Training in progress, step 100

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/README.md

@@ -0,0 +1,59 @@
+---
+datasets: /mnt/petrelfs/songmingyang/songmingyang/data/mm/reasoning/Clevr_CoGenT_TrainA_R1
+library_name: transformers
+model_name: Qwen2-VL-2B-Instruct-SFT
+tags:
+- generated_from_trainer
+- R1-V
+- trl
+- sft
+licence: license
+---
+
+# Model Card for Qwen2-VL-2B-Instruct-SFT
+
+This model is a fine-tuned version of [None](https://huggingface.co/None) on the [/mnt/petrelfs/songmingyang/songmingyang/data/mm/reasoning/Clevr_CoGenT_TrainA_R1](https://huggingface.co/datasets//mnt/petrelfs/songmingyang/songmingyang/data/mm/reasoning/Clevr_CoGenT_TrainA_R1) dataset.
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="hitsmy/Qwen2-VL-2B-Instruct-SFT", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/pjlab_moe/r1_v/runs/t2fin07w) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.14.0
+- Transformers: 4.49.0.dev0
+- Pytorch: 2.6.0
+- Datasets: 3.3.1
+- Tokenizers: 0.21.0
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/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
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9995764506565015,
+    "total_flos": 8.380405809686774e+17,
+    "train_loss": 0.5979039826635587,
+    "train_runtime": 5676.8996,
+    "train_samples": 37773,
+    "train_samples_per_second": 6.654,
+    "train_steps_per_second": 0.104
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/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 %}"
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/config.json

@@ -0,0 +1,49 @@
+{
+  "_name_or_path": "/mnt/petrelfs/songmingyang/songmingyang/model/mm/Qwen2-VL-2B-Instruct",
+  "architectures": [
+    "Qwen2VLForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_token_id": 151655,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2_vl",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": {
+    "mrope_section": [
+      16,
+      24,
+      24
+    ],
+    "rope_type": "default",
+    "type": "default"
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.49.0.dev0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "hidden_size": 1536,
+    "in_chans": 3,
+    "model_type": "qwen2_vl",
+    "spatial_patch_size": 14,
+    "torch_dtype": "bfloat16"
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 151936
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/generation_config.json

@@ -0,0 +1,15 @@
+{
+  "attn_implementation": null,
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.01,
+  "top_k": 1,
+  "top_p": 0.001,
+  "transformers_version": "4.49.0.dev0",
+  "use_cache": false
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93ba58e6c790bf087c8d5f6d7404a72eac1ce870fa43c54f9fb37e7cb4342390
+size 4884798456

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/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": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/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
+  }
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88a3a6fcb80132f76da8aa40cdc3fccd7e5d8468ef15421f5b0c2715e85217d2
+size 11420538

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/tokenizer_config.json

@@ -0,0 +1,145 @@
+{
+  "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",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "Qwen2VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9995764506565015,
+    "total_flos": 8.380405809686774e+17,
+    "train_loss": 0.5979039826635587,
+    "train_runtime": 5676.8996,
+    "train_samples": 37773,
+    "train_samples_per_second": 6.654,
+    "train_steps_per_second": 0.104
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/trainer_state.json

@@ -0,0 +1,868 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9995764506565015,
+  "eval_steps": 500,
+  "global_step": 590,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00847098686997035,
+      "grad_norm": 13.040608406066895,
+      "learning_rate": 1.6949152542372882e-06,
+      "loss": 1.8317,
+      "step": 5
+    },
+    {
+      "epoch": 0.0169419737399407,
+      "grad_norm": 7.28167200088501,
+      "learning_rate": 3.3898305084745763e-06,
+      "loss": 1.7467,
+      "step": 10
+    },
+    {
+      "epoch": 0.025412960609911054,
+      "grad_norm": 3.6362063884735107,
+      "learning_rate": 5.084745762711865e-06,
+      "loss": 1.4691,
+      "step": 15
+    },
+    {
+      "epoch": 0.0338839474798814,
+      "grad_norm": 2.443824291229248,
+      "learning_rate": 6.779661016949153e-06,
+      "loss": 1.1816,
+      "step": 20
+    },
+    {
+      "epoch": 0.042354934349851756,
+      "grad_norm": 1.9287770986557007,
+      "learning_rate": 8.47457627118644e-06,
+      "loss": 1.0028,
+      "step": 25
+    },
+    {
+      "epoch": 0.05082592121982211,
+      "grad_norm": 2.287205696105957,
+      "learning_rate": 1.016949152542373e-05,
+      "loss": 0.8979,
+      "step": 30
+    },
+    {
+      "epoch": 0.05929690808979246,
+      "grad_norm": 1.8406411409378052,
+      "learning_rate": 1.1864406779661018e-05,
+      "loss": 0.8245,
+      "step": 35
+    },
+    {
+      "epoch": 0.0677678949597628,
+      "grad_norm": 1.597153663635254,
+      "learning_rate": 1.3559322033898305e-05,
+      "loss": 0.801,
+      "step": 40
+    },
+    {
+      "epoch": 0.07623888182973317,
+      "grad_norm": 1.4034295082092285,
+      "learning_rate": 1.5254237288135594e-05,
+      "loss": 0.7545,
+      "step": 45
+    },
+    {
+      "epoch": 0.08470986869970351,
+      "grad_norm": 3.763864517211914,
+      "learning_rate": 1.694915254237288e-05,
+      "loss": 0.7336,
+      "step": 50
+    },
+    {
+      "epoch": 0.09318085556967387,
+      "grad_norm": 1.451323390007019,
+      "learning_rate": 1.864406779661017e-05,
+      "loss": 0.7121,
+      "step": 55
+    },
+    {
+      "epoch": 0.10165184243964422,
+      "grad_norm": 1.5066395998001099,
+      "learning_rate": 1.9999824983320176e-05,
+      "loss": 0.6881,
+      "step": 60
+    },
+    {
+      "epoch": 0.11012282930961458,
+      "grad_norm": 1.4491875171661377,
+      "learning_rate": 1.9993700042749937e-05,
+      "loss": 0.673,
+      "step": 65
+    },
+    {
+      "epoch": 0.11859381617958492,
+      "grad_norm": 1.27461838722229,
+      "learning_rate": 1.9978830393392338e-05,
+      "loss": 0.6507,
+      "step": 70
+    },
+    {
+      "epoch": 0.12706480304955528,
+      "grad_norm": 1.191027045249939,
+      "learning_rate": 1.995522904651977e-05,
+      "loss": 0.6384,
+      "step": 75
+    },
+    {
+      "epoch": 0.1355357899195256,
+      "grad_norm": 1.237403392791748,
+      "learning_rate": 1.992291665383325e-05,
+      "loss": 0.6637,
+      "step": 80
+    },
+    {
+      "epoch": 0.14400677678949597,
+      "grad_norm": 1.1272355318069458,
+      "learning_rate": 1.9881921489391738e-05,
+      "loss": 0.635,
+      "step": 85
+    },
+    {
+      "epoch": 0.15247776365946633,
+      "grad_norm": 1.168715476989746,
+      "learning_rate": 1.983227942487172e-05,
+      "loss": 0.6445,
+      "step": 90
+    },
+    {
+      "epoch": 0.1609487505294367,
+      "grad_norm": 1.050166130065918,
+      "learning_rate": 1.9774033898178668e-05,
+      "loss": 0.6412,
+      "step": 95
+    },
+    {
+      "epoch": 0.16941973739940702,
+      "grad_norm": 1.268314003944397,
+      "learning_rate": 1.9707235875437932e-05,
+      "loss": 0.6233,
+      "step": 100
+    },
+    {
+      "epoch": 0.17789072426937738,
+      "grad_norm": 1.3509944677352905,
+      "learning_rate": 1.963194380639825e-05,
+      "loss": 0.6138,
+      "step": 105
+    },
+    {
+      "epoch": 0.18636171113934774,
+      "grad_norm": 1.1367347240447998,
+      "learning_rate": 1.954822357328692e-05,
+      "loss": 0.6045,
+      "step": 110
+    },
+    {
+      "epoch": 0.19483269800931807,
+      "grad_norm": 1.1019567251205444,
+      "learning_rate": 1.9456148433161387e-05,
+      "loss": 0.6176,
+      "step": 115
+    },
+    {
+      "epoch": 0.20330368487928843,
+      "grad_norm": 1.0358731746673584,
+      "learning_rate": 1.9355798953807715e-05,
+      "loss": 0.5925,
+      "step": 120
+    },
+    {
+      "epoch": 0.2117746717492588,
+      "grad_norm": 1.0830167531967163,
+      "learning_rate": 1.924726294324196e-05,
+      "loss": 0.6107,
+      "step": 125
+    },
+    {
+      "epoch": 0.22024565861922915,
+      "grad_norm": 1.0958232879638672,
+      "learning_rate": 1.9130635372876245e-05,
+      "loss": 0.5953,
+      "step": 130
+    },
+    {
+      "epoch": 0.22871664548919948,
+      "grad_norm": 1.0959957838058472,
+      "learning_rate": 1.9006018294416648e-05,
+      "loss": 0.6107,
+      "step": 135
+    },
+    {
+      "epoch": 0.23718763235916984,
+      "grad_norm": 1.1108900308609009,
+      "learning_rate": 1.8873520750565716e-05,
+      "loss": 0.5817,
+      "step": 140
+    },
+    {
+      "epoch": 0.2456586192291402,
+      "grad_norm": 1.057532787322998,
+      "learning_rate": 1.8733258679607674e-05,
+      "loss": 0.6108,
+      "step": 145
+    },
+    {
+      "epoch": 0.25412960609911056,
+      "grad_norm": 1.2086259126663208,
+      "learning_rate": 1.858535481395986e-05,
+      "loss": 0.5875,
+      "step": 150
+    },
+    {
+      "epoch": 0.2626005929690809,
+      "grad_norm": 1.0949163436889648,
+      "learning_rate": 1.8429938572779154e-05,
+      "loss": 0.5951,
+      "step": 155
+    },
+    {
+      "epoch": 0.2710715798390512,
+      "grad_norm": 0.9488250017166138,
+      "learning_rate": 1.8267145948717338e-05,
+      "loss": 0.5868,
+      "step": 160
+    },
+    {
+      "epoch": 0.2795425667090216,
+      "grad_norm": 1.0208171606063843,
+      "learning_rate": 1.8097119388924524e-05,
+      "loss": 0.5835,
+      "step": 165
+    },
+    {
+      "epoch": 0.28801355357899194,
+      "grad_norm": 1.1816986799240112,
+      "learning_rate": 1.7920007670404738e-05,
+      "loss": 0.5703,
+      "step": 170
+    },
+    {
+      "epoch": 0.29648454044896233,
+      "grad_norm": 1.1194366216659546,
+      "learning_rate": 1.7735965769832754e-05,
+      "loss": 0.5752,
+      "step": 175
+    },
+    {
+      "epoch": 0.30495552731893266,
+      "grad_norm": 1.1583960056304932,
+      "learning_rate": 1.7545154727946065e-05,
+      "loss": 0.5967,
+      "step": 180
+    },
+    {
+      "epoch": 0.313426514188903,
+      "grad_norm": 0.9982605576515198,
+      "learning_rate": 1.7347741508630673e-05,
+      "loss": 0.5798,
+      "step": 185
+    },
+    {
+      "epoch": 0.3218975010588734,
+      "grad_norm": 1.17666494846344,
+      "learning_rate": 1.7143898852824005e-05,
+      "loss": 0.5647,
+      "step": 190
+    },
+    {
+      "epoch": 0.3303684879288437,
+      "grad_norm": 0.9470816254615784,
+      "learning_rate": 1.6933805127362744e-05,
+      "loss": 0.546,
+      "step": 195
+    },
+    {
+      "epoch": 0.33883947479881404,
+      "grad_norm": 0.9323899745941162,
+      "learning_rate": 1.671764416890793e-05,
+      "loss": 0.5567,
+      "step": 200
+    },
+    {
+      "epoch": 0.34731046166878443,
+      "grad_norm": 1.0073851346969604,
+      "learning_rate": 1.649560512308378e-05,
+      "loss": 0.5649,
+      "step": 205
+    },
+    {
+      "epoch": 0.35578144853875476,
+      "grad_norm": 1.2009499073028564,
+      "learning_rate": 1.6267882278971102e-05,
+      "loss": 0.5563,
+      "step": 210
+    },
+    {
+      "epoch": 0.3642524354087251,
+      "grad_norm": 1.1687507629394531,
+      "learning_rate": 1.603467489910004e-05,
+      "loss": 0.5704,
+      "step": 215
+    },
+    {
+      "epoch": 0.3727234222786955,
+      "grad_norm": 1.0257385969161987,
+      "learning_rate": 1.5796187045090943e-05,
+      "loss": 0.5473,
+      "step": 220
+    },
+    {
+      "epoch": 0.3811944091486658,
+      "grad_norm": 1.0044339895248413,
+      "learning_rate": 1.5552627399095943e-05,
+      "loss": 0.5557,
+      "step": 225
+    },
+    {
+      "epoch": 0.38966539601863615,
+      "grad_norm": 1.0678868293762207,
+      "learning_rate": 1.5304209081197425e-05,
+      "loss": 0.5487,
+      "step": 230
+    },
+    {
+      "epoch": 0.39813638288860653,
+      "grad_norm": 0.9731626510620117,
+      "learning_rate": 1.5051149462923285e-05,
+      "loss": 0.5632,
+      "step": 235
+    },
+    {
+      "epoch": 0.40660736975857686,
+      "grad_norm": 0.9739297032356262,
+      "learning_rate": 1.4793669977041978e-05,
+      "loss": 0.5604,
+      "step": 240
+    },
+    {
+      "epoch": 0.41507835662854725,
+      "grad_norm": 1.1298881769180298,
+      "learning_rate": 1.4531995923803974e-05,
+      "loss": 0.5748,
+      "step": 245
+    },
+    {
+      "epoch": 0.4235493434985176,
+      "grad_norm": 0.946855902671814,
+      "learning_rate": 1.4266356273799044e-05,
+      "loss": 0.5386,
+      "step": 250
+    },
+    {
+      "epoch": 0.4320203303684879,
+      "grad_norm": 0.8729770183563232,
+      "learning_rate": 1.3996983467601921e-05,
+      "loss": 0.5672,
+      "step": 255
+    },
+    {
+      "epoch": 0.4404913172384583,
+      "grad_norm": 0.9312366843223572,
+      "learning_rate": 1.372411321238166e-05,
+      "loss": 0.5441,
+      "step": 260
+    },
+    {
+      "epoch": 0.44896230410842863,
+      "grad_norm": 0.9651340246200562,
+      "learning_rate": 1.3447984275652638e-05,
+      "loss": 0.5487,
+      "step": 265
+    },
+    {
+      "epoch": 0.45743329097839897,
+      "grad_norm": 0.998904287815094,
+      "learning_rate": 1.3168838276347691e-05,
+      "loss": 0.5397,
+      "step": 270
+    },
+    {
+      "epoch": 0.46590427784836935,
+      "grad_norm": 1.2853810787200928,
+      "learning_rate": 1.2886919473396212e-05,
+      "loss": 0.5386,
+      "step": 275
+    },
+    {
+      "epoch": 0.4743752647183397,
+      "grad_norm": 1.006282925605774,
+      "learning_rate": 1.2602474551992165e-05,
+      "loss": 0.5501,
+      "step": 280
+    },
+    {
+      "epoch": 0.48284625158831,
+      "grad_norm": 0.9217785596847534,
+      "learning_rate": 1.2315752407739093e-05,
+      "loss": 0.529,
+      "step": 285
+    },
+    {
+      "epoch": 0.4913172384582804,
+      "grad_norm": 0.9664483070373535,
+      "learning_rate": 1.2027003928860936e-05,
+      "loss": 0.5295,
+      "step": 290
+    },
+    {
+      "epoch": 0.49978822532825073,
+      "grad_norm": 1.0580369234085083,
+      "learning_rate": 1.1736481776669307e-05,
+      "loss": 0.5504,
+      "step": 295
+    },
+    {
+      "epoch": 0.5082592121982211,
+      "grad_norm": 0.938615620136261,
+      "learning_rate": 1.1444440164479215e-05,
+      "loss": 0.5307,
+      "step": 300
+    },
+    {
+      "epoch": 0.5167301990681914,
+      "grad_norm": 0.91016685962677,
+      "learning_rate": 1.115113463516683e-05,
+      "loss": 0.5214,
+      "step": 305
+    },
+    {
+      "epoch": 0.5252011859381618,
+      "grad_norm": 0.9125617146492004,
+      "learning_rate": 1.085682183756377e-05,
+      "loss": 0.5407,
+      "step": 310
+    },
+    {
+      "epoch": 0.5336721728081322,
+      "grad_norm": 0.9857435822486877,
+      "learning_rate": 1.0561759301883714e-05,
+      "loss": 0.5508,
+      "step": 315
+    },
+    {
+      "epoch": 0.5421431596781024,
+      "grad_norm": 0.948715090751648,
+      "learning_rate": 1.026620521437775e-05,
+      "loss": 0.5275,
+      "step": 320
+    },
+    {
+      "epoch": 0.5506141465480728,
+      "grad_norm": 0.9650415778160095,
+      "learning_rate": 9.970418191415703e-06,
+      "loss": 0.5143,
+      "step": 325
+    },
+    {
+      "epoch": 0.5590851334180432,
+      "grad_norm": 0.9797161221504211,
+      "learning_rate": 9.674657053191079e-06,
+      "loss": 0.529,
+      "step": 330
+    },
+    {
+      "epoch": 0.5675561202880135,
+      "grad_norm": 0.9044686555862427,
+      "learning_rate": 9.379180597247661e-06,
+      "loss": 0.5446,
+      "step": 335
+    },
+    {
+      "epoch": 0.5760271071579839,
+      "grad_norm": 0.9637364149093628,
+      "learning_rate": 9.084247372025938e-06,
+      "loss": 0.5207,
+      "step": 340
+    },
+    {
+      "epoch": 0.5844980940279543,
+      "grad_norm": 0.883882462978363,
+      "learning_rate": 8.790115450627486e-06,
+      "loss": 0.5177,
+      "step": 345
+    },
+    {
+      "epoch": 0.5929690808979247,
+      "grad_norm": 0.9830591678619385,
+      "learning_rate": 8.497042204995299e-06,
+      "loss": 0.5386,
+      "step": 350
+    },
+    {
+      "epoch": 0.6014400677678949,
+      "grad_norm": 0.9278344511985779,
+      "learning_rate": 8.205284080707634e-06,
+      "loss": 0.5258,
+      "step": 355
+    },
+    {
+      "epoch": 0.6099110546378653,
+      "grad_norm": 0.8791617155075073,
+      "learning_rate": 7.915096372582467e-06,
+      "loss": 0.5407,
+      "step": 360
+    },
+    {
+      "epoch": 0.6183820415078357,
+      "grad_norm": 0.9185119271278381,
+      "learning_rate": 7.626733001288852e-06,
+      "loss": 0.527,
+      "step": 365
+    },
+    {
+      "epoch": 0.626853028377806,
+      "grad_norm": 0.8270648717880249,
+      "learning_rate": 7.3404462911607325e-06,
+      "loss": 0.5312,
+      "step": 370
+    },
+    {
+      "epoch": 0.6353240152477764,
+      "grad_norm": 0.919711709022522,
+      "learning_rate": 7.056486749407552e-06,
+      "loss": 0.5254,
+      "step": 375
+    },
+    {
+      "epoch": 0.6437950021177468,
+      "grad_norm": 0.8501454591751099,
+      "learning_rate": 6.775102846914912e-06,
+      "loss": 0.5159,
+      "step": 380
+    },
+    {
+      "epoch": 0.652265988987717,
+      "grad_norm": 0.9181873202323914,
+      "learning_rate": 6.4965408008270355e-06,
+      "loss": 0.5175,
+      "step": 385
+    },
+    {
+      "epoch": 0.6607369758576874,
+      "grad_norm": 0.8412730693817139,
+      "learning_rate": 6.221044359101317e-06,
+      "loss": 0.5249,
+      "step": 390
+    },
+    {
+      "epoch": 0.6692079627276578,
+      "grad_norm": 0.8827643394470215,
+      "learning_rate": 5.948854587223465e-06,
+      "loss": 0.5222,
+      "step": 395
+    },
+    {
+      "epoch": 0.6776789495976281,
+      "grad_norm": 0.7885822653770447,
+      "learning_rate": 5.680209657269871e-06,
+      "loss": 0.5122,
+      "step": 400
+    },
+    {
+      "epoch": 0.6861499364675985,
+      "grad_norm": 0.8819693922996521,
+      "learning_rate": 5.415344639501754e-06,
+      "loss": 0.5287,
+      "step": 405
+    },
+    {
+      "epoch": 0.6946209233375689,
+      "grad_norm": 0.8051272034645081,
+      "learning_rate": 5.1544912966735e-06,
+      "loss": 0.5132,
+      "step": 410
+    },
+    {
+      "epoch": 0.7030919102075391,
+      "grad_norm": 0.831628680229187,
+      "learning_rate": 4.897877881235091e-06,
+      "loss": 0.5088,
+      "step": 415
+    },
+    {
+      "epoch": 0.7115628970775095,
+      "grad_norm": 0.8426679968833923,
+      "learning_rate": 4.645728935606194e-06,
+      "loss": 0.5163,
+      "step": 420
+    },
+    {
+      "epoch": 0.7200338839474799,
+      "grad_norm": 0.8241559267044067,
+      "learning_rate": 4.398265095696539e-06,
+      "loss": 0.5174,
+      "step": 425
+    },
+    {
+      "epoch": 0.7285048708174502,
+      "grad_norm": 0.8578051924705505,
+      "learning_rate": 4.1557028978446415e-06,
+      "loss": 0.5129,
+      "step": 430
+    },
+    {
+      "epoch": 0.7369758576874206,
+      "grad_norm": 0.8619440197944641,
+      "learning_rate": 3.918254589343683e-06,
+      "loss": 0.5102,
+      "step": 435
+    },
+    {
+      "epoch": 0.745446844557391,
+      "grad_norm": 0.8140995502471924,
+      "learning_rate": 3.6861279427204634e-06,
+      "loss": 0.5052,
+      "step": 440
+    },
+    {
+      "epoch": 0.7539178314273612,
+      "grad_norm": 0.7656389474868774,
+      "learning_rate": 3.4595260739298174e-06,
+      "loss": 0.5247,
+      "step": 445
+    },
+    {
+      "epoch": 0.7623888182973316,
+      "grad_norm": 0.780764102935791,
+      "learning_rate": 3.2386472646236565e-06,
+      "loss": 0.5139,
+      "step": 450
+    },
+    {
+      "epoch": 0.770859805167302,
+      "grad_norm": 0.7912269234657288,
+      "learning_rate": 3.023684788650154e-06,
+      "loss": 0.5184,
+      "step": 455
+    },
+    {
+      "epoch": 0.7793307920372723,
+      "grad_norm": 0.7656291127204895,
+      "learning_rate": 2.814826742934823e-06,
+      "loss": 0.5168,
+      "step": 460
+    },
+    {
+      "epoch": 0.7878017789072427,
+      "grad_norm": 0.7654049396514893,
+      "learning_rate": 2.6122558828915647e-06,
+      "loss": 0.5127,
+      "step": 465
+    },
+    {
+      "epoch": 0.7962727657772131,
+      "grad_norm": 0.8781611919403076,
+      "learning_rate": 2.4161494625076164e-06,
+      "loss": 0.5068,
+      "step": 470
+    },
+    {
+      "epoch": 0.8047437526471835,
+      "grad_norm": 0.7922006249427795,
+      "learning_rate": 2.2266790792424096e-06,
+      "loss": 0.5243,
+      "step": 475
+    },
+    {
+      "epoch": 0.8132147395171537,
+      "grad_norm": 0.7805562019348145,
+      "learning_rate": 2.044010523875969e-06,
+      "loss": 0.5114,
+      "step": 480
+    },
+    {
+      "epoch": 0.8216857263871241,
+      "grad_norm": 0.8500534296035767,
+      "learning_rate": 1.868303635438332e-06,
+      "loss": 0.4978,
+      "step": 485
+    },
+    {
+      "epoch": 0.8301567132570945,
+      "grad_norm": 0.7626408934593201,
+      "learning_rate": 1.699712161346846e-06,
+      "loss": 0.5108,
+      "step": 490
+    },
+    {
+      "epoch": 0.8386277001270648,
+      "grad_norm": 0.7929341197013855,
+      "learning_rate": 1.5383836228737815e-06,
+      "loss": 0.5126,
+      "step": 495
+    },
+    {
+      "epoch": 0.8470986869970352,
+      "grad_norm": 0.7848495244979858,
+      "learning_rate": 1.3844591860619382e-06,
+      "loss": 0.5037,
+      "step": 500
+    },
+    {
+      "epoch": 0.8555696738670056,
+      "grad_norm": 0.7474762797355652,
+      "learning_rate": 1.2380735382012576e-06,
+      "loss": 0.5151,
+      "step": 505
+    },
+    {
+      "epoch": 0.8640406607369758,
+      "grad_norm": 0.7843493819236755,
+      "learning_rate": 1.0993547699744366e-06,
+      "loss": 0.5114,
+      "step": 510
+    },
+    {
+      "epoch": 0.8725116476069462,
+      "grad_norm": 0.7787851095199585,
+      "learning_rate": 9.684242633747642e-07,
+      "loss": 0.5111,
+      "step": 515
+    },
+    {
+      "epoch": 0.8809826344769166,
+      "grad_norm": 0.7845005989074707,
+      "learning_rate": 8.453965854941748e-07,
+      "loss": 0.5046,
+      "step": 520
+    },
+    {
+      "epoch": 0.8894536213468869,
+      "grad_norm": 0.7967577576637268,
+      "learning_rate": 7.303793882745181e-07,
+      "loss": 0.5016,
+      "step": 525
+    },
+    {
+      "epoch": 0.8979246082168573,
+      "grad_norm": 0.7523807883262634,
+      "learning_rate": 6.234733143097215e-07,
+      "loss": 0.4974,
+      "step": 530
+    },
+    {
+      "epoch": 0.9063955950868277,
+      "grad_norm": 0.7827950119972229,
+      "learning_rate": 5.247719087812897e-07,
+      "loss": 0.4984,
+      "step": 535
+    },
+    {
+      "epoch": 0.9148665819567979,
+      "grad_norm": 0.7315457463264465,
+      "learning_rate": 4.343615376042065e-07,
+      "loss": 0.5147,
+      "step": 540
+    },
+    {
+      "epoch": 0.9233375688267683,
+      "grad_norm": 0.8033891916275024,
+      "learning_rate": 3.5232131185484075e-07,
+      "loss": 0.5116,
+      "step": 545
+    },
+    {
+      "epoch": 0.9318085556967387,
+      "grad_norm": 0.7409123778343201,
+      "learning_rate": 2.78723018547008e-07,
+      "loss": 0.4918,
+      "step": 550
+    },
+    {
+      "epoch": 0.940279542566709,
+      "grad_norm": 0.7420827150344849,
+      "learning_rate": 2.1363105781673888e-07,
+      "loss": 0.5066,
+      "step": 555
+    },
+    {
+      "epoch": 0.9487505294366794,
+      "grad_norm": 0.7862848043441772,
+      "learning_rate": 1.5710238657074218e-07,
+      "loss": 0.5183,
+      "step": 560
+    },
+    {
+      "epoch": 0.9572215163066498,
+      "grad_norm": 0.7402486205101013,
+      "learning_rate": 1.0918646864784166e-07,
+      "loss": 0.5182,
+      "step": 565
+    },
+    {
+      "epoch": 0.96569250317662,
+      "grad_norm": 0.7287072539329529,
+      "learning_rate": 6.99252315370269e-08,
+      "loss": 0.4976,
+      "step": 570
+    },
+    {
+      "epoch": 0.9741634900465904,
+      "grad_norm": 0.7665913105010986,
+      "learning_rate": 3.9353029689974676e-08,
+      "loss": 0.5127,
+      "step": 575
+    },
+    {
+      "epoch": 0.9826344769165608,
+      "grad_norm": 0.6921188831329346,
+      "learning_rate": 1.7496614460135174e-08,
+      "loss": 0.5173,
+      "step": 580
+    },
+    {
+      "epoch": 0.9911054637865311,
+      "grad_norm": 0.7035255432128906,
+      "learning_rate": 4.375110694713192e-09,
+      "loss": 0.4961,
+      "step": 585
+    },
+    {
+      "epoch": 0.9995764506565015,
+      "grad_norm": 0.7657430768013,
+      "learning_rate": 0.0,
+      "loss": 0.5122,
+      "step": 590
+    },
+    {
+      "epoch": 0.9995764506565015,
+      "step": 590,
+      "total_flos": 8.380405809686774e+17,
+      "train_loss": 0.5979039826635587,
+      "train_runtime": 5676.8996,
+      "train_samples_per_second": 6.654,
+      "train_steps_per_second": 0.104
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 590,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.380405809686774e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

Qwen2-VL-2B-SFT-Clevr_CoGenT_TrainA_R1/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4201fbf24a6175bb8994b3536ee3da6283f8bf00816fd2000cb9ac77dfbdeec1
+size 7160

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93ba58e6c790bf087c8d5f6d7404a72eac1ce870fa43c54f9fb37e7cb4342390
+oid sha256:426010c540c6dbf0ea47ab6c0a907ada5c1d40c9075c5e1d192958c358ad85a0
 size 4884798456

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4201fbf24a6175bb8994b3536ee3da6283f8bf00816fd2000cb9ac77dfbdeec1
+oid sha256:80c546c01c4ec4fa55ba4b1925cc4ffc99669e1a841af40c3cd38158d628c7e6
 size 7160