| 0.00s - Debugger warning: It seems that frozen modules are being used, which may | |
| 0.00s - make the debugger miss breakpoints. Please pass -Xfrozen_modules=off | |
| 0.00s - to python to disable frozen modules. | |
| 0.00s - Note: Debugging will proceed. Set PYDEVD_DISABLE_FILE_VALIDATION=1 to disable this validation. | |
| [INFO|tokenization_utils_base.py:2024] 2024-01-21 14:22:37,060 >> loading file tokenizer.model | |
| [INFO|tokenization_utils_base.py:2024] 2024-01-21 14:22:37,061 >> loading file added_tokens.json | |
| [INFO|tokenization_utils_base.py:2024] 2024-01-21 14:22:37,061 >> loading file special_tokens_map.json | |
| [INFO|tokenization_utils_base.py:2024] 2024-01-21 14:22:37,061 >> loading file tokenizer_config.json | |
| [INFO|tokenization_utils_base.py:2024] 2024-01-21 14:22:37,061 >> loading file tokenizer.json | |
| [INFO|configuration_utils.py:737] 2024-01-21 14:22:37,440 >> loading configuration file ./models/dolphin-2.6-mistral-7b-dpo-laser/config.json | |
| [INFO|configuration_utils.py:802] 2024-01-21 14:22:37,443 >> Model config MistralConfig { | |
| "_name_or_path": "./models/dolphin-2.6-mistral-7b-dpo-laser", | |
| "architectures": [ | |
| "MistralForCausalLM" | |
| ], | |
| "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": 32768, | |
| "model_type": "mistral", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 8, | |
| "rms_norm_eps": 1e-05, | |
| "rope_theta": 10000.0, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.36.2", | |
| "use_cache": false, | |
| "vocab_size": 32001 | |
| } | |
| [INFO|modeling_utils.py:3341] 2024-01-21 14:22:40,379 >> loading weights file ./models/dolphin-2.6-mistral-7b-dpo-laser/model.safetensors.index.json | |
| [INFO|modeling_utils.py:1341] 2024-01-21 14:22:40,379 >> Instantiating MistralForCausalLM model under default dtype torch.bfloat16. | |
| [INFO|configuration_utils.py:826] 2024-01-21 14:22:40,382 >> Generate config GenerationConfig { | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "use_cache": false | |
| } | |
| Waiting for debugger attach | |
| Backend TkAgg is interactive backend. Turning interactive mode on. | |
| Loading checkpoint shards: 0%| | 0/3 [00:00<?, ?it/s] Loading checkpoint shards: 33%|███▎ | 1/3 [00:00<00:00, 5.50it/s] Loading checkpoint shards: 67%|██████▋ | 2/3 [00:00<00:00, 5.45it/s] Loading checkpoint shards: 100%|██████████| 3/3 [00:00<00:00, 5.57it/s] Loading checkpoint shards: 100%|██████████| 3/3 [00:00<00:00, 5.53it/s] | |
| [INFO|modeling_utils.py:4185] 2024-01-21 14:22:41,298 >> All model checkpoint weights were used when initializing MistralForCausalLM. | |
| [INFO|modeling_utils.py:4193] 2024-01-21 14:22:41,298 >> All the weights of MistralForCausalLM were initialized from the model checkpoint at ./models/dolphin-2.6-mistral-7b-dpo-laser. | |
| If your task is similar to the task the model of the checkpoint was trained on, you can already use MistralForCausalLM for predictions without further training. | |
| [INFO|configuration_utils.py:779] 2024-01-21 14:22:41,311 >> loading configuration file ./models/dolphin-2.6-mistral-7b-dpo-laser/generation_config.json | |
| [INFO|configuration_utils.py:826] 2024-01-21 14:22:41,312 >> Generate config GenerationConfig { | |
| "bos_token_id": 1, | |
| "eos_token_id": 2 | |
| } | |
| 01/21/2024 14:22:41 - INFO - llmtuner.model.adapter - Fine-tuning method: LoRA | |
| 01/21/2024 14:22:49 - INFO - llmtuner.model.adapter - Merged 1 adapter(s). | |
| 01/21/2024 14:22:49 - INFO - llmtuner.model.adapter - Loaded adapter(s): ./models/sft/dolphin-2.6-mistral-7b-dpo-laser-sft-glaive-function-calling-v2-ep1-lora | |
| 01/21/2024 14:22:49 - INFO - llmtuner.model.loader - trainable params: 0 || all params: 7241740288 || trainable%: 0.0000 | |
| 01/21/2024 14:22:49 - 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] 2024-01-21 14:22:49,748 >> Configuration saved in ./models/export/dolphin-2.6-mistral-7b-dpo-laser-sft-glaive-function-calling-v2-ep1/config.json | |
| [INFO|configuration_utils.py:594] 2024-01-21 14:22:49,749 >> Configuration saved in ./models/export/dolphin-2.6-mistral-7b-dpo-laser-sft-glaive-function-calling-v2-ep1/generation_config.json | |
| [INFO|modeling_utils.py:2390] 2024-01-21 14:23:02,915 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 3 checkpoint shards. You can find where each parameters has been saved in the index located at ./models/export/dolphin-2.6-mistral-7b-dpo-laser-sft-glaive-function-calling-v2-ep1/model.safetensors.index.json. | |
| [INFO|tokenization_utils_base.py:2432] 2024-01-21 14:23:03,311 >> tokenizer config file saved in ./models/export/dolphin-2.6-mistral-7b-dpo-laser-sft-glaive-function-calling-v2-ep1/tokenizer_config.json | |
| [INFO|tokenization_utils_base.py:2441] 2024-01-21 14:23:03,312 >> Special tokens file saved in ./models/export/dolphin-2.6-mistral-7b-dpo-laser-sft-glaive-function-calling-v2-ep1/special_tokens_map.json | |
| [INFO|tokenization_utils_base.py:2492] 2024-01-21 14:23:03,312 >> added tokens file saved in ./models/export/dolphin-2.6-mistral-7b-dpo-laser-sft-glaive-function-calling-v2-ep1/added_tokens.json | |