[INFO|parser.py:325] 2024-07-11 11:00:10,231 >> Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, compute dtype: None

[INFO|tokenization_utils_base.py:2159] 2024-07-11 11:00:10,234 >> loading file tokenizer.model

07/11/2024 11:00:10 - INFO - llamafactory.hparams.parser - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, compute dtype: None

07/11/2024 11:00:10 - INFO - llamafactory.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1, distributed training: True, compute dtype: None

[INFO|tokenization_utils_base.py:2159] 2024-07-11 11:00:10,234 >> loading file tokenizer.json

[INFO|tokenization_utils_base.py:2159] 2024-07-11 11:00:10,234 >> loading file added_tokens.json

[INFO|tokenization_utils_base.py:2159] 2024-07-11 11:00:10,234 >> loading file special_tokens_map.json

[INFO|tokenization_utils_base.py:2159] 2024-07-11 11:00:10,235 >> loading file tokenizer_config.json

[INFO|loader.py:50] 2024-07-11 11:00:10,286 >> Loading dataset dev_output.json...

07/11/2024 11:00:10 - INFO - llamafactory.hparams.parser - Process rank: 7, device: cuda:7, n_gpu: 1, distributed training: True, compute dtype: None

07/11/2024 11:00:10 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: None

07/11/2024 11:00:10 - INFO - llamafactory.hparams.parser - Process rank: 4, device: cuda:4, n_gpu: 1, distributed training: True, compute dtype: None

07/11/2024 11:00:11 - INFO - llamafactory.data.loader - Loading dataset dev_output.json...

07/11/2024 11:00:11 - INFO - llamafactory.data.loader - Loading dataset dev_output.json...

07/11/2024 11:00:11 - INFO - llamafactory.data.loader - Loading dataset dev_output.json...

07/11/2024 11:00:11 - INFO - llamafactory.data.loader - Loading dataset dev_output.json...

07/11/2024 11:00:11 - INFO - llamafactory.data.loader - Loading dataset dev_output.json...

07/11/2024 11:00:11 - INFO - llamafactory.data.loader - Loading dataset dev_output.json...

07/11/2024 11:00:11 - INFO - llamafactory.data.loader - Loading dataset dev_output.json...

[INFO|configuration_utils.py:731] 2024-07-11 11:00:12,504 >> loading configuration file saves/LLaMA2-7B-Chat/full/train_2024-07-11-09-30-54_llama2_inst_truth/config.json

[INFO|configuration_utils.py:800] 2024-07-11 11:00:12,505 >> Model config LlamaConfig {
  "_name_or_path": "saves/LLaMA2-7B-Chat/full/train_2024-07-11-09-30-54_llama2_inst_truth",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_position_embeddings": 4096,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 32,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.42.3",
  "use_cache": false,
  "vocab_size": 32000
}


[INFO|patcher.py:81] 2024-07-11 11:00:12,505 >> Using KV cache for faster generation.

[INFO|modeling_utils.py:3553] 2024-07-11 11:00:12,529 >> loading weights file saves/LLaMA2-7B-Chat/full/train_2024-07-11-09-30-54_llama2_inst_truth/model.safetensors.index.json

[INFO|modeling_utils.py:1531] 2024-07-11 11:00:12,529 >> Instantiating LlamaForCausalLM model under default dtype torch.bfloat16.

[INFO|configuration_utils.py:1000] 2024-07-11 11:00:12,531 >> Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}


07/11/2024 11:00:12 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.

07/11/2024 11:00:12 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.

07/11/2024 11:00:12 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.

07/11/2024 11:00:12 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.

07/11/2024 11:00:12 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.

07/11/2024 11:00:12 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.

07/11/2024 11:00:12 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.

07/11/2024 11:00:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.

07/11/2024 11:00:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616

[INFO|modeling_utils.py:4364] 2024-07-11 11:00:16,788 >> All model checkpoint weights were used when initializing LlamaForCausalLM.


07/11/2024 11:00:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.

[INFO|modeling_utils.py:4372] 2024-07-11 11:00:16,788 >> All the weights of LlamaForCausalLM were initialized from the model checkpoint at saves/LLaMA2-7B-Chat/full/train_2024-07-11-09-30-54_llama2_inst_truth.
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.

[INFO|configuration_utils.py:953] 2024-07-11 11:00:16,792 >> loading configuration file saves/LLaMA2-7B-Chat/full/train_2024-07-11-09-30-54_llama2_inst_truth/generation_config.json

[INFO|configuration_utils.py:1000] 2024-07-11 11:00:16,792 >> Generate config GenerationConfig {
  "bos_token_id": 1,
  "do_sample": true,
  "eos_token_id": 2,
  "max_length": 4096,
  "pad_token_id": 0,
  "temperature": 0.6,
  "top_p": 0.9
}


07/11/2024 11:00:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616

[INFO|attention.py:80] 2024-07-11 11:00:16,798 >> Using torch SDPA for faster training and inference.

07/11/2024 11:00:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.

07/11/2024 11:00:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.

[INFO|loader.py:196] 2024-07-11 11:00:16,802 >> all params: 6,738,415,616

07/11/2024 11:00:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616

07/11/2024 11:00:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616

07/11/2024 11:00:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.

07/11/2024 11:00:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616

07/11/2024 11:00:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.

07/11/2024 11:00:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.

07/11/2024 11:00:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616

07/11/2024 11:00:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616

[INFO|trainer.py:3788] 2024-07-11 11:00:16,914 >> 
***** Running Prediction *****

[INFO|trainer.py:3790] 2024-07-11 11:00:16,914 >>   Num examples = 2554

[INFO|trainer.py:3793] 2024-07-11 11:00:16,914 >>   Batch size = 2

07/11/2024 11:00:17 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

07/11/2024 11:00:17 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

[WARNING|logging.py:328] 2024-07-11 11:00:17,582 >> We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

07/11/2024 11:00:17 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

07/11/2024 11:00:17 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

07/11/2024 11:00:17 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

07/11/2024 11:00:17 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

07/11/2024 11:00:17 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)

[INFO|trainer.py:127] 2024-07-11 11:00:34,679 >> Saving prediction results to saves/LLaMA2-7B-Chat/full/eval_2024-07-11-10-49-45/generated_predictions.jsonl