| [INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file tokenizer.model from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.model |
|
|
| [INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file tokenizer.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.json |
|
|
| [INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file added_tokens.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/added_tokens.json |
|
|
| [INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file special_tokens_map.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/special_tokens_map.json |
|
|
| [INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file tokenizer_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer_config.json |
|
|
| [INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file chat_template.jinja from cache at None |
|
|
| [INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2299 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. |
|
|
| [INFO|2025-05-30 02:31:17] processing_utils.py:930 >> loading configuration file processor_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/processor_config.json |
|
|
| [WARNING|2025-05-30 02:31:17] logging.py:328 >> Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`. |
|
|
| [INFO|2025-05-30 02:31:17] image_processing_base.py:380 >> loading configuration file preprocessor_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/preprocessor_config.json |
|
|
| [WARNING|2025-05-30 02:31:17] logging.py:328 >> Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`. |
|
|
| [INFO|2025-05-30 02:31:17] image_processing_base.py:433 >> Image processor CLIPImageProcessor { |
| "crop_size": { |
| "height": 336, |
| "width": 336 |
| }, |
| "do_center_crop": true, |
| "do_convert_rgb": true, |
| "do_normalize": true, |
| "do_rescale": true, |
| "do_resize": true, |
| "image_mean": [ |
| 0.48145466, |
| 0.4578275, |
| 0.40821073 |
| ], |
| "image_processor_type": "CLIPImageProcessor", |
| "image_std": [ |
| 0.26862954, |
| 0.26130258, |
| 0.27577711 |
| ], |
| "processor_class": "LlavaProcessor", |
| "resample": 3, |
| "rescale_factor": 0.00392156862745098, |
| "size": { |
| "shortest_edge": 336 |
| } |
| } |
|
|
|
|
| [INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file tokenizer.model from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.model |
|
|
| [INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file tokenizer.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.json |
|
|
| [INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file added_tokens.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/added_tokens.json |
|
|
| [INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file special_tokens_map.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/special_tokens_map.json |
|
|
| [INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file tokenizer_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer_config.json |
|
|
| [INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file chat_template.jinja from cache at None |
|
|
| [INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2299 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. |
|
|
| [INFO|2025-05-30 02:31:19] processing_utils.py:930 >> loading configuration file processor_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/processor_config.json |
|
|
| [INFO|2025-05-30 02:31:19] processing_utils.py:990 >> Processor LlavaProcessor: |
| - image_processor: CLIPImageProcessor { |
| "crop_size": { |
| "height": 336, |
| "width": 336 |
| }, |
| "do_center_crop": true, |
| "do_convert_rgb": true, |
| "do_normalize": true, |
| "do_rescale": true, |
| "do_resize": true, |
| "image_mean": [ |
| 0.48145466, |
| 0.4578275, |
| 0.40821073 |
| ], |
| "image_processor_type": "CLIPImageProcessor", |
| "image_std": [ |
| 0.26862954, |
| 0.26130258, |
| 0.27577711 |
| ], |
| "processor_class": "LlavaProcessor", |
| "resample": 3, |
| "rescale_factor": 0.00392156862745098, |
| "size": { |
| "shortest_edge": 336 |
| } |
| } |
|
|
| - tokenizer: LlamaTokenizerFast(name_or_path='llava-hf/llava-1.5-7b-hf', vocab_size=32000, model_max_length=1000000000000000019884624838656, is_fast=True, padding_side='left', truncation_side='right', special_tokens={'bos_token': '<s>', 'eos_token': '</s>', 'unk_token': '<unk>', 'pad_token': '<pad>', 'image_token': '<image>'}, clean_up_tokenization_spaces=False, added_tokens_decoder={ |
| 0: AddedToken("<unk>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), |
| 1: AddedToken("<s>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), |
| 2: AddedToken("</s>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), |
| 32000: AddedToken("<image>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), |
| 32001: AddedToken("<pad>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), |
| } |
| ) |
|
|
| { |
| "image_token": "<image>", |
| "num_additional_image_tokens": 1, |
| "patch_size": 14, |
| "processor_class": "LlavaProcessor", |
| "vision_feature_select_strategy": "default" |
| } |
|
|
|
|
| [INFO|2025-05-30 02:31:19] logging.py:143 >> Loading dataset /home/tsinghuaair/mawz/xxe_metchee/finetune-llms/llava-1.5-7b-hf-sticker-labels/kfold_output_zh/fold_1/stickers_label_train.json... |
|
|
| [INFO|2025-05-30 02:31:24] configuration_utils.py:698 >> loading configuration file config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/config.json |
|
|
| [INFO|2025-05-30 02:31:24] configuration_utils.py:770 >> Model config LlavaConfig { |
| "architectures": [ |
| "LlavaForConditionalGeneration" |
| ], |
| "ignore_index": -100, |
| "image_seq_length": 576, |
| "image_token_index": 32000, |
| "model_type": "llava", |
| "multimodal_projector_bias": true, |
| "pad_token_id": 32001, |
| "projector_hidden_act": "gelu", |
| "text_config": { |
| "_name_or_path": "lmsys/vicuna-7b-v1.5", |
| "architectures": [ |
| "LlamaForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "head_dim": 128, |
| "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, |
| "torch_dtype": "float16", |
| "use_cache": true, |
| "vocab_size": 32064 |
| }, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.52.1", |
| "vision_config": { |
| "attention_dropout": 0.0, |
| "hidden_act": "quick_gelu", |
| "hidden_size": 1024, |
| "image_size": 336, |
| "initializer_factor": 1.0, |
| "initializer_range": 0.02, |
| "intermediate_size": 4096, |
| "layer_norm_eps": 1e-05, |
| "model_type": "clip_vision_model", |
| "num_attention_heads": 16, |
| "num_channels": 3, |
| "num_hidden_layers": 24, |
| "patch_size": 14, |
| "projection_dim": 768, |
| "vocab_size": 32000 |
| }, |
| "vision_feature_layer": -2, |
| "vision_feature_select_strategy": "default", |
| "vocab_size": 32064 |
| } |
|
|
|
|
| [INFO|2025-05-30 02:31:24] logging.py:143 >> KV cache is disabled during training. |
|
|
| [INFO|2025-05-30 02:31:24] modeling_utils.py:1149 >> loading weights file model.safetensors from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/model.safetensors.index.json |
|
|
| [INFO|2025-05-30 02:31:24] modeling_utils.py:2239 >> Instantiating LlavaForConditionalGeneration model under default dtype torch.bfloat16. |
|
|
| [INFO|2025-05-30 02:31:24] configuration_utils.py:1135 >> Generate config GenerationConfig { |
| "pad_token_id": 32001, |
| "use_cache": false |
| } |
|
|
|
|
| [INFO|2025-05-30 02:31:25] modeling_utils.py:2239 >> Instantiating CLIPVisionModel model under default dtype torch.bfloat16. |
|
|
| [INFO|2025-05-30 02:31:25] modeling_utils.py:2239 >> Instantiating LlamaModel model under default dtype torch.bfloat16. |
|
|
| [INFO|2025-05-30 02:31:29] modeling_utils.py:5170 >> All model checkpoint weights were used when initializing LlavaForConditionalGeneration. |
|
|
|
|
| [INFO|2025-05-30 02:31:29] modeling_utils.py:5178 >> All the weights of LlavaForConditionalGeneration were initialized from the model checkpoint at llava-hf/llava-1.5-7b-hf. |
| If your task is similar to the task the model of the checkpoint was trained on, you can already use LlavaForConditionalGeneration for predictions without further training. |
|
|
| [INFO|2025-05-30 02:31:29] configuration_utils.py:1090 >> loading configuration file generation_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/generation_config.json |
|
|
| [INFO|2025-05-30 02:31:29] configuration_utils.py:1135 >> Generate config GenerationConfig { |
| "bos_token_id": 1, |
| "eos_token_id": 2, |
| "pad_token_id": 32001 |
| } |
|
|
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> Gradient checkpointing enabled. |
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> Using torch SDPA for faster training and inference. |
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> Upcasting trainable params to float32. |
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> Fine-tuning method: LoRA |
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> Found linear modules: gate_proj,q_proj,up_proj,o_proj,k_proj,v_proj,down_proj |
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> Set vision model not trainable: ['vision_tower']. |
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> Set multi model projector not trainable: multi_modal_projector. |
|
|
| [INFO|2025-05-30 02:31:30] logging.py:143 >> trainable params: 19,988,480 || all params: 7,083,415,552 || trainable%: 0.2822 |
|
|
| [INFO|2025-05-30 02:31:30] trainer.py:756 >> Using auto half precision backend |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2409 >> ***** Running training ***** |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2410 >> Num examples = 489 |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2411 >> Num Epochs = 4 |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2412 >> Instantaneous batch size per device = 2 |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2415 >> Total train batch size (w. parallel, distributed & accumulation) = 32 |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2416 >> Gradient Accumulation steps = 8 |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2417 >> Total optimization steps = 64 |
|
|
| [INFO|2025-05-30 02:31:31] trainer.py:2418 >> Number of trainable parameters = 19,988,480 |
|
|
| [WARNING|2025-05-30 02:31:33] logging.py:328 >> `loss_type=None` was set in the config but it is unrecognised.Using the default loss: `ForCausalLMLoss`. |
|
|
| [WARNING|2025-05-30 02:31:33] logging.py:328 >> `loss_type=None` was set in the config but it is unrecognised.Using the default loss: `ForCausalLMLoss`. |
|
|
| [INFO|2025-05-30 02:31:59] logging.py:143 >> {'loss': 2.2304, 'learning_rate': 4.9520e-05, 'epoch': 0.33, 'throughput': 4337.62} |
|
|
| [INFO|2025-05-30 02:32:26] logging.py:143 >> {'loss': 2.1470, 'learning_rate': 4.7600e-05, 'epoch': 0.65, 'throughput': 4382.44} |
|
|
| [INFO|2025-05-30 02:32:53] logging.py:143 >> {'loss': 1.9562, 'learning_rate': 4.4325e-05, 'epoch': 0.98, 'throughput': 4393.16} |
|
|
| [INFO|2025-05-30 02:33:17] logging.py:143 >> {'loss': 1.5887, 'learning_rate': 3.9892e-05, 'epoch': 1.26, 'throughput': 4391.91} |
|
|
| [INFO|2025-05-30 02:33:45] logging.py:143 >> {'loss': 1.7952, 'learning_rate': 3.4567e-05, 'epoch': 1.59, 'throughput': 4386.97} |
|
|
| [INFO|2025-05-30 02:34:12] logging.py:143 >> {'loss': 1.7717, 'learning_rate': 2.8668e-05, 'epoch': 1.91, 'throughput': 4380.26} |
|
|
| [INFO|2025-05-30 02:34:36] logging.py:143 >> {'loss': 1.5037, 'learning_rate': 2.2550e-05, 'epoch': 2.20, 'throughput': 4378.09} |
|
|
| [INFO|2025-05-30 02:35:03] logging.py:143 >> {'loss': 1.7171, 'learning_rate': 1.6578e-05, 'epoch': 2.52, 'throughput': 4374.05} |
|
|
| [INFO|2025-05-30 02:35:31] logging.py:143 >> {'loss': 1.6151, 'learning_rate': 1.1111e-05, 'epoch': 2.85, 'throughput': 4369.62} |
|
|
| [INFO|2025-05-30 02:35:55] logging.py:143 >> {'loss': 1.4855, 'learning_rate': 6.4762e-06, 'epoch': 3.13, 'throughput': 4366.67} |
|
|
| [INFO|2025-05-30 02:36:23] logging.py:143 >> {'loss': 1.6717, 'learning_rate': 2.9520e-06, 'epoch': 3.46, 'throughput': 4365.99} |
|
|
| [INFO|2025-05-30 02:36:50] logging.py:143 >> {'loss': 1.6515, 'learning_rate': 7.4922e-07, 'epoch': 3.78, 'throughput': 4363.33} |
|
|
| [INFO|2025-05-30 02:37:09] trainer.py:3993 >> Saving model checkpoint to saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64 |
|
|
| [INFO|2025-05-30 02:37:10] configuration_utils.py:698 >> loading configuration file config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/config.json |
|
|
| [INFO|2025-05-30 02:37:10] configuration_utils.py:770 >> Model config LlavaConfig { |
| "architectures": [ |
| "LlavaForConditionalGeneration" |
| ], |
| "ignore_index": -100, |
| "image_seq_length": 576, |
| "image_token_index": 32000, |
| "model_type": "llava", |
| "multimodal_projector_bias": true, |
| "pad_token_id": 32001, |
| "projector_hidden_act": "gelu", |
| "text_config": { |
| "_name_or_path": "lmsys/vicuna-7b-v1.5", |
| "architectures": [ |
| "LlamaForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "head_dim": 128, |
| "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, |
| "torch_dtype": "float16", |
| "use_cache": true, |
| "vocab_size": 32064 |
| }, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.52.1", |
| "vision_config": { |
| "attention_dropout": 0.0, |
| "hidden_act": "quick_gelu", |
| "hidden_size": 1024, |
| "image_size": 336, |
| "initializer_factor": 1.0, |
| "initializer_range": 0.02, |
| "intermediate_size": 4096, |
| "layer_norm_eps": 1e-05, |
| "model_type": "clip_vision_model", |
| "num_attention_heads": 16, |
| "num_channels": 3, |
| "num_hidden_layers": 24, |
| "patch_size": 14, |
| "projection_dim": 768, |
| "vocab_size": 32000 |
| }, |
| "vision_feature_layer": -2, |
| "vision_feature_select_strategy": "default", |
| "vocab_size": 32064 |
| } |
|
|
|
|
| [INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/chat_template.jinja |
|
|
| [INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/tokenizer_config.json |
|
|
| [INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/special_tokens_map.json |
|
|
| [INFO|2025-05-30 02:37:10] image_processing_base.py:260 >> Image processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/preprocessor_config.json |
|
|
| [INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/chat_template.jinja |
|
|
| [INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/tokenizer_config.json |
|
|
| [INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/special_tokens_map.json |
|
|
| [INFO|2025-05-30 02:37:10] processing_utils.py:674 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/chat_template.jinja |
|
|
| [INFO|2025-05-30 02:37:11] processing_utils.py:709 >> processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/processor_config.json |
|
|
| [INFO|2025-05-30 02:37:11] trainer.py:2676 >> |
|
|
| Training completed. Do not forget to share your model on huggingface.co/models =) |
|
|
|
|
|
|
| [INFO|2025-05-30 02:37:11] image_processing_base.py:260 >> Image processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/preprocessor_config.json |
|
|
| [INFO|2025-05-30 02:37:11] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/chat_template.jinja |
|
|
| [INFO|2025-05-30 02:37:11] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/tokenizer_config.json |
|
|
| [INFO|2025-05-30 02:37:11] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/special_tokens_map.json |
|
|
| [INFO|2025-05-30 02:37:11] processing_utils.py:674 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/chat_template.jinja |
|
|
| [INFO|2025-05-30 02:37:11] processing_utils.py:709 >> processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/processor_config.json |
|
|
| [INFO|2025-05-30 02:37:11] trainer.py:3993 >> Saving model checkpoint to saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs |
|
|
| [INFO|2025-05-30 02:37:11] configuration_utils.py:698 >> loading configuration file config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/config.json |
|
|
| [INFO|2025-05-30 02:37:11] configuration_utils.py:770 >> Model config LlavaConfig { |
| "architectures": [ |
| "LlavaForConditionalGeneration" |
| ], |
| "ignore_index": -100, |
| "image_seq_length": 576, |
| "image_token_index": 32000, |
| "model_type": "llava", |
| "multimodal_projector_bias": true, |
| "pad_token_id": 32001, |
| "projector_hidden_act": "gelu", |
| "text_config": { |
| "_name_or_path": "lmsys/vicuna-7b-v1.5", |
| "architectures": [ |
| "LlamaForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "head_dim": 128, |
| "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, |
| "torch_dtype": "float16", |
| "use_cache": true, |
| "vocab_size": 32064 |
| }, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.52.1", |
| "vision_config": { |
| "attention_dropout": 0.0, |
| "hidden_act": "quick_gelu", |
| "hidden_size": 1024, |
| "image_size": 336, |
| "initializer_factor": 1.0, |
| "initializer_range": 0.02, |
| "intermediate_size": 4096, |
| "layer_norm_eps": 1e-05, |
| "model_type": "clip_vision_model", |
| "num_attention_heads": 16, |
| "num_channels": 3, |
| "num_hidden_layers": 24, |
| "patch_size": 14, |
| "projection_dim": 768, |
| "vocab_size": 32000 |
| }, |
| "vision_feature_layer": -2, |
| "vision_feature_select_strategy": "default", |
| "vocab_size": 32064 |
| } |
|
|
|
|
| [INFO|2025-05-30 02:37:12] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/chat_template.jinja |
|
|
| [INFO|2025-05-30 02:37:12] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/tokenizer_config.json |
|
|
| [INFO|2025-05-30 02:37:12] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/special_tokens_map.json |
|
|
| [WARNING|2025-05-30 02:37:12] logging.py:148 >> No metric eval_loss to plot. |
|
|
| [WARNING|2025-05-30 02:37:12] logging.py:148 >> No metric eval_accuracy to plot. |
|
|
| [INFO|2025-05-30 02:37:12] modelcard.py:450 >> Dropping the following result as it does not have all the necessary fields: |
| {'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}} |
|
|
|
|
