Upload folder using huggingface_hub

.gitattributes

@@ -36,3 +36,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 qwen2_5vl_rlhf_v_lora_dpo-7b/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 qwen2_vl_rlhf_v_lora_dpo-2b/checkpoint-180/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 qwen2_vl_rlhf_v_lora_dpo-2b/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+qwen2_5vl-7b/lora/povid/lora/dpo/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+qwen2_5vl-7b/lora/rlaif_v/lora/dpo/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+qwen2_5vl-7b/lora/rlhf_v/lora/dpo/checkpoint-180/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+qwen2_5vl-7b/lora/rlhf_v/lora/dpo/tokenizer.json filter=lfs diff=lfs merge=lfs -text

qwen2_5vl-7b/lora/povid/lora/dpo/README.md

@@ -0,0 +1,62 @@
+---
+library_name: peft
+license: other
+base_model: Qwen/Qwen2.5-VL-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: dpo
+  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. -->
+
+# dpo
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct) on the povid 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: 5e-06
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
+- total_eval_batch_size: 32
+- 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: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.15.2
+- Transformers 4.52.4
+- Pytorch 2.7.1+cu126
+- Datasets 3.6.0
+- Tokenizers 0.21.1

qwen2_5vl-7b/lora/povid/lora/dpo/adapter_config.json

@@ -0,0 +1,120 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "layers.5.mlp.gate_proj",
+    "layers.19.mlp.up_proj",
+    "layers.25.mlp.down_proj",
+    "layers.24.mlp.up_proj",
+    "layers.11.mlp.gate_proj",
+    "layers.8.mlp.gate_proj",
+    "layers.27.mlp.up_proj",
+    "layers.14.mlp.gate_proj",
+    "layers.2.mlp.down_proj",
+    "v_proj",
+    "layers.2.mlp.up_proj",
+    "layers.27.mlp.gate_proj",
+    "layers.6.mlp.gate_proj",
+    "layers.27.mlp.down_proj",
+    "layers.15.mlp.gate_proj",
+    "layers.23.mlp.up_proj",
+    "layers.0.mlp.down_proj",
+    "layers.18.mlp.down_proj",
+    "layers.6.mlp.up_proj",
+    "layers.7.mlp.up_proj",
+    "layers.4.mlp.up_proj",
+    "layers.13.mlp.gate_proj",
+    "layers.7.mlp.down_proj",
+    "layers.16.mlp.gate_proj",
+    "layers.26.mlp.up_proj",
+    "layers.20.mlp.gate_proj",
+    "layers.20.mlp.up_proj",
+    "layers.24.mlp.gate_proj",
+    "layers.21.mlp.down_proj",
+    "layers.9.mlp.up_proj",
+    "layers.8.mlp.down_proj",
+    "layers.26.mlp.gate_proj",
+    "layers.26.mlp.down_proj",
+    "layers.22.mlp.up_proj",
+    "layers.15.mlp.up_proj",
+    "layers.3.mlp.down_proj",
+    "q_proj",
+    "layers.14.mlp.up_proj",
+    "layers.19.mlp.gate_proj",
+    "layers.21.mlp.up_proj",
+    "layers.12.mlp.down_proj",
+    "layers.11.mlp.down_proj",
+    "layers.13.mlp.down_proj",
+    "layers.3.mlp.up_proj",
+    "layers.13.mlp.up_proj",
+    "layers.9.mlp.gate_proj",
+    "layers.16.mlp.down_proj",
+    "layers.17.mlp.down_proj",
+    "layers.25.mlp.gate_proj",
+    "layers.7.mlp.gate_proj",
+    "layers.1.mlp.up_proj",
+    "layers.10.mlp.up_proj",
+    "layers.0.mlp.up_proj",
+    "k_proj",
+    "layers.21.mlp.gate_proj",
+    "layers.5.mlp.up_proj",
+    "layers.14.mlp.down_proj",
+    "layers.22.mlp.gate_proj",
+    "layers.11.mlp.up_proj",
+    "layers.3.mlp.gate_proj",
+    "layers.5.mlp.down_proj",
+    "layers.0.mlp.gate_proj",
+    "o_proj",
+    "layers.23.mlp.down_proj",
+    "layers.15.mlp.down_proj",
+    "layers.4.mlp.gate_proj",
+    "layers.8.mlp.up_proj",
+    "layers.18.mlp.gate_proj",
+    "layers.17.mlp.up_proj",
+    "layers.22.mlp.down_proj",
+    "layers.12.mlp.up_proj",
+    "layers.12.mlp.gate_proj",
+    "layers.24.mlp.down_proj",
+    "layers.18.mlp.up_proj",
+    "layers.1.mlp.down_proj",
+    "layers.4.mlp.down_proj",
+    "layers.6.mlp.down_proj",
+    "layers.23.mlp.gate_proj",
+    "layers.25.mlp.up_proj",
+    "layers.16.mlp.up_proj",
+    "layers.9.mlp.down_proj",
+    "layers.2.mlp.gate_proj",
+    "layers.19.mlp.down_proj",
+    "layers.20.mlp.down_proj",
+    "layers.17.mlp.gate_proj",
+    "layers.10.mlp.down_proj",
+    "layers.10.mlp.gate_proj",
+    "layers.1.mlp.gate_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

qwen2_5vl-7b/lora/povid/lora/dpo/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21d4a55eb0dabb6270a4fbb1bb25ce9b61d469eb406c0287e177629a9cda1e8c
+size 645981584

qwen2_5vl-7b/lora/povid/lora/dpo/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

qwen2_5vl-7b/lora/povid/lora/dpo/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 8.076450034624758e+17,
+    "train_loss": 0.07847787283388707,
+    "train_runtime": 6733.1663,
+    "train_samples_per_second": 2.552,
+    "train_steps_per_second": 0.08
+}

qwen2_5vl-7b/lora/povid/lora/dpo/chat_template.jinja

@@ -0,0 +1,7 @@
+{% 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
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% 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|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen2.5-VL-7B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/adapter_config.json

@@ -0,0 +1,120 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "layers.5.mlp.gate_proj",
+    "layers.19.mlp.up_proj",
+    "layers.25.mlp.down_proj",
+    "layers.24.mlp.up_proj",
+    "layers.11.mlp.gate_proj",
+    "layers.8.mlp.gate_proj",
+    "layers.27.mlp.up_proj",
+    "layers.14.mlp.gate_proj",
+    "layers.2.mlp.down_proj",
+    "v_proj",
+    "layers.2.mlp.up_proj",
+    "layers.27.mlp.gate_proj",
+    "layers.6.mlp.gate_proj",
+    "layers.27.mlp.down_proj",
+    "layers.15.mlp.gate_proj",
+    "layers.23.mlp.up_proj",
+    "layers.0.mlp.down_proj",
+    "layers.18.mlp.down_proj",
+    "layers.6.mlp.up_proj",
+    "layers.7.mlp.up_proj",
+    "layers.4.mlp.up_proj",
+    "layers.13.mlp.gate_proj",
+    "layers.7.mlp.down_proj",
+    "layers.16.mlp.gate_proj",
+    "layers.26.mlp.up_proj",
+    "layers.20.mlp.gate_proj",
+    "layers.20.mlp.up_proj",
+    "layers.24.mlp.gate_proj",
+    "layers.21.mlp.down_proj",
+    "layers.9.mlp.up_proj",
+    "layers.8.mlp.down_proj",
+    "layers.26.mlp.gate_proj",
+    "layers.26.mlp.down_proj",
+    "layers.22.mlp.up_proj",
+    "layers.15.mlp.up_proj",
+    "layers.3.mlp.down_proj",
+    "q_proj",
+    "layers.14.mlp.up_proj",
+    "layers.19.mlp.gate_proj",
+    "layers.21.mlp.up_proj",
+    "layers.12.mlp.down_proj",
+    "layers.11.mlp.down_proj",
+    "layers.13.mlp.down_proj",
+    "layers.3.mlp.up_proj",
+    "layers.13.mlp.up_proj",
+    "layers.9.mlp.gate_proj",
+    "layers.16.mlp.down_proj",
+    "layers.17.mlp.down_proj",
+    "layers.25.mlp.gate_proj",
+    "layers.7.mlp.gate_proj",
+    "layers.1.mlp.up_proj",
+    "layers.10.mlp.up_proj",
+    "layers.0.mlp.up_proj",
+    "k_proj",
+    "layers.21.mlp.gate_proj",
+    "layers.5.mlp.up_proj",
+    "layers.14.mlp.down_proj",
+    "layers.22.mlp.gate_proj",
+    "layers.11.mlp.up_proj",
+    "layers.3.mlp.gate_proj",
+    "layers.5.mlp.down_proj",
+    "layers.0.mlp.gate_proj",
+    "o_proj",
+    "layers.23.mlp.down_proj",
+    "layers.15.mlp.down_proj",
+    "layers.4.mlp.gate_proj",
+    "layers.8.mlp.up_proj",
+    "layers.18.mlp.gate_proj",
+    "layers.17.mlp.up_proj",
+    "layers.22.mlp.down_proj",
+    "layers.12.mlp.up_proj",
+    "layers.12.mlp.gate_proj",
+    "layers.24.mlp.down_proj",
+    "layers.18.mlp.up_proj",
+    "layers.1.mlp.down_proj",
+    "layers.4.mlp.down_proj",
+    "layers.6.mlp.down_proj",
+    "layers.23.mlp.gate_proj",
+    "layers.25.mlp.up_proj",
+    "layers.16.mlp.up_proj",
+    "layers.9.mlp.down_proj",
+    "layers.2.mlp.gate_proj",
+    "layers.19.mlp.down_proj",
+    "layers.20.mlp.down_proj",
+    "layers.17.mlp.gate_proj",
+    "layers.10.mlp.down_proj",
+    "layers.10.mlp.gate_proj",
+    "layers.1.mlp.gate_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21d4a55eb0dabb6270a4fbb1bb25ce9b61d469eb406c0287e177629a9cda1e8c
+size 645981584

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/chat_template.jinja

@@ -0,0 +1,7 @@
+{% 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
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% 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|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/preprocessor_config.json

@@ -0,0 +1,36 @@
+{
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "do_center_crop": null,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessorFast",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "input_data_format": null,
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/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_5vl-7b/lora/povid/lora/dpo/checkpoint-537/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "add_bos_token": false,
+  "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
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "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,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2_5_VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/trainer_state.json

@@ -0,0 +1,829 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 537,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0186219739292365,
+      "grad_norm": 2.07263445854187,
+      "learning_rate": 8.333333333333333e-07,
+      "logits/chosen": 1.9083625078201294,
+      "logits/rejected": 1.9776146411895752,
+      "logps/chosen": -99.38154602050781,
+      "logps/rejected": -137.68942260742188,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.4156250059604645,
+      "rewards/chosen": 0.002753954380750656,
+      "rewards/margins": 0.0035315221175551414,
+      "rewards/rejected": -0.0007775675039738417,
+      "step": 10
+    },
+    {
+      "epoch": 0.037243947858473,
+      "grad_norm": 2.077322006225586,
+      "learning_rate": 1.7592592592592594e-06,
+      "logits/chosen": 1.7842262983322144,
+      "logits/rejected": 1.8155817985534668,
+      "logps/chosen": -103.61813354492188,
+      "logps/rejected": -137.05908203125,
+      "loss": 0.6911,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": 0.003043058328330517,
+      "rewards/margins": 0.004756969399750233,
+      "rewards/rejected": -0.0017139108385890722,
+      "step": 20
+    },
+    {
+      "epoch": 0.055865921787709494,
+      "grad_norm": 2.449907064437866,
+      "learning_rate": 2.6851851851851856e-06,
+      "logits/chosen": 1.9791193008422852,
+      "logits/rejected": 2.021622657775879,
+      "logps/chosen": -102.76006317138672,
+      "logps/rejected": -141.06813049316406,
+      "loss": 0.6599,
+      "rewards/accuracies": 0.7906249761581421,
+      "rewards/chosen": 0.03256123140454292,
+      "rewards/margins": 0.07064400613307953,
+      "rewards/rejected": -0.038082774728536606,
+      "step": 30
+    },
+    {
+      "epoch": 0.074487895716946,
+      "grad_norm": 1.980841875076294,
+      "learning_rate": 3.6111111111111115e-06,
+      "logits/chosen": 1.952810525894165,
+      "logits/rejected": 2.003131151199341,
+      "logps/chosen": -111.52398681640625,
+      "logps/rejected": -149.88365173339844,
+      "loss": 0.5971,
+      "rewards/accuracies": 0.8750000596046448,
+      "rewards/chosen": 0.06918186694383621,
+      "rewards/margins": 0.2191462516784668,
+      "rewards/rejected": -0.14996439218521118,
+      "step": 40
+    },
+    {
+      "epoch": 0.0931098696461825,
+      "grad_norm": 1.909422516822815,
+      "learning_rate": 4.537037037037038e-06,
+      "logits/chosen": 1.8376410007476807,
+      "logits/rejected": 1.8602056503295898,
+      "logps/chosen": -98.7303237915039,
+      "logps/rejected": -145.1428680419922,
+      "loss": 0.484,
+      "rewards/accuracies": 0.9468750357627869,
+      "rewards/chosen": 0.06579304486513138,
+      "rewards/margins": 0.5358557105064392,
+      "rewards/rejected": -0.4700627028942108,
+      "step": 50
+    },
+    {
+      "epoch": 0.11173184357541899,
+      "grad_norm": 1.4733169078826904,
+      "learning_rate": 4.998678043309151e-06,
+      "logits/chosen": 1.9728565216064453,
+      "logits/rejected": 2.0448157787323,
+      "logps/chosen": -102.9524917602539,
+      "logps/rejected": -149.02821350097656,
+      "loss": 0.3382,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": -0.0847957655787468,
+      "rewards/margins": 1.1177407503128052,
+      "rewards/rejected": -1.2025364637374878,
+      "step": 60
+    },
+    {
+      "epoch": 0.1303538175046555,
+      "grad_norm": 0.6721972823143005,
+      "learning_rate": 4.988110776637384e-06,
+      "logits/chosen": 1.727777361869812,
+      "logits/rejected": 1.7899367809295654,
+      "logps/chosen": -102.1249008178711,
+      "logps/rejected": -157.9233856201172,
+      "loss": 0.1906,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -0.261747270822525,
+      "rewards/margins": 2.095768690109253,
+      "rewards/rejected": -2.357516288757324,
+      "step": 70
+    },
+    {
+      "epoch": 0.148975791433892,
+      "grad_norm": 0.6819734573364258,
+      "learning_rate": 4.967020933775366e-06,
+      "logits/chosen": 1.8152174949645996,
+      "logits/rejected": 1.884362816810608,
+      "logps/chosen": -110.66182708740234,
+      "logps/rejected": -183.3290557861328,
+      "loss": 0.099,
+      "rewards/accuracies": 0.9906250238418579,
+      "rewards/chosen": -0.6374696493148804,
+      "rewards/margins": 3.595221519470215,
+      "rewards/rejected": -4.232691287994385,
+      "step": 80
+    },
+    {
+      "epoch": 0.16759776536312848,
+      "grad_norm": 0.6661988496780396,
+      "learning_rate": 4.935497706683698e-06,
+      "logits/chosen": 1.5606900453567505,
+      "logits/rejected": 1.627284288406372,
+      "logps/chosen": -123.47517395019531,
+      "logps/rejected": -210.38690185546875,
+      "loss": 0.0433,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.6986806392669678,
+      "rewards/margins": 5.403690814971924,
+      "rewards/rejected": -7.1023712158203125,
+      "step": 90
+    },
+    {
+      "epoch": 0.186219739292365,
+      "grad_norm": 0.3505627512931824,
+      "learning_rate": 4.893674411596507e-06,
+      "logits/chosen": 1.4444490671157837,
+      "logits/rejected": 1.496826171875,
+      "logps/chosen": -132.90512084960938,
+      "logps/rejected": -247.66297912597656,
+      "loss": 0.0202,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -3.0164995193481445,
+      "rewards/margins": 7.708347797393799,
+      "rewards/rejected": -10.724847793579102,
+      "step": 100
+    },
+    {
+      "epoch": 0.2048417132216015,
+      "grad_norm": 0.08065284043550491,
+      "learning_rate": 4.8417279252080365e-06,
+      "logits/chosen": 1.1240510940551758,
+      "logits/rejected": 1.1461639404296875,
+      "logps/chosen": -173.87986755371094,
+      "logps/rejected": -296.1276550292969,
+      "loss": 0.0187,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -5.6900763511657715,
+      "rewards/margins": 9.046419143676758,
+      "rewards/rejected": -14.736495971679688,
+      "step": 110
+    },
+    {
+      "epoch": 0.22346368715083798,
+      "grad_norm": 0.1960441768169403,
+      "learning_rate": 4.779877936635816e-06,
+      "logits/chosen": 0.9913089275360107,
+      "logits/rejected": 1.0386312007904053,
+      "logps/chosen": -169.66835021972656,
+      "logps/rejected": -305.98406982421875,
+      "loss": 0.0341,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.384186744689941,
+      "rewards/margins": 10.059325218200684,
+      "rewards/rejected": -16.443510055541992,
+      "step": 120
+    },
+    {
+      "epoch": 0.24208566108007448,
+      "grad_norm": 0.08494703471660614,
+      "learning_rate": 4.7083860183239795e-06,
+      "logits/chosen": 1.072335958480835,
+      "logits/rejected": 1.071942925453186,
+      "logps/chosen": -155.18902587890625,
+      "logps/rejected": -295.8493957519531,
+      "loss": 0.0058,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.6623382568359375,
+      "rewards/margins": 10.138053894042969,
+      "rewards/rejected": -15.800392150878906,
+      "step": 130
+    },
+    {
+      "epoch": 0.260707635009311,
+      "grad_norm": 0.40349674224853516,
+      "learning_rate": 4.627554519815999e-06,
+      "logits/chosen": 0.9601426124572754,
+      "logits/rejected": 0.9602993726730347,
+      "logps/chosen": -172.6103515625,
+      "logps/rejected": -322.26434326171875,
+      "loss": 0.0061,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.130578994750977,
+      "rewards/margins": 11.377153396606445,
+      "rewards/rejected": -17.507732391357422,
+      "step": 140
+    },
+    {
+      "epoch": 0.27932960893854747,
+      "grad_norm": 0.3065098226070404,
+      "learning_rate": 4.537725289075239e-06,
+      "logits/chosen": 1.027471661567688,
+      "logits/rejected": 0.9988879561424255,
+      "logps/chosen": -174.87049865722656,
+      "logps/rejected": -326.5740966796875,
+      "loss": 0.0062,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.688374042510986,
+      "rewards/margins": 11.755913734436035,
+      "rewards/rejected": -18.44428825378418,
+      "step": 150
+    },
+    {
+      "epoch": 0.297951582867784,
+      "grad_norm": 0.6753863096237183,
+      "learning_rate": 4.43927822676105e-06,
+      "logits/chosen": 0.8613717555999756,
+      "logits/rejected": 0.8766331076622009,
+      "logps/chosen": -178.57435607910156,
+      "logps/rejected": -329.76153564453125,
+      "loss": 0.0106,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.277381420135498,
+      "rewards/margins": 11.393813133239746,
+      "rewards/rejected": -17.67119598388672,
+      "step": 160
+    },
+    {
+      "epoch": 0.3165735567970205,
+      "grad_norm": 0.832106351852417,
+      "learning_rate": 4.332629679574566e-06,
+      "logits/chosen": 0.909426748752594,
+      "logits/rejected": 0.9274786114692688,
+      "logps/chosen": -159.40325927734375,
+      "logps/rejected": -308.1123046875,
+      "loss": 0.0107,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.586008071899414,
+      "rewards/margins": 11.150063514709473,
+      "rewards/rejected": -16.73607063293457,
+      "step": 170
+    },
+    {
+      "epoch": 0.33519553072625696,
+      "grad_norm": 0.08242648094892502,
+      "learning_rate": 4.218230679468981e-06,
+      "logits/chosen": 0.9288380742073059,
+      "logits/rejected": 0.9024308919906616,
+      "logps/chosen": -156.82220458984375,
+      "logps/rejected": -304.9267578125,
+      "loss": 0.0169,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -4.636939525604248,
+      "rewards/margins": 11.34218692779541,
+      "rewards/rejected": -15.979126930236816,
+      "step": 180
+    },
+    {
+      "epoch": 0.3538175046554935,
+      "grad_norm": 0.08803419768810272,
+      "learning_rate": 4.096565036170936e-06,
+      "logits/chosen": 0.9101991057395935,
+      "logits/rejected": 0.8393880724906921,
+      "logps/chosen": -131.576171875,
+      "logps/rejected": -280.4066162109375,
+      "loss": 0.0095,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -3.821293354034424,
+      "rewards/margins": 11.146876335144043,
+      "rewards/rejected": -14.968170166015625,
+      "step": 190
+    },
+    {
+      "epoch": 0.37243947858473,
+      "grad_norm": 0.020818937569856644,
+      "learning_rate": 3.96814729108002e-06,
+      "logits/chosen": 0.7776044607162476,
+      "logits/rejected": 0.7479174137115479,
+      "logps/chosen": -150.67852783203125,
+      "logps/rejected": -303.72247314453125,
+      "loss": 0.0051,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -4.321155071258545,
+      "rewards/margins": 11.680810928344727,
+      "rewards/rejected": -16.00196647644043,
+      "step": 200
+    },
+    {
+      "epoch": 0.39106145251396646,
+      "grad_norm": 0.5023415684700012,
+      "learning_rate": 3.8335205411996225e-06,
+      "logits/chosen": 0.838572084903717,
+      "logits/rejected": 0.7576170563697815,
+      "logps/chosen": -154.36627197265625,
+      "logps/rejected": -313.2589111328125,
+      "loss": 0.0027,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.044532775878906,
+      "rewards/margins": 12.347441673278809,
+      "rewards/rejected": -17.3919734954834,
+      "step": 210
+    },
+    {
+      "epoch": 0.409683426443203,
+      "grad_norm": 0.007361250929534435,
+      "learning_rate": 3.6932541423020707e-06,
+      "logits/chosen": 0.7803331613540649,
+      "logits/rejected": 0.7118293642997742,
+      "logps/chosen": -171.97296142578125,
+      "logps/rejected": -343.0221252441406,
+      "loss": 0.0023,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.363333702087402,
+      "rewards/margins": 13.190146446228027,
+      "rewards/rejected": -19.553478240966797,
+      "step": 220
+    },
+    {
+      "epoch": 0.42830540037243947,
+      "grad_norm": 0.04080897197127342,
+      "learning_rate": 3.5479413010416606e-06,
+      "logits/chosen": 0.6650150418281555,
+      "logits/rejected": 0.6223341822624207,
+      "logps/chosen": -160.4796142578125,
+      "logps/rejected": -337.9312744140625,
+      "loss": 0.0034,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.2509541511535645,
+      "rewards/margins": 13.757020950317383,
+      "rewards/rejected": -20.007976531982422,
+      "step": 230
+    },
+    {
+      "epoch": 0.44692737430167595,
+      "grad_norm": 0.4697687327861786,
+      "learning_rate": 3.3981965661989146e-06,
+      "logits/chosen": 0.7506241202354431,
+      "logits/rejected": 0.7021348476409912,
+      "logps/chosen": -185.2721710205078,
+      "logps/rejected": -356.75872802734375,
+      "loss": 0.0057,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -7.140838623046875,
+      "rewards/margins": 13.574752807617188,
+      "rewards/rejected": -20.71558952331543,
+      "step": 240
+    },
+    {
+      "epoch": 0.4655493482309125,
+      "grad_norm": 4.2021660804748535,
+      "learning_rate": 3.2446532296659255e-06,
+      "logits/chosen": 0.8130508661270142,
+      "logits/rejected": 0.7506972551345825,
+      "logps/chosen": -163.47772216796875,
+      "logps/rejected": -339.3100280761719,
+      "loss": 0.0108,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.541469573974609,
+      "rewards/margins": 13.666439056396484,
+      "rewards/rejected": -20.207908630371094,
+      "step": 250
+    },
+    {
+      "epoch": 0.48417132216014896,
+      "grad_norm": 1.2759469747543335,
+      "learning_rate": 3.087960648164403e-06,
+      "logits/chosen": 0.615781307220459,
+      "logits/rejected": 0.5805816650390625,
+      "logps/chosen": -166.66835021972656,
+      "logps/rejected": -341.08709716796875,
+      "loss": 0.0056,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.2712554931640625,
+      "rewards/margins": 13.863871574401855,
+      "rewards/rejected": -20.1351261138916,
+      "step": 260
+    },
+    {
+      "epoch": 0.5027932960893855,
+      "grad_norm": 0.08350568264722824,
+      "learning_rate": 2.9287814970232688e-06,
+      "logits/chosen": 0.7528300285339355,
+      "logits/rejected": 0.7390813231468201,
+      "logps/chosen": -164.2382354736328,
+      "logps/rejected": -337.09271240234375,
+      "loss": 0.0027,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.999736309051514,
+      "rewards/margins": 14.001705169677734,
+      "rewards/rejected": -20.001441955566406,
+      "step": 270
+    },
+    {
+      "epoch": 0.521415270018622,
+      "grad_norm": 0.29949527978897095,
+      "learning_rate": 2.7677889676299442e-06,
+      "logits/chosen": 0.6687740087509155,
+      "logits/rejected": 0.6045177578926086,
+      "logps/chosen": -158.6308135986328,
+      "logps/rejected": -349.30316162109375,
+      "loss": 0.0034,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.776445388793945,
+      "rewards/margins": 15.19276237487793,
+      "rewards/rejected": -20.969207763671875,
+      "step": 280
+    },
+    {
+      "epoch": 0.5400372439478585,
+      "grad_norm": 0.006012032739818096,
+      "learning_rate": 2.6056639204077293e-06,
+      "logits/chosen": 0.6841985583305359,
+      "logits/rejected": 0.6416488885879517,
+      "logps/chosen": -152.00906372070312,
+      "logps/rejected": -336.58306884765625,
+      "loss": 0.0106,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -5.997591972351074,
+      "rewards/margins": 14.582725524902344,
+      "rewards/rejected": -20.5803165435791,
+      "step": 290
+    },
+    {
+      "epoch": 0.5586592178770949,
+      "grad_norm": 0.10845273733139038,
+      "learning_rate": 2.443092005359736e-06,
+      "logits/chosen": 0.7344327569007874,
+      "logits/rejected": 0.6978190541267395,
+      "logps/chosen": -161.67117309570312,
+      "logps/rejected": -342.5047912597656,
+      "loss": 0.0023,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.260890960693359,
+      "rewards/margins": 14.143697738647461,
+      "rewards/rejected": -20.40458869934082,
+      "step": 300
+    },
+    {
+      "epoch": 0.5772811918063314,
+      "grad_norm": 4.161528587341309,
+      "learning_rate": 2.280760762356999e-06,
+      "logits/chosen": 0.6338777542114258,
+      "logits/rejected": 0.5441797375679016,
+      "logps/chosen": -177.587646484375,
+      "logps/rejected": -365.2238464355469,
+      "loss": 0.007,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.9434590339660645,
+      "rewards/margins": 14.978950500488281,
+      "rewards/rejected": -21.922409057617188,
+      "step": 310
+    },
+    {
+      "epoch": 0.595903165735568,
+      "grad_norm": 0.15292654931545258,
+      "learning_rate": 2.119356713434064e-06,
+      "logits/chosen": 0.7987104654312134,
+      "logits/rejected": 0.7266597151756287,
+      "logps/chosen": -167.17431640625,
+      "logps/rejected": -352.7719421386719,
+      "loss": 0.0029,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.418148994445801,
+      "rewards/margins": 14.750093460083008,
+      "rewards/rejected": -21.168243408203125,
+      "step": 320
+    },
+    {
+      "epoch": 0.6145251396648045,
+      "grad_norm": 0.01223732065409422,
+      "learning_rate": 1.9595624593891576e-06,
+      "logits/chosen": 0.7314679622650146,
+      "logits/rejected": 0.6928003430366516,
+      "logps/chosen": -170.919921875,
+      "logps/rejected": -359.259033203125,
+      "loss": 0.0088,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.704370021820068,
+      "rewards/margins": 15.030937194824219,
+      "rewards/rejected": -21.735305786132812,
+      "step": 330
+    },
+    {
+      "epoch": 0.633147113594041,
+      "grad_norm": 0.3380753993988037,
+      "learning_rate": 1.8020537929678192e-06,
+      "logits/chosen": 0.6550232768058777,
+      "logits/rejected": 0.6400955319404602,
+      "logps/chosen": -166.73977661132812,
+      "logps/rejected": -344.7969970703125,
+      "loss": 0.0058,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.093208312988281,
+      "rewards/margins": 14.206398010253906,
+      "rewards/rejected": -20.299606323242188,
+      "step": 340
+    },
+    {
+      "epoch": 0.6517690875232774,
+      "grad_norm": 0.03426047042012215,
+      "learning_rate": 1.647496840838767e-06,
+      "logits/chosen": 0.6896477341651917,
+      "logits/rejected": 0.6510384678840637,
+      "logps/chosen": -165.2506103515625,
+      "logps/rejected": -338.61004638671875,
+      "loss": 0.0028,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.837466716766357,
+      "rewards/margins": 13.876227378845215,
+      "rewards/rejected": -19.713693618774414,
+      "step": 350
+    },
+    {
+      "epoch": 0.6703910614525139,
+      "grad_norm": 0.845697820186615,
+      "learning_rate": 1.4965452464489817e-06,
+      "logits/chosen": 0.7214924693107605,
+      "logits/rejected": 0.6409821510314941,
+      "logps/chosen": -158.24691772460938,
+      "logps/rejected": -346.9435119628906,
+      "loss": 0.0088,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -5.867282867431641,
+      "rewards/margins": 15.158857345581055,
+      "rewards/rejected": -21.026140213012695,
+      "step": 360
+    },
+    {
+      "epoch": 0.6890130353817505,
+      "grad_norm": 0.03940608352422714,
+      "learning_rate": 1.3498374056721198e-06,
+      "logits/chosen": 0.5552241802215576,
+      "logits/rejected": 0.4949037730693817,
+      "logps/chosen": -156.23068237304688,
+      "logps/rejected": -345.18939208984375,
+      "loss": 0.008,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.261874198913574,
+      "rewards/margins": 15.237826347351074,
+      "rewards/rejected": -21.499698638916016,
+      "step": 370
+    },
+    {
+      "epoch": 0.707635009310987,
+      "grad_norm": 0.020929953083395958,
+      "learning_rate": 1.2079937669410968e-06,
+      "logits/chosen": 0.5396690368652344,
+      "logits/rejected": 0.5160779356956482,
+      "logps/chosen": -156.75762939453125,
+      "logps/rejected": -348.7319030761719,
+      "loss": 0.0027,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.514359474182129,
+      "rewards/margins": 15.283174514770508,
+      "rewards/rejected": -21.797534942626953,
+      "step": 380
+    },
+    {
+      "epoch": 0.7262569832402235,
+      "grad_norm": 0.022455064579844475,
+      "learning_rate": 1.0716142072829552e-06,
+      "logits/chosen": 0.6602338552474976,
+      "logits/rejected": 0.5932304263114929,
+      "logps/chosen": -169.5976104736328,
+      "logps/rejected": -352.7994079589844,
+      "loss": 0.001,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.581348419189453,
+      "rewards/margins": 15.190673828125,
+      "rewards/rejected": -21.772024154663086,
+      "step": 390
+    },
+    {
+      "epoch": 0.74487895716946,
+      "grad_norm": 0.008257913403213024,
+      "learning_rate": 9.412754953531664e-07,
+      "logits/chosen": 0.6103745698928833,
+      "logits/rejected": 0.5022205114364624,
+      "logps/chosen": -176.54592895507812,
+      "logps/rejected": -374.6795654296875,
+      "loss": 0.0022,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.988586902618408,
+      "rewards/margins": 15.85957145690918,
+      "rewards/rejected": -22.848159790039062,
+      "step": 400
+    },
+    {
+      "epoch": 0.7635009310986964,
+      "grad_norm": 0.027111830189824104,
+      "learning_rate": 8.17528852198515e-07,
+      "logits/chosen": 0.5302518606185913,
+      "logits/rejected": 0.4578227996826172,
+      "logps/chosen": -170.05894470214844,
+      "logps/rejected": -358.9085388183594,
+      "loss": 0.0074,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.636836051940918,
+      "rewards/margins": 15.394896507263184,
+      "rewards/rejected": -22.0317325592041,
+      "step": 410
+    },
+    {
+      "epoch": 0.7821229050279329,
+      "grad_norm": 0.07998599857091904,
+      "learning_rate": 7.008976200644962e-07,
+      "logits/chosen": 0.609848141670227,
+      "logits/rejected": 0.5438084006309509,
+      "logps/chosen": -172.95252990722656,
+      "logps/rejected": -362.3902893066406,
+      "loss": 0.0188,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.9202423095703125,
+      "rewards/margins": 15.082725524902344,
+      "rewards/rejected": -22.002967834472656,
+      "step": 420
+    },
+    {
+      "epoch": 0.8007448789571695,
+      "grad_norm": 0.13893048465251923,
+      "learning_rate": 5.918750491061323e-07,
+      "logits/chosen": 0.5778518915176392,
+      "logits/rejected": 0.5182854533195496,
+      "logps/chosen": -158.56326293945312,
+      "logps/rejected": -344.1929626464844,
+      "loss": 0.0055,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.2186408042907715,
+      "rewards/margins": 15.193116188049316,
+      "rewards/rejected": -21.411758422851562,
+      "step": 430
+    },
+    {
+      "epoch": 0.819366852886406,
+      "grad_norm": 2.4267919063568115,
+      "learning_rate": 4.909222113625545e-07,
+      "logits/chosen": 0.5850950479507446,
+      "logits/rejected": 0.5577961802482605,
+      "logps/chosen": -185.70611572265625,
+      "logps/rejected": -377.9247741699219,
+      "loss": 0.005,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -7.237918376922607,
+      "rewards/margins": 15.439810752868652,
+      "rewards/rejected": -22.6777286529541,
+      "step": 440
+    },
+    {
+      "epoch": 0.8379888268156425,
+      "grad_norm": 0.0008476414368487895,
+      "learning_rate": 3.98466050817429e-07,
+      "logits/chosen": 0.6695820689201355,
+      "logits/rejected": 0.6188743114471436,
+      "logps/chosen": -183.7545928955078,
+      "logps/rejected": -378.2325439453125,
+      "loss": 0.006,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.689547538757324,
+      "rewards/margins": 15.513374328613281,
+      "rewards/rejected": -22.202922821044922,
+      "step": 450
+    },
+    {
+      "epoch": 0.8566108007448789,
+      "grad_norm": 0.05125750973820686,
+      "learning_rate": 3.148975777918095e-07,
+      "logits/chosen": 0.7254775762557983,
+      "logits/rejected": 0.7036611437797546,
+      "logps/chosen": -183.53570556640625,
+      "logps/rejected": -364.9049987792969,
+      "loss": 0.0149,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.9107441902160645,
+      "rewards/margins": 14.929256439208984,
+      "rewards/rejected": -21.840002059936523,
+      "step": 460
+    },
+    {
+      "epoch": 0.8752327746741154,
+      "grad_norm": 0.03878343477845192,
+      "learning_rate": 2.4057021530559305e-07,
+      "logits/chosen": 0.6778849959373474,
+      "logits/rejected": 0.6134467124938965,
+      "logps/chosen": -160.04876708984375,
+      "logps/rejected": -345.55792236328125,
+      "loss": 0.0523,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.244124412536621,
+      "rewards/margins": 14.95977783203125,
+      "rewards/rejected": -21.20390510559082,
+      "step": 470
+    },
+    {
+      "epoch": 0.8938547486033519,
+      "grad_norm": 0.4984408915042877,
+      "learning_rate": 1.7579830440103806e-07,
+      "logits/chosen": 0.7279980182647705,
+      "logits/rejected": 0.6711809039115906,
+      "logps/chosen": -184.20223999023438,
+      "logps/rejected": -371.1146240234375,
+      "loss": 0.0068,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.921341896057129,
+      "rewards/margins": 15.303674697875977,
+      "rewards/rejected": -22.22501564025879,
+      "step": 480
+    },
+    {
+      "epoch": 0.9124767225325885,
+      "grad_norm": 0.01238794531673193,
+      "learning_rate": 1.2085577474955534e-07,
+      "logits/chosen": 0.6070926785469055,
+      "logits/rejected": 0.5749675035476685,
+      "logps/chosen": -143.08651733398438,
+      "logps/rejected": -331.60455322265625,
+      "loss": 0.0022,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.676121711730957,
+      "rewards/margins": 14.917220115661621,
+      "rewards/rejected": -20.593341827392578,
+      "step": 490
+    },
+    {
+      "epoch": 0.931098696461825,
+      "grad_norm": 0.2902233600616455,
+      "learning_rate": 7.597498616396887e-08,
+      "logits/chosen": 0.7550373673439026,
+      "logits/rejected": 0.6526434421539307,
+      "logps/chosen": -171.61978149414062,
+      "logps/rejected": -364.5791931152344,
+      "loss": 0.0045,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.362552642822266,
+      "rewards/margins": 15.774834632873535,
+      "rewards/rejected": -22.137386322021484,
+      "step": 500
+    },
+    {
+      "epoch": 0.9497206703910615,
+      "grad_norm": 0.8458730578422546,
+      "learning_rate": 4.134574591564494e-08,
+      "logits/chosen": 0.6793017983436584,
+      "logits/rejected": 0.6275220513343811,
+      "logps/chosen": -171.3497772216797,
+      "logps/rejected": -364.985595703125,
+      "loss": 0.0296,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": -6.516476154327393,
+      "rewards/margins": 15.53239631652832,
+      "rewards/rejected": -22.048871994018555,
+      "step": 510
+    },
+    {
+      "epoch": 0.9683426443202979,
+      "grad_norm": 0.0072677601128816605,
+      "learning_rate": 1.7114506012405608e-08,
+      "logits/chosen": 0.7302192449569702,
+      "logits/rejected": 0.6580381393432617,
+      "logps/chosen": -155.1013641357422,
+      "logps/rejected": -344.39349365234375,
+      "loss": 0.0092,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.073050022125244,
+      "rewards/margins": 14.828834533691406,
+      "rewards/rejected": -20.901885986328125,
+      "step": 520
+    },
+    {
+      "epoch": 0.9869646182495344,
+      "grad_norm": 2.421435832977295,
+      "learning_rate": 3.383743832051367e-09,
+      "logits/chosen": 0.6325838565826416,
+      "logits/rejected": 0.5940303802490234,
+      "logps/chosen": -160.53848266601562,
+      "logps/rejected": -349.3865966796875,
+      "loss": 0.008,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.37663459777832,
+      "rewards/margins": 15.258707046508789,
+      "rewards/rejected": -21.63534164428711,
+      "step": 530
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 537,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 5000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.076450034624758e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ce5546fe880bfaa3e5882ec09bb54cbb5f855d97b00d554f5d40e7acc87ffb6
+size 6161

qwen2_5vl-7b/lora/povid/lora/dpo/checkpoint-537/video_preprocessor_config.json

@@ -0,0 +1,86 @@
+{
+  "_valid_kwargs_names": [
+    "do_convert_rgb",
+    "do_resize",
+    "size",
+    "size_divisor",
+    "default_to_square",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "do_pad",
+    "do_center_crop",
+    "crop_size",
+    "data_format",
+    "input_data_format",
+    "device",
+    "min_pixels",
+    "max_pixels",
+    "patch_size",
+    "temporal_patch_size",
+    "merge_size"
+  ],
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "do_center_crop": null,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": null,
+  "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
+  ],
+  "input_data_format": null,
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "model_valid_processing_keys": [
+    "do_convert_rgb",
+    "do_resize",
+    "size",
+    "size_divisor",
+    "default_to_square",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "do_pad",
+    "do_center_crop",
+    "crop_size",
+    "data_format",
+    "input_data_format",
+    "device",
+    "min_pixels",
+    "max_pixels",
+    "patch_size",
+    "temporal_patch_size",
+    "merge_size"
+  ],
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "size_divisor": null,
+  "temporal_patch_size": 2,
+  "video_processor_type": "Qwen2VLVideoProcessor"
+}

qwen2_5vl-7b/lora/povid/lora/dpo/preprocessor_config.json

@@ -0,0 +1,36 @@
+{
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "do_center_crop": null,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessorFast",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "input_data_format": null,
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}

qwen2_5vl-7b/lora/povid/lora/dpo/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_5vl-7b/lora/povid/lora/dpo/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

qwen2_5vl-7b/lora/povid/lora/dpo/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "add_bos_token": false,
+  "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
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "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,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2_5_VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

qwen2_5vl-7b/lora/povid/lora/dpo/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 8.076450034624758e+17,
+    "train_loss": 0.07847787283388707,
+    "train_runtime": 6733.1663,
+    "train_samples_per_second": 2.552,
+    "train_steps_per_second": 0.08
+}

qwen2_5vl-7b/lora/povid/lora/dpo/trainer_log.jsonl

@@ -0,0 +1,54 @@
+{"current_steps": 10, "total_steps": 537, "loss": 0.6916, "accuracy": 0.4156250059604645, "lr": 8.333333333333333e-07, "epoch": 0.0186219739292365, "percentage": 1.86, "elapsed_time": "0:02:09", "remaining_time": "1:53:49"}
+{"current_steps": 20, "total_steps": 537, "loss": 0.6911, "accuracy": 0.53125, "lr": 1.7592592592592594e-06, "epoch": 0.037243947858473, "percentage": 3.72, "elapsed_time": "0:04:16", "remaining_time": "1:50:40"}
+{"current_steps": 30, "total_steps": 537, "loss": 0.6599, "accuracy": 0.7906249761581421, "lr": 2.6851851851851856e-06, "epoch": 0.055865921787709494, "percentage": 5.59, "elapsed_time": "0:06:24", "remaining_time": "1:48:18"}
+{"current_steps": 40, "total_steps": 537, "loss": 0.5971, "accuracy": 0.8750000596046448, "lr": 3.6111111111111115e-06, "epoch": 0.074487895716946, "percentage": 7.45, "elapsed_time": "0:08:31", "remaining_time": "1:45:50"}
+{"current_steps": 50, "total_steps": 537, "loss": 0.484, "accuracy": 0.9468750357627869, "lr": 4.537037037037038e-06, "epoch": 0.0931098696461825, "percentage": 9.31, "elapsed_time": "0:10:37", "remaining_time": "1:43:29"}
+{"current_steps": 60, "total_steps": 537, "loss": 0.3382, "accuracy": 0.9750000238418579, "lr": 4.998678043309151e-06, "epoch": 0.11173184357541899, "percentage": 11.17, "elapsed_time": "0:12:39", "remaining_time": "1:40:36"}
+{"current_steps": 70, "total_steps": 537, "loss": 0.1906, "accuracy": 0.9937499761581421, "lr": 4.988110776637384e-06, "epoch": 0.1303538175046555, "percentage": 13.04, "elapsed_time": "0:14:42", "remaining_time": "1:38:08"}
+{"current_steps": 80, "total_steps": 537, "loss": 0.099, "accuracy": 0.9906250238418579, "lr": 4.967020933775366e-06, "epoch": 0.148975791433892, "percentage": 14.9, "elapsed_time": "0:16:46", "remaining_time": "1:35:51"}
+{"current_steps": 90, "total_steps": 537, "loss": 0.0433, "accuracy": 1.0, "lr": 4.935497706683698e-06, "epoch": 0.16759776536312848, "percentage": 16.76, "elapsed_time": "0:18:51", "remaining_time": "1:33:41"}
+{"current_steps": 100, "total_steps": 537, "loss": 0.0202, "accuracy": 1.0, "lr": 4.893674411596507e-06, "epoch": 0.186219739292365, "percentage": 18.62, "elapsed_time": "0:20:52", "remaining_time": "1:31:14"}
+{"current_steps": 110, "total_steps": 537, "loss": 0.0187, "accuracy": 0.9937499761581421, "lr": 4.8417279252080365e-06, "epoch": 0.2048417132216015, "percentage": 20.48, "elapsed_time": "0:22:59", "remaining_time": "1:29:16"}
+{"current_steps": 120, "total_steps": 537, "loss": 0.0341, "accuracy": 0.9937499761581421, "lr": 4.779877936635816e-06, "epoch": 0.22346368715083798, "percentage": 22.35, "elapsed_time": "0:25:07", "remaining_time": "1:27:19"}
+{"current_steps": 130, "total_steps": 537, "loss": 0.0058, "accuracy": 1.0, "lr": 4.7083860183239795e-06, "epoch": 0.24208566108007448, "percentage": 24.21, "elapsed_time": "0:27:11", "remaining_time": "1:25:08"}
+{"current_steps": 140, "total_steps": 537, "loss": 0.0061, "accuracy": 0.996874988079071, "lr": 4.627554519815999e-06, "epoch": 0.260707635009311, "percentage": 26.07, "elapsed_time": "0:29:19", "remaining_time": "1:23:08"}
+{"current_steps": 150, "total_steps": 537, "loss": 0.0062, "accuracy": 1.0, "lr": 4.537725289075239e-06, "epoch": 0.27932960893854747, "percentage": 27.93, "elapsed_time": "0:31:24", "remaining_time": "1:21:01"}
+{"current_steps": 160, "total_steps": 537, "loss": 0.0106, "accuracy": 0.9937499761581421, "lr": 4.43927822676105e-06, "epoch": 0.297951582867784, "percentage": 29.8, "elapsed_time": "0:33:31", "remaining_time": "1:18:59"}
+{"current_steps": 170, "total_steps": 537, "loss": 0.0107, "accuracy": 1.0, "lr": 4.332629679574566e-06, "epoch": 0.3165735567970205, "percentage": 31.66, "elapsed_time": "0:35:38", "remaining_time": "1:16:55"}
+{"current_steps": 180, "total_steps": 537, "loss": 0.0169, "accuracy": 0.996874988079071, "lr": 4.218230679468981e-06, "epoch": 0.33519553072625696, "percentage": 33.52, "elapsed_time": "0:37:42", "remaining_time": "1:14:47"}
+{"current_steps": 190, "total_steps": 537, "loss": 0.0095, "accuracy": 1.0, "lr": 4.096565036170936e-06, "epoch": 0.3538175046554935, "percentage": 35.38, "elapsed_time": "0:39:44", "remaining_time": "1:12:35"}
+{"current_steps": 200, "total_steps": 537, "loss": 0.0051, "accuracy": 1.0, "lr": 3.96814729108002e-06, "epoch": 0.37243947858473, "percentage": 37.24, "elapsed_time": "0:41:51", "remaining_time": "1:10:31"}
+{"current_steps": 210, "total_steps": 537, "loss": 0.0027, "accuracy": 1.0, "lr": 3.8335205411996225e-06, "epoch": 0.39106145251396646, "percentage": 39.11, "elapsed_time": "0:43:57", "remaining_time": "1:08:27"}
+{"current_steps": 220, "total_steps": 537, "loss": 0.0023, "accuracy": 1.0, "lr": 3.6932541423020707e-06, "epoch": 0.409683426443203, "percentage": 40.97, "elapsed_time": "0:46:04", "remaining_time": "1:06:22"}
+{"current_steps": 230, "total_steps": 537, "loss": 0.0034, "accuracy": 1.0, "lr": 3.5479413010416606e-06, "epoch": 0.42830540037243947, "percentage": 42.83, "elapsed_time": "0:48:10", "remaining_time": "1:04:17"}
+{"current_steps": 240, "total_steps": 537, "loss": 0.0057, "accuracy": 1.0, "lr": 3.3981965661989146e-06, "epoch": 0.44692737430167595, "percentage": 44.69, "elapsed_time": "0:50:14", "remaining_time": "1:02:10"}
+{"current_steps": 250, "total_steps": 537, "loss": 0.0108, "accuracy": 0.996874988079071, "lr": 3.2446532296659255e-06, "epoch": 0.4655493482309125, "percentage": 46.55, "elapsed_time": "0:52:19", "remaining_time": "1:00:04"}
+{"current_steps": 260, "total_steps": 537, "loss": 0.0056, "accuracy": 1.0, "lr": 3.087960648164403e-06, "epoch": 0.48417132216014896, "percentage": 48.42, "elapsed_time": "0:54:25", "remaining_time": "0:57:58"}
+{"current_steps": 270, "total_steps": 537, "loss": 0.0027, "accuracy": 1.0, "lr": 2.9287814970232688e-06, "epoch": 0.5027932960893855, "percentage": 50.28, "elapsed_time": "0:56:29", "remaining_time": "0:55:51"}
+{"current_steps": 280, "total_steps": 537, "loss": 0.0034, "accuracy": 1.0, "lr": 2.7677889676299442e-06, "epoch": 0.521415270018622, "percentage": 52.14, "elapsed_time": "0:58:34", "remaining_time": "0:53:45"}
+{"current_steps": 290, "total_steps": 537, "loss": 0.0106, "accuracy": 0.996874988079071, "lr": 2.6056639204077293e-06, "epoch": 0.5400372439478585, "percentage": 54.0, "elapsed_time": "1:00:36", "remaining_time": "0:51:37"}
+{"current_steps": 300, "total_steps": 537, "loss": 0.0023, "accuracy": 1.0, "lr": 2.443092005359736e-06, "epoch": 0.5586592178770949, "percentage": 55.87, "elapsed_time": "1:02:39", "remaining_time": "0:49:30"}
+{"current_steps": 310, "total_steps": 537, "loss": 0.007, "accuracy": 0.996874988079071, "lr": 2.280760762356999e-06, "epoch": 0.5772811918063314, "percentage": 57.73, "elapsed_time": "1:04:46", "remaining_time": "0:47:25"}
+{"current_steps": 320, "total_steps": 537, "loss": 0.0029, "accuracy": 1.0, "lr": 2.119356713434064e-06, "epoch": 0.595903165735568, "percentage": 59.59, "elapsed_time": "1:06:51", "remaining_time": "0:45:20"}
+{"current_steps": 330, "total_steps": 537, "loss": 0.0088, "accuracy": 0.996874988079071, "lr": 1.9595624593891576e-06, "epoch": 0.6145251396648045, "percentage": 61.45, "elapsed_time": "1:08:56", "remaining_time": "0:43:14"}
+{"current_steps": 340, "total_steps": 537, "loss": 0.0058, "accuracy": 0.996874988079071, "lr": 1.8020537929678192e-06, "epoch": 0.633147113594041, "percentage": 63.31, "elapsed_time": "1:11:03", "remaining_time": "0:41:10"}
+{"current_steps": 350, "total_steps": 537, "loss": 0.0028, "accuracy": 1.0, "lr": 1.647496840838767e-06, "epoch": 0.6517690875232774, "percentage": 65.18, "elapsed_time": "1:13:08", "remaining_time": "0:39:04"}
+{"current_steps": 360, "total_steps": 537, "loss": 0.0088, "accuracy": 0.996874988079071, "lr": 1.4965452464489817e-06, "epoch": 0.6703910614525139, "percentage": 67.04, "elapsed_time": "1:15:12", "remaining_time": "0:36:58"}
+{"current_steps": 370, "total_steps": 537, "loss": 0.008, "accuracy": 0.9937499761581421, "lr": 1.3498374056721198e-06, "epoch": 0.6890130353817505, "percentage": 68.9, "elapsed_time": "1:17:16", "remaining_time": "0:34:52"}
+{"current_steps": 380, "total_steps": 537, "loss": 0.0027, "accuracy": 1.0, "lr": 1.2079937669410968e-06, "epoch": 0.707635009310987, "percentage": 70.76, "elapsed_time": "1:19:21", "remaining_time": "0:32:47"}
+{"current_steps": 390, "total_steps": 537, "loss": 0.001, "accuracy": 1.0, "lr": 1.0716142072829552e-06, "epoch": 0.7262569832402235, "percentage": 72.63, "elapsed_time": "1:21:23", "remaining_time": "0:30:40"}
+{"current_steps": 400, "total_steps": 537, "loss": 0.0022, "accuracy": 1.0, "lr": 9.412754953531664e-07, "epoch": 0.74487895716946, "percentage": 74.49, "elapsed_time": "1:23:31", "remaining_time": "0:28:36"}
+{"current_steps": 410, "total_steps": 537, "loss": 0.0074, "accuracy": 1.0, "lr": 8.17528852198515e-07, "epoch": 0.7635009310986964, "percentage": 76.35, "elapsed_time": "1:25:37", "remaining_time": "0:26:31"}
+{"current_steps": 420, "total_steps": 537, "loss": 0.0188, "accuracy": 0.996874988079071, "lr": 7.008976200644962e-07, "epoch": 0.7821229050279329, "percentage": 78.21, "elapsed_time": "1:27:41", "remaining_time": "0:24:25"}
+{"current_steps": 430, "total_steps": 537, "loss": 0.0055, "accuracy": 1.0, "lr": 5.918750491061323e-07, "epoch": 0.8007448789571695, "percentage": 80.07, "elapsed_time": "1:29:46", "remaining_time": "0:22:20"}
+{"current_steps": 440, "total_steps": 537, "loss": 0.005, "accuracy": 0.996874988079071, "lr": 4.909222113625545e-07, "epoch": 0.819366852886406, "percentage": 81.94, "elapsed_time": "1:31:51", "remaining_time": "0:20:14"}
+{"current_steps": 450, "total_steps": 537, "loss": 0.006, "accuracy": 0.996874988079071, "lr": 3.98466050817429e-07, "epoch": 0.8379888268156425, "percentage": 83.8, "elapsed_time": "1:33:58", "remaining_time": "0:18:10"}
+{"current_steps": 460, "total_steps": 537, "loss": 0.0149, "accuracy": 0.9937499761581421, "lr": 3.148975777918095e-07, "epoch": 0.8566108007448789, "percentage": 85.66, "elapsed_time": "1:36:03", "remaining_time": "0:16:04"}
+{"current_steps": 470, "total_steps": 537, "loss": 0.0523, "accuracy": 0.996874988079071, "lr": 2.4057021530559305e-07, "epoch": 0.8752327746741154, "percentage": 87.52, "elapsed_time": "1:38:11", "remaining_time": "0:13:59"}
+{"current_steps": 480, "total_steps": 537, "loss": 0.0068, "accuracy": 1.0, "lr": 1.7579830440103806e-07, "epoch": 0.8938547486033519, "percentage": 89.39, "elapsed_time": "1:40:23", "remaining_time": "0:11:55"}
+{"current_steps": 490, "total_steps": 537, "loss": 0.0022, "accuracy": 1.0, "lr": 1.2085577474955534e-07, "epoch": 0.9124767225325885, "percentage": 91.25, "elapsed_time": "1:42:23", "remaining_time": "0:09:49"}
+{"current_steps": 500, "total_steps": 537, "loss": 0.0045, "accuracy": 1.0, "lr": 7.597498616396887e-08, "epoch": 0.931098696461825, "percentage": 93.11, "elapsed_time": "1:44:27", "remaining_time": "0:07:43"}
+{"current_steps": 510, "total_steps": 537, "loss": 0.0296, "accuracy": 0.987500011920929, "lr": 4.134574591564494e-08, "epoch": 0.9497206703910615, "percentage": 94.97, "elapsed_time": "1:46:31", "remaining_time": "0:05:38"}
+{"current_steps": 520, "total_steps": 537, "loss": 0.0092, "accuracy": 0.996874988079071, "lr": 1.7114506012405608e-08, "epoch": 0.9683426443202979, "percentage": 96.83, "elapsed_time": "1:48:34", "remaining_time": "0:03:32"}
+{"current_steps": 530, "total_steps": 537, "loss": 0.008, "accuracy": 1.0, "lr": 3.383743832051367e-09, "epoch": 0.9869646182495344, "percentage": 98.7, "elapsed_time": "1:50:39", "remaining_time": "0:01:27"}
+{"current_steps": 537, "total_steps": 537, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "1:52:11", "remaining_time": "0:00:00"}

qwen2_5vl-7b/lora/povid/lora/dpo/trainer_state.json

@@ -0,0 +1,838 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 537,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0186219739292365,
+      "grad_norm": 2.07263445854187,
+      "learning_rate": 8.333333333333333e-07,
+      "logits/chosen": 1.9083625078201294,
+      "logits/rejected": 1.9776146411895752,
+      "logps/chosen": -99.38154602050781,
+      "logps/rejected": -137.68942260742188,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.4156250059604645,
+      "rewards/chosen": 0.002753954380750656,
+      "rewards/margins": 0.0035315221175551414,
+      "rewards/rejected": -0.0007775675039738417,
+      "step": 10
+    },
+    {
+      "epoch": 0.037243947858473,
+      "grad_norm": 2.077322006225586,
+      "learning_rate": 1.7592592592592594e-06,
+      "logits/chosen": 1.7842262983322144,
+      "logits/rejected": 1.8155817985534668,
+      "logps/chosen": -103.61813354492188,
+      "logps/rejected": -137.05908203125,
+      "loss": 0.6911,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": 0.003043058328330517,
+      "rewards/margins": 0.004756969399750233,
+      "rewards/rejected": -0.0017139108385890722,
+      "step": 20
+    },
+    {
+      "epoch": 0.055865921787709494,
+      "grad_norm": 2.449907064437866,
+      "learning_rate": 2.6851851851851856e-06,
+      "logits/chosen": 1.9791193008422852,
+      "logits/rejected": 2.021622657775879,
+      "logps/chosen": -102.76006317138672,
+      "logps/rejected": -141.06813049316406,
+      "loss": 0.6599,
+      "rewards/accuracies": 0.7906249761581421,
+      "rewards/chosen": 0.03256123140454292,
+      "rewards/margins": 0.07064400613307953,
+      "rewards/rejected": -0.038082774728536606,
+      "step": 30
+    },
+    {
+      "epoch": 0.074487895716946,
+      "grad_norm": 1.980841875076294,
+      "learning_rate": 3.6111111111111115e-06,
+      "logits/chosen": 1.952810525894165,
+      "logits/rejected": 2.003131151199341,
+      "logps/chosen": -111.52398681640625,
+      "logps/rejected": -149.88365173339844,
+      "loss": 0.5971,
+      "rewards/accuracies": 0.8750000596046448,
+      "rewards/chosen": 0.06918186694383621,
+      "rewards/margins": 0.2191462516784668,
+      "rewards/rejected": -0.14996439218521118,
+      "step": 40
+    },
+    {
+      "epoch": 0.0931098696461825,
+      "grad_norm": 1.909422516822815,
+      "learning_rate": 4.537037037037038e-06,
+      "logits/chosen": 1.8376410007476807,
+      "logits/rejected": 1.8602056503295898,
+      "logps/chosen": -98.7303237915039,
+      "logps/rejected": -145.1428680419922,
+      "loss": 0.484,
+      "rewards/accuracies": 0.9468750357627869,
+      "rewards/chosen": 0.06579304486513138,
+      "rewards/margins": 0.5358557105064392,
+      "rewards/rejected": -0.4700627028942108,
+      "step": 50
+    },
+    {
+      "epoch": 0.11173184357541899,
+      "grad_norm": 1.4733169078826904,
+      "learning_rate": 4.998678043309151e-06,
+      "logits/chosen": 1.9728565216064453,
+      "logits/rejected": 2.0448157787323,
+      "logps/chosen": -102.9524917602539,
+      "logps/rejected": -149.02821350097656,
+      "loss": 0.3382,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": -0.0847957655787468,
+      "rewards/margins": 1.1177407503128052,
+      "rewards/rejected": -1.2025364637374878,
+      "step": 60
+    },
+    {
+      "epoch": 0.1303538175046555,
+      "grad_norm": 0.6721972823143005,
+      "learning_rate": 4.988110776637384e-06,
+      "logits/chosen": 1.727777361869812,
+      "logits/rejected": 1.7899367809295654,
+      "logps/chosen": -102.1249008178711,
+      "logps/rejected": -157.9233856201172,
+      "loss": 0.1906,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -0.261747270822525,
+      "rewards/margins": 2.095768690109253,
+      "rewards/rejected": -2.357516288757324,
+      "step": 70
+    },
+    {
+      "epoch": 0.148975791433892,
+      "grad_norm": 0.6819734573364258,
+      "learning_rate": 4.967020933775366e-06,
+      "logits/chosen": 1.8152174949645996,
+      "logits/rejected": 1.884362816810608,
+      "logps/chosen": -110.66182708740234,
+      "logps/rejected": -183.3290557861328,
+      "loss": 0.099,
+      "rewards/accuracies": 0.9906250238418579,
+      "rewards/chosen": -0.6374696493148804,
+      "rewards/margins": 3.595221519470215,
+      "rewards/rejected": -4.232691287994385,
+      "step": 80
+    },
+    {
+      "epoch": 0.16759776536312848,
+      "grad_norm": 0.6661988496780396,
+      "learning_rate": 4.935497706683698e-06,
+      "logits/chosen": 1.5606900453567505,
+      "logits/rejected": 1.627284288406372,
+      "logps/chosen": -123.47517395019531,
+      "logps/rejected": -210.38690185546875,
+      "loss": 0.0433,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.6986806392669678,
+      "rewards/margins": 5.403690814971924,
+      "rewards/rejected": -7.1023712158203125,
+      "step": 90
+    },
+    {
+      "epoch": 0.186219739292365,
+      "grad_norm": 0.3505627512931824,
+      "learning_rate": 4.893674411596507e-06,
+      "logits/chosen": 1.4444490671157837,
+      "logits/rejected": 1.496826171875,
+      "logps/chosen": -132.90512084960938,
+      "logps/rejected": -247.66297912597656,
+      "loss": 0.0202,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -3.0164995193481445,
+      "rewards/margins": 7.708347797393799,
+      "rewards/rejected": -10.724847793579102,
+      "step": 100
+    },
+    {
+      "epoch": 0.2048417132216015,
+      "grad_norm": 0.08065284043550491,
+      "learning_rate": 4.8417279252080365e-06,
+      "logits/chosen": 1.1240510940551758,
+      "logits/rejected": 1.1461639404296875,
+      "logps/chosen": -173.87986755371094,
+      "logps/rejected": -296.1276550292969,
+      "loss": 0.0187,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -5.6900763511657715,
+      "rewards/margins": 9.046419143676758,
+      "rewards/rejected": -14.736495971679688,
+      "step": 110
+    },
+    {
+      "epoch": 0.22346368715083798,
+      "grad_norm": 0.1960441768169403,
+      "learning_rate": 4.779877936635816e-06,
+      "logits/chosen": 0.9913089275360107,
+      "logits/rejected": 1.0386312007904053,
+      "logps/chosen": -169.66835021972656,
+      "logps/rejected": -305.98406982421875,
+      "loss": 0.0341,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.384186744689941,
+      "rewards/margins": 10.059325218200684,
+      "rewards/rejected": -16.443510055541992,
+      "step": 120
+    },
+    {
+      "epoch": 0.24208566108007448,
+      "grad_norm": 0.08494703471660614,
+      "learning_rate": 4.7083860183239795e-06,
+      "logits/chosen": 1.072335958480835,
+      "logits/rejected": 1.071942925453186,
+      "logps/chosen": -155.18902587890625,
+      "logps/rejected": -295.8493957519531,
+      "loss": 0.0058,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.6623382568359375,
+      "rewards/margins": 10.138053894042969,
+      "rewards/rejected": -15.800392150878906,
+      "step": 130
+    },
+    {
+      "epoch": 0.260707635009311,
+      "grad_norm": 0.40349674224853516,
+      "learning_rate": 4.627554519815999e-06,
+      "logits/chosen": 0.9601426124572754,
+      "logits/rejected": 0.9602993726730347,
+      "logps/chosen": -172.6103515625,
+      "logps/rejected": -322.26434326171875,
+      "loss": 0.0061,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.130578994750977,
+      "rewards/margins": 11.377153396606445,
+      "rewards/rejected": -17.507732391357422,
+      "step": 140
+    },
+    {
+      "epoch": 0.27932960893854747,
+      "grad_norm": 0.3065098226070404,
+      "learning_rate": 4.537725289075239e-06,
+      "logits/chosen": 1.027471661567688,
+      "logits/rejected": 0.9988879561424255,
+      "logps/chosen": -174.87049865722656,
+      "logps/rejected": -326.5740966796875,
+      "loss": 0.0062,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.688374042510986,
+      "rewards/margins": 11.755913734436035,
+      "rewards/rejected": -18.44428825378418,
+      "step": 150
+    },
+    {
+      "epoch": 0.297951582867784,
+      "grad_norm": 0.6753863096237183,
+      "learning_rate": 4.43927822676105e-06,
+      "logits/chosen": 0.8613717555999756,
+      "logits/rejected": 0.8766331076622009,
+      "logps/chosen": -178.57435607910156,
+      "logps/rejected": -329.76153564453125,
+      "loss": 0.0106,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.277381420135498,
+      "rewards/margins": 11.393813133239746,
+      "rewards/rejected": -17.67119598388672,
+      "step": 160
+    },
+    {
+      "epoch": 0.3165735567970205,
+      "grad_norm": 0.832106351852417,
+      "learning_rate": 4.332629679574566e-06,
+      "logits/chosen": 0.909426748752594,
+      "logits/rejected": 0.9274786114692688,
+      "logps/chosen": -159.40325927734375,
+      "logps/rejected": -308.1123046875,
+      "loss": 0.0107,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.586008071899414,
+      "rewards/margins": 11.150063514709473,
+      "rewards/rejected": -16.73607063293457,
+      "step": 170
+    },
+    {
+      "epoch": 0.33519553072625696,
+      "grad_norm": 0.08242648094892502,
+      "learning_rate": 4.218230679468981e-06,
+      "logits/chosen": 0.9288380742073059,
+      "logits/rejected": 0.9024308919906616,
+      "logps/chosen": -156.82220458984375,
+      "logps/rejected": -304.9267578125,
+      "loss": 0.0169,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -4.636939525604248,
+      "rewards/margins": 11.34218692779541,
+      "rewards/rejected": -15.979126930236816,
+      "step": 180
+    },
+    {
+      "epoch": 0.3538175046554935,
+      "grad_norm": 0.08803419768810272,
+      "learning_rate": 4.096565036170936e-06,
+      "logits/chosen": 0.9101991057395935,
+      "logits/rejected": 0.8393880724906921,
+      "logps/chosen": -131.576171875,
+      "logps/rejected": -280.4066162109375,
+      "loss": 0.0095,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -3.821293354034424,
+      "rewards/margins": 11.146876335144043,
+      "rewards/rejected": -14.968170166015625,
+      "step": 190
+    },
+    {
+      "epoch": 0.37243947858473,
+      "grad_norm": 0.020818937569856644,
+      "learning_rate": 3.96814729108002e-06,
+      "logits/chosen": 0.7776044607162476,
+      "logits/rejected": 0.7479174137115479,
+      "logps/chosen": -150.67852783203125,
+      "logps/rejected": -303.72247314453125,
+      "loss": 0.0051,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -4.321155071258545,
+      "rewards/margins": 11.680810928344727,
+      "rewards/rejected": -16.00196647644043,
+      "step": 200
+    },
+    {
+      "epoch": 0.39106145251396646,
+      "grad_norm": 0.5023415684700012,
+      "learning_rate": 3.8335205411996225e-06,
+      "logits/chosen": 0.838572084903717,
+      "logits/rejected": 0.7576170563697815,
+      "logps/chosen": -154.36627197265625,
+      "logps/rejected": -313.2589111328125,
+      "loss": 0.0027,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.044532775878906,
+      "rewards/margins": 12.347441673278809,
+      "rewards/rejected": -17.3919734954834,
+      "step": 210
+    },
+    {
+      "epoch": 0.409683426443203,
+      "grad_norm": 0.007361250929534435,
+      "learning_rate": 3.6932541423020707e-06,
+      "logits/chosen": 0.7803331613540649,
+      "logits/rejected": 0.7118293642997742,
+      "logps/chosen": -171.97296142578125,
+      "logps/rejected": -343.0221252441406,
+      "loss": 0.0023,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.363333702087402,
+      "rewards/margins": 13.190146446228027,
+      "rewards/rejected": -19.553478240966797,
+      "step": 220
+    },
+    {
+      "epoch": 0.42830540037243947,
+      "grad_norm": 0.04080897197127342,
+      "learning_rate": 3.5479413010416606e-06,
+      "logits/chosen": 0.6650150418281555,
+      "logits/rejected": 0.6223341822624207,
+      "logps/chosen": -160.4796142578125,
+      "logps/rejected": -337.9312744140625,
+      "loss": 0.0034,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.2509541511535645,
+      "rewards/margins": 13.757020950317383,
+      "rewards/rejected": -20.007976531982422,
+      "step": 230
+    },
+    {
+      "epoch": 0.44692737430167595,
+      "grad_norm": 0.4697687327861786,
+      "learning_rate": 3.3981965661989146e-06,
+      "logits/chosen": 0.7506241202354431,
+      "logits/rejected": 0.7021348476409912,
+      "logps/chosen": -185.2721710205078,
+      "logps/rejected": -356.75872802734375,
+      "loss": 0.0057,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -7.140838623046875,
+      "rewards/margins": 13.574752807617188,
+      "rewards/rejected": -20.71558952331543,
+      "step": 240
+    },
+    {
+      "epoch": 0.4655493482309125,
+      "grad_norm": 4.2021660804748535,
+      "learning_rate": 3.2446532296659255e-06,
+      "logits/chosen": 0.8130508661270142,
+      "logits/rejected": 0.7506972551345825,
+      "logps/chosen": -163.47772216796875,
+      "logps/rejected": -339.3100280761719,
+      "loss": 0.0108,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.541469573974609,
+      "rewards/margins": 13.666439056396484,
+      "rewards/rejected": -20.207908630371094,
+      "step": 250
+    },
+    {
+      "epoch": 0.48417132216014896,
+      "grad_norm": 1.2759469747543335,
+      "learning_rate": 3.087960648164403e-06,
+      "logits/chosen": 0.615781307220459,
+      "logits/rejected": 0.5805816650390625,
+      "logps/chosen": -166.66835021972656,
+      "logps/rejected": -341.08709716796875,
+      "loss": 0.0056,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.2712554931640625,
+      "rewards/margins": 13.863871574401855,
+      "rewards/rejected": -20.1351261138916,
+      "step": 260
+    },
+    {
+      "epoch": 0.5027932960893855,
+      "grad_norm": 0.08350568264722824,
+      "learning_rate": 2.9287814970232688e-06,
+      "logits/chosen": 0.7528300285339355,
+      "logits/rejected": 0.7390813231468201,
+      "logps/chosen": -164.2382354736328,
+      "logps/rejected": -337.09271240234375,
+      "loss": 0.0027,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.999736309051514,
+      "rewards/margins": 14.001705169677734,
+      "rewards/rejected": -20.001441955566406,
+      "step": 270
+    },
+    {
+      "epoch": 0.521415270018622,
+      "grad_norm": 0.29949527978897095,
+      "learning_rate": 2.7677889676299442e-06,
+      "logits/chosen": 0.6687740087509155,
+      "logits/rejected": 0.6045177578926086,
+      "logps/chosen": -158.6308135986328,
+      "logps/rejected": -349.30316162109375,
+      "loss": 0.0034,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.776445388793945,
+      "rewards/margins": 15.19276237487793,
+      "rewards/rejected": -20.969207763671875,
+      "step": 280
+    },
+    {
+      "epoch": 0.5400372439478585,
+      "grad_norm": 0.006012032739818096,
+      "learning_rate": 2.6056639204077293e-06,
+      "logits/chosen": 0.6841985583305359,
+      "logits/rejected": 0.6416488885879517,
+      "logps/chosen": -152.00906372070312,
+      "logps/rejected": -336.58306884765625,
+      "loss": 0.0106,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -5.997591972351074,
+      "rewards/margins": 14.582725524902344,
+      "rewards/rejected": -20.5803165435791,
+      "step": 290
+    },
+    {
+      "epoch": 0.5586592178770949,
+      "grad_norm": 0.10845273733139038,
+      "learning_rate": 2.443092005359736e-06,
+      "logits/chosen": 0.7344327569007874,
+      "logits/rejected": 0.6978190541267395,
+      "logps/chosen": -161.67117309570312,
+      "logps/rejected": -342.5047912597656,
+      "loss": 0.0023,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.260890960693359,
+      "rewards/margins": 14.143697738647461,
+      "rewards/rejected": -20.40458869934082,
+      "step": 300
+    },
+    {
+      "epoch": 0.5772811918063314,
+      "grad_norm": 4.161528587341309,
+      "learning_rate": 2.280760762356999e-06,
+      "logits/chosen": 0.6338777542114258,
+      "logits/rejected": 0.5441797375679016,
+      "logps/chosen": -177.587646484375,
+      "logps/rejected": -365.2238464355469,
+      "loss": 0.007,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.9434590339660645,
+      "rewards/margins": 14.978950500488281,
+      "rewards/rejected": -21.922409057617188,
+      "step": 310
+    },
+    {
+      "epoch": 0.595903165735568,
+      "grad_norm": 0.15292654931545258,
+      "learning_rate": 2.119356713434064e-06,
+      "logits/chosen": 0.7987104654312134,
+      "logits/rejected": 0.7266597151756287,
+      "logps/chosen": -167.17431640625,
+      "logps/rejected": -352.7719421386719,
+      "loss": 0.0029,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.418148994445801,
+      "rewards/margins": 14.750093460083008,
+      "rewards/rejected": -21.168243408203125,
+      "step": 320
+    },
+    {
+      "epoch": 0.6145251396648045,
+      "grad_norm": 0.01223732065409422,
+      "learning_rate": 1.9595624593891576e-06,
+      "logits/chosen": 0.7314679622650146,
+      "logits/rejected": 0.6928003430366516,
+      "logps/chosen": -170.919921875,
+      "logps/rejected": -359.259033203125,
+      "loss": 0.0088,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.704370021820068,
+      "rewards/margins": 15.030937194824219,
+      "rewards/rejected": -21.735305786132812,
+      "step": 330
+    },
+    {
+      "epoch": 0.633147113594041,
+      "grad_norm": 0.3380753993988037,
+      "learning_rate": 1.8020537929678192e-06,
+      "logits/chosen": 0.6550232768058777,
+      "logits/rejected": 0.6400955319404602,
+      "logps/chosen": -166.73977661132812,
+      "logps/rejected": -344.7969970703125,
+      "loss": 0.0058,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.093208312988281,
+      "rewards/margins": 14.206398010253906,
+      "rewards/rejected": -20.299606323242188,
+      "step": 340
+    },
+    {
+      "epoch": 0.6517690875232774,
+      "grad_norm": 0.03426047042012215,
+      "learning_rate": 1.647496840838767e-06,
+      "logits/chosen": 0.6896477341651917,
+      "logits/rejected": 0.6510384678840637,
+      "logps/chosen": -165.2506103515625,
+      "logps/rejected": -338.61004638671875,
+      "loss": 0.0028,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.837466716766357,
+      "rewards/margins": 13.876227378845215,
+      "rewards/rejected": -19.713693618774414,
+      "step": 350
+    },
+    {
+      "epoch": 0.6703910614525139,
+      "grad_norm": 0.845697820186615,
+      "learning_rate": 1.4965452464489817e-06,
+      "logits/chosen": 0.7214924693107605,
+      "logits/rejected": 0.6409821510314941,
+      "logps/chosen": -158.24691772460938,
+      "logps/rejected": -346.9435119628906,
+      "loss": 0.0088,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -5.867282867431641,
+      "rewards/margins": 15.158857345581055,
+      "rewards/rejected": -21.026140213012695,
+      "step": 360
+    },
+    {
+      "epoch": 0.6890130353817505,
+      "grad_norm": 0.03940608352422714,
+      "learning_rate": 1.3498374056721198e-06,
+      "logits/chosen": 0.5552241802215576,
+      "logits/rejected": 0.4949037730693817,
+      "logps/chosen": -156.23068237304688,
+      "logps/rejected": -345.18939208984375,
+      "loss": 0.008,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.261874198913574,
+      "rewards/margins": 15.237826347351074,
+      "rewards/rejected": -21.499698638916016,
+      "step": 370
+    },
+    {
+      "epoch": 0.707635009310987,
+      "grad_norm": 0.020929953083395958,
+      "learning_rate": 1.2079937669410968e-06,
+      "logits/chosen": 0.5396690368652344,
+      "logits/rejected": 0.5160779356956482,
+      "logps/chosen": -156.75762939453125,
+      "logps/rejected": -348.7319030761719,
+      "loss": 0.0027,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.514359474182129,
+      "rewards/margins": 15.283174514770508,
+      "rewards/rejected": -21.797534942626953,
+      "step": 380
+    },
+    {
+      "epoch": 0.7262569832402235,
+      "grad_norm": 0.022455064579844475,
+      "learning_rate": 1.0716142072829552e-06,
+      "logits/chosen": 0.6602338552474976,
+      "logits/rejected": 0.5932304263114929,
+      "logps/chosen": -169.5976104736328,
+      "logps/rejected": -352.7994079589844,
+      "loss": 0.001,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.581348419189453,
+      "rewards/margins": 15.190673828125,
+      "rewards/rejected": -21.772024154663086,
+      "step": 390
+    },
+    {
+      "epoch": 0.74487895716946,
+      "grad_norm": 0.008257913403213024,
+      "learning_rate": 9.412754953531664e-07,
+      "logits/chosen": 0.6103745698928833,
+      "logits/rejected": 0.5022205114364624,
+      "logps/chosen": -176.54592895507812,
+      "logps/rejected": -374.6795654296875,
+      "loss": 0.0022,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.988586902618408,
+      "rewards/margins": 15.85957145690918,
+      "rewards/rejected": -22.848159790039062,
+      "step": 400
+    },
+    {
+      "epoch": 0.7635009310986964,
+      "grad_norm": 0.027111830189824104,
+      "learning_rate": 8.17528852198515e-07,
+      "logits/chosen": 0.5302518606185913,
+      "logits/rejected": 0.4578227996826172,
+      "logps/chosen": -170.05894470214844,
+      "logps/rejected": -358.9085388183594,
+      "loss": 0.0074,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.636836051940918,
+      "rewards/margins": 15.394896507263184,
+      "rewards/rejected": -22.0317325592041,
+      "step": 410
+    },
+    {
+      "epoch": 0.7821229050279329,
+      "grad_norm": 0.07998599857091904,
+      "learning_rate": 7.008976200644962e-07,
+      "logits/chosen": 0.609848141670227,
+      "logits/rejected": 0.5438084006309509,
+      "logps/chosen": -172.95252990722656,
+      "logps/rejected": -362.3902893066406,
+      "loss": 0.0188,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.9202423095703125,
+      "rewards/margins": 15.082725524902344,
+      "rewards/rejected": -22.002967834472656,
+      "step": 420
+    },
+    {
+      "epoch": 0.8007448789571695,
+      "grad_norm": 0.13893048465251923,
+      "learning_rate": 5.918750491061323e-07,
+      "logits/chosen": 0.5778518915176392,
+      "logits/rejected": 0.5182854533195496,
+      "logps/chosen": -158.56326293945312,
+      "logps/rejected": -344.1929626464844,
+      "loss": 0.0055,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.2186408042907715,
+      "rewards/margins": 15.193116188049316,
+      "rewards/rejected": -21.411758422851562,
+      "step": 430
+    },
+    {
+      "epoch": 0.819366852886406,
+      "grad_norm": 2.4267919063568115,
+      "learning_rate": 4.909222113625545e-07,
+      "logits/chosen": 0.5850950479507446,
+      "logits/rejected": 0.5577961802482605,
+      "logps/chosen": -185.70611572265625,
+      "logps/rejected": -377.9247741699219,
+      "loss": 0.005,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -7.237918376922607,
+      "rewards/margins": 15.439810752868652,
+      "rewards/rejected": -22.6777286529541,
+      "step": 440
+    },
+    {
+      "epoch": 0.8379888268156425,
+      "grad_norm": 0.0008476414368487895,
+      "learning_rate": 3.98466050817429e-07,
+      "logits/chosen": 0.6695820689201355,
+      "logits/rejected": 0.6188743114471436,
+      "logps/chosen": -183.7545928955078,
+      "logps/rejected": -378.2325439453125,
+      "loss": 0.006,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.689547538757324,
+      "rewards/margins": 15.513374328613281,
+      "rewards/rejected": -22.202922821044922,
+      "step": 450
+    },
+    {
+      "epoch": 0.8566108007448789,
+      "grad_norm": 0.05125750973820686,
+      "learning_rate": 3.148975777918095e-07,
+      "logits/chosen": 0.7254775762557983,
+      "logits/rejected": 0.7036611437797546,
+      "logps/chosen": -183.53570556640625,
+      "logps/rejected": -364.9049987792969,
+      "loss": 0.0149,
+      "rewards/accuracies": 0.9937499761581421,
+      "rewards/chosen": -6.9107441902160645,
+      "rewards/margins": 14.929256439208984,
+      "rewards/rejected": -21.840002059936523,
+      "step": 460
+    },
+    {
+      "epoch": 0.8752327746741154,
+      "grad_norm": 0.03878343477845192,
+      "learning_rate": 2.4057021530559305e-07,
+      "logits/chosen": 0.6778849959373474,
+      "logits/rejected": 0.6134467124938965,
+      "logps/chosen": -160.04876708984375,
+      "logps/rejected": -345.55792236328125,
+      "loss": 0.0523,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.244124412536621,
+      "rewards/margins": 14.95977783203125,
+      "rewards/rejected": -21.20390510559082,
+      "step": 470
+    },
+    {
+      "epoch": 0.8938547486033519,
+      "grad_norm": 0.4984408915042877,
+      "learning_rate": 1.7579830440103806e-07,
+      "logits/chosen": 0.7279980182647705,
+      "logits/rejected": 0.6711809039115906,
+      "logps/chosen": -184.20223999023438,
+      "logps/rejected": -371.1146240234375,
+      "loss": 0.0068,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.921341896057129,
+      "rewards/margins": 15.303674697875977,
+      "rewards/rejected": -22.22501564025879,
+      "step": 480
+    },
+    {
+      "epoch": 0.9124767225325885,
+      "grad_norm": 0.01238794531673193,
+      "learning_rate": 1.2085577474955534e-07,
+      "logits/chosen": 0.6070926785469055,
+      "logits/rejected": 0.5749675035476685,
+      "logps/chosen": -143.08651733398438,
+      "logps/rejected": -331.60455322265625,
+      "loss": 0.0022,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -5.676121711730957,
+      "rewards/margins": 14.917220115661621,
+      "rewards/rejected": -20.593341827392578,
+      "step": 490
+    },
+    {
+      "epoch": 0.931098696461825,
+      "grad_norm": 0.2902233600616455,
+      "learning_rate": 7.597498616396887e-08,
+      "logits/chosen": 0.7550373673439026,
+      "logits/rejected": 0.6526434421539307,
+      "logps/chosen": -171.61978149414062,
+      "logps/rejected": -364.5791931152344,
+      "loss": 0.0045,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.362552642822266,
+      "rewards/margins": 15.774834632873535,
+      "rewards/rejected": -22.137386322021484,
+      "step": 500
+    },
+    {
+      "epoch": 0.9497206703910615,
+      "grad_norm": 0.8458730578422546,
+      "learning_rate": 4.134574591564494e-08,
+      "logits/chosen": 0.6793017983436584,
+      "logits/rejected": 0.6275220513343811,
+      "logps/chosen": -171.3497772216797,
+      "logps/rejected": -364.985595703125,
+      "loss": 0.0296,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": -6.516476154327393,
+      "rewards/margins": 15.53239631652832,
+      "rewards/rejected": -22.048871994018555,
+      "step": 510
+    },
+    {
+      "epoch": 0.9683426443202979,
+      "grad_norm": 0.0072677601128816605,
+      "learning_rate": 1.7114506012405608e-08,
+      "logits/chosen": 0.7302192449569702,
+      "logits/rejected": 0.6580381393432617,
+      "logps/chosen": -155.1013641357422,
+      "logps/rejected": -344.39349365234375,
+      "loss": 0.0092,
+      "rewards/accuracies": 0.996874988079071,
+      "rewards/chosen": -6.073050022125244,
+      "rewards/margins": 14.828834533691406,
+      "rewards/rejected": -20.901885986328125,
+      "step": 520
+    },
+    {
+      "epoch": 0.9869646182495344,
+      "grad_norm": 2.421435832977295,
+      "learning_rate": 3.383743832051367e-09,
+      "logits/chosen": 0.6325838565826416,
+      "logits/rejected": 0.5940303802490234,
+      "logps/chosen": -160.53848266601562,
+      "logps/rejected": -349.3865966796875,
+      "loss": 0.008,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -6.37663459777832,
+      "rewards/margins": 15.258707046508789,
+      "rewards/rejected": -21.63534164428711,
+      "step": 530
+    },
+    {
+      "epoch": 1.0,
+      "step": 537,
+      "total_flos": 8.076450034624758e+17,
+      "train_loss": 0.07847787283388707,
+      "train_runtime": 6733.1663,
+      "train_samples_per_second": 2.552,
+      "train_steps_per_second": 0.08
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 537,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 5000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.076450034624758e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

qwen2_5vl-7b/lora/povid/lora/dpo/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ce5546fe880bfaa3e5882ec09bb54cbb5f855d97b00d554f5d40e7acc87ffb6
+size 6161

qwen2_5vl-7b/lora/povid/lora/dpo/video_preprocessor_config.json

@@ -0,0 +1,86 @@
+{
+  "_valid_kwargs_names": [
+    "do_convert_rgb",
+    "do_resize",
+    "size",
+    "size_divisor",
+    "default_to_square",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "do_pad",
+    "do_center_crop",
+    "crop_size",
+    "data_format",
+    "input_data_format",
+    "device",
+    "min_pixels",
+    "max_pixels",
+    "patch_size",
+    "temporal_patch_size",
+    "merge_size"
+  ],
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "do_center_crop": null,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": null,
+  "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
+  ],
+  "input_data_format": null,
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "model_valid_processing_keys": [
+    "do_convert_rgb",
+    "do_resize",
+    "size",
+    "size_divisor",
+    "default_to_square",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "do_pad",
+    "do_center_crop",
+    "crop_size",
+    "data_format",
+    "input_data_format",
+    "device",
+    "min_pixels",
+    "max_pixels",
+    "patch_size",
+    "temporal_patch_size",
+    "merge_size"
+  ],
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "size_divisor": null,
+  "temporal_patch_size": 2,
+  "video_processor_type": "Qwen2VLVideoProcessor"
+}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/README.md

@@ -0,0 +1,62 @@
+---
+library_name: peft
+license: other
+base_model: Qwen/Qwen2.5-VL-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: dpo
+  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. -->
+
+# dpo
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct) on the rlaif_v 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: 5e-06
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
+- total_eval_batch_size: 32
+- 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: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.15.2
+- Transformers 4.52.4
+- Pytorch 2.7.1+cu126
+- Datasets 3.6.0
+- Tokenizers 0.21.1

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/adapter_config.json

@@ -0,0 +1,120 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "layers.24.mlp.gate_proj",
+    "layers.16.mlp.down_proj",
+    "layers.1.mlp.up_proj",
+    "layers.6.mlp.down_proj",
+    "layers.4.mlp.down_proj",
+    "layers.8.mlp.gate_proj",
+    "layers.9.mlp.gate_proj",
+    "layers.3.mlp.down_proj",
+    "layers.17.mlp.gate_proj",
+    "layers.0.mlp.up_proj",
+    "layers.27.mlp.down_proj",
+    "layers.2.mlp.down_proj",
+    "layers.13.mlp.up_proj",
+    "layers.23.mlp.down_proj",
+    "layers.18.mlp.up_proj",
+    "layers.23.mlp.up_proj",
+    "layers.10.mlp.gate_proj",
+    "v_proj",
+    "layers.9.mlp.down_proj",
+    "layers.7.mlp.gate_proj",
+    "layers.20.mlp.down_proj",
+    "layers.14.mlp.gate_proj",
+    "layers.22.mlp.up_proj",
+    "layers.6.mlp.gate_proj",
+    "layers.12.mlp.down_proj",
+    "layers.4.mlp.up_proj",
+    "layers.19.mlp.down_proj",
+    "q_proj",
+    "o_proj",
+    "layers.25.mlp.gate_proj",
+    "layers.6.mlp.up_proj",
+    "layers.2.mlp.gate_proj",
+    "layers.25.mlp.up_proj",
+    "layers.8.mlp.up_proj",
+    "layers.20.mlp.gate_proj",
+    "layers.25.mlp.down_proj",
+    "layers.24.mlp.down_proj",
+    "layers.24.mlp.up_proj",
+    "layers.3.mlp.up_proj",
+    "layers.16.mlp.up_proj",
+    "layers.0.mlp.down_proj",
+    "layers.22.mlp.down_proj",
+    "layers.26.mlp.up_proj",
+    "k_proj",
+    "layers.26.mlp.down_proj",
+    "layers.1.mlp.down_proj",
+    "layers.21.mlp.gate_proj",
+    "layers.9.mlp.up_proj",
+    "layers.1.mlp.gate_proj",
+    "layers.5.mlp.up_proj",
+    "layers.12.mlp.gate_proj",
+    "layers.17.mlp.up_proj",
+    "layers.21.mlp.down_proj",
+    "layers.8.mlp.down_proj",
+    "layers.22.mlp.gate_proj",
+    "layers.13.mlp.down_proj",
+    "layers.27.mlp.gate_proj",
+    "layers.14.mlp.down_proj",
+    "layers.19.mlp.up_proj",
+    "layers.21.mlp.up_proj",
+    "layers.5.mlp.down_proj",
+    "layers.7.mlp.down_proj",
+    "layers.16.mlp.gate_proj",
+    "layers.10.mlp.up_proj",
+    "layers.17.mlp.down_proj",
+    "layers.15.mlp.gate_proj",
+    "layers.15.mlp.down_proj",
+    "layers.11.mlp.down_proj",
+    "layers.11.mlp.gate_proj",
+    "layers.14.mlp.up_proj",
+    "layers.2.mlp.up_proj",
+    "layers.5.mlp.gate_proj",
+    "layers.3.mlp.gate_proj",
+    "layers.18.mlp.gate_proj",
+    "layers.19.mlp.gate_proj",
+    "layers.4.mlp.gate_proj",
+    "layers.18.mlp.down_proj",
+    "layers.27.mlp.up_proj",
+    "layers.20.mlp.up_proj",
+    "layers.12.mlp.up_proj",
+    "layers.23.mlp.gate_proj",
+    "layers.13.mlp.gate_proj",
+    "layers.26.mlp.gate_proj",
+    "layers.7.mlp.up_proj",
+    "layers.11.mlp.up_proj",
+    "layers.10.mlp.down_proj",
+    "layers.0.mlp.gate_proj",
+    "layers.15.mlp.up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85f0170ca0025fe47d4a399764ec1bc4b37f3d48558744e39ff1d385fd87bd9b
+size 645981584

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 5.288202184977023e+18,
+    "train_loss": 0.636212300299864,
+    "train_runtime": 56956.7963,
+    "train_samples_per_second": 1.46,
+    "train_steps_per_second": 0.046
+}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/chat_template.jinja

@@ -0,0 +1,7 @@
+{% 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
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% 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|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen2.5-VL-7B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/adapter_config.json

@@ -0,0 +1,120 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "layers.24.mlp.gate_proj",
+    "layers.16.mlp.down_proj",
+    "layers.1.mlp.up_proj",
+    "layers.6.mlp.down_proj",
+    "layers.4.mlp.down_proj",
+    "layers.8.mlp.gate_proj",
+    "layers.9.mlp.gate_proj",
+    "layers.3.mlp.down_proj",
+    "layers.17.mlp.gate_proj",
+    "layers.0.mlp.up_proj",
+    "layers.27.mlp.down_proj",
+    "layers.2.mlp.down_proj",
+    "layers.13.mlp.up_proj",
+    "layers.23.mlp.down_proj",
+    "layers.18.mlp.up_proj",
+    "layers.23.mlp.up_proj",
+    "layers.10.mlp.gate_proj",
+    "v_proj",
+    "layers.9.mlp.down_proj",
+    "layers.7.mlp.gate_proj",
+    "layers.20.mlp.down_proj",
+    "layers.14.mlp.gate_proj",
+    "layers.22.mlp.up_proj",
+    "layers.6.mlp.gate_proj",
+    "layers.12.mlp.down_proj",
+    "layers.4.mlp.up_proj",
+    "layers.19.mlp.down_proj",
+    "q_proj",
+    "o_proj",
+    "layers.25.mlp.gate_proj",
+    "layers.6.mlp.up_proj",
+    "layers.2.mlp.gate_proj",
+    "layers.25.mlp.up_proj",
+    "layers.8.mlp.up_proj",
+    "layers.20.mlp.gate_proj",
+    "layers.25.mlp.down_proj",
+    "layers.24.mlp.down_proj",
+    "layers.24.mlp.up_proj",
+    "layers.3.mlp.up_proj",
+    "layers.16.mlp.up_proj",
+    "layers.0.mlp.down_proj",
+    "layers.22.mlp.down_proj",
+    "layers.26.mlp.up_proj",
+    "k_proj",
+    "layers.26.mlp.down_proj",
+    "layers.1.mlp.down_proj",
+    "layers.21.mlp.gate_proj",
+    "layers.9.mlp.up_proj",
+    "layers.1.mlp.gate_proj",
+    "layers.5.mlp.up_proj",
+    "layers.12.mlp.gate_proj",
+    "layers.17.mlp.up_proj",
+    "layers.21.mlp.down_proj",
+    "layers.8.mlp.down_proj",
+    "layers.22.mlp.gate_proj",
+    "layers.13.mlp.down_proj",
+    "layers.27.mlp.gate_proj",
+    "layers.14.mlp.down_proj",
+    "layers.19.mlp.up_proj",
+    "layers.21.mlp.up_proj",
+    "layers.5.mlp.down_proj",
+    "layers.7.mlp.down_proj",
+    "layers.16.mlp.gate_proj",
+    "layers.10.mlp.up_proj",
+    "layers.17.mlp.down_proj",
+    "layers.15.mlp.gate_proj",
+    "layers.15.mlp.down_proj",
+    "layers.11.mlp.down_proj",
+    "layers.11.mlp.gate_proj",
+    "layers.14.mlp.up_proj",
+    "layers.2.mlp.up_proj",
+    "layers.5.mlp.gate_proj",
+    "layers.3.mlp.gate_proj",
+    "layers.18.mlp.gate_proj",
+    "layers.19.mlp.gate_proj",
+    "layers.4.mlp.gate_proj",
+    "layers.18.mlp.down_proj",
+    "layers.27.mlp.up_proj",
+    "layers.20.mlp.up_proj",
+    "layers.12.mlp.up_proj",
+    "layers.23.mlp.gate_proj",
+    "layers.13.mlp.gate_proj",
+    "layers.26.mlp.gate_proj",
+    "layers.7.mlp.up_proj",
+    "layers.11.mlp.up_proj",
+    "layers.10.mlp.down_proj",
+    "layers.0.mlp.gate_proj",
+    "layers.15.mlp.up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85f0170ca0025fe47d4a399764ec1bc4b37f3d48558744e39ff1d385fd87bd9b
+size 645981584

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/chat_template.jinja

@@ -0,0 +1,7 @@
+{% 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
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% 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|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/preprocessor_config.json

@@ -0,0 +1,36 @@
+{
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "do_center_crop": null,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessorFast",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "input_data_format": null,
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/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_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

qwen2_5vl-7b/lora/rlaif_v/lora/dpo/checkpoint-2598/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "add_bos_token": false,
+  "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
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "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,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2_5_VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}