[INFO|tokenization_utils_base.py:2024] 2024-01-18 19:25:40,385 >> loading file tokenizer.model
[INFO|tokenization_utils_base.py:2024] 2024-01-18 19:25:40,385 >> loading file added_tokens.json
[INFO|tokenization_utils_base.py:2024] 2024-01-18 19:25:40,385 >> loading file special_tokens_map.json
[INFO|tokenization_utils_base.py:2024] 2024-01-18 19:25:40,385 >> loading file tokenizer_config.json
[INFO|tokenization_utils_base.py:2024] 2024-01-18 19:25:40,385 >> loading file tokenizer.json
[INFO|configuration_utils.py:737] 2024-01-18 19:25:40,429 >> loading configuration file ./models/LMCocktail-10.7B-v1/config.json
[INFO|configuration_utils.py:802] 2024-01-18 19:25:40,430 >> Model config LlamaConfig {
  "_name_or_path": "./models/LMCocktail-10.7B-v1",
  "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": 14336,
  "max_position_embeddings": 4096,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 48,
  "num_key_value_heads": 8,
  "pad_token_id": 2,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.36.2",
  "use_cache": true,
  "vocab_size": 32000
}

[INFO|modeling_utils.py:3341] 2024-01-18 19:25:40,446 >> loading weights file ./models/LMCocktail-10.7B-v1/model.safetensors.index.json
[INFO|modeling_utils.py:1341] 2024-01-18 19:25:40,447 >> Instantiating LlamaForCausalLM model under default dtype torch.float16.
[INFO|configuration_utils.py:826] 2024-01-18 19:25:40,447 >> Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "pad_token_id": 2
}


Loading checkpoint shards:   0%|          | 0/5 [00:00<?, ?it/s]
Loading checkpoint shards:  20%|██        | 1/5 [00:00<00:00,  6.36it/s]
Loading checkpoint shards:  40%|████      | 2/5 [00:00<00:00,  6.36it/s]
Loading checkpoint shards:  60%|██████    | 3/5 [00:00<00:00,  6.37it/s]
Loading checkpoint shards:  80%|████████  | 4/5 [00:00<00:00,  6.28it/s]
Loading checkpoint shards: 100%|██████████| 5/5 [00:00<00:00,  6.33it/s]
Loading checkpoint shards: 100%|██████████| 5/5 [00:00<00:00,  6.33it/s]
[INFO|modeling_utils.py:4185] 2024-01-18 19:25:41,404 >> All model checkpoint weights were used when initializing LlamaForCausalLM.

[INFO|modeling_utils.py:4193] 2024-01-18 19:25:41,404 >> All the weights of LlamaForCausalLM were initialized from the model checkpoint at ./models/LMCocktail-10.7B-v1.
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:779] 2024-01-18 19:25:41,407 >> loading configuration file ./models/LMCocktail-10.7B-v1/generation_config.json
[INFO|configuration_utils.py:826] 2024-01-18 19:25:41,408 >> Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "pad_token_id": 2,
  "use_cache": false
}

01/18/2024 19:25:41 - INFO - llmtuner.model.adapter - Fine-tuning method: LoRA
01/18/2024 19:25:43 - INFO - llmtuner.model.adapter - Merged 1 adapter(s).
01/18/2024 19:25:43 - INFO - llmtuner.model.adapter - Loaded adapter(s): ./models/sft/LMCocktail-10.7B-v1-sft-glaive-function-calling-v2-ep1-lora
01/18/2024 19:25:43 - INFO - llmtuner.model.loader - trainable params: 0 || all params: 10731524096 || trainable%: 0.0000
01/18/2024 19:25:43 - INFO - llmtuner.model.loader - This IS expected that the trainable params is 0 if you are using model for inference only.
[INFO|configuration_utils.py:483] 2024-01-18 19:25:43,941 >> Configuration saved in ./models/export/LMCocktail-10.7B-v1-sft-glaive-function-calling-v2-ep1/config.json
[INFO|configuration_utils.py:594] 2024-01-18 19:25:43,941 >> Configuration saved in ./models/export/LMCocktail-10.7B-v1-sft-glaive-function-calling-v2-ep1/generation_config.json
[INFO|modeling_utils.py:2390] 2024-01-18 19:26:02,405 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 5 checkpoint shards. You can find where each parameters has been saved in the index located at ./models/export/LMCocktail-10.7B-v1-sft-glaive-function-calling-v2-ep1/model.safetensors.index.json.
[INFO|tokenization_utils_base.py:2432] 2024-01-18 19:26:02,406 >> tokenizer config file saved in ./models/export/LMCocktail-10.7B-v1-sft-glaive-function-calling-v2-ep1/tokenizer_config.json
[INFO|tokenization_utils_base.py:2441] 2024-01-18 19:26:02,406 >> Special tokens file saved in ./models/export/LMCocktail-10.7B-v1-sft-glaive-function-calling-v2-ep1/special_tokens_map.json