Upload folder using huggingface_hub

.gitattributes

@@ -34,3 +34,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 checkpoints/grpo-step-50/checkpoint-50/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/grpo-step-100/checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/grpo-step-100/checkpoint-50/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/grpo-step-100/README.md

@@ -0,0 +1,67 @@
+---
+base_model: Jackrong/Qwopus3.5-9B-v3
+library_name: transformers
+model_name: v4-grpo
+tags:
+- generated_from_trainer
+- grpo
+- trl
+licence: license
+---
+
+# Model Card for v4-grpo
+
+This model is a fine-tuned version of [Jackrong/Qwopus3.5-9B-v3](https://huggingface.co/Jackrong/Qwopus3.5-9B-v3).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+ 
+
+
+
+This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
+
+### Framework versions
+
+- TRL: 1.0.0
+- Transformers: 5.5.0
+- Pytorch: 2.10.0+cu128
+- Datasets: 4.8.4
+- Tokenizers: 0.22.2
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{shao2024deepseekmath,
+    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
+    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
+    year         = 2024,
+    eprint       = {arXiv:2402.03300},
+}
+```
+
+Cite TRL as:
+    
+```bibtex
+@software{vonwerra2020trl,
+  title   = {{TRL: Transformers Reinforcement Learning}},
+  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
+  license = {Apache-2.0},
+  url     = {https://github.com/huggingface/trl},
+  year    = {2020}
+}
+```

checkpoints/grpo-step-100/checkpoint-100/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Jackrong/Qwopus3.5-9B-v3
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Jackrong/Qwopus3.5-9B-v3
+- grpo
+- lora
+- transformers
+- trl
+---
+
+# 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 0.18.1

checkpoints/grpo-step-100/checkpoint-100/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Jackrong/Qwopus3.5-9B-v3",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "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": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "down_proj",
+    "v_proj",
+    "gate_proj",
+    "k_proj",
+    "up_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": true
+}

checkpoints/grpo-step-100/checkpoint-100/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98c1ffb415fdab545676413e189c8662851384492805d0dc17dcc5ca05e31af7
+size 465602056

checkpoints/grpo-step-100/checkpoint-100/chat_template.jinja

@@ -0,0 +1,88 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant
+<think>
+' }}
+{%- endif %}

checkpoints/grpo-step-100/checkpoint-100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5d635356e49d7334fcc6f63bfeee85011fa08259541774a3f695367920d413c
+size 236725541

checkpoints/grpo-step-100/checkpoint-100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f636873e3493040dc8151702cecaab845fd289d6ef239f3f2f3aa58bb7ecfdeb
+size 14645

checkpoints/grpo-step-100/checkpoint-100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d409847397718e7ea4e37312980fc24718ed3e95ea2129a95f3bb5e0f30fdbf5
+size 1465

checkpoints/grpo-step-100/checkpoint-100/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0848a62b3efcaac1ef8e29bea66fd083ef76d8d6fb2bdd483cb4dba81dc1c9a8
+size 19989512

checkpoints/grpo-step-100/checkpoint-100/tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": false,
+  "model_max_length": 8192,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|vision_pad|>",
+  "padding_side": "right",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

checkpoints/grpo-step-100/checkpoint-100/trainer_state.json

@@ -0,0 +1,574 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.09345794392523364,
+  "eval_steps": 500,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 455.3333333333333,
+      "completions/max_terminated_length": 455.3333333333333,
+      "completions/mean_length": 370.1666666666667,
+      "completions/mean_terminated_length": 370.1666666666667,
+      "completions/min_length": 241.33333333333334,
+      "completions/min_terminated_length": 241.33333333333334,
+      "entropy": 0.20471833571791648,
+      "epoch": 0.004672897196261682,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.5874403715133667,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": -0.0174385666847229,
+      "num_tokens": 10556.0,
+      "reward": 1.1229166984558105,
+      "reward_std": 0.21075093746185303,
+      "rewards/multiplicative_reward/mean": 1.1229166984558105,
+      "rewards/multiplicative_reward/std": 0.21075095236301422,
+      "step": 5,
+      "step_time": 175.64732578959973
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0001853372377809137,
+      "clip_ratio/low_min": 0.0001853372377809137,
+      "clip_ratio/region_mean": 0.0001853372377809137,
+      "completions/clipped_ratio": 0.0625,
+      "completions/max_length": 853.5,
+      "completions/max_terminated_length": 763.0,
+      "completions/mean_length": 642.9375,
+      "completions/mean_terminated_length": 625.2053833007812,
+      "completions/min_length": 511.0,
+      "completions/min_terminated_length": 511.0,
+      "entropy": 0.20119954496622086,
+      "epoch": 0.009345794392523364,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09654995054006577,
+      "learning_rate": 4.5e-06,
+      "loss": -0.09893683791160583,
+      "num_tokens": 21963.0,
+      "reward": 1.149999976158142,
+      "reward_std": 0.2208700068295002,
+      "rewards/multiplicative_reward/mean": 1.149999976158142,
+      "rewards/multiplicative_reward/std": 0.2208700142800808,
+      "step": 10,
+      "step_time": 315.6367059516
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0004878048785030842,
+      "clip_ratio/low_min": 0.0004878048785030842,
+      "clip_ratio/region_mean": 0.0004878048785030842,
+      "completions/clipped_ratio": 0.041666666666666664,
+      "completions/max_length": 762.3333333333334,
+      "completions/max_terminated_length": 761.3333333333334,
+      "completions/mean_length": 515.1666666666666,
+      "completions/mean_terminated_length": 502.4107259114583,
+      "completions/min_length": 296.0,
+      "completions/min_terminated_length": 296.0,
+      "entropy": 0.24918381869792938,
+      "epoch": 0.014018691588785047,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12115617096424103,
+      "learning_rate": 4.989935734988098e-06,
+      "loss": -0.14156837463378907,
+      "num_tokens": 35943.0,
+      "reward": 1.0958333412806194,
+      "reward_std": 0.35721248139937717,
+      "rewards/multiplicative_reward/mean": 1.0958333412806194,
+      "rewards/multiplicative_reward/std": 0.35721250623464584,
+      "step": 15,
+      "step_time": 391.1071364838002
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.1875,
+      "completions/max_length": 871.0,
+      "completions/max_terminated_length": 823.0,
+      "completions/mean_length": 706.0,
+      "completions/mean_terminated_length": 658.7124938964844,
+      "completions/min_length": 527.5,
+      "completions/min_terminated_length": 527.5,
+      "entropy": 0.18833256512880325,
+      "epoch": 0.018691588785046728,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.16035032272338867,
+      "learning_rate": 4.949188496058089e-06,
+      "loss": -0.09449982643127441,
+      "num_tokens": 48487.0,
+      "reward": 1.0437499582767487,
+      "reward_std": 0.3550429940223694,
+      "rewards/multiplicative_reward/mean": 1.0437499582767487,
+      "rewards/multiplicative_reward/std": 0.3550429902970791,
+      "step": 20,
+      "step_time": 327.72734730440004
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 8.510638144798577e-05,
+      "clip_ratio/low_min": 8.510638144798577e-05,
+      "clip_ratio/region_mean": 8.510638144798577e-05,
+      "completions/clipped_ratio": 0.041666666666666664,
+      "completions/max_length": 807.3333333333334,
+      "completions/max_terminated_length": 803.6666666666666,
+      "completions/mean_length": 596.9166666666666,
+      "completions/mean_terminated_length": 584.202392578125,
+      "completions/min_length": 427.3333333333333,
+      "completions/min_terminated_length": 427.3333333333333,
+      "entropy": 0.24548179060220718,
+      "epoch": 0.02336448598130841,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.26090770959854126,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": -0.1003144383430481,
+      "num_tokens": 64301.0,
+      "reward": 1.18749996026357,
+      "reward_std": 0.20679950838287672,
+      "rewards/multiplicative_reward/mean": 1.18749996026357,
+      "rewards/multiplicative_reward/std": 0.20679953570167223,
+      "step": 25,
+      "step_time": 423.7583607568005
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.75,
+      "completions/max_length": 1024.0,
+      "completions/max_terminated_length": 474.0,
+      "completions/mean_length": 970.1875,
+      "completions/mean_terminated_length": 404.375,
+      "completions/min_length": 850.0,
+      "completions/min_terminated_length": 338.0,
+      "entropy": 0.15614256262779236,
+      "epoch": 0.028037383177570093,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0,
+      "learning_rate": 4.7761938666470405e-06,
+      "loss": -0.005558112263679504,
+      "num_tokens": 81048.0,
+      "reward": 0.4312499687075615,
+      "reward_std": 0.3087201565504074,
+      "rewards/multiplicative_reward/mean": 0.4312499687075615,
+      "rewards/multiplicative_reward/std": 0.3087201565504074,
+      "step": 30,
+      "step_time": 415.75604495680017
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.25,
+      "completions/max_length": 618.6666666666666,
+      "completions/max_terminated_length": 514.6666666666666,
+      "completions/mean_length": 536.375,
+      "completions/mean_terminated_length": 451.625,
+      "completions/min_length": 405.0,
+      "completions/min_terminated_length": 405.0,
+      "entropy": 0.1147035963833332,
+      "epoch": 0.03271028037383177,
+      "frac_reward_zero_std": 0.3333333333333333,
+      "grad_norm": 0.4312528967857361,
+      "learning_rate": 4.646121984004666e-06,
+      "loss": -0.06840584874153137,
+      "num_tokens": 95513.0,
+      "reward": 0.8583333392937978,
+      "reward_std": 0.23224478214979172,
+      "rewards/multiplicative_reward/mean": 0.8583333392937978,
+      "rewards/multiplicative_reward/std": 0.23224476476510367,
+      "step": 35,
+      "step_time": 313.75861934560055
+    },
+    {
+      "clip_ratio/high_max": 6.939625018276274e-05,
+      "clip_ratio/high_mean": 6.939625018276274e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 6.939625018276274e-05,
+      "completions/clipped_ratio": 0.0625,
+      "completions/max_length": 870.5,
+      "completions/max_terminated_length": 797.5,
+      "completions/mean_length": 694.375,
+      "completions/mean_terminated_length": 675.4642944335938,
+      "completions/min_length": 566.5,
+      "completions/min_terminated_length": 566.5,
+      "entropy": 0.1549898698925972,
+      "epoch": 0.037383177570093455,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.2594030499458313,
+      "learning_rate": 4.4890613722044526e-06,
+      "loss": 0.012206726521253587,
+      "num_tokens": 108071.0,
+      "reward": 1.2062499523162842,
+      "reward_std": 0.021777553483843803,
+      "rewards/multiplicative_reward/mean": 1.2062499523162842,
+      "rewards/multiplicative_reward/std": 0.021777544170618057,
+      "step": 40,
+      "step_time": 330.83437446999994
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.6666666666666666,
+      "completions/max_length": 1016.6666666666666,
+      "completions/max_terminated_length": 334.0,
+      "completions/mean_length": 909.625,
+      "completions/mean_terminated_length": 226.95833333333334,
+      "completions/min_length": 862.3333333333334,
+      "completions/min_terminated_length": 179.66666666666666,
+      "entropy": 0.0735762745141983,
+      "epoch": 0.04205607476635514,
+      "frac_reward_zero_std": 0.3333333333333333,
+      "grad_norm": 0.0,
+      "learning_rate": 4.3069871595684795e-06,
+      "loss": -0.00706263780593872,
+      "num_tokens": 131198.0,
+      "reward": 0.5687499915560087,
+      "reward_std": 0.0635174810886383,
+      "rewards/multiplicative_reward/mean": 0.5687499915560087,
+      "rewards/multiplicative_reward/std": 0.06351748605569203,
+      "step": 45,
+      "step_time": 596.8530479572007
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.25,
+      "completions/max_length": 877.0,
+      "completions/max_terminated_length": 829.0,
+      "completions/mean_length": 752.625,
+      "completions/mean_terminated_length": 690.875,
+      "completions/min_length": 610.5,
+      "completions/min_terminated_length": 610.5,
+      "entropy": 0.12532927691936493,
+      "epoch": 0.04672897196261682,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.2574126422405243,
+      "learning_rate": 4.102189034962561e-06,
+      "loss": -0.08543808460235595,
+      "num_tokens": 144320.0,
+      "reward": 0.9312499165534973,
+      "reward_std": 0.4012690559029579,
+      "rewards/multiplicative_reward/mean": 0.9312499165534973,
+      "rewards/multiplicative_reward/std": 0.4012690484523773,
+      "step": 50,
+      "step_time": 325.3835452261992
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 7.434944272972643e-05,
+      "clip_ratio/low_min": 7.434944272972643e-05,
+      "clip_ratio/region_mean": 7.434944272972643e-05,
+      "completions/clipped_ratio": 0.3333333333333333,
+      "completions/max_length": 898.3333333333334,
+      "completions/max_terminated_length": 866.0,
+      "completions/mean_length": 781.125,
+      "completions/mean_terminated_length": 765.9345296223959,
+      "completions/min_length": 675.6666666666666,
+      "completions/min_terminated_length": 675.6666666666666,
+      "entropy": 0.22954411506652833,
+      "epoch": 0.0514018691588785,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07312487065792084,
+      "learning_rate": 3.8772424536302565e-06,
+      "loss": 0.0017251893877983093,
+      "num_tokens": 164275.0,
+      "reward": 1.012499988079071,
+      "reward_std": 0.0874652589360873,
+      "rewards/multiplicative_reward/mean": 1.012499988079071,
+      "rewards/multiplicative_reward/std": 0.08746525645256042,
+      "step": 55,
+      "step_time": 495.2329136221997
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 7.812500116415322e-05,
+      "clip_ratio/low_min": 7.812500116415322e-05,
+      "clip_ratio/region_mean": 7.812500116415322e-05,
+      "completions/clipped_ratio": 0.1875,
+      "completions/max_length": 833.5,
+      "completions/max_terminated_length": 824.0,
+      "completions/mean_length": 681.6875,
+      "completions/mean_terminated_length": 621.9500122070312,
+      "completions/min_length": 473.0,
+      "completions/min_terminated_length": 473.0,
+      "entropy": 0.14068461135029792,
+      "epoch": 0.056074766355140186,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.33382320404052734,
+      "learning_rate": 3.634976249348867e-06,
+      "loss": -0.054555124044418334,
+      "num_tokens": 176318.0,
+      "reward": 1.128125011920929,
+      "reward_std": 0.27171332482248545,
+      "rewards/multiplicative_reward/mean": 1.128125011920929,
+      "rewards/multiplicative_reward/std": 0.2717133266851306,
+      "step": 60,
+      "step_time": 307.00902329860037
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.3333333333333333,
+      "completions/max_length": 655.0,
+      "completions/max_terminated_length": 313.6666666666667,
+      "completions/mean_length": 576.625,
+      "completions/mean_terminated_length": 235.29166666666666,
+      "completions/min_length": 528.6666666666666,
+      "completions/min_terminated_length": 187.33333333333334,
+      "entropy": 0.08125153928995132,
+      "epoch": 0.06074766355140187,
+      "frac_reward_zero_std": 0.3333333333333333,
+      "grad_norm": 0.12123720347881317,
+      "learning_rate": 3.3784370602033572e-06,
+      "loss": -0.022451573610305788,
+      "num_tokens": 191021.0,
+      "reward": 0.9020833273728689,
+      "reward_std": 0.19084744403759638,
+      "rewards/multiplicative_reward/mean": 0.9020833273728689,
+      "rewards/multiplicative_reward/std": 0.19084744900465012,
+      "step": 65,
+      "step_time": 317.94532901660205
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0004962049017194659,
+      "clip_ratio/low_min": 0.0004962049017194659,
+      "clip_ratio/region_mean": 0.0004962049017194659,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 480.5,
+      "completions/max_terminated_length": 480.5,
+      "completions/mean_length": 406.875,
+      "completions/mean_terminated_length": 406.875,
+      "completions/min_length": 329.5,
+      "completions/min_terminated_length": 329.5,
+      "entropy": 0.14233311265707016,
+      "epoch": 0.06542056074766354,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.17829293012619019,
+      "learning_rate": 3.1108510153447352e-06,
+      "loss": 0.050252318382263184,
+      "num_tokens": 198723.0,
+      "reward": 1.3250000476837158,
+      "reward_std": 0.048718432895839214,
+      "rewards/multiplicative_reward/mean": 1.3250000476837158,
+      "rewards/multiplicative_reward/std": 0.04871843010187149,
+      "step": 70,
+      "step_time": 130.87347414019968
+    },
+    {
+      "clip_ratio/high_max": 7.33675726223737e-05,
+      "clip_ratio/high_mean": 7.33675726223737e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 7.33675726223737e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 826.6666666666666,
+      "completions/max_terminated_length": 826.6666666666666,
+      "completions/mean_length": 554.75,
+      "completions/mean_terminated_length": 554.75,
+      "completions/min_length": 365.6666666666667,
+      "completions/min_terminated_length": 365.6666666666667,
+      "entropy": 0.20384509563446046,
+      "epoch": 0.07009345794392523,
+      "frac_reward_zero_std": 0.3333333333333333,
+      "grad_norm": 0.4420561194419861,
+      "learning_rate": 2.835583164544139e-06,
+      "loss": -0.0002071455121040344,
+      "num_tokens": 213605.0,
+      "reward": 1.2916666666666667,
+      "reward_std": 0.04977604250113169,
+      "rewards/multiplicative_reward/mean": 1.2916666666666667,
+      "rewards/multiplicative_reward/std": 0.049776035050551094,
+      "step": 75,
+      "step_time": 435.94155339179997
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 528.0,
+      "completions/max_terminated_length": 528.0,
+      "completions/mean_length": 443.3125,
+      "completions/mean_terminated_length": 443.3125,
+      "completions/min_length": 331.0,
+      "completions/min_terminated_length": 331.0,
+      "entropy": 0.18517069816589354,
+      "epoch": 0.07476635514018691,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.34526562690734863,
+      "learning_rate": 2.556095160739513e-06,
+      "loss": 0.012152573466300965,
+      "num_tokens": 221994.0,
+      "reward": 1.3250000476837158,
+      "reward_std": 0.02314553502947092,
+      "rewards/multiplicative_reward/mean": 1.3250000476837158,
+      "rewards/multiplicative_reward/std": 0.023145535960793495,
+      "step": 80,
+      "step_time": 147.25621184560003
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.25,
+      "completions/max_length": 798.0,
+      "completions/max_terminated_length": 750.0,
+      "completions/mean_length": 613.375,
+      "completions/mean_terminated_length": 532.25,
+      "completions/min_length": 396.0,
+      "completions/min_terminated_length": 396.0,
+      "entropy": 0.13030193448066713,
+      "epoch": 0.0794392523364486,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.3123658001422882,
+      "learning_rate": 2.2759017277414165e-06,
+      "loss": -0.06673300266265869,
+      "num_tokens": 238027.0,
+      "reward": 1.006249984105428,
+      "reward_std": 0.2273760692526897,
+      "rewards/multiplicative_reward/mean": 1.006249984105428,
+      "rewards/multiplicative_reward/std": 0.22737608291208744,
+      "step": 85,
+      "step_time": 411.0016910238002
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.1875,
+      "completions/max_length": 719.0,
+      "completions/max_terminated_length": 677.0,
+      "completions/mean_length": 600.3125,
+      "completions/mean_terminated_length": 550.8125,
+      "completions/min_length": 397.0,
+      "completions/min_terminated_length": 397.0,
+      "entropy": 0.2051653802394867,
+      "epoch": 0.08411214953271028,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10376756638288498,
+      "learning_rate": 1.9985264605418185e-06,
+      "loss": -0.0393097311258316,
+      "num_tokens": 248736.0,
+      "reward": 1.1343749463558197,
+      "reward_std": 0.33943885937333107,
+      "rewards/multiplicative_reward/mean": 1.1343749463558197,
+      "rewards/multiplicative_reward/std": 0.33943886309862137,
+      "step": 90,
+      "step_time": 259.8986448414013
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.125,
+      "completions/max_length": 857.3333333333334,
+      "completions/max_terminated_length": 787.6666666666666,
+      "completions/mean_length": 660.0416666666666,
+      "completions/mean_terminated_length": 618.6567586263021,
+      "completions/min_length": 473.3333333333333,
+      "completions/min_terminated_length": 473.3333333333333,
+      "entropy": 0.16872838214039804,
+      "epoch": 0.08878504672897196,
+      "frac_reward_zero_std": 0.6666666666666666,
+      "grad_norm": 0.0,
+      "learning_rate": 1.7274575140626318e-06,
+      "loss": 0.005825211107730865,
+      "num_tokens": 266105.0,
+      "reward": 1.256249984105428,
+      "reward_std": 0.11883610486984253,
+      "rewards/multiplicative_reward/mean": 1.256249984105428,
+      "rewards/multiplicative_reward/std": 0.11883612473805745,
+      "step": 95,
+      "step_time": 477.2306483531996
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.1875,
+      "completions/max_length": 747.5,
+      "completions/max_terminated_length": 725.0,
+      "completions/mean_length": 617.4375,
+      "completions/mean_terminated_length": 581.9250183105469,
+      "completions/min_length": 507.5,
+      "completions/min_terminated_length": 507.5,
+      "entropy": 0.2064228281378746,
+      "epoch": 0.09345794392523364,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.7731993198394775,
+      "learning_rate": 1.466103737583699e-06,
+      "loss": 0.06998323202133179,
+      "num_tokens": 277072.0,
+      "reward": 0.8937499970197678,
+      "reward_std": 0.04274389147758484,
+      "rewards/multiplicative_reward/mean": 0.8937499970197678,
+      "rewards/multiplicative_reward/std": 0.0427439184859395,
+      "step": 100,
+      "step_time": 260.3944303373995
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 150,
+  "num_input_tokens_seen": 277072,
+  "num_train_epochs": 1,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/grpo-step-100/checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa659081e27b753a6caa3755ba28e42290b91f133fde6c340bf9b2bd455a4caa
+size 7057

checkpoints/grpo-step-100/checkpoint-50/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Jackrong/Qwopus3.5-9B-v3
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Jackrong/Qwopus3.5-9B-v3
+- grpo
+- lora
+- transformers
+- trl
+---
+
+# 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 0.18.1

checkpoints/grpo-step-100/checkpoint-50/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Jackrong/Qwopus3.5-9B-v3",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "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": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "down_proj",
+    "v_proj",
+    "gate_proj",
+    "k_proj",
+    "up_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": true
+}

checkpoints/grpo-step-100/checkpoint-50/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7148cb64299b6a82eebb522c77aa7e1fe01ac728fae7b065adfdf75d3936a0c9
+size 465602056

checkpoints/grpo-step-100/checkpoint-50/chat_template.jinja

@@ -0,0 +1,88 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant
+<think>
+' }}
+{%- endif %}

checkpoints/grpo-step-100/checkpoint-50/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e12433bfee7b2190f9e446fe5f0300ac5457aa6a44a68fe2239bd4d4a7e45269
+size 236725541

checkpoints/grpo-step-100/checkpoint-50/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfe1fa6a7e95621f14fcc10829716c918b3fabf863db7c45bc55c3cb6bb5e8e2
+size 14645

checkpoints/grpo-step-100/checkpoint-50/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:093bcd59fe0e63c6c73ebf24a5f7d752365277b8e18f4119492628371c6f27db
+size 1465

checkpoints/grpo-step-100/checkpoint-50/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0848a62b3efcaac1ef8e29bea66fd083ef76d8d6fb2bdd483cb4dba81dc1c9a8
+size 19989512

checkpoints/grpo-step-100/checkpoint-50/tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": false,
+  "model_max_length": 8192,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|vision_pad|>",
+  "padding_side": "right",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

checkpoints/grpo-step-100/checkpoint-50/trainer_state.json

@@ -0,0 +1,304 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.04672897196261682,
+  "eval_steps": 500,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 455.3333333333333,
+      "completions/max_terminated_length": 455.3333333333333,
+      "completions/mean_length": 370.1666666666667,
+      "completions/mean_terminated_length": 370.1666666666667,
+      "completions/min_length": 241.33333333333334,
+      "completions/min_terminated_length": 241.33333333333334,
+      "entropy": 0.20471833571791648,
+      "epoch": 0.004672897196261682,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.5874403715133667,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": -0.0174385666847229,
+      "num_tokens": 10556.0,
+      "reward": 1.1229166984558105,
+      "reward_std": 0.21075093746185303,
+      "rewards/multiplicative_reward/mean": 1.1229166984558105,
+      "rewards/multiplicative_reward/std": 0.21075095236301422,
+      "step": 5,
+      "step_time": 175.64732578959973
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0001853372377809137,
+      "clip_ratio/low_min": 0.0001853372377809137,
+      "clip_ratio/region_mean": 0.0001853372377809137,
+      "completions/clipped_ratio": 0.0625,
+      "completions/max_length": 853.5,
+      "completions/max_terminated_length": 763.0,
+      "completions/mean_length": 642.9375,
+      "completions/mean_terminated_length": 625.2053833007812,
+      "completions/min_length": 511.0,
+      "completions/min_terminated_length": 511.0,
+      "entropy": 0.20119954496622086,
+      "epoch": 0.009345794392523364,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09654995054006577,
+      "learning_rate": 4.5e-06,
+      "loss": -0.09893683791160583,
+      "num_tokens": 21963.0,
+      "reward": 1.149999976158142,
+      "reward_std": 0.2208700068295002,
+      "rewards/multiplicative_reward/mean": 1.149999976158142,
+      "rewards/multiplicative_reward/std": 0.2208700142800808,
+      "step": 10,
+      "step_time": 315.6367059516
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0004878048785030842,
+      "clip_ratio/low_min": 0.0004878048785030842,
+      "clip_ratio/region_mean": 0.0004878048785030842,
+      "completions/clipped_ratio": 0.041666666666666664,
+      "completions/max_length": 762.3333333333334,
+      "completions/max_terminated_length": 761.3333333333334,
+      "completions/mean_length": 515.1666666666666,
+      "completions/mean_terminated_length": 502.4107259114583,
+      "completions/min_length": 296.0,
+      "completions/min_terminated_length": 296.0,
+      "entropy": 0.24918381869792938,
+      "epoch": 0.014018691588785047,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12115617096424103,
+      "learning_rate": 4.989935734988098e-06,
+      "loss": -0.14156837463378907,
+      "num_tokens": 35943.0,
+      "reward": 1.0958333412806194,
+      "reward_std": 0.35721248139937717,
+      "rewards/multiplicative_reward/mean": 1.0958333412806194,
+      "rewards/multiplicative_reward/std": 0.35721250623464584,
+      "step": 15,
+      "step_time": 391.1071364838002
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.1875,
+      "completions/max_length": 871.0,
+      "completions/max_terminated_length": 823.0,
+      "completions/mean_length": 706.0,
+      "completions/mean_terminated_length": 658.7124938964844,
+      "completions/min_length": 527.5,
+      "completions/min_terminated_length": 527.5,
+      "entropy": 0.18833256512880325,
+      "epoch": 0.018691588785046728,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.16035032272338867,
+      "learning_rate": 4.949188496058089e-06,
+      "loss": -0.09449982643127441,
+      "num_tokens": 48487.0,
+      "reward": 1.0437499582767487,
+      "reward_std": 0.3550429940223694,
+      "rewards/multiplicative_reward/mean": 1.0437499582767487,
+      "rewards/multiplicative_reward/std": 0.3550429902970791,
+      "step": 20,
+      "step_time": 327.72734730440004
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 8.510638144798577e-05,
+      "clip_ratio/low_min": 8.510638144798577e-05,
+      "clip_ratio/region_mean": 8.510638144798577e-05,
+      "completions/clipped_ratio": 0.041666666666666664,
+      "completions/max_length": 807.3333333333334,
+      "completions/max_terminated_length": 803.6666666666666,
+      "completions/mean_length": 596.9166666666666,
+      "completions/mean_terminated_length": 584.202392578125,
+      "completions/min_length": 427.3333333333333,
+      "completions/min_terminated_length": 427.3333333333333,
+      "entropy": 0.24548179060220718,
+      "epoch": 0.02336448598130841,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.26090770959854126,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": -0.1003144383430481,
+      "num_tokens": 64301.0,
+      "reward": 1.18749996026357,
+      "reward_std": 0.20679950838287672,
+      "rewards/multiplicative_reward/mean": 1.18749996026357,
+      "rewards/multiplicative_reward/std": 0.20679953570167223,
+      "step": 25,
+      "step_time": 423.7583607568005
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.75,
+      "completions/max_length": 1024.0,
+      "completions/max_terminated_length": 474.0,
+      "completions/mean_length": 970.1875,
+      "completions/mean_terminated_length": 404.375,
+      "completions/min_length": 850.0,
+      "completions/min_terminated_length": 338.0,
+      "entropy": 0.15614256262779236,
+      "epoch": 0.028037383177570093,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0,
+      "learning_rate": 4.7761938666470405e-06,
+      "loss": -0.005558112263679504,
+      "num_tokens": 81048.0,
+      "reward": 0.4312499687075615,
+      "reward_std": 0.3087201565504074,
+      "rewards/multiplicative_reward/mean": 0.4312499687075615,
+      "rewards/multiplicative_reward/std": 0.3087201565504074,
+      "step": 30,
+      "step_time": 415.75604495680017
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.25,
+      "completions/max_length": 618.6666666666666,
+      "completions/max_terminated_length": 514.6666666666666,
+      "completions/mean_length": 536.375,
+      "completions/mean_terminated_length": 451.625,
+      "completions/min_length": 405.0,
+      "completions/min_terminated_length": 405.0,
+      "entropy": 0.1147035963833332,
+      "epoch": 0.03271028037383177,
+      "frac_reward_zero_std": 0.3333333333333333,
+      "grad_norm": 0.4312528967857361,
+      "learning_rate": 4.646121984004666e-06,
+      "loss": -0.06840584874153137,
+      "num_tokens": 95513.0,
+      "reward": 0.8583333392937978,
+      "reward_std": 0.23224478214979172,
+      "rewards/multiplicative_reward/mean": 0.8583333392937978,
+      "rewards/multiplicative_reward/std": 0.23224476476510367,
+      "step": 35,
+      "step_time": 313.75861934560055
+    },
+    {
+      "clip_ratio/high_max": 6.939625018276274e-05,
+      "clip_ratio/high_mean": 6.939625018276274e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 6.939625018276274e-05,
+      "completions/clipped_ratio": 0.0625,
+      "completions/max_length": 870.5,
+      "completions/max_terminated_length": 797.5,
+      "completions/mean_length": 694.375,
+      "completions/mean_terminated_length": 675.4642944335938,
+      "completions/min_length": 566.5,
+      "completions/min_terminated_length": 566.5,
+      "entropy": 0.1549898698925972,
+      "epoch": 0.037383177570093455,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.2594030499458313,
+      "learning_rate": 4.4890613722044526e-06,
+      "loss": 0.012206726521253587,
+      "num_tokens": 108071.0,
+      "reward": 1.2062499523162842,
+      "reward_std": 0.021777553483843803,
+      "rewards/multiplicative_reward/mean": 1.2062499523162842,
+      "rewards/multiplicative_reward/std": 0.021777544170618057,
+      "step": 40,
+      "step_time": 330.83437446999994
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.6666666666666666,
+      "completions/max_length": 1016.6666666666666,
+      "completions/max_terminated_length": 334.0,
+      "completions/mean_length": 909.625,
+      "completions/mean_terminated_length": 226.95833333333334,
+      "completions/min_length": 862.3333333333334,
+      "completions/min_terminated_length": 179.66666666666666,
+      "entropy": 0.0735762745141983,
+      "epoch": 0.04205607476635514,
+      "frac_reward_zero_std": 0.3333333333333333,
+      "grad_norm": 0.0,
+      "learning_rate": 4.3069871595684795e-06,
+      "loss": -0.00706263780593872,
+      "num_tokens": 131198.0,
+      "reward": 0.5687499915560087,
+      "reward_std": 0.0635174810886383,
+      "rewards/multiplicative_reward/mean": 0.5687499915560087,
+      "rewards/multiplicative_reward/std": 0.06351748605569203,
+      "step": 45,
+      "step_time": 596.8530479572007
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.25,
+      "completions/max_length": 877.0,
+      "completions/max_terminated_length": 829.0,
+      "completions/mean_length": 752.625,
+      "completions/mean_terminated_length": 690.875,
+      "completions/min_length": 610.5,
+      "completions/min_terminated_length": 610.5,
+      "entropy": 0.12532927691936493,
+      "epoch": 0.04672897196261682,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.2574126422405243,
+      "learning_rate": 4.102189034962561e-06,
+      "loss": -0.08543808460235595,
+      "num_tokens": 144320.0,
+      "reward": 0.9312499165534973,
+      "reward_std": 0.4012690559029579,
+      "rewards/multiplicative_reward/mean": 0.9312499165534973,
+      "rewards/multiplicative_reward/std": 0.4012690484523773,
+      "step": 50,
+      "step_time": 325.3835452261992
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 150,
+  "num_input_tokens_seen": 144320,
+  "num_train_epochs": 1,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/grpo-step-100/checkpoint-50/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa659081e27b753a6caa3755ba28e42290b91f133fde6c340bf9b2bd455a4caa
+size 7057