Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+llama_factory/v2-20241220/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+llama_factory/v2-20241220-pos/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+llama_factory/v3-20241225/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+llama_factory/v3-20241225-1_10/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+llama_factory/v3-20241225-1_5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+llama_factory/v3-20241225-pos/tokenizer.json filter=lfs diff=lfs merge=lfs -text

llama_factory/v2-20241220-pos/README.md

@@ -0,0 +1,62 @@
+---
+library_name: peft
+license: other
+base_model: /disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: v2-20241220-pos
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# v2-20241220-pos
+
+This model is a fine-tuned version of [/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct](https://huggingface.co//disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct) on the M^2RAR-20241220-MLLM-Pos dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.46.1
+- Pytorch 2.3.1+cu121
+- Datasets 3.1.0
+- Tokenizers 0.20.3

llama_factory/v2-20241220-pos/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(?!.*visual).*(?:q_proj|down_proj|v_proj|o_proj|k_proj|up_proj|gate_proj).*",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

llama_factory/v2-20241220-pos/added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

llama_factory/v2-20241220-pos/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.56,
+    "total_flos": 4.173594118670254e+17,
+    "train_loss": 1.6055979132652283,
+    "train_runtime": 1821.5091,
+    "train_samples_per_second": 0.494,
+    "train_steps_per_second": 0.007
+}

llama_factory/v2-20241220-pos/chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

llama_factory/v2-20241220-pos/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

llama_factory/v2-20241220-pos/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

llama_factory/v2-20241220-pos/tokenizer_config.json

@@ -0,0 +1,144 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

llama_factory/v2-20241220-pos/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.56,
+    "total_flos": 4.173594118670254e+17,
+    "train_loss": 1.6055979132652283,
+    "train_runtime": 1821.5091,
+    "train_samples_per_second": 0.494,
+    "train_steps_per_second": 0.007
+}

llama_factory/v2-20241220-pos/trainer_log.jsonl

@@ -0,0 +1,13 @@
+{"current_steps": 1, "total_steps": 12, "loss": 1.7861, "lr": 5e-05, "epoch": 0.21333333333333335, "percentage": 8.33, "elapsed_time": "0:02:26", "remaining_time": "0:26:52"}
+{"current_steps": 2, "total_steps": 12, "loss": 1.7271, "lr": 0.0001, "epoch": 0.4266666666666667, "percentage": 16.67, "elapsed_time": "0:05:12", "remaining_time": "0:26:04"}
+{"current_steps": 3, "total_steps": 12, "loss": 1.6749, "lr": 9.755282581475769e-05, "epoch": 0.64, "percentage": 25.0, "elapsed_time": "0:07:38", "remaining_time": "0:22:55"}
+{"current_steps": 4, "total_steps": 12, "loss": 1.6561, "lr": 9.045084971874738e-05, "epoch": 0.8533333333333334, "percentage": 33.33, "elapsed_time": "0:10:17", "remaining_time": "0:20:35"}
+{"current_steps": 5, "total_steps": 12, "loss": 1.6393, "lr": 7.938926261462366e-05, "epoch": 1.0666666666666667, "percentage": 41.67, "elapsed_time": "0:12:23", "remaining_time": "0:17:20"}
+{"current_steps": 6, "total_steps": 12, "loss": 1.5852, "lr": 6.545084971874738e-05, "epoch": 1.28, "percentage": 50.0, "elapsed_time": "0:14:45", "remaining_time": "0:14:45"}
+{"current_steps": 7, "total_steps": 12, "loss": 1.5696, "lr": 5e-05, "epoch": 1.4933333333333334, "percentage": 58.33, "elapsed_time": "0:17:15", "remaining_time": "0:12:19"}
+{"current_steps": 8, "total_steps": 12, "loss": 1.5286, "lr": 3.4549150281252636e-05, "epoch": 1.7066666666666666, "percentage": 66.67, "elapsed_time": "0:19:31", "remaining_time": "0:09:45"}
+{"current_steps": 9, "total_steps": 12, "loss": 1.5246, "lr": 2.061073738537635e-05, "epoch": 1.92, "percentage": 75.0, "elapsed_time": "0:21:51", "remaining_time": "0:07:17"}
+{"current_steps": 10, "total_steps": 12, "loss": 1.5174, "lr": 9.549150281252633e-06, "epoch": 2.1333333333333333, "percentage": 83.33, "elapsed_time": "0:24:56", "remaining_time": "0:04:59"}
+{"current_steps": 11, "total_steps": 12, "loss": 1.525, "lr": 2.4471741852423237e-06, "epoch": 2.3466666666666667, "percentage": 91.67, "elapsed_time": "0:26:57", "remaining_time": "0:02:27"}
+{"current_steps": 12, "total_steps": 12, "loss": 1.5334, "lr": 0.0, "epoch": 2.56, "percentage": 100.0, "elapsed_time": "0:29:49", "remaining_time": "0:00:00"}
+{"current_steps": 12, "total_steps": 12, "epoch": 2.56, "percentage": 100.0, "elapsed_time": "0:30:21", "remaining_time": "0:00:00"}

llama_factory/v2-20241220-pos/trainer_state.json

@@ -0,0 +1,126 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.56,
+  "eval_steps": 500,
+  "global_step": 12,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.21333333333333335,
+      "grad_norm": 0.5428073406219482,
+      "learning_rate": 5e-05,
+      "loss": 1.7861,
+      "step": 1
+    },
+    {
+      "epoch": 0.4266666666666667,
+      "grad_norm": 0.5242983102798462,
+      "learning_rate": 0.0001,
+      "loss": 1.7271,
+      "step": 2
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.36863189935684204,
+      "learning_rate": 9.755282581475769e-05,
+      "loss": 1.6749,
+      "step": 3
+    },
+    {
+      "epoch": 0.8533333333333334,
+      "grad_norm": 0.4955863058567047,
+      "learning_rate": 9.045084971874738e-05,
+      "loss": 1.6561,
+      "step": 4
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.3081843852996826,
+      "learning_rate": 7.938926261462366e-05,
+      "loss": 1.6393,
+      "step": 5
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.23544219136238098,
+      "learning_rate": 6.545084971874738e-05,
+      "loss": 1.5852,
+      "step": 6
+    },
+    {
+      "epoch": 1.4933333333333334,
+      "grad_norm": 0.22464187443256378,
+      "learning_rate": 5e-05,
+      "loss": 1.5696,
+      "step": 7
+    },
+    {
+      "epoch": 1.7066666666666666,
+      "grad_norm": 0.188608318567276,
+      "learning_rate": 3.4549150281252636e-05,
+      "loss": 1.5286,
+      "step": 8
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.1828916221857071,
+      "learning_rate": 2.061073738537635e-05,
+      "loss": 1.5246,
+      "step": 9
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 0.18487173318862915,
+      "learning_rate": 9.549150281252633e-06,
+      "loss": 1.5174,
+      "step": 10
+    },
+    {
+      "epoch": 2.3466666666666667,
+      "grad_norm": 0.17777486145496368,
+      "learning_rate": 2.4471741852423237e-06,
+      "loss": 1.525,
+      "step": 11
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.1764741837978363,
+      "learning_rate": 0.0,
+      "loss": 1.5334,
+      "step": 12
+    },
+    {
+      "epoch": 2.56,
+      "step": 12,
+      "total_flos": 4.173594118670254e+17,
+      "train_loss": 1.6055979132652283,
+      "train_runtime": 1821.5091,
+      "train_samples_per_second": 0.494,
+      "train_steps_per_second": 0.007
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 12,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.173594118670254e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

llama_factory/v2-20241220/README.md

@@ -0,0 +1,62 @@
+---
+library_name: peft
+license: other
+base_model: /disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: llama_factory
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# llama_factory
+
+This model is a fine-tuned version of [/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct](https://huggingface.co//disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct) on the M^2RAR-20241220-MLLM dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.46.1
+- Pytorch 2.3.1+cu121
+- Datasets 3.1.0
+- Tokenizers 0.20.3

llama_factory/v2-20241220/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(?!.*visual).*(?:up_proj|q_proj|o_proj|gate_proj|v_proj|down_proj|k_proj).*",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

llama_factory/v2-20241220/added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

llama_factory/v2-20241220/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.88,
+    "total_flos": 5.97913280723288e+17,
+    "train_loss": 1.5128402180141873,
+    "train_runtime": 2593.9911,
+    "train_samples_per_second": 0.463,
+    "train_steps_per_second": 0.007
+}

llama_factory/v2-20241220/chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

llama_factory/v2-20241220/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

llama_factory/v2-20241220/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

llama_factory/v2-20241220/tokenizer_config.json

@@ -0,0 +1,144 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

llama_factory/v2-20241220/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.88,
+    "total_flos": 5.97913280723288e+17,
+    "train_loss": 1.5128402180141873,
+    "train_runtime": 2593.9911,
+    "train_samples_per_second": 0.463,
+    "train_steps_per_second": 0.007
+}

llama_factory/v2-20241220/trainer_log.jsonl

@@ -0,0 +1,19 @@
+{"current_steps": 1, "total_steps": 18, "loss": 1.6395, "lr": 5e-05, "epoch": 0.16, "percentage": 5.56, "elapsed_time": "0:02:35", "remaining_time": "0:44:07"}
+{"current_steps": 2, "total_steps": 18, "loss": 1.6454, "lr": 0.0001, "epoch": 0.32, "percentage": 11.11, "elapsed_time": "0:05:01", "remaining_time": "0:40:08"}
+{"current_steps": 3, "total_steps": 18, "loss": 1.6031, "lr": 9.903926402016153e-05, "epoch": 0.48, "percentage": 16.67, "elapsed_time": "0:07:13", "remaining_time": "0:36:09"}
+{"current_steps": 4, "total_steps": 18, "loss": 1.6443, "lr": 9.619397662556435e-05, "epoch": 0.64, "percentage": 22.22, "elapsed_time": "0:09:34", "remaining_time": "0:33:31"}
+{"current_steps": 5, "total_steps": 18, "loss": 1.5788, "lr": 9.157348061512727e-05, "epoch": 0.8, "percentage": 27.78, "elapsed_time": "0:12:14", "remaining_time": "0:31:49"}
+{"current_steps": 6, "total_steps": 18, "loss": 1.586, "lr": 8.535533905932738e-05, "epoch": 0.96, "percentage": 33.33, "elapsed_time": "0:14:59", "remaining_time": "0:29:59"}
+{"current_steps": 7, "total_steps": 18, "loss": 1.5621, "lr": 7.777851165098012e-05, "epoch": 1.12, "percentage": 38.89, "elapsed_time": "0:17:18", "remaining_time": "0:27:11"}
+{"current_steps": 8, "total_steps": 18, "loss": 1.4846, "lr": 6.91341716182545e-05, "epoch": 1.28, "percentage": 44.44, "elapsed_time": "0:20:05", "remaining_time": "0:25:07"}
+{"current_steps": 9, "total_steps": 18, "loss": 1.488, "lr": 5.9754516100806423e-05, "epoch": 1.44, "percentage": 50.0, "elapsed_time": "0:22:24", "remaining_time": "0:22:24"}
+{"current_steps": 10, "total_steps": 18, "loss": 1.4901, "lr": 5e-05, "epoch": 1.6, "percentage": 55.56, "elapsed_time": "0:24:58", "remaining_time": "0:19:58"}
+{"current_steps": 11, "total_steps": 18, "loss": 1.4207, "lr": 4.0245483899193595e-05, "epoch": 1.76, "percentage": 61.11, "elapsed_time": "0:27:21", "remaining_time": "0:17:24"}
+{"current_steps": 12, "total_steps": 18, "loss": 1.4992, "lr": 3.086582838174551e-05, "epoch": 1.92, "percentage": 66.67, "elapsed_time": "0:29:33", "remaining_time": "0:14:46"}
+{"current_steps": 13, "total_steps": 18, "loss": 1.4965, "lr": 2.2221488349019903e-05, "epoch": 2.08, "percentage": 72.22, "elapsed_time": "0:31:52", "remaining_time": "0:12:15"}
+{"current_steps": 14, "total_steps": 18, "loss": 1.4594, "lr": 1.4644660940672627e-05, "epoch": 2.24, "percentage": 77.78, "elapsed_time": "0:33:51", "remaining_time": "0:09:40"}
+{"current_steps": 15, "total_steps": 18, "loss": 1.42, "lr": 8.426519384872733e-06, "epoch": 2.4, "percentage": 83.33, "elapsed_time": "0:35:57", "remaining_time": "0:07:11"}
+{"current_steps": 16, "total_steps": 18, "loss": 1.411, "lr": 3.8060233744356633e-06, "epoch": 2.56, "percentage": 88.89, "elapsed_time": "0:38:11", "remaining_time": "0:04:46"}
+{"current_steps": 17, "total_steps": 18, "loss": 1.4063, "lr": 9.607359798384785e-07, "epoch": 2.7199999999999998, "percentage": 94.44, "elapsed_time": "0:40:31", "remaining_time": "0:02:23"}
+{"current_steps": 18, "total_steps": 18, "loss": 1.3964, "lr": 0.0, "epoch": 2.88, "percentage": 100.0, "elapsed_time": "0:42:45", "remaining_time": "0:00:00"}
+{"current_steps": 18, "total_steps": 18, "epoch": 2.88, "percentage": 100.0, "elapsed_time": "0:43:13", "remaining_time": "0:00:00"}

llama_factory/v2-20241220/trainer_state.json

@@ -0,0 +1,168 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.88,
+  "eval_steps": 500,
+  "global_step": 18,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.49046701192855835,
+      "learning_rate": 5e-05,
+      "loss": 1.6395,
+      "step": 1
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.47982165217399597,
+      "learning_rate": 0.0001,
+      "loss": 1.6454,
+      "step": 2
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3478511571884155,
+      "learning_rate": 9.903926402016153e-05,
+      "loss": 1.6031,
+      "step": 3
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.465070515871048,
+      "learning_rate": 9.619397662556435e-05,
+      "loss": 1.6443,
+      "step": 4
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.28730812668800354,
+      "learning_rate": 9.157348061512727e-05,
+      "loss": 1.5788,
+      "step": 5
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.2595618665218353,
+      "learning_rate": 8.535533905932738e-05,
+      "loss": 1.586,
+      "step": 6
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.22310923039913177,
+      "learning_rate": 7.777851165098012e-05,
+      "loss": 1.5621,
+      "step": 7
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.20716898143291473,
+      "learning_rate": 6.91341716182545e-05,
+      "loss": 1.4846,
+      "step": 8
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.20206721127033234,
+      "learning_rate": 5.9754516100806423e-05,
+      "loss": 1.488,
+      "step": 9
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.1985960155725479,
+      "learning_rate": 5e-05,
+      "loss": 1.4901,
+      "step": 10
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.18855780363082886,
+      "learning_rate": 4.0245483899193595e-05,
+      "loss": 1.4207,
+      "step": 11
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.18942952156066895,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 1.4992,
+      "step": 12
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.19068744778633118,
+      "learning_rate": 2.2221488349019903e-05,
+      "loss": 1.4965,
+      "step": 13
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.1741170436143875,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 1.4594,
+      "step": 14
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.16722294688224792,
+      "learning_rate": 8.426519384872733e-06,
+      "loss": 1.42,
+      "step": 15
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.1697998344898224,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 1.411,
+      "step": 16
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.17374582588672638,
+      "learning_rate": 9.607359798384785e-07,
+      "loss": 1.4063,
+      "step": 17
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.17088964581489563,
+      "learning_rate": 0.0,
+      "loss": 1.3964,
+      "step": 18
+    },
+    {
+      "epoch": 2.88,
+      "step": 18,
+      "total_flos": 5.97913280723288e+17,
+      "train_loss": 1.5128402180141873,
+      "train_runtime": 2593.9911,
+      "train_samples_per_second": 0.463,
+      "train_steps_per_second": 0.007
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 18,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.97913280723288e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

llama_factory/v3-20241225-1_10/README.md

@@ -0,0 +1,62 @@
+---
+library_name: peft
+license: other
+base_model: /disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: v3-20241225-1_10
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# v3-20241225-1_10
+
+This model is a fine-tuned version of [/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct](https://huggingface.co//disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct) on the M^2RAR-20241225-MLLM-1_10 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.46.1
+- Pytorch 2.3.1+cu121
+- Datasets 3.1.0
+- Tokenizers 0.20.3

llama_factory/v3-20241225-1_10/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(?!.*visual).*(?:o_proj|down_proj|q_proj|gate_proj|up_proj|v_proj|k_proj).*",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

llama_factory/v3-20241225-1_10/added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

llama_factory/v3-20241225-1_10/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9547511312217196,
+    "total_flos": 1.9929234183615939e+18,
+    "train_loss": 1.4086421551527801,
+    "train_runtime": 8849.9709,
+    "train_samples_per_second": 0.598,
+    "train_steps_per_second": 0.009
+}

llama_factory/v3-20241225-1_10/chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

llama_factory/v3-20241225-1_10/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

llama_factory/v3-20241225-1_10/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

llama_factory/v3-20241225-1_10/tokenizer_config.json

@@ -0,0 +1,144 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

llama_factory/v3-20241225-1_10/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9547511312217196,
+    "total_flos": 1.9929234183615939e+18,
+    "train_loss": 1.4086421551527801,
+    "train_runtime": 8849.9709,
+    "train_samples_per_second": 0.598,
+    "train_steps_per_second": 0.009
+}

llama_factory/v3-20241225-1_10/trainer_log.jsonl

@@ -0,0 +1,82 @@
+{"current_steps": 1, "total_steps": 81, "loss": 1.6301, "lr": 1.1111111111111112e-05, "epoch": 0.03619909502262444, "percentage": 1.23, "elapsed_time": "0:01:57", "remaining_time": "2:36:51"}
+{"current_steps": 2, "total_steps": 81, "loss": 1.6279, "lr": 2.2222222222222223e-05, "epoch": 0.07239819004524888, "percentage": 2.47, "elapsed_time": "0:03:51", "remaining_time": "2:32:14"}
+{"current_steps": 3, "total_steps": 81, "loss": 1.5965, "lr": 3.3333333333333335e-05, "epoch": 0.1085972850678733, "percentage": 3.7, "elapsed_time": "0:05:22", "remaining_time": "2:19:46"}
+{"current_steps": 4, "total_steps": 81, "loss": 1.6076, "lr": 4.4444444444444447e-05, "epoch": 0.14479638009049775, "percentage": 4.94, "elapsed_time": "0:07:16", "remaining_time": "2:19:57"}
+{"current_steps": 5, "total_steps": 81, "loss": 1.6144, "lr": 5.555555555555556e-05, "epoch": 0.18099547511312217, "percentage": 6.17, "elapsed_time": "0:09:17", "remaining_time": "2:21:09"}
+{"current_steps": 6, "total_steps": 81, "loss": 1.5657, "lr": 6.666666666666667e-05, "epoch": 0.2171945701357466, "percentage": 7.41, "elapsed_time": "0:11:14", "remaining_time": "2:20:31"}
+{"current_steps": 7, "total_steps": 81, "loss": 1.5651, "lr": 7.777777777777778e-05, "epoch": 0.25339366515837103, "percentage": 8.64, "elapsed_time": "0:12:54", "remaining_time": "2:16:29"}
+{"current_steps": 8, "total_steps": 81, "loss": 1.5488, "lr": 8.888888888888889e-05, "epoch": 0.2895927601809955, "percentage": 9.88, "elapsed_time": "0:14:46", "remaining_time": "2:14:46"}
+{"current_steps": 9, "total_steps": 81, "loss": 1.5898, "lr": 0.0001, "epoch": 0.3257918552036199, "percentage": 11.11, "elapsed_time": "0:16:59", "remaining_time": "2:15:59"}
+{"current_steps": 10, "total_steps": 81, "loss": 1.5448, "lr": 9.99524110790929e-05, "epoch": 0.36199095022624433, "percentage": 12.35, "elapsed_time": "0:18:38", "remaining_time": "2:12:19"}
+{"current_steps": 11, "total_steps": 81, "loss": 1.5285, "lr": 9.980973490458728e-05, "epoch": 0.39819004524886875, "percentage": 13.58, "elapsed_time": "0:20:38", "remaining_time": "2:11:19"}
+{"current_steps": 12, "total_steps": 81, "loss": 1.5645, "lr": 9.957224306869053e-05, "epoch": 0.4343891402714932, "percentage": 14.81, "elapsed_time": "0:21:58", "remaining_time": "2:06:21"}
+{"current_steps": 13, "total_steps": 81, "loss": 1.4987, "lr": 9.924038765061042e-05, "epoch": 0.47058823529411764, "percentage": 16.05, "elapsed_time": "0:23:58", "remaining_time": "2:05:23"}
+{"current_steps": 14, "total_steps": 81, "loss": 1.4982, "lr": 9.881480035599667e-05, "epoch": 0.5067873303167421, "percentage": 17.28, "elapsed_time": "0:25:56", "remaining_time": "2:04:08"}
+{"current_steps": 15, "total_steps": 81, "loss": 1.5061, "lr": 9.829629131445342e-05, "epoch": 0.5429864253393665, "percentage": 18.52, "elapsed_time": "0:27:36", "remaining_time": "2:01:29"}
+{"current_steps": 16, "total_steps": 81, "loss": 1.4963, "lr": 9.768584753741134e-05, "epoch": 0.579185520361991, "percentage": 19.75, "elapsed_time": "0:29:04", "remaining_time": "1:58:08"}
+{"current_steps": 17, "total_steps": 81, "loss": 1.5242, "lr": 9.698463103929542e-05, "epoch": 0.6153846153846154, "percentage": 20.99, "elapsed_time": "0:31:04", "remaining_time": "1:56:57"}
+{"current_steps": 18, "total_steps": 81, "loss": 1.49, "lr": 9.619397662556435e-05, "epoch": 0.6515837104072398, "percentage": 22.22, "elapsed_time": "0:33:00", "remaining_time": "1:55:32"}
+{"current_steps": 19, "total_steps": 81, "loss": 1.5342, "lr": 9.53153893518325e-05, "epoch": 0.6877828054298643, "percentage": 23.46, "elapsed_time": "0:34:40", "remaining_time": "1:53:07"}
+{"current_steps": 20, "total_steps": 81, "loss": 1.5256, "lr": 9.435054165891109e-05, "epoch": 0.7239819004524887, "percentage": 24.69, "elapsed_time": "0:36:06", "remaining_time": "1:50:07"}
+{"current_steps": 21, "total_steps": 81, "loss": 1.5167, "lr": 9.330127018922194e-05, "epoch": 0.7601809954751131, "percentage": 25.93, "elapsed_time": "0:37:59", "remaining_time": "1:48:33"}
+{"current_steps": 22, "total_steps": 81, "loss": 1.5255, "lr": 9.21695722906443e-05, "epoch": 0.7963800904977375, "percentage": 27.16, "elapsed_time": "0:39:30", "remaining_time": "1:45:57"}
+{"current_steps": 23, "total_steps": 81, "loss": 1.5062, "lr": 9.09576022144496e-05, "epoch": 0.832579185520362, "percentage": 28.4, "elapsed_time": "0:41:33", "remaining_time": "1:44:47"}
+{"current_steps": 24, "total_steps": 81, "loss": 1.4894, "lr": 8.966766701456177e-05, "epoch": 0.8687782805429864, "percentage": 29.63, "elapsed_time": "0:43:16", "remaining_time": "1:42:46"}
+{"current_steps": 25, "total_steps": 81, "loss": 1.4553, "lr": 8.83022221559489e-05, "epoch": 0.9049773755656109, "percentage": 30.86, "elapsed_time": "0:45:18", "remaining_time": "1:41:29"}
+{"current_steps": 26, "total_steps": 81, "loss": 1.4824, "lr": 8.68638668405062e-05, "epoch": 0.9411764705882353, "percentage": 32.1, "elapsed_time": "0:47:17", "remaining_time": "1:40:03"}
+{"current_steps": 27, "total_steps": 81, "loss": 1.5244, "lr": 8.535533905932738e-05, "epoch": 0.9773755656108597, "percentage": 33.33, "elapsed_time": "0:49:09", "remaining_time": "1:38:18"}
+{"current_steps": 28, "total_steps": 81, "loss": 1.4473, "lr": 8.377951038078302e-05, "epoch": 1.0248868778280542, "percentage": 34.57, "elapsed_time": "0:50:46", "remaining_time": "1:36:06"}
+{"current_steps": 29, "total_steps": 81, "loss": 1.4101, "lr": 8.213938048432697e-05, "epoch": 1.0610859728506787, "percentage": 35.8, "elapsed_time": "0:52:32", "remaining_time": "1:34:13"}
+{"current_steps": 30, "total_steps": 81, "loss": 1.4621, "lr": 8.043807145043604e-05, "epoch": 1.0972850678733033, "percentage": 37.04, "elapsed_time": "0:54:57", "remaining_time": "1:33:26"}
+{"current_steps": 31, "total_steps": 81, "loss": 1.4011, "lr": 7.86788218175523e-05, "epoch": 1.1334841628959276, "percentage": 38.27, "elapsed_time": "0:56:27", "remaining_time": "1:31:03"}
+{"current_steps": 32, "total_steps": 81, "loss": 1.4529, "lr": 7.68649804173412e-05, "epoch": 1.169683257918552, "percentage": 39.51, "elapsed_time": "0:58:11", "remaining_time": "1:29:06"}
+{"current_steps": 33, "total_steps": 81, "loss": 1.4195, "lr": 7.500000000000001e-05, "epoch": 1.2058823529411764, "percentage": 40.74, "elapsed_time": "0:59:41", "remaining_time": "1:26:48"}
+{"current_steps": 34, "total_steps": 81, "loss": 1.4106, "lr": 7.308743066175172e-05, "epoch": 1.242081447963801, "percentage": 41.98, "elapsed_time": "1:01:39", "remaining_time": "1:25:14"}
+{"current_steps": 35, "total_steps": 81, "loss": 1.3482, "lr": 7.113091308703498e-05, "epoch": 1.2782805429864252, "percentage": 43.21, "elapsed_time": "1:03:15", "remaining_time": "1:23:08"}
+{"current_steps": 36, "total_steps": 81, "loss": 1.4463, "lr": 6.91341716182545e-05, "epoch": 1.3144796380090498, "percentage": 44.44, "elapsed_time": "1:05:39", "remaining_time": "1:22:03"}
+{"current_steps": 37, "total_steps": 81, "loss": 1.3528, "lr": 6.710100716628344e-05, "epoch": 1.3506787330316743, "percentage": 45.68, "elapsed_time": "1:07:30", "remaining_time": "1:20:17"}
+{"current_steps": 38, "total_steps": 81, "loss": 1.3881, "lr": 6.503528997521366e-05, "epoch": 1.3868778280542986, "percentage": 46.91, "elapsed_time": "1:09:10", "remaining_time": "1:18:16"}
+{"current_steps": 39, "total_steps": 81, "loss": 1.374, "lr": 6.294095225512603e-05, "epoch": 1.4230769230769231, "percentage": 48.15, "elapsed_time": "1:10:59", "remaining_time": "1:16:27"}
+{"current_steps": 40, "total_steps": 81, "loss": 1.3717, "lr": 6.0821980696905146e-05, "epoch": 1.4592760180995474, "percentage": 49.38, "elapsed_time": "1:12:39", "remaining_time": "1:14:28"}
+{"current_steps": 41, "total_steps": 81, "loss": 1.37, "lr": 5.868240888334653e-05, "epoch": 1.495475113122172, "percentage": 50.62, "elapsed_time": "1:14:09", "remaining_time": "1:12:21"}
+{"current_steps": 42, "total_steps": 81, "loss": 1.359, "lr": 5.6526309611002594e-05, "epoch": 1.5316742081447963, "percentage": 51.85, "elapsed_time": "1:15:42", "remaining_time": "1:10:17"}
+{"current_steps": 43, "total_steps": 81, "loss": 1.421, "lr": 5.435778713738292e-05, "epoch": 1.5678733031674208, "percentage": 53.09, "elapsed_time": "1:17:41", "remaining_time": "1:08:39"}
+{"current_steps": 44, "total_steps": 81, "loss": 1.4027, "lr": 5.218096936826681e-05, "epoch": 1.6040723981900453, "percentage": 54.32, "elapsed_time": "1:19:38", "remaining_time": "1:06:58"}
+{"current_steps": 45, "total_steps": 81, "loss": 1.3634, "lr": 5e-05, "epoch": 1.6402714932126696, "percentage": 55.56, "elapsed_time": "1:21:20", "remaining_time": "1:05:04"}
+{"current_steps": 46, "total_steps": 81, "loss": 1.361, "lr": 4.781903063173321e-05, "epoch": 1.6764705882352942, "percentage": 56.79, "elapsed_time": "1:22:56", "remaining_time": "1:03:06"}
+{"current_steps": 47, "total_steps": 81, "loss": 1.3613, "lr": 4.564221286261709e-05, "epoch": 1.7126696832579187, "percentage": 58.02, "elapsed_time": "1:24:43", "remaining_time": "1:01:17"}
+{"current_steps": 48, "total_steps": 81, "loss": 1.367, "lr": 4.347369038899744e-05, "epoch": 1.748868778280543, "percentage": 59.26, "elapsed_time": "1:26:23", "remaining_time": "0:59:23"}
+{"current_steps": 49, "total_steps": 81, "loss": 1.3626, "lr": 4.131759111665349e-05, "epoch": 1.7850678733031673, "percentage": 60.49, "elapsed_time": "1:28:19", "remaining_time": "0:57:40"}
+{"current_steps": 50, "total_steps": 81, "loss": 1.3847, "lr": 3.917801930309486e-05, "epoch": 1.8212669683257918, "percentage": 61.73, "elapsed_time": "1:30:31", "remaining_time": "0:56:07"}
+{"current_steps": 51, "total_steps": 81, "loss": 1.3584, "lr": 3.705904774487396e-05, "epoch": 1.8574660633484164, "percentage": 62.96, "elapsed_time": "1:32:36", "remaining_time": "0:54:28"}
+{"current_steps": 52, "total_steps": 81, "loss": 1.393, "lr": 3.4964710024786354e-05, "epoch": 1.8936651583710407, "percentage": 64.2, "elapsed_time": "1:34:40", "remaining_time": "0:52:47"}
+{"current_steps": 53, "total_steps": 81, "loss": 1.3333, "lr": 3.289899283371657e-05, "epoch": 1.9298642533936652, "percentage": 65.43, "elapsed_time": "1:36:20", "remaining_time": "0:50:53"}
+{"current_steps": 54, "total_steps": 81, "loss": 1.3567, "lr": 3.086582838174551e-05, "epoch": 1.9660633484162897, "percentage": 66.67, "elapsed_time": "1:38:07", "remaining_time": "0:49:03"}
+{"current_steps": 55, "total_steps": 81, "loss": 1.3388, "lr": 2.886908691296504e-05, "epoch": 2.013574660633484, "percentage": 67.9, "elapsed_time": "1:40:11", "remaining_time": "0:47:21"}
+{"current_steps": 56, "total_steps": 81, "loss": 1.3288, "lr": 2.6912569338248315e-05, "epoch": 2.0497737556561084, "percentage": 69.14, "elapsed_time": "1:41:57", "remaining_time": "0:45:31"}
+{"current_steps": 57, "total_steps": 81, "loss": 1.2998, "lr": 2.500000000000001e-05, "epoch": 2.085972850678733, "percentage": 70.37, "elapsed_time": "1:43:55", "remaining_time": "0:43:45"}
+{"current_steps": 58, "total_steps": 81, "loss": 1.2406, "lr": 2.3135019582658802e-05, "epoch": 2.1221719457013575, "percentage": 71.6, "elapsed_time": "1:45:51", "remaining_time": "0:41:58"}
+{"current_steps": 59, "total_steps": 81, "loss": 1.2579, "lr": 2.132117818244771e-05, "epoch": 2.158371040723982, "percentage": 72.84, "elapsed_time": "1:47:46", "remaining_time": "0:40:11"}
+{"current_steps": 60, "total_steps": 81, "loss": 1.2737, "lr": 1.9561928549563968e-05, "epoch": 2.1945701357466065, "percentage": 74.07, "elapsed_time": "1:49:12", "remaining_time": "0:38:13"}
+{"current_steps": 61, "total_steps": 81, "loss": 1.3014, "lr": 1.7860619515673033e-05, "epoch": 2.230769230769231, "percentage": 75.31, "elapsed_time": "1:51:16", "remaining_time": "0:36:28"}
+{"current_steps": 62, "total_steps": 81, "loss": 1.3528, "lr": 1.622048961921699e-05, "epoch": 2.266968325791855, "percentage": 76.54, "elapsed_time": "1:53:05", "remaining_time": "0:34:39"}
+{"current_steps": 63, "total_steps": 81, "loss": 1.2639, "lr": 1.4644660940672627e-05, "epoch": 2.3031674208144794, "percentage": 77.78, "elapsed_time": "1:54:54", "remaining_time": "0:32:49"}
+{"current_steps": 64, "total_steps": 81, "loss": 1.3244, "lr": 1.3136133159493802e-05, "epoch": 2.339366515837104, "percentage": 79.01, "elapsed_time": "1:57:14", "remaining_time": "0:31:08"}
+{"current_steps": 65, "total_steps": 81, "loss": 1.3534, "lr": 1.1697777844051105e-05, "epoch": 2.3755656108597285, "percentage": 80.25, "elapsed_time": "1:59:18", "remaining_time": "0:29:22"}
+{"current_steps": 66, "total_steps": 81, "loss": 1.2637, "lr": 1.0332332985438248e-05, "epoch": 2.411764705882353, "percentage": 81.48, "elapsed_time": "2:00:39", "remaining_time": "0:27:25"}
+{"current_steps": 67, "total_steps": 81, "loss": 1.3538, "lr": 9.042397785550405e-06, "epoch": 2.4479638009049776, "percentage": 82.72, "elapsed_time": "2:02:40", "remaining_time": "0:25:37"}
+{"current_steps": 68, "total_steps": 81, "loss": 1.3247, "lr": 7.830427709355725e-06, "epoch": 2.484162895927602, "percentage": 83.95, "elapsed_time": "2:04:17", "remaining_time": "0:23:45"}
+{"current_steps": 69, "total_steps": 81, "loss": 1.2721, "lr": 6.698729810778065e-06, "epoch": 2.520361990950226, "percentage": 85.19, "elapsed_time": "2:06:07", "remaining_time": "0:21:56"}
+{"current_steps": 70, "total_steps": 81, "loss": 1.2966, "lr": 5.649458341088915e-06, "epoch": 2.5565610859728505, "percentage": 86.42, "elapsed_time": "2:07:32", "remaining_time": "0:20:02"}
+{"current_steps": 71, "total_steps": 81, "loss": 1.2594, "lr": 4.684610648167503e-06, "epoch": 2.5927601809954752, "percentage": 87.65, "elapsed_time": "2:09:29", "remaining_time": "0:18:14"}
+{"current_steps": 72, "total_steps": 81, "loss": 1.2951, "lr": 3.8060233744356633e-06, "epoch": 2.6289592760180995, "percentage": 88.89, "elapsed_time": "2:11:02", "remaining_time": "0:16:22"}
+{"current_steps": 73, "total_steps": 81, "loss": 1.3408, "lr": 3.0153689607045845e-06, "epoch": 2.665158371040724, "percentage": 90.12, "elapsed_time": "2:13:04", "remaining_time": "0:14:35"}
+{"current_steps": 74, "total_steps": 81, "loss": 1.2951, "lr": 2.314152462588659e-06, "epoch": 2.7013574660633486, "percentage": 91.36, "elapsed_time": "2:15:00", "remaining_time": "0:12:46"}
+{"current_steps": 75, "total_steps": 81, "loss": 1.2927, "lr": 1.70370868554659e-06, "epoch": 2.737556561085973, "percentage": 92.59, "elapsed_time": "2:16:56", "remaining_time": "0:10:57"}
+{"current_steps": 76, "total_steps": 81, "loss": 1.3133, "lr": 1.1851996440033319e-06, "epoch": 2.773755656108597, "percentage": 93.83, "elapsed_time": "2:18:48", "remaining_time": "0:09:07"}
+{"current_steps": 77, "total_steps": 81, "loss": 1.3135, "lr": 7.596123493895991e-07, "epoch": 2.8099547511312215, "percentage": 95.06, "elapsed_time": "2:20:21", "remaining_time": "0:07:17"}
+{"current_steps": 78, "total_steps": 81, "loss": 1.2631, "lr": 4.277569313094809e-07, "epoch": 2.8461538461538463, "percentage": 96.3, "elapsed_time": "2:22:03", "remaining_time": "0:05:27"}
+{"current_steps": 79, "total_steps": 81, "loss": 1.2773, "lr": 1.9026509541272275e-07, "epoch": 2.8823529411764706, "percentage": 97.53, "elapsed_time": "2:23:50", "remaining_time": "0:03:38"}
+{"current_steps": 80, "total_steps": 81, "loss": 1.2997, "lr": 4.7588920907110094e-08, "epoch": 2.918552036199095, "percentage": 98.77, "elapsed_time": "2:25:32", "remaining_time": "0:01:49"}
+{"current_steps": 81, "total_steps": 81, "loss": 1.268, "lr": 0.0, "epoch": 2.9547511312217196, "percentage": 100.0, "elapsed_time": "2:26:50", "remaining_time": "0:00:00"}
+{"current_steps": 81, "total_steps": 81, "epoch": 2.9547511312217196, "percentage": 100.0, "elapsed_time": "2:27:29", "remaining_time": "0:00:00"}

llama_factory/v3-20241225-1_10/trainer_state.json

@@ -0,0 +1,609 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9547511312217196,
+  "eval_steps": 500,
+  "global_step": 81,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03619909502262444,
+      "grad_norm": 0.4522567093372345,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 1.6301,
+      "step": 1
+    },
+    {
+      "epoch": 0.07239819004524888,
+      "grad_norm": 0.4915740489959717,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 1.6279,
+      "step": 2
+    },
+    {
+      "epoch": 0.1085972850678733,
+      "grad_norm": 0.4293298125267029,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 1.5965,
+      "step": 3
+    },
+    {
+      "epoch": 0.14479638009049775,
+      "grad_norm": 0.30257710814476013,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 1.6076,
+      "step": 4
+    },
+    {
+      "epoch": 0.18099547511312217,
+      "grad_norm": 0.338920921087265,
+      "learning_rate": 5.555555555555556e-05,
+      "loss": 1.6144,
+      "step": 5
+    },
+    {
+      "epoch": 0.2171945701357466,
+      "grad_norm": 0.393881618976593,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 1.5657,
+      "step": 6
+    },
+    {
+      "epoch": 0.25339366515837103,
+      "grad_norm": 0.2913854718208313,
+      "learning_rate": 7.777777777777778e-05,
+      "loss": 1.5651,
+      "step": 7
+    },
+    {
+      "epoch": 0.2895927601809955,
+      "grad_norm": 0.2615966498851776,
+      "learning_rate": 8.888888888888889e-05,
+      "loss": 1.5488,
+      "step": 8
+    },
+    {
+      "epoch": 0.3257918552036199,
+      "grad_norm": 0.2777751386165619,
+      "learning_rate": 0.0001,
+      "loss": 1.5898,
+      "step": 9
+    },
+    {
+      "epoch": 0.36199095022624433,
+      "grad_norm": 0.24253539741039276,
+      "learning_rate": 9.99524110790929e-05,
+      "loss": 1.5448,
+      "step": 10
+    },
+    {
+      "epoch": 0.39819004524886875,
+      "grad_norm": 0.21929268538951874,
+      "learning_rate": 9.980973490458728e-05,
+      "loss": 1.5285,
+      "step": 11
+    },
+    {
+      "epoch": 0.4343891402714932,
+      "grad_norm": 0.2206713855266571,
+      "learning_rate": 9.957224306869053e-05,
+      "loss": 1.5645,
+      "step": 12
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.19960802793502808,
+      "learning_rate": 9.924038765061042e-05,
+      "loss": 1.4987,
+      "step": 13
+    },
+    {
+      "epoch": 0.5067873303167421,
+      "grad_norm": 0.20623129606246948,
+      "learning_rate": 9.881480035599667e-05,
+      "loss": 1.4982,
+      "step": 14
+    },
+    {
+      "epoch": 0.5429864253393665,
+      "grad_norm": 0.2134510725736618,
+      "learning_rate": 9.829629131445342e-05,
+      "loss": 1.5061,
+      "step": 15
+    },
+    {
+      "epoch": 0.579185520361991,
+      "grad_norm": 0.20414741337299347,
+      "learning_rate": 9.768584753741134e-05,
+      "loss": 1.4963,
+      "step": 16
+    },
+    {
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.19349144399166107,
+      "learning_rate": 9.698463103929542e-05,
+      "loss": 1.5242,
+      "step": 17
+    },
+    {
+      "epoch": 0.6515837104072398,
+      "grad_norm": 0.19233722984790802,
+      "learning_rate": 9.619397662556435e-05,
+      "loss": 1.49,
+      "step": 18
+    },
+    {
+      "epoch": 0.6877828054298643,
+      "grad_norm": 0.1920696496963501,
+      "learning_rate": 9.53153893518325e-05,
+      "loss": 1.5342,
+      "step": 19
+    },
+    {
+      "epoch": 0.7239819004524887,
+      "grad_norm": 0.18896055221557617,
+      "learning_rate": 9.435054165891109e-05,
+      "loss": 1.5256,
+      "step": 20
+    },
+    {
+      "epoch": 0.7601809954751131,
+      "grad_norm": 0.18282486498355865,
+      "learning_rate": 9.330127018922194e-05,
+      "loss": 1.5167,
+      "step": 21
+    },
+    {
+      "epoch": 0.7963800904977375,
+      "grad_norm": 0.19715480506420135,
+      "learning_rate": 9.21695722906443e-05,
+      "loss": 1.5255,
+      "step": 22
+    },
+    {
+      "epoch": 0.832579185520362,
+      "grad_norm": 0.17317330837249756,
+      "learning_rate": 9.09576022144496e-05,
+      "loss": 1.5062,
+      "step": 23
+    },
+    {
+      "epoch": 0.8687782805429864,
+      "grad_norm": 0.18779075145721436,
+      "learning_rate": 8.966766701456177e-05,
+      "loss": 1.4894,
+      "step": 24
+    },
+    {
+      "epoch": 0.9049773755656109,
+      "grad_norm": 0.1841159462928772,
+      "learning_rate": 8.83022221559489e-05,
+      "loss": 1.4553,
+      "step": 25
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.1778552532196045,
+      "learning_rate": 8.68638668405062e-05,
+      "loss": 1.4824,
+      "step": 26
+    },
+    {
+      "epoch": 0.9773755656108597,
+      "grad_norm": 0.18053092062473297,
+      "learning_rate": 8.535533905932738e-05,
+      "loss": 1.5244,
+      "step": 27
+    },
+    {
+      "epoch": 1.0248868778280542,
+      "grad_norm": 0.17003145813941956,
+      "learning_rate": 8.377951038078302e-05,
+      "loss": 1.4473,
+      "step": 28
+    },
+    {
+      "epoch": 1.0610859728506787,
+      "grad_norm": 0.1815570592880249,
+      "learning_rate": 8.213938048432697e-05,
+      "loss": 1.4101,
+      "step": 29
+    },
+    {
+      "epoch": 1.0972850678733033,
+      "grad_norm": 0.17359577119350433,
+      "learning_rate": 8.043807145043604e-05,
+      "loss": 1.4621,
+      "step": 30
+    },
+    {
+      "epoch": 1.1334841628959276,
+      "grad_norm": 0.18072284758090973,
+      "learning_rate": 7.86788218175523e-05,
+      "loss": 1.4011,
+      "step": 31
+    },
+    {
+      "epoch": 1.169683257918552,
+      "grad_norm": 0.18027463555335999,
+      "learning_rate": 7.68649804173412e-05,
+      "loss": 1.4529,
+      "step": 32
+    },
+    {
+      "epoch": 1.2058823529411764,
+      "grad_norm": 0.18900279700756073,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 1.4195,
+      "step": 33
+    },
+    {
+      "epoch": 1.242081447963801,
+      "grad_norm": 0.17625436186790466,
+      "learning_rate": 7.308743066175172e-05,
+      "loss": 1.4106,
+      "step": 34
+    },
+    {
+      "epoch": 1.2782805429864252,
+      "grad_norm": 0.18460573256015778,
+      "learning_rate": 7.113091308703498e-05,
+      "loss": 1.3482,
+      "step": 35
+    },
+    {
+      "epoch": 1.3144796380090498,
+      "grad_norm": 0.1863405853509903,
+      "learning_rate": 6.91341716182545e-05,
+      "loss": 1.4463,
+      "step": 36
+    },
+    {
+      "epoch": 1.3506787330316743,
+      "grad_norm": 0.188517764210701,
+      "learning_rate": 6.710100716628344e-05,
+      "loss": 1.3528,
+      "step": 37
+    },
+    {
+      "epoch": 1.3868778280542986,
+      "grad_norm": 0.1890035718679428,
+      "learning_rate": 6.503528997521366e-05,
+      "loss": 1.3881,
+      "step": 38
+    },
+    {
+      "epoch": 1.4230769230769231,
+      "grad_norm": 0.19418419897556305,
+      "learning_rate": 6.294095225512603e-05,
+      "loss": 1.374,
+      "step": 39
+    },
+    {
+      "epoch": 1.4592760180995474,
+      "grad_norm": 0.1959460824728012,
+      "learning_rate": 6.0821980696905146e-05,
+      "loss": 1.3717,
+      "step": 40
+    },
+    {
+      "epoch": 1.495475113122172,
+      "grad_norm": 0.20840130746364594,
+      "learning_rate": 5.868240888334653e-05,
+      "loss": 1.37,
+      "step": 41
+    },
+    {
+      "epoch": 1.5316742081447963,
+      "grad_norm": 0.21801875531673431,
+      "learning_rate": 5.6526309611002594e-05,
+      "loss": 1.359,
+      "step": 42
+    },
+    {
+      "epoch": 1.5678733031674208,
+      "grad_norm": 0.20987379550933838,
+      "learning_rate": 5.435778713738292e-05,
+      "loss": 1.421,
+      "step": 43
+    },
+    {
+      "epoch": 1.6040723981900453,
+      "grad_norm": 0.19880157709121704,
+      "learning_rate": 5.218096936826681e-05,
+      "loss": 1.4027,
+      "step": 44
+    },
+    {
+      "epoch": 1.6402714932126696,
+      "grad_norm": 0.20954720675945282,
+      "learning_rate": 5e-05,
+      "loss": 1.3634,
+      "step": 45
+    },
+    {
+      "epoch": 1.6764705882352942,
+      "grad_norm": 0.20119519531726837,
+      "learning_rate": 4.781903063173321e-05,
+      "loss": 1.361,
+      "step": 46
+    },
+    {
+      "epoch": 1.7126696832579187,
+      "grad_norm": 0.20504230260849,
+      "learning_rate": 4.564221286261709e-05,
+      "loss": 1.3613,
+      "step": 47
+    },
+    {
+      "epoch": 1.748868778280543,
+      "grad_norm": 0.20827656984329224,
+      "learning_rate": 4.347369038899744e-05,
+      "loss": 1.367,
+      "step": 48
+    },
+    {
+      "epoch": 1.7850678733031673,
+      "grad_norm": 0.21221719682216644,
+      "learning_rate": 4.131759111665349e-05,
+      "loss": 1.3626,
+      "step": 49
+    },
+    {
+      "epoch": 1.8212669683257918,
+      "grad_norm": 0.21247024834156036,
+      "learning_rate": 3.917801930309486e-05,
+      "loss": 1.3847,
+      "step": 50
+    },
+    {
+      "epoch": 1.8574660633484164,
+      "grad_norm": 0.21873697638511658,
+      "learning_rate": 3.705904774487396e-05,
+      "loss": 1.3584,
+      "step": 51
+    },
+    {
+      "epoch": 1.8936651583710407,
+      "grad_norm": 0.21476612985134125,
+      "learning_rate": 3.4964710024786354e-05,
+      "loss": 1.393,
+      "step": 52
+    },
+    {
+      "epoch": 1.9298642533936652,
+      "grad_norm": 0.21541458368301392,
+      "learning_rate": 3.289899283371657e-05,
+      "loss": 1.3333,
+      "step": 53
+    },
+    {
+      "epoch": 1.9660633484162897,
+      "grad_norm": 0.2195931226015091,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 1.3567,
+      "step": 54
+    },
+    {
+      "epoch": 2.013574660633484,
+      "grad_norm": 0.22267796099185944,
+      "learning_rate": 2.886908691296504e-05,
+      "loss": 1.3388,
+      "step": 55
+    },
+    {
+      "epoch": 2.0497737556561084,
+      "grad_norm": 0.2236703336238861,
+      "learning_rate": 2.6912569338248315e-05,
+      "loss": 1.3288,
+      "step": 56
+    },
+    {
+      "epoch": 2.085972850678733,
+      "grad_norm": 0.22568458318710327,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 1.2998,
+      "step": 57
+    },
+    {
+      "epoch": 2.1221719457013575,
+      "grad_norm": 0.22315552830696106,
+      "learning_rate": 2.3135019582658802e-05,
+      "loss": 1.2406,
+      "step": 58
+    },
+    {
+      "epoch": 2.158371040723982,
+      "grad_norm": 0.22059525549411774,
+      "learning_rate": 2.132117818244771e-05,
+      "loss": 1.2579,
+      "step": 59
+    },
+    {
+      "epoch": 2.1945701357466065,
+      "grad_norm": 0.24967165291309357,
+      "learning_rate": 1.9561928549563968e-05,
+      "loss": 1.2737,
+      "step": 60
+    },
+    {
+      "epoch": 2.230769230769231,
+      "grad_norm": 0.2367183119058609,
+      "learning_rate": 1.7860619515673033e-05,
+      "loss": 1.3014,
+      "step": 61
+    },
+    {
+      "epoch": 2.266968325791855,
+      "grad_norm": 0.2517261207103729,
+      "learning_rate": 1.622048961921699e-05,
+      "loss": 1.3528,
+      "step": 62
+    },
+    {
+      "epoch": 2.3031674208144794,
+      "grad_norm": 0.235427126288414,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 1.2639,
+      "step": 63
+    },
+    {
+      "epoch": 2.339366515837104,
+      "grad_norm": 0.22943593561649323,
+      "learning_rate": 1.3136133159493802e-05,
+      "loss": 1.3244,
+      "step": 64
+    },
+    {
+      "epoch": 2.3755656108597285,
+      "grad_norm": 0.22623376548290253,
+      "learning_rate": 1.1697777844051105e-05,
+      "loss": 1.3534,
+      "step": 65
+    },
+    {
+      "epoch": 2.411764705882353,
+      "grad_norm": 0.240826815366745,
+      "learning_rate": 1.0332332985438248e-05,
+      "loss": 1.2637,
+      "step": 66
+    },
+    {
+      "epoch": 2.4479638009049776,
+      "grad_norm": 0.2174130380153656,
+      "learning_rate": 9.042397785550405e-06,
+      "loss": 1.3538,
+      "step": 67
+    },
+    {
+      "epoch": 2.484162895927602,
+      "grad_norm": 0.23589667677879333,
+      "learning_rate": 7.830427709355725e-06,
+      "loss": 1.3247,
+      "step": 68
+    },
+    {
+      "epoch": 2.520361990950226,
+      "grad_norm": 0.24112889170646667,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 1.2721,
+      "step": 69
+    },
+    {
+      "epoch": 2.5565610859728505,
+      "grad_norm": 0.23095567524433136,
+      "learning_rate": 5.649458341088915e-06,
+      "loss": 1.2966,
+      "step": 70
+    },
+    {
+      "epoch": 2.5927601809954752,
+      "grad_norm": 0.2425680309534073,
+      "learning_rate": 4.684610648167503e-06,
+      "loss": 1.2594,
+      "step": 71
+    },
+    {
+      "epoch": 2.6289592760180995,
+      "grad_norm": 0.2282722145318985,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 1.2951,
+      "step": 72
+    },
+    {
+      "epoch": 2.665158371040724,
+      "grad_norm": 0.2365112006664276,
+      "learning_rate": 3.0153689607045845e-06,
+      "loss": 1.3408,
+      "step": 73
+    },
+    {
+      "epoch": 2.7013574660633486,
+      "grad_norm": 0.23960916697978973,
+      "learning_rate": 2.314152462588659e-06,
+      "loss": 1.2951,
+      "step": 74
+    },
+    {
+      "epoch": 2.737556561085973,
+      "grad_norm": 0.23407232761383057,
+      "learning_rate": 1.70370868554659e-06,
+      "loss": 1.2927,
+      "step": 75
+    },
+    {
+      "epoch": 2.773755656108597,
+      "grad_norm": 0.23612752556800842,
+      "learning_rate": 1.1851996440033319e-06,
+      "loss": 1.3133,
+      "step": 76
+    },
+    {
+      "epoch": 2.8099547511312215,
+      "grad_norm": 0.24510255455970764,
+      "learning_rate": 7.596123493895991e-07,
+      "loss": 1.3135,
+      "step": 77
+    },
+    {
+      "epoch": 2.8461538461538463,
+      "grad_norm": 0.25504744052886963,
+      "learning_rate": 4.277569313094809e-07,
+      "loss": 1.2631,
+      "step": 78
+    },
+    {
+      "epoch": 2.8823529411764706,
+      "grad_norm": 0.22783517837524414,
+      "learning_rate": 1.9026509541272275e-07,
+      "loss": 1.2773,
+      "step": 79
+    },
+    {
+      "epoch": 2.918552036199095,
+      "grad_norm": 0.23234006762504578,
+      "learning_rate": 4.7588920907110094e-08,
+      "loss": 1.2997,
+      "step": 80
+    },
+    {
+      "epoch": 2.9547511312217196,
+      "grad_norm": 0.24794724583625793,
+      "learning_rate": 0.0,
+      "loss": 1.268,
+      "step": 81
+    },
+    {
+      "epoch": 2.9547511312217196,
+      "step": 81,
+      "total_flos": 1.9929234183615939e+18,
+      "train_loss": 1.4086421551527801,
+      "train_runtime": 8849.9709,
+      "train_samples_per_second": 0.598,
+      "train_steps_per_second": 0.009
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 81,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.9929234183615939e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

llama_factory/v3-20241225-1_5/README.md

@@ -0,0 +1,62 @@
+---
+library_name: peft
+license: other
+base_model: /disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: v3-20241225-1_5
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# v3-20241225-1_5
+
+This model is a fine-tuned version of [/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct](https://huggingface.co//disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct) on the M^2RAR-20241225-MLLM-1_5 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.46.1
+- Pytorch 2.3.1+cu121
+- Datasets 3.1.0
+- Tokenizers 0.20.3

llama_factory/v3-20241225-1_5/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/disk/maziao/model-zoo/hf-repo/models--Qwen--Qwen2-VL-7B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(?!.*visual).*(?:q_proj|down_proj|k_proj|up_proj|gate_proj|v_proj|o_proj).*",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

llama_factory/v3-20241225-1_5/added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

llama_factory/v3-20241225-1_5/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.987551867219917,
+    "total_flos": 2.252458950505005e+18,
+    "train_loss": 1.4155288457870483,
+    "train_runtime": 9888.3803,
+    "train_samples_per_second": 0.584,
+    "train_steps_per_second": 0.009
+}

llama_factory/v3-20241225-1_5/chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

llama_factory/v3-20241225-1_5/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}