File size: 4,843 Bytes
db187ba |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
[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
|