316):  53%|█████▎    | 499/938 [03:32<02:52,  2.55it/s]
Training 1/1 epoch (loss 1.3316):  53%|█████▎    | 500/938 [03:32<02:50,  2.56it/s]
Training 1/1 epoch (loss 1.5156):  53%|█████▎    | 500/938 [03:32<02:50,  2.56it/s]
Training 1/1 epoch (loss 1.5156):  53%|█████▎    | 501/938 [03:32<02:50,  2.57it/s]
Training 1/1 epoch (loss 1.5175):  53%|█████▎    | 501/938 [03:33<02:50,  2.57it/s]
Training 1/1 epoch (loss 1.5175):  54%|█████▎    | 502/938 [03:33<02:51,  2.54it/s]
Training 1/1 epoch (loss 1.4548):  54%|█████▎    | 502/938 [03:33<02:51,  2.54it/s]
Training 1/1 epoch (loss 1.4548):  54%|█████▎    | 503/938 [03:33<02:48,  2.58it/s]
Training 1/1 epoch (loss 1.3606):  54%|█████▎    | 503/938 [03:33<02:48,  2.58it/s]
Training 1/1 epoch (loss 1.3606):  54%|█████▎    | 504/938 [03:33<02:47,  2.59it/s]
Training 1/1 epoch (loss 1.4042):  54%|█████▎    | 504/938 [03:34<02:47,  2.59it/s]
Training 1/1 epoch (loss 1.4042):  54%|█████▍    | 505/938 [03:34<02:48,  2.56it/s]
Training 1/1 epoch (loss 1.4014):  54%|█████▍    | 505/938 [03:34<02:48,  2.56it/s]
Training 1/1 epoch (loss 1.4014):  54%|█████▍    | 506/938 [03:34<02:50,  2.53it/s]
Training 1/1 epoch (loss 1.4774):  54%|█████▍    | 506/938 [03:35<02:50,  2.53it/s]
Training 1/1 epoch (loss 1.4774):  54%|█████▍    | 507/938 [03:35<03:16,  2.19it/s]
Training 1/1 epoch (loss 1.4478):  54%|█████▍    | 507/938 [03:35<03:16,  2.19it/s]
Training 1/1 epoch (loss 1.4478):  54%|█████▍    | 508/938 [03:35<03:19,  2.16it/s]
Training 1/1 epoch (loss 1.4450):  54%|█████▍    | 508/938 [03:36<03:19,  2.16it/s]
Training 1/1 epoch (loss 1.4450):  54%|█████▍    | 509/938 [03:36<03:33,  2.01it/s]
Training 1/1 epoch (loss 1.4957):  54%|█████▍    | 509/938 [03:36<03:33,  2.01it/s]
Training 1/1 epoch (loss 1.4957):  54%|█████▍    | 510/938 [03:36<03:19,  2.14it/s]
Training 1/1 epoch (loss 1.6024):  54%|█████▍    | 510/938 [03:37<03:19,  2.14it/s]
Training 1/1 epoch (loss 1.6024):  54%|█████▍    | 511/938 [03:37<03:06,  2.29it/s]
Training 1/1 epoch (loss 1.3900):  54%|█████▍    | 511/938 [03:37<03:06,  2.29it/s]
Training 1/1 epoch (loss 1.3900):  55%|█████▍    | 512/938 [03:37<03:00,  2.36it/s]
Training 1/1 epoch (loss 1.4610):  55%|█████▍    | 512/938 [03:37<03:00,  2.36it/s]
Training 1/1 epoch (loss 1.4610):  55%|█████▍    | 513/938 [03:37<02:54,  2.43it/s]
Training 1/1 epoch (loss 1.4921):  55%|█████▍    | 513/938 [03:38<02:54,  2.43it/s]
Training 1/1 epoch (loss 1.4921):  55%|█████▍    | 514/938 [03:38<02:46,  2.54it/s]
Training 1/1 epoch (loss 1.3990):  55%|█████▍    | 514/938 [03:38<02:46,  2.54it/s]
Training 1/1 epoch (loss 1.3990):  55%|█████▍    | 515/938 [03:38<02:52,  2.45it/s]
Training 1/1 epoch (loss 1.4454):  55%|█████▍    | 515/938 [03:39<02:52,  2.45it/s]
Training 1/1 epoch (loss 1.4454):  55%|█████▌    | 516/938 [03:39<02:51,  2.46it/s]
Training 1/1 epoch (loss 1.4819):  55%|█████▌    | 516/938 [03:39<02:51,  2.46it/s]
Training 1/1 epoch (loss 1.4819):  55%|█████▌    | 517/938 [03:39<02:47,  2.51it/s]
Training 1/1 epoch (loss 1.4307):  55%|█████▌    | 517/938 [03:39<02:47,  2.51it/s]
Training 1/1 epoch (loss 1.4307):  55%|█████▌    | 518/938 [03:39<02:42,  2.59it/s]
Training 1/1 epoch (loss 1.5075):  55%|█████▌    | 518/938 [03:40<02:42,  2.59it/s]
Training 1/1 epoch (loss 1.5075):  55%|█████▌    | 519/938 [03:40<02:40,  2.61it/s]
Training 1/1 epoch (loss 1.5315):  55%|█████▌    | 519/938 [03:40<02:40,  2.61it/s]
Training 1/1 epoch (loss 1.5315):  55%|█████▌    | 520/938 [03:40<02:57,  2.35it/s]
Training 1/1 epoch (loss 1.3255):  55%|█████▌    | 520/938 [03:41<02:57,  2.35it/s]
Training 1/1 epoch (loss 1.3255):  56%|█████▌    | 521/938 [03:41<02:52,  2.42it/s]
Training 1/1 epoch (loss 1.4286):  56%|█████▌    | 521/938 [03:41<02:52,  2.42it/s]
Training 1/1 epoch (loss 1.4286):  56%|█████▌    | 522/938 [03:41<02:48,  2.46it/s]
Training 1/1 epoch (loss 1.4392):  56%|█████▌    | 522/938 [03:41<02:48,  2.46it/s]
Training 1/1 epoch (loss 1.4392):  56%|█████▌    | 523/938 [03:41<02:50,  2.44it/s]
Training 1/1 epoch (loss 1.3217):  56%|█████▌    | 523/938 [03:42<02:50,  2.44it/s]
Training 1/1 epoch (loss 1.3217):  56%|█████▌    | 524/938 [03:42<02:42,  2.55it/s]
Training 1/1 epoch (loss 1.4022):  56%|█████▌    | 524/938 [03:42<02:42,  2.55it/s]
Training 1/1 epoch (loss 1.4022):  56%|█████▌    | 525/938 [03:42<02:40,  2.58it/s]
Training 1/1 epoch (loss 1.4504):  56%|█████▌    | 525/938 [03:42<02:40,  2.58it/s]
Training 1/1 epoch (loss 1.4504):  56%|█████▌    | 526/938 [03:42<02:39,  2.58it/s]
Training 1/1 epoch (loss 1.4327):  56%|█████▌    | 526/938 [03:43<02:39,  2.58it/s]
Training 1/1 epoch (loss 1.4327):  56%|█████▌    | 527/938 [03:43<02:42,  2.53it/s]
Training 1/1 epoch (loss 1.4568):  56%|█████▌    | 527/938 [03:43<02:42,  2.53it/s]
Training 1/1 epoch (loss 1.4568):  56%|█████▋    | 528/938 [03:43<02:41,  2.54it/s]
Training 1/1 epoch (loss 1.3612):  56%|█████▋    | 528/938 [03:44<02:41,  2.54it/s]
Training 1/1 epoch (loss 1.3612):  56%|█████▋    | 529/938 [03:44<02:40,  2.55it/s]
Training 1/1 epoch (loss 1.4102):  56%|█████▋    | 529/938 [03:44<02:40,  2.55it/s]
Training 1/1 epoch (loss 1.4102):  57%|█████▋    | 530/938 [03:44<02:39,  2.56it/s]
Training 1/1 epoch (loss 1.3848):  57%|█████▋    | 530/938 [03:45<02:39,  2.56it/s]
Training 1/1 epoch (loss 1.3848):  57%|█████▋    | 531/938 [03:45<02:51,  2.37it/s]
Training 1/1 epoch (loss 1.5194):  57%|█████▋    | 531/938 [03:45<02:51,  2.37it/s]
Training 1/1 epoch (loss 1.5194):  57%|█████▋    | 532/938 [03:45<02:50,  2.38it/s]
Training 1/1 epoch (loss 1.4782):  57%|█████▋    | 532/938 [03:45<02:50,  2.38it/s]
Training 1/1 epoch (loss 1.4782):  57%|█████▋    | 533/938 [03:45<02:44,  2.47it/s]
Training 1/1 epoch (loss 1.4074):  57%|█████▋    | 533/938 [03:46<02:44,  2.47it/s]
Training 1/1 epoch (loss 1.4074):  57%|█████▋    | 534/938 [03:46<02:46,  2.43it/s]
Training 1/1 epoch (loss 1.3901):  57%|█████▋    | 534/938 [03:46<02:46,  2.43it/s]
Training 1/1 epoch (loss 1.3901):  57%|█████▋    | 535/938 [03:46<02:50,  2.36it/s]
Training 1/1 epoch (loss 1.3537):  57%|█████▋    | 535/938 [03:47<02:50,  2.36it/s]
Training 1/1 epoch (loss 1.3537):  57%|█████▋    | 536/938 [03:47<02:50,  2.35it/s]
Training 1/1 epoch (loss 1.4780):  57%|█████▋    | 536/938 [03:47<02:50,  2.35it/s]
Training 1/1 epoch (loss 1.4780):  57%|█████▋    | 537/938 [03:47<02:47,  2.40it/s]
Training 1/1 epoch (loss 1.3681):  57%|█████▋    | 537/938 [03:47<02:47,  2.40it/s]
Training 1/1 epoch (loss 1.3681):  57%|█████▋    | 538/938 [03:47<02:38,  2.52it/s]
Training 1/1 epoch (loss 1.4344):  57%|█████▋    | 538/938 [03:48<02:38,  2.52it/s]
Training 1/1 epoch (loss 1.4344):  57%|█████▋    | 539/938 [03:48<02:50,  2.35it/s]
Training 1/1 epoch (loss 1.4409):  57%|█████▋    | 539/938 [03:48<02:50,  2.35it/s]
Training 1/1 epoch (loss 1.4409):  58%|█████▊    | 540/938 [03:48<02:48,  2.36it/s]
Training 1/1 epoch (loss 1.3000):  58%|█████▊    | 540/938 [03:49<02:48,  2.36it/s]
Training 1/1 epoch (loss 1.3000):  58%|█████▊    | 541/938 [03:49<02:41,  2.45it/s]
Training 1/1 epoch (loss 1.4010):  58%|█████▊    | 541/938 [03:49<02:41,  2.45it/s]
Training 1/1 epoch (loss 1.4010):  58%|█████▊    | 542/938 [03:49<02:43,  2.43it/s]
Training 1/1 epoch (loss 1.4578):  58%|█████▊    | 542/938 [03:49<02:43,  2.43it/s]
Training 1/1 epoch (loss 1.4578):  58%|█████▊    | 543/938 [03:49<02:38,  2.49it/s]
Training 1/1 epoch (loss 1.3741):  58%|█████▊    | 543/938 [03:50<02:38,  2.49it/s]
Training 1/1 epoch (loss 1.3741):  58%|█████▊    | 544/938 [03:50<02:37,  2.50it/s]
Training 1/1 epoch (loss 1.3385):  58%|█████▊    | 544/938 [03:50<02:37,  2.50it/s]
Training 1/1 epoch (loss 1.3385):  58%|█████▊    | 545/938 [03:50<02:39,  2.47it/s]
Training 1/1 epoch (loss 1.3472):  58%|█████▊    | 545/938 [03:51<02:39,  2.47it/s]
Training 1/1 epoch (loss 1.3472):  58%|█████▊    | 546/938 [03:51<02:36,  2.51it/s]
Training 1/1 epoch (loss 1.3616):  58%|█████▊    | 546/938 [03:51<02:36,  2.51it/s]
Training 1/1 epoch (loss 1.3616):  58%|█████▊    | 547/938 [03:51<02:41,  2.42it/s]
Training 1/1 epoch (loss 1.5096):  58%|█████▊    | 547/938 [03:52<02:41,  2.42it/s]
Training 1/1 epoch (loss 1.5096):  58%|█████▊    | 548/938 [03:52<02:38,  2.46it/s]
Training 1/1 epoch (loss 1.3692):  58%|█████▊    | 548/938 [03:52<02:38,  2.46it/s]
Training 1/1 epoch (loss 1.3692):  59%|█████▊    | 549/938 [03:52<02:41,  2.41it/s]
Training 1/1 epoch (loss 1.4574):  59%|█████▊    | 549/938 [03:52<02:41,  2.41it/s]
Training 1/1 epoch (loss 1.4574):  59%|█████▊    | 550/938 [03:52<02:38,  2.45it/s]
Training 1/1 epoch (loss 1.4303):  59%|█████▊    | 550/938 [03:53<02:38,  2.45it/s]
Training 1/1 epoch (loss 1.4303):  59%|█████▊    | 551/938 [03:53<02:35,  2.49it/s]
Training 1/1 epoch (loss 1.4070):  59%|█████▊    | 551/938 [03:53<02:35,  2.49it/s]
Training 1/1 epoch (loss 1.4070):  59%|█████▉    | 552/938 [03:53<02:40,  2.40it/s]
Training 1/1 epoch (loss 1.4771):  59%|█████▉    | 552/938 [03:54<02:40,  2.40it/s]
Training 1/1 epoch (loss 1.4771):  59%|█████▉    | 553/938 [03:54<02:35,  2.48it/s]
Training 1/1 epoch (loss 1.4594):  59%|█████▉    | 553/938 [03:54<02:35,  2.48it/s]
Training 1/1 epoch (loss 1.4594):  59%|█████▉    | 554/938 [03:54<02:34,  2.49it/s]
Training 1/1 epoch (loss 1.3736):  59%|█████▉    | 554/938 [03:54<02:34,  2.49it/s]
Training 1/1 epoch (loss 1.3736):  59%|█████▉    | 555/938 [03:54<02:36,  2.45it/s]
Training 1/1 epoch (loss 1.4771):  59%|█████▉    | 555/938 [03:55<02:36,  2.45it/s]
Training 1/1 epoch (loss 1.4771):  59%|█████▉    | 556/938 [03:55<02:33,  2.48it/s]
Training 1/1 epoch (loss 1.4699):  59%|█████▉    | 556/938 [03:55<02:33,  2.48it/s]
Training 1/1 epoch (loss 1.4699):  59%|█████▉    | 557/938 [03:55<02:32,  2.50it/s]
Training 1/1 epoch (loss 1.4556):  59%|█████▉    | 557/938 [03:56<02:32,  2.50it/s]
Training 1/1 epoch (loss 1.4556):  59%|█████▉    | 558/938 [03:56<02:27,  2.57it/s]
Training 1/1 epoch (loss 1.4552):  59%|█████▉    | 558/938 [03:56<02:27,  2.57it/s]
Training 1/1 epoch (loss 1.4552):  60%|█████▉    | 559/938 [03:56<02:28,  2.55it/s]
Training 1/1 epoch (loss 1.3534):  60%|█████▉    | 559/938 [03:56<02:28,  2.55it/s]
Training 1/1 epoch (loss 1.3534):  60%|█████▉    | 560/938 [03:56<02:32,  2.48it/s]
Training 1/1 epoch (loss 1.4651):  60%|█████▉    | 560/938 [03:57<02:32,  2.48it/s]
Training 1/1 epoch (loss 1.4651):  60%|█████▉    | 561/938 [03:57<02:30,  2.51it/s]
Training 1/1 epoch (loss 1.5284):  60%|█████▉    | 561/938 [03:57<02:30,  2.51it/s]
Training 1/1 epoch (loss 1.5284):  60%|█████▉    | 562/938 [03:57<02:28,  2.53it/s]
Training 1/1 epoch (loss 1.3448):  60%|█████▉    | 562/938 [03:58<02:28,  2.53it/s]
Training 1/1 epoch (loss 1.3448):  60%|██████    | 563/938 [03:58<02:31,  2.47it/s]
Training 1/1 epoch (loss 1.4598):  60%|██████    | 563/938 [03:58<02:31,  2.47it/s]
Training 1/1 epoch (loss 1.4598):  60%|██████    | 564/938 [03:58<02:31,  2.47it/s]
Training 1/1 epoch (loss 1.4176):  60%|██████    | 564/938 [03:58<02:31,  2.47it/s]
Training 1/1 epoch (loss 1.4176):  60%|██████    | 565/938 [03:58<02:29,  2.50it/s]
Training 1/1 epoch (loss 1.3988):  60%|██████    | 565/938 [03:59<02:29,  2.50it/s]
Training 1/1 epoch (loss 1.3988):  60%|██████    | 566/938 [03:59<02:25,  2.56it/s]
Training 1/1 epoch (loss 1.4618):  60%|██████    | 566/938 [03:59<02:25,  2.56it/s]
Training 1/1 epoch (loss 1.4618):  60%|██████    | 567/938 [03:59<02:22,  2.61it/s]
Training 1/1 epoch (loss 1.5128):  60%|██████    | 567/938 [03:59<02:22,  2.61it/s]
Training 1/1 epoch (loss 1.5128):  61%|██████    | 568/938 [03:59<02:24,  2.56it/s]
Training 1/1 epoch (loss 1.5869):  61%|██████    | 568/938 [04:00<02:24,  2.56it/s]
Training 1/1 epoch (loss 1.5869):  61%|██████    | 569/938 [04:00<02:23,  2.58it/s]
Training 1/1 epoch (loss 1.4363):  61%|██████    | 569/938 [04:00<02:23,  2.58it/s]
Training 1/1 epoch (loss 1.4363):  61%|██████    | 570/938 [04:00<02:24,  2.55it/s]
Training 1/1 epoch (loss 1.4779):  61%|██████    | 570/938 [04:01<02:24,  2.55it/s]
Training 1/1 epoch (loss 1.4779):  61%|██████    | 571/938 [04:01<02:23,  2.56it/s]
Training 1/1 epoch (loss 1.4521):  61%|██████    | 571/938 [04:01<02:23,  2.56it/s]
Training 1/1 epoch (loss 1.4521):  61%|██████    | 572/938 [04:01<02:23,  2.55it/s]
Training 1/1 epoch (loss 1.4669):  61%|██████    | 572/938 [04:01<02:23,  2.55it/s]
Training 1/1 epoch (loss 1.4669):  61%|██████    | 573/938 [04:01<02:20,  2.59it/s]
Training 1/1 epoch (loss 1.4337):  61%|██████    | 573/938 [04:02<02:20,  2.59it/s]
Training 1/1 epoch (loss 1.4337):  61%|██████    | 574/938 [04:02<02:26,  2.48it/s]
Training 1/1 epoch (loss 1.4187):  61%|██████    | 574/938 [04:02<02:26,  2.48it/s]
Training 1/1 epoch (loss 1.4187):  61%|██████▏   | 575/938 [04:02<02:27,  2.47it/s]
Training 1/1 epoch (loss 1.3987):  61%|██████▏   | 575/938 [04:03<02:27,  2.47it/s]
Training 1/1 epoch (loss 1.3987):  61%|██████▏   | 576/938 [04:03<02:29,  2.43it/s]
Training 1/1 epoch (loss 1.3968):  61%|██████▏   | 576/938 [04:03<02:29,  2.43it/s]
Training 1/1 epoch (loss 1.3968):  62%|██████▏   | 577/938 [04:03<02:26,  2.47it/s]
Training 1/1 epoch (loss 1.5279):  62%|██████▏   | 577/938 [04:04<02:26,  2.47it/s]
Training 1/1 epoch (loss 1.5279):  62%|██████▏   | 578/938 [04:04<02:25,  2.47it/s]
Training 1/1 epoch (loss 1.4752):  62%|██████▏   | 578/938 [04:04<02:25,  2.47it/s]
Training 1/1 epoch (loss 1.4752):  62%|██████▏   | 579/938 [04:04<02:25,  2.47it/s]
Training 1/1 epoch (loss 1.4625):  62%|██████▏   | 579/938 [04:04<02:25,  2.47it/s]
Training 1/1 epoch (loss 1.4625):  62%|██████▏   | 580/938 [04:04<02:33,  2.33it/s]
Training 1/1 epoch (loss 1.4332):  62%|██████▏   | 580/938 [04:05<02:33,  2.33it/s]
Training 1/1 epoch (loss 1.4332):  62%|██████▏   | 581/938 [04:05<02:42,  2.20it/s]
Training 1/1 epoch (loss 1.3807):  62%|██████▏   | 581/938 [04:06<02:42,  2.20it/s]
Training 1/1 epoch (loss 1.3807):  62%|██████▏   | 582/938 [04:06<03:01,  1.96it/s]
Training 1/1 epoch (loss 1.4794):  62%|██████▏   | 582/938 [04:06<03:01,  1.96it/s]
Training 1/1 epoch (loss 1.4794):  62%|██████▏   | 583/938 [04:06<02:50,  2.09it/s]
Training 1/1 epoch (loss 1.3210):  62%|██████▏   | 583/938 [04:06<02:50,  2.09it/s]
Training 1/1 epoch (loss 1.3210):  62%|██████▏   | 584/938 [04:06<02:43,  2.16it/s]
Training 1/1 epoch (loss 1.3330):  62%|██████▏   | 584/938 [04:07<02:43,  2.16it/s]
Training 1/1 epoch (loss 1.3330):  62%|██████▏   | 585/938 [04:07<02:33,  2.30it/s]
Training 1/1 epoch (loss 1.4809):  62%|██████▏   | 585/938 [04:07<02:33,  2.30it/s]
Training 1/1 epoch (loss 1.4809):  62%|██████▏   | 586/938 [04:07<02:29,  2.35it/s]
Training 1/1 epoch (loss 1.4550):  62%|██████▏   | 586/938 [04:08<02:29,  2.35it/s]
Training 1/1 epoch (loss 1.4550):  63%|██████▎   | 587/938 [04:08<02:28,  2.37it/s]
Training 1/1 epoch (loss 1.4898):  63%|██████▎   | 587/938 [04:08<02:28,  2.37it/s]
Training 1/1 epoch (loss 1.4898):  63%|██████▎   | 588/938 [04:08<02:22,  2.45it/s]
Training 1/1 epoch (loss 1.4710):  63%|██████▎   | 588/938 [04:08<02:22,  2.45it/s]
Training 1/1 epoch (loss 1.4710):  63%|██████▎   | 589/938 [04:08<02:19,  2.50it/s]
Training 1/1 epoch (loss 1.4837):  63%|██████▎   | 589/938 [04:09<02:19,  2.50it/s]
Training 1/1 epoch (loss 1.4837):  63%|██████▎   | 590/938 [04:09<02:18,  2.52it/s]
Training 1/1 epoch (loss 1.4034):  63%|██████▎   | 590/938 [04:09<02:18,  2.52it/s]
Training 1/1 epoch (loss 1.4034):  63%|██████▎   | 591/938 [04:09<02:16,  2.55it/s]
Training 1/1 epoch (loss 1.3146):  63%|██████▎   | 591/938 [04:10<02:16,  2.55it/s]
Training 1/1 epoch (loss 1.3146):  63%|██████▎   | 592/938 [04:10<02:20,  2.45it/s]
Training 1/1 epoch (loss 1.5097):  63%|██████▎   | 592/938 [04:10<02:20,  2.45it/s]
Training 1/1 epoch (loss 1.5097):  63%|██████▎   | 593/938 [04:10<02:19,  2.48it/s]
Training 1/1 epoch (loss 1.4368):  63%|██████▎   | 593/938 [04:10<02:19,  2.48it/s]
Training 1/1 epoch (loss 1.4368):  63%|██████▎   | 594/938 [04:10<02:19,  2.47it/s]
Training 1/1 epoch (loss 1.3773):  63%|██████▎   | 594/938 [04:11<02:19,  2.47it/s]
Training 1/1 epoch (loss 1.3773):  63%|██████▎   | 595/938 [04:11<02:17,  2.49it/s]
Training 1/1 epoch (loss 1.4964):  63%|██████▎   | 595/938 [04:11<02:17,  2.49it/s]
Training 1/1 epoch (loss 1.4964):  64%|██████▎   | 596/938 [04:11<02:22,  2.41it/s]
Training 1/1 epoch (loss 1.4193):  64%|██████▎   | 596/938 [04:12<02:22,  2.41it/s]
Training 1/1 epoch (loss 1.4193):  64%|██████▎   | 597/938 [04:12<02:19,  2.44it/s]
Training 1/1 epoch (loss 1.5206):  64%|██████▎   | 597/938 [04:12<02:19,  2.44it/s]
Training 1/1 epoch (loss 1.5206):  64%|██████▍   | 598/938 [04:12<02:15,  2.50it/s]
Training 1/1 epoch (loss 1.4747):  64%|██████▍   | 598/938 [04:12<02:15,  2.50it/s]
Training 1/1 epoch (loss 1.4747):  64%|██████▍   | 599/938 [04:12<02:17,  2.46it/s]
Training 1/1 epoch (loss 1.3846):  64%|██████▍   | 599/938 [04:13<02:17,  2.46it/s]
Training 1/1 epoch (loss 1.3846):  64%|██████▍   | 600/938 [04:13<02:16,  2.47it/s]
Training 1/1 epoch (loss 1.4954):  64%|██████▍   | 600/938 [04:13<02:16,  2.47it/s]
Training 1/1 epoch (loss 1.4954):  64%|██████▍   | 601/938 [04:13<02:14,  2.50it/s]
Training 1/1 epoch (loss 1.3438):  64%|██████▍   | 601/938 [04:14<02:14,  2.50it/s]
Training 1/1 epoch (loss 1.3438):  64%|██████▍   | 602/938 [04:14<02:14,  2.50it/s]
Training 1/1 epoch (loss 1.4715):  64%|██████▍   | 602/938 [04:14<02:14,  2.50it/s]
Training 1/1 epoch (loss 1.4715):  64%|██████▍   | 603/938 [04:14<02:14,  2.48it/s]
Training 1/1 epoch (loss 1.4930):  64%|██████▍   | 603/938 [04:14<02:14,  2.48it/s]
Training 1/1 epoch (loss 1.4930):  64%|██████▍   | 604/938 [04:14<02:20,  2.37it/s]
Training 1/1 epoch (loss 1.4025):  64%|██████▍   | 604/938 [04:15<02:20,  2.37it/s]
Training 1/1 epoch (loss 1.4025):  64%|██████▍   | 605/938 [04:15<02:14,  2.48it/s]
Training 1/1 epoch (loss 1.4946):  64%|██████▍   | 605/938 [04:15<02:14,  2.48it/s]
Training 1/1 epoch (loss 1.4946):  65%|██████▍   | 606/938 [04:15<02:16,  2.42it/s]
Training 1/1 epoch (loss 1.4157):  65%|██████▍   | 606/938 [04:16<02:16,  2.42it/s]
Training 1/1 epoch (loss 1.4157):  65%|██████▍   | 607/938 [04:16<02:13,  2.47it/s]
Training 1/1 epoch (loss 1.5149):  65%|██████▍   | 607/938 [04:16<02:13,  2.47it/s]
Training 1/1 epoch (loss 1.5149):  65%|██████▍   | 608/938 [04:16<02:17,  2.39it/s]
Training 1/1 epoch (loss 1.4091):  65%|██████▍   | 608/938 [04:16<02:17,  2.39it/s]
Training 1/1 epoch (loss 1.4091):  65%|██████▍   | 609/938 [04:16<02:16,  2.41it/s]
Training 1/1 epoch (loss 1.3924):  65%|██████▍   | 609/938 [04:17<02:16,  2.41it/s]
Training 1/1 epoch (loss 1.3924):  65%|██████▌   | 610/938 [04:17<02:12,  2.47it/s]
Training 1/1 epoch (loss 1.4786):  65%|██████▌   | 610/938 [04:17<02:12,  2.47it/s]
Training 1/1 epoch (loss 1.4786):  65%|██████▌   | 611/938 [04:17<02:08,  2.55it/s]
Training 1/1 epoch (loss 1.3736):  65%|██████▌   | 611/938 [04:18<02:08,  2.55it/s]
Training 1/1 epoch (loss 1.3736):  65%|██████▌   | 612/938 [04:18<02:12,  2.45it/s]
Training 1/1 epoch (loss 1.2647):  65%|██████▌   | 612/938 [04:18<02:12,  2.45it/s]
Training 1/1 epoch (loss 1.2647):  65%|██████▌   | 613/938 [04:18<02:13,  2.44it/s]
Training 1/1 epoch (loss 1.5651):  65%|██████▌   | 613/938 [04:18<02:13,  2.44it/s]
Training 1/1 epoch (loss 1.5651):  65%|██████▌   | 614/938 [04:18<02:11,  2.47it/s]
Training 1/1 epoch (loss 1.3193):  65%|██████▌   | 614/938 [04:19<02:11,  2.47it/s]
Training 1/1 epoch (loss 1.3193):  66%|██████▌   | 615/938 [04:19<02:10,  2.47it/s]
Training 1/1 epoch (loss 1.4646):  66%|██████▌   | 615/938 [04:19<02:10,  2.47it/s]
Training 1/1 epoch (loss 1.4646):  66%|██████▌   | 616/938 [04:19<02:09,  2.49it/s]
Training 1/1 epoch (loss 1.3352):  66%|██████▌   | 616/938 [04:20<02:09,  2.49it/s]
Training 1/1 epoch (loss 1.3352):  66%|██████▌   | 617/938 [04:20<02:05,  2.56it/s]
Training 1/1 epoch (loss 1.3857):  66%|██████▌   | 617/938 [04:20<02:05,  2.56it/s]
Training 1/1 epoch (loss 1.3857):  66%|██████▌   | 618/938 [04:20<02:09,  2.48it/s]
Training 1/1 epoch (loss 1.4323):  66%|██████▌   | 618/938 [04:20<02:09,  2.48it/s]
Training 1/1 epoch (loss 1.4323):  66%|██████▌   | 619/938 [04:20<02:07,  2.51it/s]
Training 1/1 epoch (loss 1.3219):  66%|██████▌   | 619/938 [04:21<02:07,  2.51it/s]
Training 1/1 epoch (loss 1.3219):  66%|██████▌   | 620/938 [04:21<02:02,  2.59it/s]
Training 1/1 epoch (loss 1.3018):  66%|██████▌   | 620/938 [04:21<02:02,  2.59it/s]
Training 1/1 epoch (loss 1.3018):  66%|██████▌   | 621/938 [04:21<02:06,  2.51it/s]
Training 1/1 epoch (loss 1.5541):  66%|██████▌   | 621/938 [04:22<02:06,  2.51it/s]
Training 1/1 epoch (loss 1.5541):  66%|██████▋   | 622/938 [04:22<02:02,  2.58it/s]
Training 1/1 epoch (loss 1.3655):  66%|██████▋   | 622/938 [04:22<02:02,  2.58it/s]
Training 1/1 epoch (loss 1.3655):  66%|██████▋   | 623/938 [04:22<02:07,  2.47it/s]
Training 1/1 epoch (loss 1.3774):  66%|██████▋   | 623/938 [04:22<02:07,  2.47it/s]
Training 1/1 epoch (loss 1.3774):  67%|██████▋   | 624/938 [04:22<02:09,  2.43it/s]
Training 1/1 epoch (loss 1.2429):  67%|██████▋   | 624/938 [04:23<02:09,  2.43it/s]
Training 1/1 epoch (loss 1.2429):  67%|██████▋   | 625/938 [04:23<02:07,  2.46it/s]
Training 1/1 epoch (loss 1.4347):  67%|██████▋   | 625/938 [04:23<02:07,  2.46it/s]
Training 1/1 epoch (loss 1.4347):  67%|██████▋   | 626/938 [04:23<02:05,  2.49it/s]
Training 1/1 epoch (loss 1.4701):  67%|██████▋   | 626/938 [04:24<02:05,  2.49it/s]
Training 1/1 epoch (loss 1.4701):  67%|██████▋   | 627/938 [04:24<02:02,  2.55it/s]
Training 1/1 epoch (loss 1.4212):  67%|██████▋   | 627/938 [04:24<02:02,  2.55it/s]
Training 1/1 epoch (loss 1.4212):  67%|██████▋   | 628/938 [04:24<01:58,  2.61it/s]
Training 1/1 epoch (loss 1.4026):  67%|██████▋   | 628/938 [04:24<01:58,  2.61it/s]
Training 1/1 epoch (loss 1.4026):  67%|██████▋   | 629/938 [04:24<02:01,  2.54it/s]
Training 1/1 epoch (loss 1.4472):  67%|██████▋   | 629/938 [04:25<02:01,  2.54it/s]
Training 1/1 epoch (loss 1.4472):  67%|██████▋   | 630/938 [04:25<02:03,  2.50it/s]
Training 1/1 epoch (loss 1.3325):  67%|██████▋   | 630/938 [04:25<02:03,  2.50it/s]
Training 1/1 epoch (loss 1.3325):  67%|██████▋   | 631/938 [04:25<02:04,  2.47it/s]
Training 1/1 epoch (loss 1.4928):  67%|██████▋   | 631/938 [04:26<02:04,  2.47it/s]
Training 1/1 epoch (loss 1.4928):  67%|██████▋   | 632/938 [04:26<02:03,  2.48it/s]
Training 1/1 epoch (loss 1.3055):  67%|██████▋   | 632/938 [04:26<02:03,  2.48it/s]
Training 1/1 epoch (loss 1.3055):  67%|██████▋   | 633/938 [04:26<02:01,  2.51it/s]
Training 1/1 epoch (loss 1.4178):  67%|██████▋   | 633/938 [04:26<02:01,  2.51it/s]
Training 1/1 epoch (loss 1.4178):  68%|██████▊   | 634/938 [04:26<01:59,  2.54it/s]
Training 1/1 epoch (loss 1.4152):  68%|██████▊   | 634/938 [04:27<01:59,  2.54it/s]
Training 1/1 epoch (loss 1.4152):  68%|██████▊   | 635/938 [04:27<02:04,  2.43it/s]
Training 1/1 epoch (loss 1.3715):  68%|██████▊   | 635/938 [04:27<02:04,  2.43it/s]
Training 1/1 epoch (loss 1.3715):  68%|██████▊   | 636/938 [04:27<02:00,  2.50it/s]
Training 1/1 epoch (loss 1.4981):  68%|██████▊   | 636/938 [04:28<02:00,  2.50it/s]
Training 1/1 epoch (loss 1.4981):  68%|██████▊   | 637/938 [04:28<01:58,  2.55it/s]
Training 1/1 epoch (loss 1.4219):  68%|██████▊   | 637/938 [04:28<01:58,  2.55it/s]
Training 1/1 epoch (loss 1.4219):  68%|██████▊   | 638/938 [04:28<01:56,  2.58it/s]
Training 1/1 epoch (loss 1.3188):  68%|██████▊   | 638/938 [04:28<01:56,  2.58it/s]
Training 1/1 epoch (loss 1.3188):  68%|██████▊   | 639/938 [04:28<02:01,  2.46it/s]
Training 1/1 epoch (loss 1.4781):  68%|██████▊   | 639/938 [04:29<02:01,  2.46it/s]
Training 1/1 epoch (loss 1.4781):  68%|██████▊   | 640/938 [04:29<02:01,  2.44it/s]
Training 1/1 epoch (loss 1.3213):  68%|██████▊   | 640/938 [04:29<02:01,  2.44it/s]
Training 1/1 epoch (loss 1.3213):  68%|██████▊   | 641/938 [04:29<01:57,  2.52it/s]
Training 1/1 epoch (loss 1.3270):  68%|██████▊   | 641/938 [04:30<01:57,  2.52it/s]
Training 1/1 epoch (loss 1.3270):  68%|██████▊   | 642/938 [04:30<02:00,  2.45it/s]
Training 1/1 epoch (loss 1.4478):  68%|██████▊   | 642/938 [04:30<02:00,  2.45it/s]
Training 1/1 epoch (loss 1.4478):  69%|██████▊   | 643/938 [04:30<01:59,  2.47it/s]
Training 1/1 epoch (loss 1.3158):  69%|██████▊   | 643/938 [04:30<01:59,  2.47it/s]
Training 1/1 epoch (loss 1.3158):  69%|██████▊   | 644/938 [04:30<01:57,  2.50it/s]
Training 1/1 epoch (loss 1.4322):  69%|██████▊   | 644/938 [04:31<01:57,  2.50it/s]
Training 1/1 epoch (loss 1.4322):  69%|██████▉   | 645/938 [04:31<01:57,  2.49it/s]
Training 1/1 epoch (loss 1.4980):  69%|██████▉   | 645/938 [04:31<01:57,  2.49it/s]
Training 1/1 epoch (loss 1.4980):  69%|██████▉   | 646/938 [04:31<01:54,  2.55it/s]
Training 1/1 epoch (loss 1.4898):  69%|██████▉   | 646/938 [04:32<01:54,  2.55it/s]
Training 1/1 epoch (loss 1.4898):  69%|██████▉   | 647/938 [04:32<01:51,  2.61it/s]
Training 1/1 epoch (loss 1.4786):  69%|██████▉   | 647/938 [04:32<01:51,  2.61it/s]
Training 1/1 epoch (loss 1.4786):  69%|██████▉   | 648/938 [04:32<01:54,  2.54it/s]
Training 1/1 epoch (loss 1.4262):  69%|██████▉   | 648/938 [04:32<01:54,  2.54it/s]
Training 1/1 epoch (loss 1.4262):  69%|██████▉   | 649/938 [04:32<01:56,  2.49it/s]
Training 1/1 epoch (loss 1.3514):  69%|██████▉   | 649/938 [04:33<01:56,  2.49it/s]
Training 1/1 epoch (loss 1.3514):  69%|██████▉   | 650/938 [04:33<01:59,  2.42it/s]
Training 1/1 epoch (loss 1.4159):  69%|██████▉   | 650/938 [04:33<01:59,  2.42it/s]
Training 1/1 epoch (loss 1.4159):  69%|██████▉   | 651/938 [04:33<01:55,  2.50it/s]
Training 1/1 epoch (loss 1.3796):  69%|██████▉   | 651/938 [04:34<01:55,  2.50it/s]
Training 1/1 epoch (loss 1.3796):  70%|██████▉   | 652/938 [04:34<01:56,  2.46it/s]
Training 1/1 epoch (loss 1.5353):  70%|██████▉   | 652/938 [04:34<01:56,  2.46it/s]
Training 1/1 epoch (loss 1.5353):  70%|██████▉   | 653/938 [04:34<01:54,  2.48it/s]
Training 1/1 epoch (loss 1.5088):  70%|██████▉   | 653/938 [04:34<01:54,  2.48it/s]
Training 1/1 epoch (loss 1.5088):  70%|██████▉   | 654/938 [04:34<01:53,  2.51it/s]
Training 1/1 epoch (loss 1.3820):  70%|██████▉   | 654/938 [04:35<01:53,  2.51it/s]
Training 1/1 epoch (loss 1.3820):  70%|██████▉   | 655/938 [04:35<01:59,  2.36it/s]
Training 1/1 epoch (loss 1.4236):  70%|██████▉   | 655/938 [04:36<01:59,  2.36it/s]
Training 1/1 epoch (loss 1.4236):  70%|██████▉   | 656/938 [04:36<02:17,  2.05it/s]
Training 1/1 epoch (loss 1.4872):  70%|██████▉   | 656/938 [04:36<02:17,  2.05it/s]
Training 1/1 epoch (loss 1.4872):  70%|███████   | 657/938 [04:36<02:12,  2.13it/s]
Training 1/1 epoch (loss 1.4189):  70%|███████   | 657/938 [04:36<02:12,  2.13it/s]
Training 1/1 epoch (loss 1.4189):  70%|███████   | 658/938 [04:36<02:06,  2.21it/s]
Training 1/1 epoch (loss 1.4803):  70%|███████   | 658/938 [04:37<02:06,  2.21it/s]
Training 1/1 epoch (loss 1.4803):  70%|███████   | 659/938 [04:37<02:00,  2.32it/s]
Training 1/1 epoch (loss 1.4382):  70%|███████   | 659/938 [04:37<02:00,  2.32it/s]
Training 1/1 epoch (loss 1.4382):  70%|███████   | 660/938 [04:37<01:57,  2.36it/s]
Training 1/1 epoch (loss 1.4592):  70%|███████   | 660/938 [04:38<01:57,  2.36it/s]
Training 1/1 epoch (loss 1.4592):  70%|███████   | 661/938 [04:38<01:54,  2.43it/s]
Training 1/1 epoch (loss 1.4172):  70%|███████   | 661/938 [04:38<01:54,  2.43it/s]
Training 1/1 epoch (loss 1.4172):  71%|███████   | 662/938 [04:38<01:50,  2.49it/s]
Training 1/1 epoch (loss 1.3271):  71%|███████   | 662/938 [04:38<01:50,  2.49it/s]
Training 1/1 epoch (loss 1.3271):  71%|███████   | 663/938 [04:38<01:51,  2.46it/s]
Training 1/1 epoch (loss 1.4438):  71%|███████   | 663/938 [04:39<01:51,  2.46it/s]
Training 1/1 epoch (loss 1.4438):  71%|███████   | 664/938 [04:39<01:52,  2.44it/s]
Training 1/1 epoch (loss 1.4533):  71%|███████   | 664/938 [04:39<01:52,  2.44it/s]
Training 1/1 epoch (loss 1.4533):  71%|███████   | 665/938 [04:39<01:50,  2.47it/s]
Training 1/1 epoch (loss 1.3928):  71%|███████   | 665/938 [04:40<01:50,  2.47it/s]
Training 1/1 epoch (loss 1.3928):  71%|███████   | 666/938 [04:40<01:49,  2.47it/s]
Training 1/1 epoch (loss 1.3925):  71%|███████   | 666/938 [04:40<01:49,  2.47it/s]
Training 1/1 epoch (loss 1.3925):  71%|███████   | 667/938 [04:40<01:48,  2.49it/s]
Training 1/1 epoch (loss 1.2964):  71%|███████   | 667/938 [04:40<01:48,  2.49it/s]
Training 1/1 epoch (loss 1.2964):  71%|███████   | 668/938 [04:40<01:46,  2.55it/s]
Training 1/1 epoch (loss 1.3570):  71%|███████   | 668/938 [04:41<01:46,  2.55it/s]
Training 1/1 epoch (loss 1.3570):  71%|███████▏  | 669/938 [04:41<01:42,  2.62it/s]
Training 1/1 epoch (loss 1.3157):  71%|███████▏  | 669/938 [04:41<01:42,  2.62it/s]
Training 1/1 epoch (loss 1.3157):  71%|███████▏  | 670/938 [04:41<01:42,  2.61it/s]
Training 1/1 epoch (loss 1.4791):  71%|███████▏  | 670/938 [04:41<01:42,  2.61it/s]
Training 1/1 epoch (loss 1.4791):  72%|███████▏  | 671/938 [04:41<01:40,  2.65it/s]
Training 1/1 epoch (loss 1.5764):  72%|███████▏  | 671/938 [04:42<01:40,  2.65it/s]
Training 1/1 epoch (loss 1.5764):  72%|███████▏  | 672/938 [04:42<01:44,  2.54it/s]
Training 1/1 epoch (loss 1.3959):  72%|███████▏  | 672/938 [04:42<01:44,  2.54it/s]
Training 1/1 epoch (loss 1.3959):  72%|███████▏  | 673/938 [04:42<01:42,  2.59it/s]
Training 1/1 epoch (loss 1.4215):  72%|███████▏  | 673/938 [04:43<01:42,  2.59it/s]
Training 1/1 epoch (loss 1.4215):  72%|███████▏  | 674/938 [04:43<01:41,  2.60it/s]
Training 1/1 epoch (loss 1.3360):  72%|███████▏  | 674/938 [04:43<01:41,  2.60it/s]
Training 1/1 epoch (loss 1.3360):  72%|███████▏  | 675/938 [04:43<01:40,  2.62it/s]
Training 1/1 epoch (loss 1.4123):  72%|███████▏  | 675/938 [04:43<01:40,  2.62it/s]
Training 1/1 epoch (loss 1.4123):  72%|███████▏  | 676/938 [04:43<01:44,  2.50it/s]
Training 1/1 epoch (loss 1.5267):  72%|███████▏  | 676/938 [04:44<01:44,  2.50it/s]
Training 1/1 epoch (loss 1.5267):  72%|███████▏  | 677/938 [04:44<01:43,  2.52it/s]
Training 1/1 epoch (loss 1.3370):  72%|███████▏  | 677/938 [04:44<01:43,  2.52it/s]
Training 1/1 epoch (loss 1.3370):  72%|███████▏  | 678/938 [04:44<01:51,  2.34it/s]
Training 1/1 epoch (loss 1.3895):  72%|███████▏  | 678/938 [04:45<01:51,  2.34it/s]
Training 1/1 epoch (loss 1.3895):  72%|███████▏  | 679/938 [04:45<01:50,  2.35it/s]
Training 1/1 epoch (loss 1.4430):  72%|███████▏  | 679/938 [04:45<01:50,  2.35it/s]
Training 1/1 epoch (loss 1.4430):  72%|███████▏  | 680/938 [04:45<01:49,  2.36it/s]
Training 1/1 epoch (loss 1.4054):  72%|███████▏  | 680/938 [04:46<01:49,  2.36it/s]
Training 1/1 epoch (loss 1.4054):  73%|███████▎  | 681/938 [04:46<01:48,  2.36it/s]
Training 1/1 epoch (loss 1.4632):  73%|███████▎  | 681/938 [04:46<01:48,  2.36it/s]
Training 1/1 epoch (loss 1.4632):  73%|███████▎  | 682/938 [04:46<01:45,  2.42it/s]
Training 1/1 epoch (loss 1.4338):  73%|███████▎  | 682/938 [04:46<01:45,  2.42it/s]
Training 1/1 epoch (loss 1.4338):  73%|███████▎  | 683/938 [04:46<01:44,  2.45it/s]
Training 1/1 epoch (loss 1.4547):  73%|███████▎  | 683/938 [04:47<01:44,  2.45it/s]
Training 1/1 epoch (loss 1.4547):  73%|███████▎  | 684/938 [04:47<01:42,  2.48it/s]
Training 1/1 epoch (loss 1.4786):  73%|███████▎  | 684/938 [04:47<01:42,  2.48it/s]
Training 1/1 epoch (loss 1.4786):  73%|███████▎  | 685/938 [04:47<01:40,  2.53it/s]
Training 1/1 epoch (loss 1.5015):  73%|███████▎  | 685/938 [04:48<01:40,  2.53it/s]
Training 1/1 epoch (loss 1.5015):  73%|███████▎  | 686/938 [04:48<01:46,  2.37it/s]
Training 1/1 epoch (loss 1.3905):  73%|███████▎  | 686/938 [04:48<01:46,  2.37it/s]
Training 1/1 epoch (loss 1.3905):  73%|███████▎  | 687/938 [04:48<01:43,  2.42it/s]
Training 1/1 epoch (loss 1.4509):  73%|███████▎  | 687/938 [04:48<01:43,  2.42it/s]
Training 1/1 epoch (loss 1.4509):  73%|███████▎  | 688/938 [04:48<01:42,  2.44it/s]
Training 1/1 epoch (loss 1.4738):  73%|███████▎  | 688/938 [04:49<01:42,  2.44it/s]
Training 1/1 epoch (loss 1.4738):  73%|███████▎  | 689/938 [04:49<01:44,  2.39it/s]
Training 1/1 epoch (loss 1.4620):  73%|███████▎  | 689/938 [04:49<01:44,  2.39it/s]
Training 1/1 epoch (loss 1.4620):  74%|███████▎  | 690/938 [04:49<01:42,  2.41it/s]
Training 1/1 epoch (loss 1.5188):  74%|███████▎  | 690/938 [04:50<01:42,  2.41it/s]
Training 1/1 epoch (loss 1.5188):  74%|███████▎  | 691/938 [04:50<01:41,  2.44it/s]
Training 1/1 epoch (loss 1.4448):  74%|███████▎  | 691/938 [04:50<01:41,  2.44it/s]
Training 1/1 epoch (loss 1.4448):  74%|███████▍  | 692/938 [04:50<01:43,  2.37it/s]
Training 1/1 epoch (loss 1.4126):  74%|███████▍  | 692/938 [04:51<01:43,  2.37it/s]
Training 1/1 epoch (loss 1.4126):  74%|███████▍  | 693/938 [04:51<01:40,  2.44it/s]
Training 1/1 epoch (loss 1.3541):  74%|███████▍  | 693/938 [04:51<01:40,  2.44it/s]
Training 1/1 epoch (loss 1.3541):  74%|███████▍  | 694/938 [04:51<01:39,  2.44it/s]
Training 1/1 epoch (loss 1.4373):  74%|███████▍  | 694/938 [04:51<01:39,  2.44it/s]
Training 1/1 epoch (loss 1.4373):  74%|███████▍  | 695/938 [04:51<01:36,  2.52it/s]
Training 1/1 epoch (loss 1.2442):  74%|███████▍  | 695/938 [04:52<01:36,  2.52it/s]
Training 1/1 epoch (loss 1.2442):  74%|███████▍  | 696/938 [04:52<01:37,  2.48it/s]
Training 1/1 epoch (loss 1.4684):  74%|███████▍  | 696/938 [04:52<01:37,  2.48it/s]
Training 1/1 epoch (loss 1.4684):  74%|███████▍  | 697/938 [04:52<01:38,  2.45it/s]
Training 1/1 epoch (loss 1.3573):  74%|███████▍  | 697/938 [04:53<01:38,  2.45it/s]
Training 1/1 epoch (loss 1.3573):  74%|███████▍  | 698/938 [04:53<01:36,  2.48it/s]
Training 1/1 epoch (loss 1.3015):  74%|███████▍  | 698/938 [04:53<01:36,  2.48it/s]
Training 1/1 epoch (loss 1.3015):  75%|███████▍  | 699/938 [04:53<01:35,  2.50it/s]
Training 1/1 epoch (loss 1.4357):  75%|███████▍  | 699/938 [04:53<01:35,  2.50it/s]
Training 1/1 epoch (loss 1.4357):  75%|███████▍  | 700/938 [04:53<01:32,  2.57it/s]
Training 1/1 epoch (loss 1.3594):  75%|███████▍  | 700/938 [04:54<01:32,  2.57it/s]
Training 1/1 epoch (loss 1.3594):  75%|███████▍  | 701/938 [04:54<01:33,  2.54it/s]
Training 1/1 epoch (loss 1.3315):  75%|███████▍  | 701/938 [04:54<01:33,  2.54it/s]
Training 1/1 epoch (loss 1.3315):  75%|███████▍  | 702/938 [04:54<01:31,  2.58it/s]
Training 1/1 epoch (loss 1.4739):  75%|███████▍  | 702/938 [04:54<01:31,  2.58it/s]
Training 1/1 epoch (loss 1.4739):  75%|███████▍  | 703/938 [04:54<01:33,  2.51it/s]
Training 1/1 epoch (loss 1.3683):  75%|███████▍  | 703/938 [04:55<01:33,  2.51it/s]
Training 1/1 epoch (loss 1.3683):  75%|███████▌  | 704/938 [04:55<01:40,  2.33it/s]
Training 1/1 epoch (loss 1.4203):  75%|███████▌  | 704/938 [04:55<01:40,  2.33it/s]
Training 1/1 epoch (loss 1.4203):  75%|███████▌  | 705/938 [04:55<01:38,  2.36it/s]
Training 1/1 epoch (loss 1.3593):  75%|███████▌  | 705/938 [04:56<01:38,  2.36it/s]
Training 1/1 epoch (loss 1.3593):  75%|███████▌  | 706/938 [04:56<01:35,  2.44it/s]
Training 1/1 epoch (loss 1.4233):  75%|███████▌  | 706/938 [04:56<01:35,  2.44it/s]
Training 1/1 epoch (loss 1.4233):  75%|███████▌  | 707/938 [04:56<01:32,  2.50it/s]
Training 1/1 epoch (loss 1.4395):  75%|███████▌  | 707/938 [04:57<01:32,  2.50it/s]
Training 1/1 epoch (loss 1.4395):  75%|███████▌  | 708/938 [04:57<01:31,  2.52it/s]
Training 1/1 epoch (loss 1.5097):  75%|███████▌  | 708/938 [04:57<01:31,  2.52it/s]
Training 1/1 epoch (loss 1.5097):  76%|███████▌  | 709/938 [04:57<01:29,  2.55it/s]
Training 1/1 epoch (loss 1.3613):  76%|███████▌  | 709/938 [04:57<01:29,  2.55it/s]
Training 1/1 epoch (loss 1.3613):  76%|███████▌  | 710/938 [04:57<01:27,  2.60it/s]
Training 1/1 epoch (loss 1.4762):  76%|███████▌  | 710/938 [04:58<01:27,  2.60it/s]
Training 1/1 epoch (loss 1.4762):  76%|███████▌  | 711/938 [04:58<01:28,  2.57it/s]
Training 1/1 epoch (loss 1.3622):  76%|███████▌  | 711/938 [04:58<01:28,  2.57it/s]
Training 1/1 epoch (loss 1.3622):  76%|███████▌  | 712/938 [04:58<01:29,  2.51it/s]
Training 1/1 epoch (loss 1.4447):  76%|███████▌  | 712/938 [04:58<01:29,  2.51it/s]
Training 1/1 epoch (loss 1.4447):  76%|███████▌  | 713/938 [04:58<01:28,  2.53it/s]
Training 1/1 epoch (loss 1.3983):  76%|███████▌  | 713/938 [04:59<01:28,  2.53it/s]
Training 1/1 epoch (loss 1.3983):  76%|███████▌  | 714/938 [04:59<01:27,  2.55it/s]
Training 1/1 epoch (loss 1.4320):  76%|███████▌  | 714/938 [04:59<01:27,  2.55it/s]
Training 1/1 epoch (loss 1.4320):  76%|███████▌  | 715/938 [04:59<01:27,  2.55it/s]
Training 1/1 epoch (loss 1.3344):  76%|███████▌  | 715/938 [05:00<01:27,  2.55it/s]
Training 1/1 epoch (loss 1.3344):  76%|███████▋  | 716/938 [05:00<01:30,  2.44it/s]
Training 1/1 epoch (loss 1.4136):  76%|███████▋  | 716/938 [05:00<01:30,  2.44it/s]
Training 1/1 epoch (loss 1.4136):  76%|███████▋  | 717/938 [05:00<01:30,  2.44it/s]
Training 1/1 epoch (loss 1.3638):  76%|███████▋  | 717/938 [05:01<01:30,  2.44it/s]
Training 1/1 epoch (loss 1.3638):  77%|███████▋  | 718/938 [05:01<01:28,  2.48it/s]
Training 1/1 epoch (loss 1.4472):  77%|███████▋  | 718/938 [05:01<01:28,  2.48it/s]
Training 1/1 epoch (loss 1.4472):  77%|███████▋  | 719/938 [05:01<01:26,  2.52it/s]
Training 1/1 epoch (loss 1.3605):  77%|███████▋  | 719/938 [05:01<01:26,  2.52it/s]
Training 1/1 epoch (loss 1.3605):  77%|███████▋  | 720/938 [05:01<01:25,  2.54it/s]
Training 1/1 epoch (loss 1.3218):  77%|███████▋  | 720/938 [05:02<01:25,  2.54it/s]
Training 1/1 epoch (loss 1.3218):  77%|███████▋  | 721/938 [05:02<01:26,  2.50it/s]
Training 1/1 epoch (loss 1.3436):  77%|███████▋  | 721/938 [05:02<01:26,  2.50it/s]
Training 1/1 epoch (loss 1.3436):  77%|███████▋  | 722/938 [05:02<01:25,  2.53it/s]
Training 1/1 epoch (loss 1.2589):  77%|███████▋  | 722/938 [05:02<01:25,  2.53it/s]
Training 1/1 epoch (loss 1.2589):  77%|███████▋  | 723/938 [05:02<01:24,  2.56it/s]
Training 1/1 epoch (loss 1.3220):  77%|███████▋  | 723/938 [05:03<01:24,  2.56it/s]
Training 1/1 epoch (loss 1.3220):  77%|███████▋  | 724/938 [05:03<01:22,  2.59it/s]
Training 1/1 epoch (loss 1.3663):  77%|███████▋  | 724/938 [05:03<01:22,  2.59it/s]
Training 1/1 epoch (loss 1.3663):  77%|███████▋  | 725/938 [05:03<01:24,  2.51it/s]
Training 1/1 epoch (loss 1.3039):  77%|███████▋  | 725/938 [05:04<01:24,  2.51it/s]
Training 1/1 epoch (loss 1.3039):  77%|███████▋  | 726/938 [05:04<01:24,  2.50it/s]
Training 1/1 epoch (loss 1.4317):  77%|███████▋  | 726/938 [05:04<01:24,  2.50it/s]
Training 1/1 epoch (loss 1.4317):  78%|███████▊  | 727/938 [05:04<01:23,  2.53it/s]
Training 1/1 epoch (loss 1.2703):  78%|███████▊  | 727/938 [05:05<01:23,  2.53it/s]
Training 1/1 epoch (loss 1.2703):  78%|███████▊  | 728/938 [05:05<01:35,  2.21it/s]
Training 1/1 epoch (loss 1.3306):  78%|███████▊  | 728/938 [05:05<01:35,  2.21it/s]
Training 1/1 epoch (loss 1.3306):  78%|███████▊  | 729/938 [05:05<01:34,  2.21it/s]
Training 1/1 epoch (loss 1.4215):  78%|███████▊  | 729/938 [05:06<01:34,  2.21it/s]
Training 1/1 epoch (loss 1.4215):  78%|███████▊  | 730/938 [05:06<01:46,  1.95it/s]
Training 1/1 epoch (loss 1.4456):  78%|███████▊  | 730/938 [05:06<01:46,  1.95it/s]
Training 1/1 epoch (loss 1.4456):  78%|███████▊  | 731/938 [05:06<01:40,  2.06it/s]
Training 1/1 epoch (loss 1.3847):  78%|███████▊  | 731/938 [05:07<01:40,  2.06it/s]
Training 1/1 epoch (loss 1.3847):  78%|███████▊  | 732/938 [05:07<01:34,  2.18it/s]
Training 1/1 epoch (loss 1.4539):  78%|███████▊  | 732/938 [05:07<01:34,  2.18it/s]
Training 1/1 epoch (loss 1.4539):  78%|███████▊  | 733/938 [05:07<01:29,  2.30it/s]
Training 1/1 epoch (loss 1.4272):  78%|███████▊  | 733/938 [05:07<01:29,  2.30it/s]
Training 1/1 epoch (loss 1.4272):  78%|███████▊  | 734/938 [05:07<01:24,  2.40it/s]
Training 1/1 epoch (loss 1.4538):  78%|███████▊  | 734/938 [05:08<01:24,  2.40it/s]
Training 1/1 epoch (loss 1.4538):  78%|███████▊  | 735/938 [05:08<01:23,  2.44it/s]
Training 1/1 epoch (loss 1.3981):  78%|███████▊  | 735/938 [05:08<01:23,  2.44it/s]
Training 1/1 epoch (loss 1.3981):  78%|███████▊  | 736/938 [05:08<01:23,  2.41it/s]
Training 1/1 epoch (loss 1.3895):  78%|███████▊  | 736/938 [05:09<01:23,  2.41it/s]
Training 1/1 epoch (loss 1.3895):  79%|███████▊  | 737/938 [05:09<01:23,  2.42it/s]
Training 1/1 epoch (loss 1.3480):  79%|███████▊  | 737/938 [05:09<01:23,  2.42it/s]
Training 1/1 epoch (loss 1.3480):  79%|███████▊  | 738/938 [05:09<01:19,  2.51it/s]
Training 1/1 epoch (loss 1.3297):  79%|███████▊  | 738/938 [05:09<01:19,  2.51it/s]
Training 1/1 epoch (loss 1.3297):  79%|███████▉  | 739/938 [05:09<01:18,  2.54it/s]
Training 1/1 epoch (loss 1.4349):  79%|███████▉  | 739/938 [05:10<01:18,  2.54it/s]
Training 1/1 epoch (loss 1.4349):  79%|███████▉  | 740/938 [05:10<01:18,  2.53it/s]
Training 1/1 epoch (loss 1.4453):  79%|███████▉  | 740/938 [05:10<01:18,  2.53it/s]
Training 1/1 epoch (loss 1.4453):  79%|███████▉  | 741/938 [05:10<01:16,  2.58it/s]
Training 1/1 epoch (loss 1.4072):  79%|███████▉  | 741/938 [05:10<01:16,  2.58it/s]
Training 1/1 epoch (loss 1.4072):  79%|███████▉  | 742/938 [05:10<01:15,  2.60it/s]
Training 1/1 epoch (loss 1.4113):  79%|███████▉  | 742/938 [05:11<01:15,  2.60it/s]
Training 1/1 epoch (loss 1.4113):  79%|███████▉  | 743/938 [05:11<01:14,  2.60it/s]
Training 1/1 epoch (loss 1.3668):  79%|███████▉  | 743/938 [05:11<01:14,  2.60it/s]
Training 1/1 epoch (loss 1.3668):  79%|███████▉  | 744/938 [05:11<01:19,  2.45it/s]
Training 1/1 epoch (loss 1.4664):  79%|███████▉  | 744/938 [05:12<01:19,  2.45it/s]
Training 1/1 epoch (loss 1.4664):  79%|███████▉  | 745/938 [05:12<01:17,  2.49it/s]
Training 1/1 epoch (loss 1.3700):  79%|███████▉  | 745/938 [05:12<01:17,  2.49it/s]
Training 1/1 epoch (loss 1.3700):  80%|███████▉  | 746/938 [05:12<01:15,  2.55it/s]
Training 1/1 epoch (loss 1.4895):  80%|███████▉  | 746/938 [05:13<01:15,  2.55it/s]
Training 1/1 epoch (loss 1.4895):  80%|███████▉  | 747/938 [05:13<01:20,  2.37it/s]
Training 1/1 epoch (loss 1.3520):  80%|███████▉  | 747/938 [05:13<01:20,  2.37it/s]
Training 1/1 epoch (loss 1.3520):  80%|███████▉  | 748/938 [05:13<01:17,  2.46it/s]
Training 1/1 epoch (loss 1.3584):  80%|███████▉  | 748/938 [05:13<01:17,  2.46it/s]
Training 1/1 epoch (loss 1.3584):  80%|███████▉  | 749/938 [05:13<01:17,  2.43it/s]
Training 1/1 epoch (loss 1.3440):  80%|███████▉  | 749/938 [05:14<01:17,  2.43it/s]
Training 1/1 epoch (loss 1.3440):  80%|███████▉  | 750/938 [05:14<01:16,  2.47it/s]
Training 1/1 epoch (loss 1.3620):  80%|███████▉  | 750/938 [05:14<01:16,  2.47it/s]
Training 1/1 epoch (loss 1.3620):  80%|████████  | 751/938 [05:14<01:14,  2.51it/s]
Training 1/1 epoch (loss 1.3784):  80%|████████  | 751/938 [05:15<01:14,  2.51it/s]
Training 1/1 epoch (loss 1.3784):  80%|████████  | 752/938 [05:15<01:17,  2.39it/s]
Training 1/1 epoch (loss 1.4466):  80%|████████  | 752/938 [05:15<01:17,  2.39it/s]
Training 1/1 epoch (loss 1.4466):  80%|████████  | 753/938 [05:15<01:15,  2.46it/s]
Training 1/1 epoch (loss 1.4629):  80%|████████  | 753/938 [05:15<01:15,  2.46it/s]
Training 1/1 epoch (loss 1.4629):  80%|████████  | 754/938 [05:15<01:14,  2.47it/s]
Training 1/1 epoch (loss 1.4066):  80%|████████  | 754/938 [05:16<01:14,  2.47it/s]
Training 1/1 epoch (loss 1.4066):  80%|████████  | 755/938 [05:16<01:13,  2.50it/s]
Training 1/1 epoch (loss 1.4283):  80%|████████  | 755/938 [05:16<01:13,  2.50it/s]
Training 1/1 epoch (loss 1.4283):  81%|████████  | 756/938 [05:16<01:14,  2.46it/s]
Training 1/1 epoch (loss 1.3993):  81%|████████  | 756/938 [05:17<01:14,  2.46it/s]
Training 1/1 epoch (loss 1.3993):  81%|████████  | 757/938 [05:17<01:14,  2.42it/s]
Training 1/1 epoch (loss 1.5349):  81%|████████  | 757/938 [05:17<01:14,  2.42it/s]
Training 1/1 epoch (loss 1.5349):  81%|████████  | 758/938 [05:17<01:11,  2.51it/s]
Training 1/1 epoch (loss 1.3726):  81%|████████  | 758/938 [05:17<01:11,  2.51it/s]
Training 1/1 epoch (loss 1.3726):  81%|████████  | 759/938 [05:17<01:11,  2.52it/s]
Training 1/1 epoch (loss 1.2733):  81%|████████  | 759/938 [05:18<01:11,  2.52it/s]
Training 1/1 epoch (loss 1.2733):  81%|████████  | 760/938 [05:18<01:17,  2.30it/s]
Training 1/1 epoch (loss 1.5541):  81%|████████  | 760/938 [05:18<01:17,  2.30it/s]
Training 1/1 epoch (loss 1.5541):  81%|████████  | 761/938 [05:18<01:13,  2.41it/s]
Training 1/1 epoch (loss 1.3840):  81%|████████  | 761/938 [05:19<01:13,  2.41it/s]
Training 1/1 epoch (loss 1.3840):  81%|████████  | 762/938 [05:19<01:10,  2.50it/s]
Training 1/1 epoch (loss 1.5095):  81%|████████  | 762/938 [05:19<01:10,  2.50it/s]
Training 1/1 epoch (loss 1.5095):  81%|████████▏ | 763/938 [05:19<01:11,  2.46it/s]
Training 1/1 epoch (loss 1.2816):  81%|████████▏ | 763/938 [05:19<01:11,  2.46it/s]
Training 1/1 epoch (loss 1.2816):  81%|████████▏ | 764/938 [05:19<01:10,  2.46it/s]
Training 1/1 epoch (loss 1.4703):  81%|████████▏ | 764/938 [05:20<01:10,  2.46it/s]
Training 1/1 epoch (loss 1.4703):  82%|████████▏ | 765/938 [05:20<01:08,  2.53it/s]
Training 1/1 epoch (loss 1.3712):  82%|████████▏ | 765/938 [05:20<01:08,  2.53it/s]
Training 1/1 epoch (loss 1.3712):  82%|████████▏ | 766/938 [05:20<01:08,  2.51it/s]
Training 1/1 epoch (loss 1.4840):  82%|████████▏ | 766/938 [05:21<01:08,  2.51it/s]
Training 1/1 epoch (loss 1.4840):  82%|████████▏ | 767/938 [05:21<01:07,  2.52it/s]
Training 1/1 epoch (loss 1.3549):  82%|████████▏ | 767/938 [05:21<01:07,  2.52it/s]
Training 1/1 epoch (loss 1.3549):  82%|████████▏ | 768/938 [05:21<01:05,  2.58it/s]
Training 1/1 epoch (loss 1.5747):  82%|████████▏ | 768/938 [05:21<01:05,  2.58it/s]
Training 1/1 epoch (loss 1.5747):  82%|████████▏ | 769/938 [05:21<01:06,  2.54it/s]
Training 1/1 epoch (loss 1.3923):  82%|████████▏ | 769/938 [05:22<01:06,  2.54it/s]
Training 1/1 epoch (loss 1.3923):  82%|████████▏ | 770/938 [05:22<01:05,  2.56it/s]
Training 1/1 epoch (loss 1.3280):  82%|████████▏ | 770/938 [05:22<01:05,  2.56it/s]
Training 1/1 epoch (loss 1.3280):  82%|████████▏ | 771/938 [05:22<01:03,  2.62it/s]
Training 1/1 epoch (loss 1.3406):  82%|████████▏ | 771/938 [05:23<01:03,  2.62it/s]
Training 1/1 epoch (loss 1.3406):  82%|████████▏ | 772/938 [05:23<01:04,  2.55it/s]
Training 1/1 epoch (loss 1.3483):  82%|████████▏ | 772/938 [05:23<01:04,  2.55it/s]
Training 1/1 epoch (loss 1.3483):  82%|████████▏ | 773/938 [05:23<01:03,  2.61it/s]
Training 1/1 epoch (loss 1.2872):  82%|████████▏ | 773/938 [05:23<01:03,  2.61it/s]
Training 1/1 epoch (loss 1.2872):  83%|████████▎ | 774/938 [05:23<01:03,  2.57it/s]
Training 1/1 epoch (loss 1.3299):  83%|████████▎ | 774/938 [05:24<01:03,  2.57it/s]
Training 1/1 epoch (loss 1.3299):  83%|████████▎ | 775/938 [05:24<01:03,  2.58it/s]
Training 1/1 epoch (loss 1.4621):  83%|████████▎ | 775/938 [05:24<01:03,  2.58it/s]
Training 1/1 epoch (loss 1.4621):  83%|████████▎ | 776/938 [05:24<01:03,  2.55it/s]
Training 1/1 epoch (loss 1.4238):  83%|████████▎ | 776/938 [05:24<01:03,  2.55it/s]
Training 1/1 epoch (loss 1.4238):  83%|████████▎ | 777/938 [05:24<01:03,  2.54it/s]
Training 1/1 epoch (loss 1.3755):  83%|████████▎ | 777/938 [05:25<01:03,  2.54it/s]
Training 1/1 epoch (loss 1.3755):  83%|████████▎ | 778/938 [05:25<01:04,  2.46it/s]
Training 1/1 epoch (loss 1.3005):  83%|████████▎ | 778/938 [05:25<01:04,  2.46it/s]
Training 1/1 epoch (loss 1.3005):  83%|████████▎ | 779/938 [05:25<01:05,  2.44it/s]
Training 1/1 epoch (loss 1.4695):  83%|████████▎ | 779/938 [05:26<01:05,  2.44it/s]
Training 1/1 epoch (loss 1.4695):  83%|████████▎ | 780/938 [05:26<01:04,  2.47it/s]
Training 1/1 epoch (loss 1.4693):  83%|████████▎ | 780/938 [05:26<01:04,  2.47it/s]
Training 1/1 epoch (loss 1.4693):  83%|████████▎ | 781/938 [05:26<01:02,  2.53it/s]
Training 1/1 epoch (loss 1.3386):  83%|████████▎ | 781/938 [05:27<01:02,  2.53it/s]
Training 1/1 epoch (loss 1.3386):  83%|████████▎ | 782/938 [05:27<01:03,  2.47it/s]
Training 1/1 epoch (loss 1.4340):  83%|████████▎ | 782/938 [05:27<01:03,  2.47it/s]
Training 1/1 epoch (loss 1.4340):  83%|████████▎ | 783/938 [05:27<01:01,  2.51it/s]
Training 1/1 epoch (loss 1.3781):  83%|████████▎ | 783/938 [05:27<01:01,  2.51it/s]
Training 1/1 epoch (loss 1.3781):  84%|████████▎ | 784/938 [05:27<01:01,  2.50it/s]
Training 1/1 epoch (loss 1.4109):  84%|████████▎ | 784/938 [05:28<01:01,  2.50it/s]
Training 1/1 epoch (loss 1.4109):  84%|████████▎ | 785/938 [05:28<01:00,  2.54it/s]
Training 1/1 epoch (loss 1.4292):  84%|████████▎ | 785/938 [05:28<01:00,  2.54it/s]
Training 1/1 epoch (loss 1.4292):  84%|████████▍ | 786/938 [05:28<00:59,  2.54it/s]
Training 1/1 epoch (loss 1.3912):  84%|████████▍ | 786/938 [05:28<00:59,  2.54it/s]
Training 1/1 epoch (loss 1.3912):  84%|████████▍ | 787/938 [05:28<00:57,  2.63it/s]
Training 1/1 epoch (loss 1.5159):  84%|████████▍ | 787/938 [05:29<00:57,  2.63it/s]
Training 1/1 epoch (loss 1.5159):  84%|████████▍ | 788/938 [05:29<00:57,  2.63it/s]
Training 1/1 epoch (loss 1.2799):  84%|████████▍ | 788/938 [05:29<00:57,  2.63it/s]
Training 1/1 epoch (loss 1.2799):  84%|████████▍ | 789/938 [05:29<00:58,  2.53it/s]
Training 1/1 epoch (loss 1.2874):  84%|████████▍ | 789/938 [05:30<00:58,  2.53it/s]
Training 1/1 epoch (loss 1.2874):  84%|████████▍ | 790/938 [05:30<00:58,  2.53it/s]
Training 1/1 epoch (loss 1.4524):  84%|████████▍ | 790/938 [05:30<00:58,  2.53it/s]
Training 1/1 epoch (loss 1.4524):  84%|████████▍ | 791/938 [05:30<00:58,  2.52it/s]
Training 1/1 epoch (loss 1.3427):  84%|████████▍ | 791/938 [05:30<00:58,  2.52it/s]
Training 1/1 epoch (loss 1.3427):  84%|████████▍ | 792/938 [05:30<00:57,  2.52it/s]
Training 1/1 epoch (loss 1.4570):  84%|████████▍ | 792/938 [05:31<00:57,  2.52it/s]
Training 1/1 epoch (loss 1.4570):  85%|████████▍ | 793/938 [05:31<01:00,  2.39it/s]
Training 1/1 epoch (loss 1.4562):  85%|████████▍ | 793/938 [05:31<01:00,  2.39it/s]
Training 1/1 epoch (loss 1.4562):  85%|████████▍ | 794/938 [05:31<00:58,  2.44it/s]
Training 1/1 epoch (loss 1.3881):  85%|████████▍ | 794/938 [05:32<00:58,  2.44it/s]
Training 1/1 epoch (loss 1.3881):  85%|████████▍ | 795/938 [05:32<00:58,  2.45it/s]
Training 1/1 epoch (loss 1.3780):  85%|████████▍ | 795/938 [05:32<00:58,  2.45it/s]
Training 1/1 epoch (loss 1.3780):  85%|████████▍ | 796/938 [05:32<01:00,  2.36it/s]
Training 1/1 epoch (loss 1.3430):  85%|████████▍ | 796/938 [05:33<01:00,  2.36it/s]
Training 1/1 epoch (loss 1.3430):  85%|████████▍ | 797/938 [05:33<00:57,  2.44it/s]
Training 1/1 epoch (loss 1.4737):  85%|████████▍ | 797/938 [05:33<00:57,  2.44it/s]
Training 1/1 epoch (loss 1.4737):  85%|████████▌ | 798/938 [05:33<00:57,  2.42it/s]
Training 1/1 epoch (loss 1.4330):  85%|████████▌ | 798/938 [05:33<00:57,  2.42it/s]
Training 1/1 epoch (loss 1.4330):  85%|████████▌ | 799/938 [05:33<00:57,  2.42it/s]
Training 1/1 epoch (loss 1.3727):  85%|████████▌ | 799/938 [05:34<00:57,  2.42it/s]
Training 1/1 epoch (loss 1.3727):  85%|████████▌ | 800/938 [05:34<00:58,  2.35it/s]
Training 1/1 epoch (loss 1.4603):  85%|████████▌ | 800/938 [05:34<00:58,  2.35it/s]
Training 1/1 epoch (loss 1.4603):  85%|████████▌ | 801/938 [05:34<00:58,  2.35it/s]
Training 1/1 epoch (loss 1.4940):  85%|████████▌ | 801/938 [05:35<00:58,  2.35it/s]
Training 1/1 epoch (loss 1.4940):  86%|████████▌ | 802/938 [05:35<01:01,  2.21it/s]
Training 1/1 epoch (loss 1.4308):  86%|████████▌ | 802/938 [05:35<01:01,  2.21it/s]
Training 1/1 epoch (loss 1.4308):  86%|████████▌ | 803/938 [05:35<01:03,  2.13it/s]
Training 1/1 epoch (loss 1.3787):  86%|████████▌ | 803/938 [05:36<01:03,  2.13it/s]
Training 1/1 epoch (loss 1.3787):  86%|████████▌ | 804/938 [05:36<01:06,  2.01it/s]
Training 1/1 epoch (loss 1.3808):  86%|████████▌ | 804/938 [05:36<01:06,  2.01it/s]
Training 1/1 epoch (loss 1.3808):  86%|████████▌ | 805/938 [05:36<01:02,  2.14it/s]
Training 1/1 epoch (loss 1.3776):  86%|████████▌ | 805/938 [05:37<01:02,  2.14it/s]
Training 1/1 epoch (loss 1.3776):  86%|████████▌ | 806/938 [05:37<01:00,  2.18it/s]
Training 1/1 epoch (loss 1.3633):  86%|████████▌ | 806/938 [05:37<01:00,  2.18it/s]
Training 1/1 epoch (loss 1.3633):  86%|████████▌ | 807/938 [05:37<00:57,  2.28it/s]
Training 1/1 epoch (loss 1.4126):  86%|████████▌ | 807/938 [05:37<00:57,  2.28it/s]
Training 1/1 epoch (loss 1.4126):  86%|████████▌ | 808/938 [05:37<00:55,  2.35it/s]
Training 1/1 epoch (loss 1.3316):  86%|████████▌ | 808/938 [05:38<00:55,  2.35it/s]
Training 1/1 epoch (loss 1.3316):  86%|████████▌ | 809/938 [05:38<00:52,  2.44it/s]
Training 1/1 epoch (loss 1.4222):  86%|████████▌ | 809/938 [05:38<00:52,  2.44it/s]
Training 1/1 epoch (loss 1.4222):  86%|████████▋ | 810/938 [05:38<00:52,  2.45it/s]
Training 1/1 epoch (loss 1.2999):  86%|████████▋ | 810/938 [05:39<00:52,  2.45it/s]
Training 1/1 epoch (loss 1.2999):  86%|████████▋ | 811/938 [05:39<00:50,  2.51it/s]
Training 1/1 epoch (loss 1.3898):  86%|████████▋ | 811/938 [05:39<00:50,  2.51it/s]
Training 1/1 epoch (loss 1.3898):  87%|████████▋ | 812/938 [05:39<00:49,  2.56it/s]
Training 1/1 epoch (loss 1.4494):  87%|████████▋ | 812/938 [05:39<00:49,  2.56it/s]
Training 1/1 epoch (loss 1.4494):  87%|████████▋ | 813/938 [05:39<00:49,  2.55it/s]
Training 1/1 epoch (loss 1.3817):  87%|████████▋ | 813/938 [05:40<00:49,  2.55it/s]
Training 1/1 epoch (loss 1.3817):  87%|████████▋ | 814/938 [05:40<00:47,  2.61it/s]
Training 1/1 epoch (loss 1.3954):  87%|████████▋ | 814/938 [05:40<00:47,  2.61it/s]
Training 1/1 epoch (loss 1.3954):  87%|████████▋ | 815/938 [05:40<00:48,  2.53it/s]
Training 1/1 epoch (loss 1.4282):  87%|████████▋ | 815/938 [05:41<00:48,  2.53it/s]
Training 1/1 epoch (loss 1.4282):  87%|████████▋ | 816/938 [05:41<00:48,  2.52it/s]
Training 1/1 epoch (loss 1.4028):  87%|████████▋ | 816/938 [05:41<00:48,  2.52it/s]
Training 1/1 epoch (loss 1.4028):  87%|████████▋ | 817/938 [05:41<00:46,  2.61it/s]
Training 1/1 epoch (loss 1.3436):  87%|████████▋ | 817/938 [05:41<00:46,  2.61it/s]
Training 1/1 epoch (loss 1.3436):  87%|████████▋ | 818/938 [05:41<00:45,  2.64it/s]
Training 1/1 epoch (loss 1.5247):  87%|████████▋ | 818/938 [05:42<00:45,  2.64it/s]
Training 1/1 epoch (loss 1.5247):  87%|████████▋ | 819/938 [05:42<00:45,  2.62it/s]
Training 1/1 epoch (loss 1.3505):  87%|████████▋ | 819/938 [05:42<00:45,  2.62it/s]
Training 1/1 epoch (loss 1.3505):  87%|████████▋ | 820/938 [05:42<00:44,  2.65it/s]
Training 1/1 epoch (loss 1.3999):  87%|████████▋ | 820/938 [05:43<00:44,  2.65it/s]
Training 1/1 epoch (loss 1.3999):  88%|████████▊ | 821/938 [05:43<00:47,  2.46it/s]
Training 1/1 epoch (loss 1.5262):  88%|████████▊ | 821/938 [05:43<00:47,  2.46it/s]
Training 1/1 epoch (loss 1.5262):  88%|████████▊ | 822/938 [05:43<00:46,  2.49it/s]
Training 1/1 epoch (loss 1.3645):  88%|████████▊ | 822/938 [05:43<00:46,  2.49it/s]
Training 1/1 epoch (loss 1.3645):  88%|████████▊ | 823/938 [05:43<00:45,  2.55it/s]
Training 1/1 epoch (loss 1.3908):  88%|████████▊ | 823/938 [05:44<00:45,  2.55it/s]
Training 1/1 epoch (loss 1.3908):  88%|████████▊ | 824/938 [05:44<00:45,  2.52it/s]
Training 1/1 epoch (loss 1.3037):  88%|████████▊ | 824/938 [05:44<00:45,  2.52it/s]
Training 1/1 epoch (loss 1.3037):  88%|████████▊ | 825/938 [05:44<00:44,  2.56it/s]
Training 1/1 epoch (loss 1.4308):  88%|████████▊ | 825/938 [05:45<00:44,  2.56it/s]
Training 1/1 epoch (loss 1.4308):  88%|████████▊ | 826/938 [05:45<00:45,  2.48it/s]
Training 1/1 epoch (loss 1.4077):  88%|████████▊ | 826/938 [05:45<00:45,  2.48it/s]
Training 1/1 epoch (loss 1.4077):  88%|████████▊ | 827/938 [05:45<00:43,  2.56it/s]
Training 1/1 epoch (loss 1.5200):  88%|████████▊ | 827/938 [05:45<00:43,  2.56it/s]
Training 1/1 epoch (loss 1.5200):  88%|████████▊ | 828/938 [05:45<00:42,  2.57it/s]
Training 1/1 epoch (loss 1.3508):  88%|████████▊ | 828/938 [05:46<00:42,  2.57it/s]
Training 1/1 epoch (loss 1.3508):  88%|████████▊ | 829/938 [05:46<00:42,  2.58it/s]
Training 1/1 epoch (loss 1.4548):  88%|████████▊ | 829/938 [05:46<00:42,  2.58it/s]
Training 1/1 epoch (loss 1.4548):  88%|████████▊ | 830/938 [05:46<00:41,  2.58it/s]
Training 1/1 epoch (loss 1.5039):  88%|████████▊ | 830/938 [05:46<00:41,  2.58it/s]
Training 1/1 epoch (loss 1.5039):  89%|████████▊ | 831/938 [05:46<00:41,  2.59it/s]
Training 1/1 epoch (loss 1.4359):  89%|████████▊ | 831/938 [05:47<00:41,  2.59it/s]
Training 1/1 epoch (loss 1.4359):  89%|████████▊ | 832/938 [05:47<00:42,  2.51it/s]
Training 1/1 epoch (loss 1.4135):  89%|████████▊ | 832/938 [05:47<00:42,  2.51it/s]
Training 1/1 epoch (loss 1.4135):  89%|████████▉ | 833/938 [05:47<00:41,  2.54it/s]
Training 1/1 epoch (loss 1.3888):  89%|████████▉ | 833/938 [05:48<00:41,  2.54it/s]
Training 1/1 epoch (loss 1.3888):  89%|████████▉ | 834/938 [05:48<00:43,  2.40it/s]
Training 1/1 epoch (loss 1.3691):  89%|████████▉ | 834/938 [05:48<00:43,  2.40it/s]
Training 1/1 epoch (loss 1.3691):  89%|████████▉ | 835/938 [05:48<00:42,  2.44it/s]
Training 1/1 epoch (loss 1.3504):  89%|████████▉ | 835/938 [05:48<00:42,  2.44it/s]
Training 1/1 epoch (loss 1.3504):  89%|████████▉ | 836/938 [05:48<00:40,  2.49it/s]
Training 1/1 epoch (loss 1.4833):  89%|████████▉ | 836/938 [05:49<00:40,  2.49it/s]
Training 1/1 epoch (loss 1.4833):  89%|████████▉ | 837/938 [05:49<00:40,  2.48it/s]
Training 1/1 epoch (loss 1.4190):  89%|████████▉ | 837/938 [05:49<00:40,  2.48it/s]
Training 1/1 epoch (loss 1.4190):  89%|████████▉ | 838/938 [05:49<00:39,  2.55it/s]
Training 1/1 epoch (loss 1.3475):  89%|████████▉ | 838/938 [05:50<00:39,  2.55it/s]
Training 1/1 epoch (loss 1.3475):  89%|████████▉ | 839/938 [05:50<00:38,  2.60it/s]
Training 1/1 epoch (loss 1.3313):  89%|████████▉ | 839/938 [05:50<00:38,  2.60it/s]
Training 1/1 epoch (loss 1.3313):  90%|████████▉ | 840/938 [05:50<00:38,  2.58it/s]
Training 1/1 epoch (loss 1.4595):  90%|████████▉ | 840/938 [05:50<00:38,  2.58it/s]
Training 1/1 epoch (loss 1.4595):  90%|████████▉ | 841/938 [05:50<00:37,  2.56it/s]
Training 1/1 epoch (loss 1.3738):  90%|████████▉ | 841/938 [05:51<00:37,  2.56it/s]
Training 1/1 epoch (loss 1.3738):  90%|████████▉ | 842/938 [05:51<00:39,  2.45it/s]
Training 1/1 epoch (loss 1.5037):  90%|████████▉ | 842/938 [05:51<00:39,  2.45it/s]
Training 1/1 epoch (loss 1.5037):  90%|████████▉ | 843/938 [05:51<00:36,  2.57it/s]
Training 1/1 epoch (loss 1.3260):  90%|████████▉ | 843/938 [05:52<00:36,  2.57it/s]
Training 1/1 epoch (loss 1.3260):  90%|████████▉ | 844/938 [05:52<00:36,  2.56it/s]
Training 1/1 epoch (loss 1.4462):  90%|████████▉ | 844/938 [05:52<00:36,  2.56it/s]
Training 1/1 epoch (loss 1.4462):  90%|█████████ | 845/938 [05:52<00:36,  2.57it/s]
Training 1/1 epoch (loss 1.5137):  90%|█████████ | 845/938 [05:52<00:36,  2.57it/s]
Training 1/1 epoch (loss 1.5137):  90%|█████████ | 846/938 [05:52<00:35,  2.57it/s]
Training 1/1 epoch (loss 1.3882):  90%|█████████ | 846/938 [05:53<00:35,  2.57it/s]
Training 1/1 epoch (loss 1.3882):  90%|█████████ | 847/938 [05:53<00:36,  2.49it/s]
Training 1/1 epoch (loss 1.3431):  90%|█████████ | 847/938 [05:53<00:36,  2.49it/s]
Training 1/1 epoch (loss 1.3431):  90%|█████████ | 848/938 [05:53<00:36,  2.46it/s]
Training 1/1 epoch (loss 1.4140):  90%|█████████ | 848/938 [05:54<00:36,  2.46it/s]
Training 1/1 epoch (loss 1.4140):  91%|█████████ | 849/938 [05:54<00:35,  2.48it/s]
Training 1/1 epoch (loss 1.3167):  91%|█████████ | 849/938 [05:54<00:35,  2.48it/s]
Training 1/1 epoch (loss 1.3167):  91%|█████████ | 850/938 [05:54<00:36,  2.42it/s]
Training 1/1 epoch (loss 1.5012):  91%|█████████ | 850/938 [05:54<00:36,  2.42it/s]
Training 1/1 epoch (loss 1.5012):  91%|█████████ | 851/938 [05:54<00:35,  2.46it/s]
Training 1/1 epoch (loss 1.4618):  91%|█████████ | 851/938 [05:55<00:35,  2.46it/s]
Training 1/1 epoch (loss 1.4618):  91%|█████████ | 852/938 [05:55<00:34,  2.49it/s]
Training 1/1 epoch (loss 1.4353):  91%|█████████ | 852/938 [05:55<00:34,  2.49it/s]
Training 1/1 epoch (loss 1.4353):  91%|█████████ | 853/938 [05:55<00:35,  2.40it/s]
Training 1/1 epoch (loss 1.4079):  91%|█████████ | 853/938 [05:56<00:35,  2.40it/s]
Training 1/1 epoch (loss 1.4079):  91%|█████████ | 854/938 [05:56<00:34,  2.47it/s]
Training 1/1 epoch (loss 1.4603):  91%|█████████ | 854/938 [05:56<00:34,  2.47it/s]
Training 1/1 epoch (loss 1.4603):  91%|█████████ | 855/938 [05:56<00:33,  2.48it/s]
Training 1/1 epoch (loss 1.4194):  91%|█████████ | 855/938 [05:56<00:33,  2.48it/s]
Training 1/1 epoch (loss 1.4194):  91%|█████████▏| 856/938 [05:56<00:32,  2.50it/s]
Training 1/1 epoch (loss 1.4601):  91%|█████████▏| 856/938 [05:57<00:32,  2.50it/s]
Training 1/1 epoch (loss 1.4601):  91%|█████████▏| 857/938 [05:57<00:32,  2.51it/s]
Training 1/1 epoch (loss 1.3853):  91%|█████████▏| 857/938 [05:57<00:32,  2.51it/s]
Training 1/1 epoch (loss 1.3853):  91%|█████████▏| 858/938 [05:57<00:31,  2.54it/s]
Training 1/1 epoch (loss 1.4101):  91%|█████████▏| 858/938 [05:58<00:31,  2.54it/s]
Training 1/1 epoch (loss 1.4101):  92%|█████████▏| 859/938 [05:58<00:30,  2.59it/s]
Training 1/1 epoch (loss 1.3287):  92%|█████████▏| 859/938 [05:58<00:30,  2.59it/s]
Training 1/1 epoch (loss 1.3287):  92%|█████████▏| 860/938 [05:58<00:29,  2.62it/s]
Training 1/1 epoch (loss 1.4031):  92%|█████████▏| 860/938 [05:58<00:29,  2.62it/s]
Training 1/1 epoch (loss 1.4031):  92%|█████████▏| 861/938 [05:58<00:30,  2.52it/s]
Training 1/1 epoch (loss 1.4975):  92%|█████████▏| 861/938 [05:59<00:30,  2.52it/s]
Training 1/1 epoch (loss 1.4975):  92%|█████████▏| 862/938 [05:59<00:29,  2.55it/s]
Training 1/1 epoch (loss 1.5202):  92%|█████████▏| 862/938 [05:59<00:29,  2.55it/s]
Training 1/1 epoch (loss 1.5202):  92%|█████████▏| 863/938 [05:59<00:29,  2.53it/s]
Training 1/1 epoch (loss 1.3898):  92%|█████████▏| 863/938 [06:00<00:29,  2.53it/s]
Training 1/1 epoch (loss 1.3898):  92%|█████████▏| 864/938 [06:00<00:31,  2.34it/s]
Training 1/1 epoch (loss 1.4608):  92%|█████████▏| 864/938 [06:00<00:31,  2.34it/s]
Training 1/1 epoch (loss 1.4608):  92%|█████████▏| 865/938 [06:00<00:31,  2.35it/s]
Training 1/1 epoch (loss 1.2776):  92%|█████████▏| 865/938 [06:00<00:31,  2.35it/s]
Training 1/1 epoch (loss 1.2776):  92%|█████████▏| 866/938 [06:00<00:29,  2.40it/s]
Training 1/1 epoch (loss 1.2095):  92%|█████████▏| 866/938 [06:01<00:29,  2.40it/s]
Training 1/1 epoch (loss 1.2095):  92%|█████████▏| 867/938 [06:01<00:29,  2.43it/s]
Training 1/1 epoch (loss 1.3537):  92%|█████████▏| 867/938 [06:01<00:29,  2.43it/s]
Training 1/1 epoch (loss 1.3537):  93%|█████████▎| 868/938 [06:01<00:28,  2.48it/s]
Training 1/1 epoch (loss 1.3781):  93%|█████████▎| 868/938 [06:02<00:28,  2.48it/s]
Training 1/1 epoch (loss 1.3781):  93%|█████████▎| 869/938 [06:02<00:27,  2.49it/s]
Training 1/1 epoch (loss 1.4681):  93%|█████████▎| 869/938 [06:02<00:27,  2.49it/s]
Training 1/1 epoch (loss 1.4681):  93%|█████████▎| 870/938 [06:02<00:26,  2.54it/s]
Training 1/1 epoch (loss 1.3456):  93%|█████████▎| 870/938 [06:02<00:26,  2.54it/s]
Training 1/1 epoch (loss 1.3456):  93%|█████████▎| 871/938 [06:02<00:26,  2.55it/s]
Training 1/1 epoch (loss 1.3910):  93%|█████████▎| 871/938 [06:03<00:26,  2.55it/s]
Training 1/1 epoch (loss 1.3910):  93%|█████████▎| 872/938 [06:03<00:27,  2.44it/s]
Training 1/1 epoch (loss 1.5047):  93%|█████████▎| 872/938 [06:03<00:27,  2.44it/s]
Training 1/1 epoch (loss 1.5047):  93%|█████████▎| 873/938 [06:03<00:26,  2.48it/s]
Training 1/1 epoch (loss 1.4312):  93%|█████████▎| 873/938 [06:04<00:26,  2.48it/s]
Training 1/1 epoch (loss 1.4312):  93%|█████████▎| 874/938 [06:04<00:26,  2.40it/s]
Training 1/1 epoch (loss 1.4616):  93%|█████████▎| 874/938 [06:04<00:26,  2.40it/s]
Training 1/1 epoch (loss 1.4616):  93%|█████████▎| 875/938 [06:04<00:26,  2.41it/s]
Training 1/1 epoch (loss 1.3844):  93%|█████████▎| 875/938 [06:05<00:26,  2.41it/s]
Training 1/1 epoch (loss 1.3844):  93%|█████████▎| 876/938 [06:05<00:25,  2.39it/s]
Training 1/1 epoch (loss 1.3567):  93%|█████████▎| 876/938 [06:05<00:25,  2.39it/s]
Training 1/1 epoch (loss 1.3567):  93%|█████████▎| 877/938 [06:05<00:27,  2.22it/s]
Training 1/1 epoch (loss 1.3621):  93%|█████████▎| 877/938 [06:06<00:27,  2.22it/s]
Training 1/1 epoch (loss 1.3621):  94%|█████████▎| 878/938 [06:06<00:30,  1.97it/s]
Training 1/1 epoch (loss 1.3306):  94%|█████████▎| 878/938 [06:06<00:30,  1.97it/s]
Training 1/1 epoch (loss 1.3306):  94%|█████████▎| 879/938 [06:06<00:28,  2.07it/s]
Training 1/1 epoch (loss 1.5588):  94%|█████████▎| 879/938 [06:07<00:28,  2.07it/s]
Training 1/1 epoch (loss 1.5588):  94%|█████████▍| 880/938 [06:07<00:26,  2.20it/s]
Training 1/1 epoch (loss 1.4320):  94%|█████████▍| 880/938 [06:07<00:26,  2.20it/s]
Training 1/1 epoch (loss 1.4320):  94%|█████████▍| 881/938 [06:07<00:25,  2.27it/s]
Training 1/1 epoch (loss 1.4221):  94%|█████████▍| 881/938 [06:07<00:25,  2.27it/s]
Training 1/1 epoch (loss 1.4221):  94%|█████████▍| 882/938 [06:07<00:24,  2.32it/s]
Training 1/1 epoch (loss 1.4939):  94%|█████████▍| 882/938 [06:08<00:24,  2.32it/s]
Training 1/1 epoch (loss 1.4939):  94%|█████████▍| 883/938 [06:08<00:23,  2.36it/s]
Training 1/1 epoch (loss 1.4519):  94%|█████████▍| 883/938 [06:08<00:23,  2.36it/s]
Training 1/1 epoch (loss 1.4519):  94%|█████████▍| 884/938 [06:08<00:22,  2.40it/s]
Training 1/1 epoch (loss 1.4336):  94%|█████████▍| 884/938 [06:09<00:22,  2.40it/s]
Training 1/1 epoch (loss 1.4336):  94%|█████████▍| 885/938 [06:09<00:21,  2.47it/s]
Training 1/1 epoch (loss 1.3369):  94%|█████████▍| 885/938 [06:09<00:21,  2.47it/s]
Training 1/1 epoch (loss 1.3369):  94%|█████████▍| 886/938 [06:09<00:21,  2.47it/s]
Training 1/1 epoch (loss 1.4194):  94%|█████████▍| 886/938 [06:09<00:21,  2.47it/s]
Training 1/1 epoch (loss 1.4194):  95%|█████████▍| 887/938 [06:09<00:20,  2.54it/s]
Training 1/1 epoch (loss 1.3670):  95%|█████████▍| 887/938 [06:10<00:20,  2.54it/s]
Training 1/1 epoch (loss 1.3670):  95%|█████████▍| 888/938 [06:10<00:19,  2.55it/s]
Training 1/1 epoch (loss 1.4161):  95%|█████████▍| 888/938 [06:10<00:19,  2.55it/s]
Training 1/1 epoch (loss 1.4161):  95%|█████████▍| 889/938 [06:10<00:19,  2.58it/s]
Training 1/1 epoch (loss 1.2729):  95%|█████████▍| 889/938 [06:10<00:19,  2.58it/s]
Training 1/1 epoch (loss 1.2729):  95%|█████████▍| 890/938 [06:10<00:18,  2.56it/s]
Training 1/1 epoch (loss 1.4136):  95%|█████████▍| 890/938 [06:11<00:18,  2.56it/s]
Training 1/1 epoch (loss 1.4136):  95%|█████████▍| 891/938 [06:11<00:17,  2.62it/s]
Training 1/1 epoch (loss 1.4520):  95%|█████████▍| 891/938 [06:11<00:17,  2.62it/s]
Training 1/1 epoch (loss 1.4520):  95%|█████████▌| 892/938 [06:11<00:17,  2.59it/s]
Training 1/1 epoch (loss 1.5104):  95%|█████████▌| 892/938 [06:12<00:17,  2.59it/s]
Training 1/1 epoch (loss 1.5104):  95%|█████████▌| 893/938 [06:12<00:17,  2.56it/s]
Training 1/1 epoch (loss 1.3210):  95%|█████████▌| 893/938 [06:12<00:17,  2.56it/s]
Training 1/1 epoch (loss 1.3210):  95%|█████████▌| 894/938 [06:12<00:17,  2.57it/s]
Training 1/1 epoch (loss 1.3850):  95%|█████████▌| 894/938 [06:12<00:17,  2.57it/s]
Training 1/1 epoch (loss 1.3850):  95%|█████████▌| 895/938 [06:12<00:16,  2.55it/s]
Training 1/1 epoch (loss 1.3641):  95%|█████████▌| 895/938 [06:13<00:16,  2.55it/s]
Training 1/1 epoch (loss 1.3641):  96%|█████████▌| 896/938 [06:13<00:16,  2.51it/s]
Training 1/1 epoch (loss 1.3617):  96%|█████████▌| 896/938 [06:13<00:16,  2.51it/s]
Training 1/1 epoch (loss 1.3617):  96%|█████████▌| 897/938 [06:13<00:16,  2.48it/s]
Training 1/1 epoch (loss 1.4344):  96%|█████████▌| 897/938 [06:14<00:16,  2.48it/s]
Training 1/1 epoch (loss 1.4344):  96%|█████████▌| 898/938 [06:14<00:16,  2.45it/s]
Training 1/1 epoch (loss 1.3797):  96%|█████████▌| 898/938 [06:14<00:16,  2.45it/s]
Training 1/1 epoch (loss 1.3797):  96%|█████████▌| 899/938 [06:14<00:16,  2.38it/s]
Training 1/1 epoch (loss 1.3837):  96%|█████████▌| 899/938 [06:15<00:16,  2.38it/s]
Training 1/1 epoch (loss 1.3837):  96%|█████████▌| 900/938 [06:15<00:15,  2.39it/s]
Training 1/1 epoch (loss 1.3018):  96%|█████████▌| 900/938 [06:15<00:15,  2.39it/s]
Training 1/1 epoch (loss 1.3018):  96%|█████████▌| 901/938 [06:15<00:15,  2.46it/s]
Training 1/1 epoch (loss 1.3036):  96%|█████████▌| 901/938 [06:15<00:15,  2.46it/s]
Training 1/1 epoch (loss 1.3036):  96%|█████████▌| 902/938 [06:15<00:14,  2.46it/s]
Training 1/1 epoch (loss 1.4066):  96%|█████████▌| 902/938 [06:16<00:14,  2.46it/s]
Training 1/1 epoch (loss 1.4066):  96%|█████████▋| 903/938 [06:16<00:13,  2.50it/s]
Training 1/1 epoch (loss 1.5011):  96%|█████████▋| 903/938 [06:16<00:13,  2.50it/s]
Training 1/1 epoch (loss 1.5011):  96%|█████████▋| 904/938 [06:16<00:13,  2.49it/s]
Training 1/1 epoch (loss 1.4453):  96%|█████████▋| 904/938 [06:16<00:13,  2.49it/s]
Training 1/1 epoch (loss 1.4453):  96%|█████████▋| 905/938 [06:16<00:13,  2.53it/s]
Training 1/1 epoch (loss 1.4301):  96%|█████████▋| 905/938 [06:17<00:13,  2.53it/s]
Training 1/1 epoch (loss 1.4301):  97%|█████████▋| 906/938 [06:17<00:12,  2.48it/s]
Training 1/1 epoch (loss 1.3429):  97%|█████████▋| 906/938 [06:17<00:12,  2.48it/s]
Training 1/1 epoch (loss 1.3429):  97%|█████████▋| 907/938 [06:17<00:12,  2.45it/s]
Training 1/1 epoch (loss 1.3784):  97%|█████████▋| 907/938 [06:18<00:12,  2.45it/s]
Training 1/1 epoch (loss 1.3784):  97%|█████████▋| 908/938 [06:18<00:12,  2.37it/s]
Training 1/1 epoch (loss 1.3607):  97%|█████████▋| 908/938 [06:18<00:12,  2.37it/s]
Training 1/1 epoch (loss 1.3607):  97%|█████████▋| 909/938 [06:18<00:12,  2.40it/s]
Training 1/1 epoch (loss 1.4616):  97%|█████████▋| 909/938 [06:19<00:12,  2.40it/s]
Training 1/1 epoch (loss 1.4616):  97%|█████████▋| 910/938 [06:19<00:11,  2.46it/s]
Training 1/1 epoch (loss 1.4125):  97%|█████████▋| 910/938 [06:19<00:11,  2.46it/s]
Training 1/1 epoch (loss 1.4125):  97%|█████████▋| 911/938 [06:19<00:10,  2.50it/s]
Training 1/1 epoch (loss 1.4056):  97%|█████████▋| 911/938 [06:19<00:10,  2.50it/s]
Training 1/1 epoch (loss 1.4056):  97%|█████████▋| 912/938 [06:19<00:10,  2.49it/s]
Training 1/1 epoch (loss 1.4123):  97%|█████████▋| 912/938 [06:20<00:10,  2.49it/s]
Training 1/1 epoch (loss 1.4123):  97%|█████████▋| 913/938 [06:20<00:09,  2.50it/s]
Training 1/1 epoch (loss 1.4569):  97%|█████████▋| 913/938 [06:20<00:09,  2.50it/s]
Training 1/1 epoch (loss 1.4569):  97%|█████████▋| 914/938 [06:20<00:09,  2.54it/s]
Training 1/1 epoch (loss 1.4257):  97%|█████████▋| 914/938 [06:21<00:09,  2.54it/s]
Training 1/1 epoch (loss 1.4257):  98%|█████████▊| 915/938 [06:21<00:09,  2.51it/s]
Training 1/1 epoch (loss 1.4759):  98%|█████████▊| 915/938 [06:21<00:09,  2.51it/s]
Training 1/1 epoch (loss 1.4759):  98%|█████████▊| 916/938 [06:21<00:08,  2.50it/s]
Training 1/1 epoch (loss 1.3801):  98%|█████████▊| 916/938 [06:21<00:08,  2.50it/s]
Training 1/1 epoch (loss 1.3801):  98%|█████████▊| 917/938 [06:21<00:08,  2.54it/s]
Training 1/1 epoch (loss 1.3763):  98%|█████████▊| 917/938 [06:22<00:08,  2.54it/s]
Training 1/1 epoch (loss 1.3763):  98%|█████████▊| 918/938 [06:22<00:07,  2.55it/s]
Training 1/1 epoch (loss 1.3907):  98%|█████████▊| 918/938 [06:22<00:07,  2.55it/s]
Training 1/1 epoch (loss 1.3907):  98%|█████████▊| 919/938 [06:22<00:07,  2.56it/s]
Training 1/1 epoch (loss 1.3770):  98%|█████████▊| 919/938 [06:22<00:07,  2.56it/s]
Training 1/1 epoch (loss 1.3770):  98%|█████████▊| 920/938 [06:22<00:07,  2.55it/s]
Training 1/1 epoch (loss 1.3672):  98%|█████████▊| 920/938 [06:23<00:07,  2.55it/s]
Training 1/1 epoch (loss 1.3672):  98%|█████████▊| 921/938 [06:23<00:06,  2.57it/s]
Training 1/1 epoch (loss 1.3476):  98%|█████████▊| 921/938 [06:23<00:06,  2.57it/s]
Training 1/1 epoch (loss 1.3476):  98%|█████████▊| 922/938 [06:23<00:06,  2.56it/s]
Training 1/1 epoch (loss 1.3112):  98%|█████████▊| 922/938 [06:24<00:06,  2.56it/s]
Training 1/1 epoch (loss 1.3112):  98%|█████████▊| 923/938 [06:24<00:05,  2.52it/s]
Training 1/1 epoch (loss 1.3869):  98%|█████████▊| 923/938 [06:24<00:05,  2.52it/s]
Training 1/1 epoch (loss 1.3869):  99%|█████████▊| 924/938 [06:24<00:05,  2.51it/s]
Training 1/1 epoch (loss 1.4174):  99%|█████████▊| 924/938 [06:25<00:05,  2.51it/s]
Training 1/1 epoch (loss 1.4174):  99%|█████████▊| 925/938 [06:25<00:05,  2.41it/s]
Training 1/1 epoch (loss 1.2937):  99%|█████████▊| 925/938 [06:25<00:05,  2.41it/s]
Training 1/1 epoch (loss 1.2937):  99%|█████████▊| 926/938 [06:25<00:04,  2.50it/s]
Training 1/1 epoch (loss 1.4355):  99%|█████████▊| 926/938 [06:25<00:04,  2.50it/s]
Training 1/1 epoch (loss 1.4355):  99%|█████████▉| 927/938 [06:25<00:04,  2.45it/s]
Training 1/1 epoch (loss 1.3572):  99%|█████████▉| 927/938 [06:26<00:04,  2.45it/s]
Training 1/1 epoch (loss 1.3572):  99%|█████████▉| 928/938 [06:26<00:04,  2.42it/s]
Training 1/1 epoch (loss 1.3109):  99%|█████████▉| 928/938 [06:26<00:04,  2.42it/s]
Training 1/1 epoch (loss 1.3109):  99%|█████████▉| 929/938 [06:26<00:03,  2.47it/s]
Training 1/1 epoch (loss 1.3701):  99%|█████████▉| 929/938 [06:27<00:03,  2.47it/s]
Training 1/1 epoch (loss 1.3701):  99%|█████████▉| 930/938 [06:27<00:03,  2.45it/s]
Training 1/1 epoch (loss 1.3276):  99%|█████████▉| 930/938 [06:27<00:03,  2.45it/s]
Training 1/1 epoch (loss 1.3276):  99%|█████████▉| 931/938 [06:27<00:02,  2.49it/s]
Training 1/1 epoch (loss 1.3996):  99%|█████████▉| 931/938 [06:27<00:02,  2.49it/s]
Training 1/1 epoch (loss 1.3996):  99%|█████████▉| 932/938 [06:27<00:02,  2.51it/s]
Training 1/1 epoch (loss 1.4141):  99%|█████████▉| 932/938 [06:28<00:02,  2.51it/s]
Training 1/1 epoch (loss 1.4141):  99%|█████████▉| 933/938 [06:28<00:01,  2.54it/s]
Training 1/1 epoch (loss 1.4167):  99%|█████████▉| 933/938 [06:28<00:01,  2.54it/s]
Training 1/1 epoch (loss 1.4167): 100%|█████████▉| 934/938 [06:28<00:01,  2.58it/s]
Training 1/1 epoch (loss 1.5263): 100%|█████████▉| 934/938 [06:28<00:01,  2.58it/s]
Training 1/1 epoch (loss 1.5263): 100%|█████████▉| 935/938 [06:28<00:01,  2.62it/s]
Training 1/1 epoch (loss 1.4366): 100%|█████████▉| 935/938 [06:29<00:01,  2.62it/s]
Training 1/1 epoch (loss 1.4366): 100%|█████████▉| 936/938 [06:29<00:00,  2.47it/s]
Training 1/1 epoch (loss 1.3975): 100%|█████████▉| 936/938 [06:29<00:00,  2.47it/s]
Training 1/1 epoch (loss 1.3975): 100%|█████████▉| 937/938 [06:29<00:00,  2.51it/s]
Training 1/1 epoch (loss 1.4480): 100%|█████████▉| 937/938 [06:30<00:00,  2.51it/s]
Training 1/1 epoch (loss 1.4480): 100%|██████████| 938/938 [06:30<00:00,  2.40it/s]
Training 1/1 epoch (loss 1.4480): 100%|██████████| 938/938 [06:30<00:00,  2.40it/s]
tokenizer config file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-3T/tinyllama-3T-s3-Q1-30k/tokenizer_config.json
Special tokens file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-3T/tinyllama-3T-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 0x15505936e050>>
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