Upload GRPO-trained humanlike essay model

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-150/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-250/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-50/tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,68 @@
+---
+base_model: Qwen/Qwen3-14B
+library_name: transformers
+model_name: qwen3grpo_humanlike2
+tags:
+- generated_from_trainer
+- grpo
+- trl
+licence: license
+---
+
+# Model Card for qwen3grpo_humanlike2
+
+This model is a fine-tuned version of [Qwen/Qwen3-14B](https://huggingface.co/Qwen/Qwen3-14B).
+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: 0.19.0
+- Transformers: 4.53.0
+- Pytorch: 2.8.0.dev20250605+cu128
+- Datasets: 3.6.0
+- Tokenizers: 0.21.2
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{zhihong2024deepseekmath,
+    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
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

checkpoint-100/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen3-14B
+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

checkpoint-100/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-100/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a35a8cabe39461870377385fea5f55e71b19cf6b26b002ee7a8cfcb7a5506382
+size 513877864

checkpoint-100/added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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
+}

checkpoint-100/chat_template.jinja

@@ -0,0 +1,89 @@
+{%- 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\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

checkpoint-100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:077c99dc2d17214dab84b7b04bfa97eaf5a643ec5c6d97101dea300f2221f9f6
+size 1028078067

checkpoint-100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0bd58d85c322dd61c8790fb097b34d015312b627d02ce2732e86d135d34ba5a
+size 14645

checkpoint-100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff23d2c8d0500b3c5785fce2b8a14536ee63969efedcfdeca4bc909dea166101
+size 1465

checkpoint-100/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "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": "<|im_end|>"
+}

checkpoint-100/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de053c72a4ae289224ac988558897303006ceea1db22f30b97d0d856969ea6b9
+size 11422818

checkpoint-100/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "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": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-100/trainer_state.json

@@ -0,0 +1,694 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.0018244845831052727,
+  "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": 620.2,
+      "completions/max_terminated_length": 620.2,
+      "completions/mean_length": 536.3875,
+      "completions/mean_terminated_length": 536.3875,
+      "completions/min_length": 455.6,
+      "completions/min_terminated_length": 455.6,
+      "epoch": 9.122422915526363e-05,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.06800810992717743,
+      "learning_rate": 2e-05,
+      "loss": -0.0026,
+      "num_tokens": 48111.0,
+      "reward": 2.393585443496704,
+      "reward_std": 0.14288972318172455,
+      "rewards/reward_humanlike/mean": -0.003280111402273178,
+      "rewards/reward_humanlike/std": 0.002425189537461847,
+      "rewards/reward_nonslop/mean": 0.45034117698669435,
+      "rewards/reward_nonslop/std": 0.021841670200228692,
+      "rewards/reward_paragraph_structure/mean": 0.9998971700668335,
+      "rewards/reward_paragraph_structure/std": 0.00036894355434924364,
+      "rewards/reward_readability_flesch/mean": 0.5846633613109589,
+      "rewards/reward_readability_flesch/std": 0.11768133006989956,
+      "rewards/reward_variance_humanlike/mean": 0.36196385622024535,
+      "rewards/reward_variance_humanlike/std": 0.036638665944337845,
+      "step": 5
+    },
+    {
+      "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": 646.0,
+      "completions/max_terminated_length": 646.0,
+      "completions/mean_length": 551.5,
+      "completions/mean_terminated_length": 551.5,
+      "completions/min_length": 446.2,
+      "completions/min_terminated_length": 446.2,
+      "epoch": 0.00018244845831052726,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07684256136417389,
+      "learning_rate": 2e-05,
+      "loss": 0.0074,
+      "num_tokens": 97303.0,
+      "reward": 2.391454339027405,
+      "reward_std": 0.1808932200074196,
+      "rewards/reward_humanlike/mean": -0.0016780316829681397,
+      "rewards/reward_humanlike/std": 0.014642862777691334,
+      "rewards/reward_nonslop/mean": 0.44268596172332764,
+      "rewards/reward_nonslop/std": 0.028079804033041,
+      "rewards/reward_paragraph_structure/mean": 1.0,
+      "rewards/reward_paragraph_structure/std": 0.0,
+      "rewards/reward_readability_flesch/mean": 0.57281613945961,
+      "rewards/reward_readability_flesch/std": 0.16779391318559647,
+      "rewards/reward_variance_humanlike/mean": 0.3776302635669708,
+      "rewards/reward_variance_humanlike/std": 0.045258138328790665,
+      "step": 10
+    },
+    {
+      "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": 598.0,
+      "completions/max_terminated_length": 598.0,
+      "completions/mean_length": 516.1125,
+      "completions/mean_terminated_length": 516.1125,
+      "completions/min_length": 431.6,
+      "completions/min_terminated_length": 431.6,
+      "epoch": 0.0002736726874657909,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07912639528512955,
+      "learning_rate": 2e-05,
+      "loss": 0.0026,
+      "num_tokens": 143712.0,
+      "reward": 2.121282529830933,
+      "reward_std": 0.19594088345766067,
+      "rewards/reward_humanlike/mean": -0.005437415838241577,
+      "rewards/reward_humanlike/std": 0.0020546794403344393,
+      "rewards/reward_nonslop/mean": 0.44375383853912354,
+      "rewards/reward_nonslop/std": 0.025456106290221214,
+      "rewards/reward_paragraph_structure/mean": 0.9997280359268188,
+      "rewards/reward_paragraph_structure/std": 0.001087829563766718,
+      "rewards/reward_readability_flesch/mean": 0.340261435508728,
+      "rewards/reward_readability_flesch/std": 0.17545561231672763,
+      "rewards/reward_variance_humanlike/mean": 0.3429765820503235,
+      "rewards/reward_variance_humanlike/std": 0.04073530249297619,
+      "step": 15
+    },
+    {
+      "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": 632.4,
+      "completions/max_terminated_length": 632.4,
+      "completions/mean_length": 511.4,
+      "completions/mean_terminated_length": 511.4,
+      "completions/min_length": 387.0,
+      "completions/min_terminated_length": 387.0,
+      "epoch": 0.00036489691662105453,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08999043703079224,
+      "learning_rate": 2e-05,
+      "loss": 0.0183,
+      "num_tokens": 189696.0,
+      "reward": 2.6498445510864257,
+      "reward_std": 0.2029375523328781,
+      "rewards/reward_humanlike/mean": 0.00019632279872894287,
+      "rewards/reward_humanlike/std": 0.006884497962892056,
+      "rewards/reward_nonslop/mean": 0.4884443640708923,
+      "rewards/reward_nonslop/std": 0.0355165209621191,
+      "rewards/reward_paragraph_structure/mean": 0.9939258456230163,
+      "rewards/reward_paragraph_structure/std": 0.023617163486778735,
+      "rewards/reward_readability_flesch/mean": 0.8020971655845642,
+      "rewards/reward_readability_flesch/std": 0.17971589416265488,
+      "rewards/reward_variance_humanlike/mean": 0.3651808500289917,
+      "rewards/reward_variance_humanlike/std": 0.042153649032115936,
+      "step": 20
+    },
+    {
+      "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": 587.0,
+      "completions/max_terminated_length": 587.0,
+      "completions/mean_length": 457.2375,
+      "completions/mean_terminated_length": 457.2375,
+      "completions/min_length": 343.8,
+      "completions/min_terminated_length": 343.8,
+      "epoch": 0.0004561211457763182,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09677507728338242,
+      "learning_rate": 2e-05,
+      "loss": -0.0021,
+      "num_tokens": 231427.0,
+      "reward": 2.66627516746521,
+      "reward_std": 0.18921945840120316,
+      "rewards/reward_humanlike/mean": 0.020775890350341795,
+      "rewards/reward_humanlike/std": 0.04260522716213018,
+      "rewards/reward_nonslop/mean": 0.4985057771205902,
+      "rewards/reward_nonslop/std": 0.03064997084438801,
+      "rewards/reward_paragraph_structure/mean": 0.977533757686615,
+      "rewards/reward_paragraph_structure/std": 0.059077487885952,
+      "rewards/reward_readability_flesch/mean": 0.7888182640075684,
+      "rewards/reward_readability_flesch/std": 0.1475653052330017,
+      "rewards/reward_variance_humanlike/mean": 0.380641496181488,
+      "rewards/reward_variance_humanlike/std": 0.04412947595119476,
+      "step": 25
+    },
+    {
+      "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": 597.6,
+      "completions/max_terminated_length": 597.6,
+      "completions/mean_length": 497.3375,
+      "completions/mean_terminated_length": 497.3375,
+      "completions/min_length": 395.4,
+      "completions/min_terminated_length": 395.4,
+      "epoch": 0.0005473453749315818,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08108917623758316,
+      "learning_rate": 2e-05,
+      "loss": 0.0017,
+      "num_tokens": 276318.0,
+      "reward": 2.7883635997772216,
+      "reward_std": 0.17138139009475709,
+      "rewards/reward_humanlike/mean": 0.11761673018336297,
+      "rewards/reward_humanlike/std": 0.08830970991402864,
+      "rewards/reward_nonslop/mean": 0.5173340082168579,
+      "rewards/reward_nonslop/std": 0.029454480484128,
+      "rewards/reward_paragraph_structure/mean": 0.9722555875778198,
+      "rewards/reward_paragraph_structure/std": 0.04261327683925629,
+      "rewards/reward_readability_flesch/mean": 0.7764949321746826,
+      "rewards/reward_readability_flesch/std": 0.10386961698532104,
+      "rewards/reward_variance_humanlike/mean": 0.4046623885631561,
+      "rewards/reward_variance_humanlike/std": 0.045019736886024474,
+      "step": 30
+    },
+    {
+      "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": 617.4,
+      "completions/max_terminated_length": 617.4,
+      "completions/mean_length": 523.825,
+      "completions/mean_terminated_length": 523.825,
+      "completions/min_length": 409.6,
+      "completions/min_terminated_length": 409.6,
+      "epoch": 0.0006385696040868455,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08894547820091248,
+      "learning_rate": 2e-05,
+      "loss": -0.0081,
+      "num_tokens": 323376.0,
+      "reward": 2.7335947513580323,
+      "reward_std": 0.12913600355386734,
+      "rewards/reward_humanlike/mean": 0.044519555568695066,
+      "rewards/reward_humanlike/std": 0.051969482377171514,
+      "rewards/reward_nonslop/mean": 0.5143645763397217,
+      "rewards/reward_nonslop/std": 0.024226398020982743,
+      "rewards/reward_paragraph_structure/mean": 0.9907094240188599,
+      "rewards/reward_paragraph_structure/std": 0.023819129168987273,
+      "rewards/reward_readability_flesch/mean": 0.7974974989891053,
+      "rewards/reward_readability_flesch/std": 0.12351922243833542,
+      "rewards/reward_variance_humanlike/mean": 0.38650373220443723,
+      "rewards/reward_variance_humanlike/std": 0.05399425402283668,
+      "step": 35
+    },
+    {
+      "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": 846.8,
+      "completions/max_terminated_length": 846.8,
+      "completions/mean_length": 555.05,
+      "completions/mean_terminated_length": 555.05,
+      "completions/min_length": 432.2,
+      "completions/min_terminated_length": 432.2,
+      "epoch": 0.0007297938332421091,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.089969202876091,
+      "learning_rate": 2e-05,
+      "loss": 0.0135,
+      "num_tokens": 372868.0,
+      "reward": 2.6368720293045045,
+      "reward_std": 0.17332236915826799,
+      "rewards/reward_humanlike/mean": 0.05930730104446411,
+      "rewards/reward_humanlike/std": 0.04687353018671274,
+      "rewards/reward_nonslop/mean": 0.4885416626930237,
+      "rewards/reward_nonslop/std": 0.03745536170899868,
+      "rewards/reward_paragraph_structure/mean": 0.890105378627777,
+      "rewards/reward_paragraph_structure/std": 0.09538187831640244,
+      "rewards/reward_readability_flesch/mean": 0.7819342494010926,
+      "rewards/reward_readability_flesch/std": 0.11071507334709167,
+      "rewards/reward_variance_humanlike/mean": 0.4169835090637207,
+      "rewards/reward_variance_humanlike/std": 0.0439291026443243,
+      "step": 40
+    },
+    {
+      "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": 613.2,
+      "completions/max_terminated_length": 613.2,
+      "completions/mean_length": 505.075,
+      "completions/mean_terminated_length": 505.075,
+      "completions/min_length": 374.0,
+      "completions/min_terminated_length": 374.0,
+      "epoch": 0.0008210180623973727,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0890246331691742,
+      "learning_rate": 2e-05,
+      "loss": 0.0102,
+      "num_tokens": 418490.0,
+      "reward": 2.6527084350585937,
+      "reward_std": 0.16761653274297714,
+      "rewards/reward_humanlike/mean": 0.08072967529296875,
+      "rewards/reward_humanlike/std": 0.06607948099263013,
+      "rewards/reward_nonslop/mean": 0.5106609463691711,
+      "rewards/reward_nonslop/std": 0.0351794958114624,
+      "rewards/reward_paragraph_structure/mean": 0.9813281297683716,
+      "rewards/reward_paragraph_structure/std": 0.05892824307084084,
+      "rewards/reward_readability_flesch/mean": 0.6913772821426392,
+      "rewards/reward_readability_flesch/std": 0.13855507373809814,
+      "rewards/reward_variance_humanlike/mean": 0.38861238956451416,
+      "rewards/reward_variance_humanlike/std": 0.039696966856718065,
+      "step": 45
+    },
+    {
+      "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": 607.2,
+      "completions/max_terminated_length": 607.2,
+      "completions/mean_length": 500.2125,
+      "completions/mean_terminated_length": 500.2125,
+      "completions/min_length": 407.0,
+      "completions/min_terminated_length": 407.0,
+      "epoch": 0.0009122422915526364,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07912778109312057,
+      "learning_rate": 2e-05,
+      "loss": 0.005,
+      "num_tokens": 463659.0,
+      "reward": 2.753040885925293,
+      "reward_std": 0.136872561275959,
+      "rewards/reward_humanlike/mean": 0.03859691172838211,
+      "rewards/reward_humanlike/std": 0.035087807662785055,
+      "rewards/reward_nonslop/mean": 0.5123575925827026,
+      "rewards/reward_nonslop/std": 0.02690323181450367,
+      "rewards/reward_paragraph_structure/mean": 0.992300295829773,
+      "rewards/reward_paragraph_structure/std": 0.030798836052417754,
+      "rewards/reward_readability_flesch/mean": 0.8399829745292664,
+      "rewards/reward_readability_flesch/std": 0.11433027684688568,
+      "rewards/reward_variance_humanlike/mean": 0.3698031187057495,
+      "rewards/reward_variance_humanlike/std": 0.04352555349469185,
+      "step": 50
+    },
+    {
+      "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": 562.0,
+      "completions/max_terminated_length": 562.0,
+      "completions/mean_length": 444.2875,
+      "completions/mean_terminated_length": 444.2875,
+      "completions/min_length": 313.0,
+      "completions/min_terminated_length": 313.0,
+      "epoch": 0.0010034665207079,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0981728583574295,
+      "learning_rate": 2e-05,
+      "loss": -0.0035,
+      "num_tokens": 504402.0,
+      "reward": 2.7707664489746096,
+      "reward_std": 0.15653531849384308,
+      "rewards/reward_humanlike/mean": 0.06527388915419578,
+      "rewards/reward_humanlike/std": 0.038607833255082366,
+      "rewards/reward_nonslop/mean": 0.5381535172462464,
+      "rewards/reward_nonslop/std": 0.030977290123701096,
+      "rewards/reward_paragraph_structure/mean": 0.9606678128242493,
+      "rewards/reward_paragraph_structure/std": 0.11089053153991699,
+      "rewards/reward_readability_flesch/mean": 0.7880517601966858,
+      "rewards/reward_readability_flesch/std": 0.14046915918588637,
+      "rewards/reward_variance_humanlike/mean": 0.4186194181442261,
+      "rewards/reward_variance_humanlike/std": 0.0556840606033802,
+      "step": 55
+    },
+    {
+      "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": 576.2,
+      "completions/max_terminated_length": 576.2,
+      "completions/mean_length": 473.1625,
+      "completions/mean_terminated_length": 473.1625,
+      "completions/min_length": 373.8,
+      "completions/min_terminated_length": 373.8,
+      "epoch": 0.0010946907498631637,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0779431164264679,
+      "learning_rate": 2e-05,
+      "loss": 0.0033,
+      "num_tokens": 547375.0,
+      "reward": 2.721493101119995,
+      "reward_std": 0.16468897759914397,
+      "rewards/reward_humanlike/mean": 0.06399692744016647,
+      "rewards/reward_humanlike/std": 0.06561810113489627,
+      "rewards/reward_nonslop/mean": 0.526066243648529,
+      "rewards/reward_nonslop/std": 0.03075612857937813,
+      "rewards/reward_paragraph_structure/mean": 0.9890687108039856,
+      "rewards/reward_paragraph_structure/std": 0.043725161254405974,
+      "rewards/reward_readability_flesch/mean": 0.7518861651420593,
+      "rewards/reward_readability_flesch/std": 0.13989782482385635,
+      "rewards/reward_variance_humanlike/mean": 0.39047507643699647,
+      "rewards/reward_variance_humanlike/std": 0.04762849882245064,
+      "step": 60
+    },
+    {
+      "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": 578.8,
+      "completions/max_terminated_length": 578.8,
+      "completions/mean_length": 483.25,
+      "completions/mean_terminated_length": 483.25,
+      "completions/min_length": 392.0,
+      "completions/min_terminated_length": 392.0,
+      "epoch": 0.0011859149790184272,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08613153547048569,
+      "learning_rate": 2e-05,
+      "loss": -0.0079,
+      "num_tokens": 591107.0,
+      "reward": 2.756886434555054,
+      "reward_std": 0.15308991372585296,
+      "rewards/reward_humanlike/mean": 0.03657452017068863,
+      "rewards/reward_humanlike/std": 0.030589387845247985,
+      "rewards/reward_nonslop/mean": 0.5334658443927764,
+      "rewards/reward_nonslop/std": 0.02915770635008812,
+      "rewards/reward_paragraph_structure/mean": 0.9779379844665528,
+      "rewards/reward_paragraph_structure/std": 0.07469315528869629,
+      "rewards/reward_readability_flesch/mean": 0.8177438259124756,
+      "rewards/reward_readability_flesch/std": 0.12392935305833816,
+      "rewards/reward_variance_humanlike/mean": 0.3911642551422119,
+      "rewards/reward_variance_humanlike/std": 0.04321202002465725,
+      "step": 65
+    },
+    {
+      "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": 607.0,
+      "completions/max_terminated_length": 607.0,
+      "completions/mean_length": 494.2,
+      "completions/mean_terminated_length": 494.2,
+      "completions/min_length": 396.0,
+      "completions/min_terminated_length": 396.0,
+      "epoch": 0.001277139208173691,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08903255313634872,
+      "learning_rate": 2e-05,
+      "loss": 0.0097,
+      "num_tokens": 635795.0,
+      "reward": 2.869403123855591,
+      "reward_std": 0.12632807344198227,
+      "rewards/reward_humanlike/mean": 0.04444831982254982,
+      "rewards/reward_humanlike/std": 0.036120910011231896,
+      "rewards/reward_nonslop/mean": 0.5243099689483642,
+      "rewards/reward_nonslop/std": 0.027720437198877335,
+      "rewards/reward_paragraph_structure/mean": 0.9873847007751465,
+      "rewards/reward_paragraph_structure/std": 0.04481239095330238,
+      "rewards/reward_readability_flesch/mean": 0.9202077388763428,
+      "rewards/reward_readability_flesch/std": 0.09855079799890518,
+      "rewards/reward_variance_humanlike/mean": 0.39305235743522643,
+      "rewards/reward_variance_humanlike/std": 0.04738411866128445,
+      "step": 70
+    },
+    {
+      "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": 554.2,
+      "completions/max_terminated_length": 554.2,
+      "completions/mean_length": 441.2,
+      "completions/mean_terminated_length": 441.2,
+      "completions/min_length": 343.0,
+      "completions/min_terminated_length": 343.0,
+      "epoch": 0.0013683634373289546,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08839516341686249,
+      "learning_rate": 2e-05,
+      "loss": 0.006,
+      "num_tokens": 676227.0,
+      "reward": 2.7607101917266847,
+      "reward_std": 0.14567108005285262,
+      "rewards/reward_humanlike/mean": 0.0049436971545219425,
+      "rewards/reward_humanlike/std": 0.009033319354057313,
+      "rewards/reward_nonslop/mean": 0.5163205862045288,
+      "rewards/reward_nonslop/std": 0.034804318100214005,
+      "rewards/reward_paragraph_structure/mean": 0.9944571018218994,
+      "rewards/reward_paragraph_structure/std": 0.022171604633331298,
+      "rewards/reward_readability_flesch/mean": 0.8619415044784546,
+      "rewards/reward_readability_flesch/std": 0.12174343764781952,
+      "rewards/reward_variance_humanlike/mean": 0.38304726481437684,
+      "rewards/reward_variance_humanlike/std": 0.05802188515663147,
+      "step": 75
+    },
+    {
+      "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": 717.2,
+      "completions/max_terminated_length": 717.2,
+      "completions/mean_length": 521.9,
+      "completions/mean_terminated_length": 521.9,
+      "completions/min_length": 412.2,
+      "completions/min_terminated_length": 412.2,
+      "epoch": 0.0014595876664842181,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0736362412571907,
+      "learning_rate": 2e-05,
+      "loss": 0.0207,
+      "num_tokens": 723019.0,
+      "reward": 2.7467071056365966,
+      "reward_std": 0.13217758685350417,
+      "rewards/reward_humanlike/mean": 0.04404948055744171,
+      "rewards/reward_humanlike/std": 0.03210974643006921,
+      "rewards/reward_nonslop/mean": 0.5157673060894012,
+      "rewards/reward_nonslop/std": 0.03307870253920555,
+      "rewards/reward_paragraph_structure/mean": 0.9820503711700439,
+      "rewards/reward_paragraph_structure/std": 0.06460869461297988,
+      "rewards/reward_readability_flesch/mean": 0.8017061710357666,
+      "rewards/reward_readability_flesch/std": 0.09596328362822533,
+      "rewards/reward_variance_humanlike/mean": 0.4031337320804596,
+      "rewards/reward_variance_humanlike/std": 0.038825053721666336,
+      "step": 80
+    },
+    {
+      "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": 568.2,
+      "completions/max_terminated_length": 568.2,
+      "completions/mean_length": 492.875,
+      "completions/mean_terminated_length": 492.875,
+      "completions/min_length": 405.2,
+      "completions/min_terminated_length": 405.2,
+      "epoch": 0.0015508118956394819,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09337656199932098,
+      "learning_rate": 2e-05,
+      "loss": 0.0205,
+      "num_tokens": 767585.0,
+      "reward": 2.8245923042297365,
+      "reward_std": 0.12125235199928283,
+      "rewards/reward_humanlike/mean": 0.04374306052923203,
+      "rewards/reward_humanlike/std": 0.05065076029859483,
+      "rewards/reward_nonslop/mean": 0.5121960043907166,
+      "rewards/reward_nonslop/std": 0.024268346652388573,
+      "rewards/reward_paragraph_structure/mean": 0.9952531695365906,
+      "rewards/reward_paragraph_structure/std": 0.01898730993270874,
+      "rewards/reward_readability_flesch/mean": 0.8540295481681823,
+      "rewards/reward_readability_flesch/std": 0.11251321136951446,
+      "rewards/reward_variance_humanlike/mean": 0.4193705081939697,
+      "rewards/reward_variance_humanlike/std": 0.0530940480530262,
+      "step": 85
+    },
+    {
+      "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": 529.2,
+      "completions/max_terminated_length": 529.2,
+      "completions/mean_length": 429.45,
+      "completions/mean_terminated_length": 429.45,
+      "completions/min_length": 340.6,
+      "completions/min_terminated_length": 340.6,
+      "epoch": 0.0016420361247947454,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10793354362249374,
+      "learning_rate": 2e-05,
+      "loss": 0.0084,
+      "num_tokens": 807109.0,
+      "reward": 2.8659895420074464,
+      "reward_std": 0.22321247607469558,
+      "rewards/reward_humanlike/mean": 0.11830640286207199,
+      "rewards/reward_humanlike/std": 0.10621838448569179,
+      "rewards/reward_nonslop/mean": 0.5382362484931946,
+      "rewards/reward_nonslop/std": 0.02679550349712372,
+      "rewards/reward_paragraph_structure/mean": 0.9388323545455932,
+      "rewards/reward_paragraph_structure/std": 0.08712138235569,
+      "rewards/reward_readability_flesch/mean": 0.8695886135101318,
+      "rewards/reward_readability_flesch/std": 0.11793182790279388,
+      "rewards/reward_variance_humanlike/mean": 0.4010259807109833,
+      "rewards/reward_variance_humanlike/std": 0.05023952871561051,
+      "step": 90
+    },
+    {
+      "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": 516.2,
+      "completions/max_terminated_length": 516.2,
+      "completions/mean_length": 427.1625,
+      "completions/mean_terminated_length": 427.1625,
+      "completions/min_length": 327.4,
+      "completions/min_terminated_length": 327.4,
+      "epoch": 0.0017332603539500092,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10153933614492416,
+      "learning_rate": 2e-05,
+      "loss": 0.0076,
+      "num_tokens": 846354.0,
+      "reward": 2.9194432735443114,
+      "reward_std": 0.1527949333190918,
+      "rewards/reward_humanlike/mean": 0.13011178895831108,
+      "rewards/reward_humanlike/std": 0.1040344807319343,
+      "rewards/reward_nonslop/mean": 0.539959442615509,
+      "rewards/reward_nonslop/std": 0.027887169644236563,
+      "rewards/reward_paragraph_structure/mean": 0.97805335521698,
+      "rewards/reward_paragraph_structure/std": 0.0572780579328537,
+      "rewards/reward_readability_flesch/mean": 0.8915346980094909,
+      "rewards/reward_readability_flesch/std": 0.10210664793848992,
+      "rewards/reward_variance_humanlike/mean": 0.37978398203849795,
+      "rewards/reward_variance_humanlike/std": 0.04532681703567505,
+      "step": 95
+    },
+    {
+      "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": 507.2,
+      "completions/max_terminated_length": 507.2,
+      "completions/mean_length": 416.6125,
+      "completions/mean_terminated_length": 416.6125,
+      "completions/min_length": 341.0,
+      "completions/min_terminated_length": 341.0,
+      "epoch": 0.0018244845831052727,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11056620627641678,
+      "learning_rate": 2e-05,
+      "loss": -0.0238,
+      "num_tokens": 884675.0,
+      "reward": 2.8297982215881348,
+      "reward_std": 0.1837340384721756,
+      "rewards/reward_humanlike/mean": 0.029251274466514588,
+      "rewards/reward_humanlike/std": 0.030034957686439157,
+      "rewards/reward_nonslop/mean": 0.5330106735229492,
+      "rewards/reward_nonslop/std": 0.02519891820847988,
+      "rewards/reward_paragraph_structure/mean": 0.9299438714981079,
+      "rewards/reward_paragraph_structure/std": 0.11956685781478882,
+      "rewards/reward_readability_flesch/mean": 0.9390206336975098,
+      "rewards/reward_readability_flesch/std": 0.1079161748290062,
+      "rewards/reward_variance_humanlike/mean": 0.3985717177391052,
+      "rewards/reward_variance_humanlike/std": 0.046227993071079256,
+      "step": 100
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 250,
+  "num_input_tokens_seen": 884675,
+  "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
+}

checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d6d1dbe50a50058c947771a7fd7c03a443069be77d949e0bd75335cc07d1af9
+size 6865

checkpoint-150/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen3-14B
+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

checkpoint-150/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-150/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b043f3ed1ecad2d74ae61f65bc553b874406c50527afbcf3a90a4ac8a88aef86
+size 513877864

checkpoint-150/added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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
+}

checkpoint-150/chat_template.jinja

@@ -0,0 +1,89 @@
+{%- 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\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

checkpoint-150/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e0a63c17e21c14c2311bad804f31dcb5395748cb54b3ac867c19dc915d95c5d
+size 1028078067

checkpoint-150/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c806fd612d0bb9895337dd2e7aa950823e97b44f3b9752eea8999dd38c306e8
+size 14645

checkpoint-150/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad292b5d14b8f8d85544d7d54740ffc013a4923be9b6dca5a08e1479e52628cc
+size 1465

checkpoint-150/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "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": "<|im_end|>"
+}

checkpoint-150/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de053c72a4ae289224ac988558897303006ceea1db22f30b97d0d856969ea6b9
+size 11422818

checkpoint-150/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "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": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-150/trainer_state.json

@@ -0,0 +1,1024 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.002736726874657909,
+  "eval_steps": 500,
+  "global_step": 150,
+  "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": 620.2,
+      "completions/max_terminated_length": 620.2,
+      "completions/mean_length": 536.3875,
+      "completions/mean_terminated_length": 536.3875,
+      "completions/min_length": 455.6,
+      "completions/min_terminated_length": 455.6,
+      "epoch": 9.122422915526363e-05,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.06800810992717743,
+      "learning_rate": 2e-05,
+      "loss": -0.0026,
+      "num_tokens": 48111.0,
+      "reward": 2.393585443496704,
+      "reward_std": 0.14288972318172455,
+      "rewards/reward_humanlike/mean": -0.003280111402273178,
+      "rewards/reward_humanlike/std": 0.002425189537461847,
+      "rewards/reward_nonslop/mean": 0.45034117698669435,
+      "rewards/reward_nonslop/std": 0.021841670200228692,
+      "rewards/reward_paragraph_structure/mean": 0.9998971700668335,
+      "rewards/reward_paragraph_structure/std": 0.00036894355434924364,
+      "rewards/reward_readability_flesch/mean": 0.5846633613109589,
+      "rewards/reward_readability_flesch/std": 0.11768133006989956,
+      "rewards/reward_variance_humanlike/mean": 0.36196385622024535,
+      "rewards/reward_variance_humanlike/std": 0.036638665944337845,
+      "step": 5
+    },
+    {
+      "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": 646.0,
+      "completions/max_terminated_length": 646.0,
+      "completions/mean_length": 551.5,
+      "completions/mean_terminated_length": 551.5,
+      "completions/min_length": 446.2,
+      "completions/min_terminated_length": 446.2,
+      "epoch": 0.00018244845831052726,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07684256136417389,
+      "learning_rate": 2e-05,
+      "loss": 0.0074,
+      "num_tokens": 97303.0,
+      "reward": 2.391454339027405,
+      "reward_std": 0.1808932200074196,
+      "rewards/reward_humanlike/mean": -0.0016780316829681397,
+      "rewards/reward_humanlike/std": 0.014642862777691334,
+      "rewards/reward_nonslop/mean": 0.44268596172332764,
+      "rewards/reward_nonslop/std": 0.028079804033041,
+      "rewards/reward_paragraph_structure/mean": 1.0,
+      "rewards/reward_paragraph_structure/std": 0.0,
+      "rewards/reward_readability_flesch/mean": 0.57281613945961,
+      "rewards/reward_readability_flesch/std": 0.16779391318559647,
+      "rewards/reward_variance_humanlike/mean": 0.3776302635669708,
+      "rewards/reward_variance_humanlike/std": 0.045258138328790665,
+      "step": 10
+    },
+    {
+      "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": 598.0,
+      "completions/max_terminated_length": 598.0,
+      "completions/mean_length": 516.1125,
+      "completions/mean_terminated_length": 516.1125,
+      "completions/min_length": 431.6,
+      "completions/min_terminated_length": 431.6,
+      "epoch": 0.0002736726874657909,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07912639528512955,
+      "learning_rate": 2e-05,
+      "loss": 0.0026,
+      "num_tokens": 143712.0,
+      "reward": 2.121282529830933,
+      "reward_std": 0.19594088345766067,
+      "rewards/reward_humanlike/mean": -0.005437415838241577,
+      "rewards/reward_humanlike/std": 0.0020546794403344393,
+      "rewards/reward_nonslop/mean": 0.44375383853912354,
+      "rewards/reward_nonslop/std": 0.025456106290221214,
+      "rewards/reward_paragraph_structure/mean": 0.9997280359268188,
+      "rewards/reward_paragraph_structure/std": 0.001087829563766718,
+      "rewards/reward_readability_flesch/mean": 0.340261435508728,
+      "rewards/reward_readability_flesch/std": 0.17545561231672763,
+      "rewards/reward_variance_humanlike/mean": 0.3429765820503235,
+      "rewards/reward_variance_humanlike/std": 0.04073530249297619,
+      "step": 15
+    },
+    {
+      "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": 632.4,
+      "completions/max_terminated_length": 632.4,
+      "completions/mean_length": 511.4,
+      "completions/mean_terminated_length": 511.4,
+      "completions/min_length": 387.0,
+      "completions/min_terminated_length": 387.0,
+      "epoch": 0.00036489691662105453,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08999043703079224,
+      "learning_rate": 2e-05,
+      "loss": 0.0183,
+      "num_tokens": 189696.0,
+      "reward": 2.6498445510864257,
+      "reward_std": 0.2029375523328781,
+      "rewards/reward_humanlike/mean": 0.00019632279872894287,
+      "rewards/reward_humanlike/std": 0.006884497962892056,
+      "rewards/reward_nonslop/mean": 0.4884443640708923,
+      "rewards/reward_nonslop/std": 0.0355165209621191,
+      "rewards/reward_paragraph_structure/mean": 0.9939258456230163,
+      "rewards/reward_paragraph_structure/std": 0.023617163486778735,
+      "rewards/reward_readability_flesch/mean": 0.8020971655845642,
+      "rewards/reward_readability_flesch/std": 0.17971589416265488,
+      "rewards/reward_variance_humanlike/mean": 0.3651808500289917,
+      "rewards/reward_variance_humanlike/std": 0.042153649032115936,
+      "step": 20
+    },
+    {
+      "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": 587.0,
+      "completions/max_terminated_length": 587.0,
+      "completions/mean_length": 457.2375,
+      "completions/mean_terminated_length": 457.2375,
+      "completions/min_length": 343.8,
+      "completions/min_terminated_length": 343.8,
+      "epoch": 0.0004561211457763182,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09677507728338242,
+      "learning_rate": 2e-05,
+      "loss": -0.0021,
+      "num_tokens": 231427.0,
+      "reward": 2.66627516746521,
+      "reward_std": 0.18921945840120316,
+      "rewards/reward_humanlike/mean": 0.020775890350341795,
+      "rewards/reward_humanlike/std": 0.04260522716213018,
+      "rewards/reward_nonslop/mean": 0.4985057771205902,
+      "rewards/reward_nonslop/std": 0.03064997084438801,
+      "rewards/reward_paragraph_structure/mean": 0.977533757686615,
+      "rewards/reward_paragraph_structure/std": 0.059077487885952,
+      "rewards/reward_readability_flesch/mean": 0.7888182640075684,
+      "rewards/reward_readability_flesch/std": 0.1475653052330017,
+      "rewards/reward_variance_humanlike/mean": 0.380641496181488,
+      "rewards/reward_variance_humanlike/std": 0.04412947595119476,
+      "step": 25
+    },
+    {
+      "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": 597.6,
+      "completions/max_terminated_length": 597.6,
+      "completions/mean_length": 497.3375,
+      "completions/mean_terminated_length": 497.3375,
+      "completions/min_length": 395.4,
+      "completions/min_terminated_length": 395.4,
+      "epoch": 0.0005473453749315818,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08108917623758316,
+      "learning_rate": 2e-05,
+      "loss": 0.0017,
+      "num_tokens": 276318.0,
+      "reward": 2.7883635997772216,
+      "reward_std": 0.17138139009475709,
+      "rewards/reward_humanlike/mean": 0.11761673018336297,
+      "rewards/reward_humanlike/std": 0.08830970991402864,
+      "rewards/reward_nonslop/mean": 0.5173340082168579,
+      "rewards/reward_nonslop/std": 0.029454480484128,
+      "rewards/reward_paragraph_structure/mean": 0.9722555875778198,
+      "rewards/reward_paragraph_structure/std": 0.04261327683925629,
+      "rewards/reward_readability_flesch/mean": 0.7764949321746826,
+      "rewards/reward_readability_flesch/std": 0.10386961698532104,
+      "rewards/reward_variance_humanlike/mean": 0.4046623885631561,
+      "rewards/reward_variance_humanlike/std": 0.045019736886024474,
+      "step": 30
+    },
+    {
+      "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": 617.4,
+      "completions/max_terminated_length": 617.4,
+      "completions/mean_length": 523.825,
+      "completions/mean_terminated_length": 523.825,
+      "completions/min_length": 409.6,
+      "completions/min_terminated_length": 409.6,
+      "epoch": 0.0006385696040868455,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08894547820091248,
+      "learning_rate": 2e-05,
+      "loss": -0.0081,
+      "num_tokens": 323376.0,
+      "reward": 2.7335947513580323,
+      "reward_std": 0.12913600355386734,
+      "rewards/reward_humanlike/mean": 0.044519555568695066,
+      "rewards/reward_humanlike/std": 0.051969482377171514,
+      "rewards/reward_nonslop/mean": 0.5143645763397217,
+      "rewards/reward_nonslop/std": 0.024226398020982743,
+      "rewards/reward_paragraph_structure/mean": 0.9907094240188599,
+      "rewards/reward_paragraph_structure/std": 0.023819129168987273,
+      "rewards/reward_readability_flesch/mean": 0.7974974989891053,
+      "rewards/reward_readability_flesch/std": 0.12351922243833542,
+      "rewards/reward_variance_humanlike/mean": 0.38650373220443723,
+      "rewards/reward_variance_humanlike/std": 0.05399425402283668,
+      "step": 35
+    },
+    {
+      "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": 846.8,
+      "completions/max_terminated_length": 846.8,
+      "completions/mean_length": 555.05,
+      "completions/mean_terminated_length": 555.05,
+      "completions/min_length": 432.2,
+      "completions/min_terminated_length": 432.2,
+      "epoch": 0.0007297938332421091,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.089969202876091,
+      "learning_rate": 2e-05,
+      "loss": 0.0135,
+      "num_tokens": 372868.0,
+      "reward": 2.6368720293045045,
+      "reward_std": 0.17332236915826799,
+      "rewards/reward_humanlike/mean": 0.05930730104446411,
+      "rewards/reward_humanlike/std": 0.04687353018671274,
+      "rewards/reward_nonslop/mean": 0.4885416626930237,
+      "rewards/reward_nonslop/std": 0.03745536170899868,
+      "rewards/reward_paragraph_structure/mean": 0.890105378627777,
+      "rewards/reward_paragraph_structure/std": 0.09538187831640244,
+      "rewards/reward_readability_flesch/mean": 0.7819342494010926,
+      "rewards/reward_readability_flesch/std": 0.11071507334709167,
+      "rewards/reward_variance_humanlike/mean": 0.4169835090637207,
+      "rewards/reward_variance_humanlike/std": 0.0439291026443243,
+      "step": 40
+    },
+    {
+      "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": 613.2,
+      "completions/max_terminated_length": 613.2,
+      "completions/mean_length": 505.075,
+      "completions/mean_terminated_length": 505.075,
+      "completions/min_length": 374.0,
+      "completions/min_terminated_length": 374.0,
+      "epoch": 0.0008210180623973727,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0890246331691742,
+      "learning_rate": 2e-05,
+      "loss": 0.0102,
+      "num_tokens": 418490.0,
+      "reward": 2.6527084350585937,
+      "reward_std": 0.16761653274297714,
+      "rewards/reward_humanlike/mean": 0.08072967529296875,
+      "rewards/reward_humanlike/std": 0.06607948099263013,
+      "rewards/reward_nonslop/mean": 0.5106609463691711,
+      "rewards/reward_nonslop/std": 0.0351794958114624,
+      "rewards/reward_paragraph_structure/mean": 0.9813281297683716,
+      "rewards/reward_paragraph_structure/std": 0.05892824307084084,
+      "rewards/reward_readability_flesch/mean": 0.6913772821426392,
+      "rewards/reward_readability_flesch/std": 0.13855507373809814,
+      "rewards/reward_variance_humanlike/mean": 0.38861238956451416,
+      "rewards/reward_variance_humanlike/std": 0.039696966856718065,
+      "step": 45
+    },
+    {
+      "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": 607.2,
+      "completions/max_terminated_length": 607.2,
+      "completions/mean_length": 500.2125,
+      "completions/mean_terminated_length": 500.2125,
+      "completions/min_length": 407.0,
+      "completions/min_terminated_length": 407.0,
+      "epoch": 0.0009122422915526364,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07912778109312057,
+      "learning_rate": 2e-05,
+      "loss": 0.005,
+      "num_tokens": 463659.0,
+      "reward": 2.753040885925293,
+      "reward_std": 0.136872561275959,
+      "rewards/reward_humanlike/mean": 0.03859691172838211,
+      "rewards/reward_humanlike/std": 0.035087807662785055,
+      "rewards/reward_nonslop/mean": 0.5123575925827026,
+      "rewards/reward_nonslop/std": 0.02690323181450367,
+      "rewards/reward_paragraph_structure/mean": 0.992300295829773,
+      "rewards/reward_paragraph_structure/std": 0.030798836052417754,
+      "rewards/reward_readability_flesch/mean": 0.8399829745292664,
+      "rewards/reward_readability_flesch/std": 0.11433027684688568,
+      "rewards/reward_variance_humanlike/mean": 0.3698031187057495,
+      "rewards/reward_variance_humanlike/std": 0.04352555349469185,
+      "step": 50
+    },
+    {
+      "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": 562.0,
+      "completions/max_terminated_length": 562.0,
+      "completions/mean_length": 444.2875,
+      "completions/mean_terminated_length": 444.2875,
+      "completions/min_length": 313.0,
+      "completions/min_terminated_length": 313.0,
+      "epoch": 0.0010034665207079,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0981728583574295,
+      "learning_rate": 2e-05,
+      "loss": -0.0035,
+      "num_tokens": 504402.0,
+      "reward": 2.7707664489746096,
+      "reward_std": 0.15653531849384308,
+      "rewards/reward_humanlike/mean": 0.06527388915419578,
+      "rewards/reward_humanlike/std": 0.038607833255082366,
+      "rewards/reward_nonslop/mean": 0.5381535172462464,
+      "rewards/reward_nonslop/std": 0.030977290123701096,
+      "rewards/reward_paragraph_structure/mean": 0.9606678128242493,
+      "rewards/reward_paragraph_structure/std": 0.11089053153991699,
+      "rewards/reward_readability_flesch/mean": 0.7880517601966858,
+      "rewards/reward_readability_flesch/std": 0.14046915918588637,
+      "rewards/reward_variance_humanlike/mean": 0.4186194181442261,
+      "rewards/reward_variance_humanlike/std": 0.0556840606033802,
+      "step": 55
+    },
+    {
+      "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": 576.2,
+      "completions/max_terminated_length": 576.2,
+      "completions/mean_length": 473.1625,
+      "completions/mean_terminated_length": 473.1625,
+      "completions/min_length": 373.8,
+      "completions/min_terminated_length": 373.8,
+      "epoch": 0.0010946907498631637,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0779431164264679,
+      "learning_rate": 2e-05,
+      "loss": 0.0033,
+      "num_tokens": 547375.0,
+      "reward": 2.721493101119995,
+      "reward_std": 0.16468897759914397,
+      "rewards/reward_humanlike/mean": 0.06399692744016647,
+      "rewards/reward_humanlike/std": 0.06561810113489627,
+      "rewards/reward_nonslop/mean": 0.526066243648529,
+      "rewards/reward_nonslop/std": 0.03075612857937813,
+      "rewards/reward_paragraph_structure/mean": 0.9890687108039856,
+      "rewards/reward_paragraph_structure/std": 0.043725161254405974,
+      "rewards/reward_readability_flesch/mean": 0.7518861651420593,
+      "rewards/reward_readability_flesch/std": 0.13989782482385635,
+      "rewards/reward_variance_humanlike/mean": 0.39047507643699647,
+      "rewards/reward_variance_humanlike/std": 0.04762849882245064,
+      "step": 60
+    },
+    {
+      "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": 578.8,
+      "completions/max_terminated_length": 578.8,
+      "completions/mean_length": 483.25,
+      "completions/mean_terminated_length": 483.25,
+      "completions/min_length": 392.0,
+      "completions/min_terminated_length": 392.0,
+      "epoch": 0.0011859149790184272,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08613153547048569,
+      "learning_rate": 2e-05,
+      "loss": -0.0079,
+      "num_tokens": 591107.0,
+      "reward": 2.756886434555054,
+      "reward_std": 0.15308991372585296,
+      "rewards/reward_humanlike/mean": 0.03657452017068863,
+      "rewards/reward_humanlike/std": 0.030589387845247985,
+      "rewards/reward_nonslop/mean": 0.5334658443927764,
+      "rewards/reward_nonslop/std": 0.02915770635008812,
+      "rewards/reward_paragraph_structure/mean": 0.9779379844665528,
+      "rewards/reward_paragraph_structure/std": 0.07469315528869629,
+      "rewards/reward_readability_flesch/mean": 0.8177438259124756,
+      "rewards/reward_readability_flesch/std": 0.12392935305833816,
+      "rewards/reward_variance_humanlike/mean": 0.3911642551422119,
+      "rewards/reward_variance_humanlike/std": 0.04321202002465725,
+      "step": 65
+    },
+    {
+      "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": 607.0,
+      "completions/max_terminated_length": 607.0,
+      "completions/mean_length": 494.2,
+      "completions/mean_terminated_length": 494.2,
+      "completions/min_length": 396.0,
+      "completions/min_terminated_length": 396.0,
+      "epoch": 0.001277139208173691,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08903255313634872,
+      "learning_rate": 2e-05,
+      "loss": 0.0097,
+      "num_tokens": 635795.0,
+      "reward": 2.869403123855591,
+      "reward_std": 0.12632807344198227,
+      "rewards/reward_humanlike/mean": 0.04444831982254982,
+      "rewards/reward_humanlike/std": 0.036120910011231896,
+      "rewards/reward_nonslop/mean": 0.5243099689483642,
+      "rewards/reward_nonslop/std": 0.027720437198877335,
+      "rewards/reward_paragraph_structure/mean": 0.9873847007751465,
+      "rewards/reward_paragraph_structure/std": 0.04481239095330238,
+      "rewards/reward_readability_flesch/mean": 0.9202077388763428,
+      "rewards/reward_readability_flesch/std": 0.09855079799890518,
+      "rewards/reward_variance_humanlike/mean": 0.39305235743522643,
+      "rewards/reward_variance_humanlike/std": 0.04738411866128445,
+      "step": 70
+    },
+    {
+      "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": 554.2,
+      "completions/max_terminated_length": 554.2,
+      "completions/mean_length": 441.2,
+      "completions/mean_terminated_length": 441.2,
+      "completions/min_length": 343.0,
+      "completions/min_terminated_length": 343.0,
+      "epoch": 0.0013683634373289546,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08839516341686249,
+      "learning_rate": 2e-05,
+      "loss": 0.006,
+      "num_tokens": 676227.0,
+      "reward": 2.7607101917266847,
+      "reward_std": 0.14567108005285262,
+      "rewards/reward_humanlike/mean": 0.0049436971545219425,
+      "rewards/reward_humanlike/std": 0.009033319354057313,
+      "rewards/reward_nonslop/mean": 0.5163205862045288,
+      "rewards/reward_nonslop/std": 0.034804318100214005,
+      "rewards/reward_paragraph_structure/mean": 0.9944571018218994,
+      "rewards/reward_paragraph_structure/std": 0.022171604633331298,
+      "rewards/reward_readability_flesch/mean": 0.8619415044784546,
+      "rewards/reward_readability_flesch/std": 0.12174343764781952,
+      "rewards/reward_variance_humanlike/mean": 0.38304726481437684,
+      "rewards/reward_variance_humanlike/std": 0.05802188515663147,
+      "step": 75
+    },
+    {
+      "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": 717.2,
+      "completions/max_terminated_length": 717.2,
+      "completions/mean_length": 521.9,
+      "completions/mean_terminated_length": 521.9,
+      "completions/min_length": 412.2,
+      "completions/min_terminated_length": 412.2,
+      "epoch": 0.0014595876664842181,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0736362412571907,
+      "learning_rate": 2e-05,
+      "loss": 0.0207,
+      "num_tokens": 723019.0,
+      "reward": 2.7467071056365966,
+      "reward_std": 0.13217758685350417,
+      "rewards/reward_humanlike/mean": 0.04404948055744171,
+      "rewards/reward_humanlike/std": 0.03210974643006921,
+      "rewards/reward_nonslop/mean": 0.5157673060894012,
+      "rewards/reward_nonslop/std": 0.03307870253920555,
+      "rewards/reward_paragraph_structure/mean": 0.9820503711700439,
+      "rewards/reward_paragraph_structure/std": 0.06460869461297988,
+      "rewards/reward_readability_flesch/mean": 0.8017061710357666,
+      "rewards/reward_readability_flesch/std": 0.09596328362822533,
+      "rewards/reward_variance_humanlike/mean": 0.4031337320804596,
+      "rewards/reward_variance_humanlike/std": 0.038825053721666336,
+      "step": 80
+    },
+    {
+      "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": 568.2,
+      "completions/max_terminated_length": 568.2,
+      "completions/mean_length": 492.875,
+      "completions/mean_terminated_length": 492.875,
+      "completions/min_length": 405.2,
+      "completions/min_terminated_length": 405.2,
+      "epoch": 0.0015508118956394819,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09337656199932098,
+      "learning_rate": 2e-05,
+      "loss": 0.0205,
+      "num_tokens": 767585.0,
+      "reward": 2.8245923042297365,
+      "reward_std": 0.12125235199928283,
+      "rewards/reward_humanlike/mean": 0.04374306052923203,
+      "rewards/reward_humanlike/std": 0.05065076029859483,
+      "rewards/reward_nonslop/mean": 0.5121960043907166,
+      "rewards/reward_nonslop/std": 0.024268346652388573,
+      "rewards/reward_paragraph_structure/mean": 0.9952531695365906,
+      "rewards/reward_paragraph_structure/std": 0.01898730993270874,
+      "rewards/reward_readability_flesch/mean": 0.8540295481681823,
+      "rewards/reward_readability_flesch/std": 0.11251321136951446,
+      "rewards/reward_variance_humanlike/mean": 0.4193705081939697,
+      "rewards/reward_variance_humanlike/std": 0.0530940480530262,
+      "step": 85
+    },
+    {
+      "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": 529.2,
+      "completions/max_terminated_length": 529.2,
+      "completions/mean_length": 429.45,
+      "completions/mean_terminated_length": 429.45,
+      "completions/min_length": 340.6,
+      "completions/min_terminated_length": 340.6,
+      "epoch": 0.0016420361247947454,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10793354362249374,
+      "learning_rate": 2e-05,
+      "loss": 0.0084,
+      "num_tokens": 807109.0,
+      "reward": 2.8659895420074464,
+      "reward_std": 0.22321247607469558,
+      "rewards/reward_humanlike/mean": 0.11830640286207199,
+      "rewards/reward_humanlike/std": 0.10621838448569179,
+      "rewards/reward_nonslop/mean": 0.5382362484931946,
+      "rewards/reward_nonslop/std": 0.02679550349712372,
+      "rewards/reward_paragraph_structure/mean": 0.9388323545455932,
+      "rewards/reward_paragraph_structure/std": 0.08712138235569,
+      "rewards/reward_readability_flesch/mean": 0.8695886135101318,
+      "rewards/reward_readability_flesch/std": 0.11793182790279388,
+      "rewards/reward_variance_humanlike/mean": 0.4010259807109833,
+      "rewards/reward_variance_humanlike/std": 0.05023952871561051,
+      "step": 90
+    },
+    {
+      "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": 516.2,
+      "completions/max_terminated_length": 516.2,
+      "completions/mean_length": 427.1625,
+      "completions/mean_terminated_length": 427.1625,
+      "completions/min_length": 327.4,
+      "completions/min_terminated_length": 327.4,
+      "epoch": 0.0017332603539500092,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10153933614492416,
+      "learning_rate": 2e-05,
+      "loss": 0.0076,
+      "num_tokens": 846354.0,
+      "reward": 2.9194432735443114,
+      "reward_std": 0.1527949333190918,
+      "rewards/reward_humanlike/mean": 0.13011178895831108,
+      "rewards/reward_humanlike/std": 0.1040344807319343,
+      "rewards/reward_nonslop/mean": 0.539959442615509,
+      "rewards/reward_nonslop/std": 0.027887169644236563,
+      "rewards/reward_paragraph_structure/mean": 0.97805335521698,
+      "rewards/reward_paragraph_structure/std": 0.0572780579328537,
+      "rewards/reward_readability_flesch/mean": 0.8915346980094909,
+      "rewards/reward_readability_flesch/std": 0.10210664793848992,
+      "rewards/reward_variance_humanlike/mean": 0.37978398203849795,
+      "rewards/reward_variance_humanlike/std": 0.04532681703567505,
+      "step": 95
+    },
+    {
+      "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": 507.2,
+      "completions/max_terminated_length": 507.2,
+      "completions/mean_length": 416.6125,
+      "completions/mean_terminated_length": 416.6125,
+      "completions/min_length": 341.0,
+      "completions/min_terminated_length": 341.0,
+      "epoch": 0.0018244845831052727,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11056620627641678,
+      "learning_rate": 2e-05,
+      "loss": -0.0238,
+      "num_tokens": 884675.0,
+      "reward": 2.8297982215881348,
+      "reward_std": 0.1837340384721756,
+      "rewards/reward_humanlike/mean": 0.029251274466514588,
+      "rewards/reward_humanlike/std": 0.030034957686439157,
+      "rewards/reward_nonslop/mean": 0.5330106735229492,
+      "rewards/reward_nonslop/std": 0.02519891820847988,
+      "rewards/reward_paragraph_structure/mean": 0.9299438714981079,
+      "rewards/reward_paragraph_structure/std": 0.11956685781478882,
+      "rewards/reward_readability_flesch/mean": 0.9390206336975098,
+      "rewards/reward_readability_flesch/std": 0.1079161748290062,
+      "rewards/reward_variance_humanlike/mean": 0.3985717177391052,
+      "rewards/reward_variance_humanlike/std": 0.046227993071079256,
+      "step": 100
+    },
+    {
+      "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": 484.8,
+      "completions/max_terminated_length": 484.8,
+      "completions/mean_length": 397.0,
+      "completions/mean_terminated_length": 397.0,
+      "completions/min_length": 311.6,
+      "completions/min_terminated_length": 311.6,
+      "epoch": 0.0019157088122605363,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12234635651111603,
+      "learning_rate": 2e-05,
+      "loss": -0.0084,
+      "num_tokens": 921523.0,
+      "reward": 2.7906761169433594,
+      "reward_std": 0.18639333844184874,
+      "rewards/reward_humanlike/mean": 0.03967294096946716,
+      "rewards/reward_humanlike/std": 0.047859178856015205,
+      "rewards/reward_nonslop/mean": 0.5455231070518494,
+      "rewards/reward_nonslop/std": 0.02572299391031265,
+      "rewards/reward_paragraph_structure/mean": 0.9380416989326477,
+      "rewards/reward_paragraph_structure/std": 0.14332405030727385,
+      "rewards/reward_readability_flesch/mean": 0.8651458501815796,
+      "rewards/reward_readability_flesch/std": 0.10486775860190392,
+      "rewards/reward_variance_humanlike/mean": 0.40229247212409974,
+      "rewards/reward_variance_humanlike/std": 0.05056553930044174,
+      "step": 105
+    },
+    {
+      "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": 535.0,
+      "completions/max_terminated_length": 535.0,
+      "completions/mean_length": 407.4,
+      "completions/mean_terminated_length": 407.4,
+      "completions/min_length": 323.6,
+      "completions/min_terminated_length": 323.6,
+      "epoch": 0.0020069330414158,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12049026042222977,
+      "learning_rate": 2e-05,
+      "loss": -0.0015,
+      "num_tokens": 959235.0,
+      "reward": 2.8142509937286375,
+      "reward_std": 0.20959972888231276,
+      "rewards/reward_humanlike/mean": 0.10315801277756691,
+      "rewards/reward_humanlike/std": 0.06646046359091998,
+      "rewards/reward_nonslop/mean": 0.549621844291687,
+      "rewards/reward_nonslop/std": 0.02784626930952072,
+      "rewards/reward_paragraph_structure/mean": 0.9427387475967407,
+      "rewards/reward_paragraph_structure/std": 0.0946300521492958,
+      "rewards/reward_readability_flesch/mean": 0.8157400369644165,
+      "rewards/reward_readability_flesch/std": 0.12322266846895218,
+      "rewards/reward_variance_humanlike/mean": 0.40299235582351683,
+      "rewards/reward_variance_humanlike/std": 0.03882267326116562,
+      "step": 110
+    },
+    {
+      "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": 546.2,
+      "completions/max_terminated_length": 546.2,
+      "completions/mean_length": 453.1875,
+      "completions/mean_terminated_length": 453.1875,
+      "completions/min_length": 365.0,
+      "completions/min_terminated_length": 365.0,
+      "epoch": 0.0020981572705710636,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10685638338327408,
+      "learning_rate": 2e-05,
+      "loss": -0.0023,
+      "num_tokens": 1000562.0,
+      "reward": 2.7050733089447023,
+      "reward_std": 0.17174948751926422,
+      "rewards/reward_humanlike/mean": 0.03010150045156479,
+      "rewards/reward_humanlike/std": 0.02183089666068554,
+      "rewards/reward_nonslop/mean": 0.537644910812378,
+      "rewards/reward_nonslop/std": 0.022509072348475455,
+      "rewards/reward_paragraph_structure/mean": 0.9359247803688049,
+      "rewards/reward_paragraph_structure/std": 0.13663459122180938,
+      "rewards/reward_readability_flesch/mean": 0.7878659129142761,
+      "rewards/reward_readability_flesch/std": 0.1104049876332283,
+      "rewards/reward_variance_humanlike/mean": 0.4135362088680267,
+      "rewards/reward_variance_humanlike/std": 0.04251410141587257,
+      "step": 115
+    },
+    {
+      "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": 563.4,
+      "completions/max_terminated_length": 563.4,
+      "completions/mean_length": 454.7875,
+      "completions/mean_terminated_length": 454.7875,
+      "completions/min_length": 365.0,
+      "completions/min_terminated_length": 365.0,
+      "epoch": 0.0021893814997263274,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10802900046110153,
+      "learning_rate": 2e-05,
+      "loss": 0.0097,
+      "num_tokens": 1042177.0,
+      "reward": 2.8397992134094237,
+      "reward_std": 0.15860502868890763,
+      "rewards/reward_humanlike/mean": 0.04295992851257324,
+      "rewards/reward_humanlike/std": 0.05235542356967926,
+      "rewards/reward_nonslop/mean": 0.5383237481117249,
+      "rewards/reward_nonslop/std": 0.029689770936965943,
+      "rewards/reward_paragraph_structure/mean": 1.0,
+      "rewards/reward_paragraph_structure/std": 0.0,
+      "rewards/reward_readability_flesch/mean": 0.8279854476451873,
+      "rewards/reward_readability_flesch/std": 0.11809881925582885,
+      "rewards/reward_variance_humanlike/mean": 0.43053008913993834,
+      "rewards/reward_variance_humanlike/std": 0.04717860147356987,
+      "step": 120
+    },
+    {
+      "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": 586.2,
+      "completions/max_terminated_length": 586.2,
+      "completions/mean_length": 476.3125,
+      "completions/mean_terminated_length": 476.3125,
+      "completions/min_length": 379.4,
+      "completions/min_terminated_length": 379.4,
+      "epoch": 0.002280605728881591,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11722852289676666,
+      "learning_rate": 2e-05,
+      "loss": 0.0084,
+      "num_tokens": 1085338.0,
+      "reward": 2.8460022449493407,
+      "reward_std": 0.12902745008468627,
+      "rewards/reward_humanlike/mean": 0.08100048899650573,
+      "rewards/reward_humanlike/std": 0.0518460288643837,
+      "rewards/reward_nonslop/mean": 0.5345206022262573,
+      "rewards/reward_nonslop/std": 0.028127148374915124,
+      "rewards/reward_paragraph_structure/mean": 0.9582128286361694,
+      "rewards/reward_paragraph_structure/std": 0.052062156796455386,
+      "rewards/reward_readability_flesch/mean": 0.8341620564460754,
+      "rewards/reward_readability_flesch/std": 0.11833389699459076,
+      "rewards/reward_variance_humanlike/mean": 0.43810625076293946,
+      "rewards/reward_variance_humanlike/std": 0.06058142632246018,
+      "step": 125
+    },
+    {
+      "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": 637.2,
+      "completions/max_terminated_length": 637.2,
+      "completions/mean_length": 525.2625,
+      "completions/mean_terminated_length": 525.2625,
+      "completions/min_length": 415.6,
+      "completions/min_terminated_length": 415.6,
+      "epoch": 0.0023718299580368545,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.1034693568944931,
+      "learning_rate": 2e-05,
+      "loss": -0.0015,
+      "num_tokens": 1132447.0,
+      "reward": 2.9677632331848143,
+      "reward_std": 0.13178672641515732,
+      "rewards/reward_humanlike/mean": 0.11690731346607208,
+      "rewards/reward_humanlike/std": 0.07614295929670334,
+      "rewards/reward_nonslop/mean": 0.5198043942451477,
+      "rewards/reward_nonslop/std": 0.027982910722494127,
+      "rewards/reward_paragraph_structure/mean": 0.9756827473640441,
+      "rewards/reward_paragraph_structure/std": 0.044346729014068845,
+      "rewards/reward_readability_flesch/mean": 0.8949468612670899,
+      "rewards/reward_readability_flesch/std": 0.08357911184430122,
+      "rewards/reward_variance_humanlike/mean": 0.4604219257831573,
+      "rewards/reward_variance_humanlike/std": 0.05337681397795677,
+      "step": 130
+    },
+    {
+      "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": 588.0,
+      "completions/max_terminated_length": 588.0,
+      "completions/mean_length": 498.6875,
+      "completions/mean_terminated_length": 498.6875,
+      "completions/min_length": 399.0,
+      "completions/min_terminated_length": 399.0,
+      "epoch": 0.0024630541871921183,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.1009395495057106,
+      "learning_rate": 2e-05,
+      "loss": -0.0043,
+      "num_tokens": 1177382.0,
+      "reward": 2.822851276397705,
+      "reward_std": 0.13398206382989883,
+      "rewards/reward_humanlike/mean": 0.06381046175956726,
+      "rewards/reward_humanlike/std": 0.0503330685198307,
+      "rewards/reward_nonslop/mean": 0.5348796844482422,
+      "rewards/reward_nonslop/std": 0.025143631175160407,
+      "rewards/reward_paragraph_structure/mean": 0.9924662351608277,
+      "rewards/reward_paragraph_structure/std": 0.03013508329167962,
+      "rewards/reward_readability_flesch/mean": 0.7840398788452149,
+      "rewards/reward_readability_flesch/std": 0.11849776804447174,
+      "rewards/reward_variance_humanlike/mean": 0.44765504002571105,
+      "rewards/reward_variance_humanlike/std": 0.045124633610248564,
+      "step": 135
+    },
+    {
+      "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": 667.8,
+      "completions/max_terminated_length": 667.8,
+      "completions/mean_length": 552.2125,
+      "completions/mean_terminated_length": 552.2125,
+      "completions/min_length": 431.8,
+      "completions/min_terminated_length": 431.8,
+      "epoch": 0.002554278416347382,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10111426562070847,
+      "learning_rate": 2e-05,
+      "loss": -0.017,
+      "num_tokens": 1226679.0,
+      "reward": 2.9173266887664795,
+      "reward_std": 0.1144576370716095,
+      "rewards/reward_humanlike/mean": 0.11132953837513923,
+      "rewards/reward_humanlike/std": 0.05890989601612091,
+      "rewards/reward_nonslop/mean": 0.4942618072032928,
+      "rewards/reward_nonslop/std": 0.02765493802726269,
+      "rewards/reward_paragraph_structure/mean": 0.9944768905639648,
+      "rewards/reward_paragraph_structure/std": 0.02089656200259924,
+      "rewards/reward_readability_flesch/mean": 0.8778629779815674,
+      "rewards/reward_readability_flesch/std": 0.08600130677223206,
+      "rewards/reward_variance_humanlike/mean": 0.4393955051898956,
+      "rewards/reward_variance_humanlike/std": 0.04750688448548317,
+      "step": 140
+    },
+    {
+      "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": 640.6,
+      "completions/max_terminated_length": 640.6,
+      "completions/mean_length": 534.1125,
+      "completions/mean_terminated_length": 534.1125,
+      "completions/min_length": 438.2,
+      "completions/min_terminated_length": 438.2,
+      "epoch": 0.0026455026455026454,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11239214986562729,
+      "learning_rate": 2e-05,
+      "loss": 0.0035,
+      "num_tokens": 1274592.0,
+      "reward": 3.053114986419678,
+      "reward_std": 0.11381979882717133,
+      "rewards/reward_humanlike/mean": 0.12257216051220894,
+      "rewards/reward_humanlike/std": 0.08092293813824654,
+      "rewards/reward_nonslop/mean": 0.5075180411338807,
+      "rewards/reward_nonslop/std": 0.026819777116179465,
+      "rewards/reward_paragraph_structure/mean": 0.9958648920059204,
+      "rewards/reward_paragraph_structure/std": 0.015135037153959275,
+      "rewards/reward_readability_flesch/mean": 0.9384488105773926,
+      "rewards/reward_readability_flesch/std": 0.07422738522291183,
+      "rewards/reward_variance_humanlike/mean": 0.4887110412120819,
+      "rewards/reward_variance_humanlike/std": 0.044287893921136856,
+      "step": 145
+    },
+    {
+      "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": 625.8,
+      "completions/max_terminated_length": 625.8,
+      "completions/mean_length": 526.2,
+      "completions/mean_terminated_length": 526.2,
+      "completions/min_length": 432.6,
+      "completions/min_terminated_length": 432.6,
+      "epoch": 0.002736726874657909,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09958786517381668,
+      "learning_rate": 2e-05,
+      "loss": 0.0085,
+      "num_tokens": 1321840.0,
+      "reward": 2.912756013870239,
+      "reward_std": 0.12361426800489425,
+      "rewards/reward_humanlike/mean": 0.05477507412433624,
+      "rewards/reward_humanlike/std": 0.04108944982290268,
+      "rewards/reward_nonslop/mean": 0.5123514175415039,
+      "rewards/reward_nonslop/std": 0.027717596665024758,
+      "rewards/reward_paragraph_structure/mean": 0.9881778597831726,
+      "rewards/reward_paragraph_structure/std": 0.04215195477008819,
+      "rewards/reward_readability_flesch/mean": 0.8884697198867798,
+      "rewards/reward_readability_flesch/std": 0.09501284211874009,
+      "rewards/reward_variance_humanlike/mean": 0.4689819157123566,
+      "rewards/reward_variance_humanlike/std": 0.04468975365161896,
+      "step": 150
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 250,
+  "num_input_tokens_seen": 1321840,
+  "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
+}

checkpoint-150/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d6d1dbe50a50058c947771a7fd7c03a443069be77d949e0bd75335cc07d1af9
+size 6865

checkpoint-200/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen3-14B
+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

checkpoint-200/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-200/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7012d37b144b1be4b97d5a7d3b44abbcbbe5937c5c2c5fda16484c87cec1a27
+size 513877864

checkpoint-200/added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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
+}

checkpoint-200/chat_template.jinja

@@ -0,0 +1,89 @@
+{%- 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\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

checkpoint-200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c69338a6dae8f8aaff2eb85dca23e5f2667c4b2bd754690e0eafdbfb5185362b
+size 1028078067

checkpoint-200/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b68a925ba8c76a0adbb4736675a87f15dd852afb04a31e061101c18d65d63df
+size 14645

checkpoint-200/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6081ec7dd1df8dbae82c29a2535e4e2a776621f54df52246eada222f4cfa3133
+size 1465

checkpoint-200/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "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": "<|im_end|>"
+}

checkpoint-200/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de053c72a4ae289224ac988558897303006ceea1db22f30b97d0d856969ea6b9
+size 11422818

checkpoint-200/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "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": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-200/trainer_state.json

@@ -0,0 +1,1354 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.0036489691662105455,
+  "eval_steps": 500,
+  "global_step": 200,
+  "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": 620.2,
+      "completions/max_terminated_length": 620.2,
+      "completions/mean_length": 536.3875,
+      "completions/mean_terminated_length": 536.3875,
+      "completions/min_length": 455.6,
+      "completions/min_terminated_length": 455.6,
+      "epoch": 9.122422915526363e-05,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.06800810992717743,
+      "learning_rate": 2e-05,
+      "loss": -0.0026,
+      "num_tokens": 48111.0,
+      "reward": 2.393585443496704,
+      "reward_std": 0.14288972318172455,
+      "rewards/reward_humanlike/mean": -0.003280111402273178,
+      "rewards/reward_humanlike/std": 0.002425189537461847,
+      "rewards/reward_nonslop/mean": 0.45034117698669435,
+      "rewards/reward_nonslop/std": 0.021841670200228692,
+      "rewards/reward_paragraph_structure/mean": 0.9998971700668335,
+      "rewards/reward_paragraph_structure/std": 0.00036894355434924364,
+      "rewards/reward_readability_flesch/mean": 0.5846633613109589,
+      "rewards/reward_readability_flesch/std": 0.11768133006989956,
+      "rewards/reward_variance_humanlike/mean": 0.36196385622024535,
+      "rewards/reward_variance_humanlike/std": 0.036638665944337845,
+      "step": 5
+    },
+    {
+      "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": 646.0,
+      "completions/max_terminated_length": 646.0,
+      "completions/mean_length": 551.5,
+      "completions/mean_terminated_length": 551.5,
+      "completions/min_length": 446.2,
+      "completions/min_terminated_length": 446.2,
+      "epoch": 0.00018244845831052726,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07684256136417389,
+      "learning_rate": 2e-05,
+      "loss": 0.0074,
+      "num_tokens": 97303.0,
+      "reward": 2.391454339027405,
+      "reward_std": 0.1808932200074196,
+      "rewards/reward_humanlike/mean": -0.0016780316829681397,
+      "rewards/reward_humanlike/std": 0.014642862777691334,
+      "rewards/reward_nonslop/mean": 0.44268596172332764,
+      "rewards/reward_nonslop/std": 0.028079804033041,
+      "rewards/reward_paragraph_structure/mean": 1.0,
+      "rewards/reward_paragraph_structure/std": 0.0,
+      "rewards/reward_readability_flesch/mean": 0.57281613945961,
+      "rewards/reward_readability_flesch/std": 0.16779391318559647,
+      "rewards/reward_variance_humanlike/mean": 0.3776302635669708,
+      "rewards/reward_variance_humanlike/std": 0.045258138328790665,
+      "step": 10
+    },
+    {
+      "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": 598.0,
+      "completions/max_terminated_length": 598.0,
+      "completions/mean_length": 516.1125,
+      "completions/mean_terminated_length": 516.1125,
+      "completions/min_length": 431.6,
+      "completions/min_terminated_length": 431.6,
+      "epoch": 0.0002736726874657909,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07912639528512955,
+      "learning_rate": 2e-05,
+      "loss": 0.0026,
+      "num_tokens": 143712.0,
+      "reward": 2.121282529830933,
+      "reward_std": 0.19594088345766067,
+      "rewards/reward_humanlike/mean": -0.005437415838241577,
+      "rewards/reward_humanlike/std": 0.0020546794403344393,
+      "rewards/reward_nonslop/mean": 0.44375383853912354,
+      "rewards/reward_nonslop/std": 0.025456106290221214,
+      "rewards/reward_paragraph_structure/mean": 0.9997280359268188,
+      "rewards/reward_paragraph_structure/std": 0.001087829563766718,
+      "rewards/reward_readability_flesch/mean": 0.340261435508728,
+      "rewards/reward_readability_flesch/std": 0.17545561231672763,
+      "rewards/reward_variance_humanlike/mean": 0.3429765820503235,
+      "rewards/reward_variance_humanlike/std": 0.04073530249297619,
+      "step": 15
+    },
+    {
+      "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": 632.4,
+      "completions/max_terminated_length": 632.4,
+      "completions/mean_length": 511.4,
+      "completions/mean_terminated_length": 511.4,
+      "completions/min_length": 387.0,
+      "completions/min_terminated_length": 387.0,
+      "epoch": 0.00036489691662105453,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08999043703079224,
+      "learning_rate": 2e-05,
+      "loss": 0.0183,
+      "num_tokens": 189696.0,
+      "reward": 2.6498445510864257,
+      "reward_std": 0.2029375523328781,
+      "rewards/reward_humanlike/mean": 0.00019632279872894287,
+      "rewards/reward_humanlike/std": 0.006884497962892056,
+      "rewards/reward_nonslop/mean": 0.4884443640708923,
+      "rewards/reward_nonslop/std": 0.0355165209621191,
+      "rewards/reward_paragraph_structure/mean": 0.9939258456230163,
+      "rewards/reward_paragraph_structure/std": 0.023617163486778735,
+      "rewards/reward_readability_flesch/mean": 0.8020971655845642,
+      "rewards/reward_readability_flesch/std": 0.17971589416265488,
+      "rewards/reward_variance_humanlike/mean": 0.3651808500289917,
+      "rewards/reward_variance_humanlike/std": 0.042153649032115936,
+      "step": 20
+    },
+    {
+      "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": 587.0,
+      "completions/max_terminated_length": 587.0,
+      "completions/mean_length": 457.2375,
+      "completions/mean_terminated_length": 457.2375,
+      "completions/min_length": 343.8,
+      "completions/min_terminated_length": 343.8,
+      "epoch": 0.0004561211457763182,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09677507728338242,
+      "learning_rate": 2e-05,
+      "loss": -0.0021,
+      "num_tokens": 231427.0,
+      "reward": 2.66627516746521,
+      "reward_std": 0.18921945840120316,
+      "rewards/reward_humanlike/mean": 0.020775890350341795,
+      "rewards/reward_humanlike/std": 0.04260522716213018,
+      "rewards/reward_nonslop/mean": 0.4985057771205902,
+      "rewards/reward_nonslop/std": 0.03064997084438801,
+      "rewards/reward_paragraph_structure/mean": 0.977533757686615,
+      "rewards/reward_paragraph_structure/std": 0.059077487885952,
+      "rewards/reward_readability_flesch/mean": 0.7888182640075684,
+      "rewards/reward_readability_flesch/std": 0.1475653052330017,
+      "rewards/reward_variance_humanlike/mean": 0.380641496181488,
+      "rewards/reward_variance_humanlike/std": 0.04412947595119476,
+      "step": 25
+    },
+    {
+      "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": 597.6,
+      "completions/max_terminated_length": 597.6,
+      "completions/mean_length": 497.3375,
+      "completions/mean_terminated_length": 497.3375,
+      "completions/min_length": 395.4,
+      "completions/min_terminated_length": 395.4,
+      "epoch": 0.0005473453749315818,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08108917623758316,
+      "learning_rate": 2e-05,
+      "loss": 0.0017,
+      "num_tokens": 276318.0,
+      "reward": 2.7883635997772216,
+      "reward_std": 0.17138139009475709,
+      "rewards/reward_humanlike/mean": 0.11761673018336297,
+      "rewards/reward_humanlike/std": 0.08830970991402864,
+      "rewards/reward_nonslop/mean": 0.5173340082168579,
+      "rewards/reward_nonslop/std": 0.029454480484128,
+      "rewards/reward_paragraph_structure/mean": 0.9722555875778198,
+      "rewards/reward_paragraph_structure/std": 0.04261327683925629,
+      "rewards/reward_readability_flesch/mean": 0.7764949321746826,
+      "rewards/reward_readability_flesch/std": 0.10386961698532104,
+      "rewards/reward_variance_humanlike/mean": 0.4046623885631561,
+      "rewards/reward_variance_humanlike/std": 0.045019736886024474,
+      "step": 30
+    },
+    {
+      "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": 617.4,
+      "completions/max_terminated_length": 617.4,
+      "completions/mean_length": 523.825,
+      "completions/mean_terminated_length": 523.825,
+      "completions/min_length": 409.6,
+      "completions/min_terminated_length": 409.6,
+      "epoch": 0.0006385696040868455,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08894547820091248,
+      "learning_rate": 2e-05,
+      "loss": -0.0081,
+      "num_tokens": 323376.0,
+      "reward": 2.7335947513580323,
+      "reward_std": 0.12913600355386734,
+      "rewards/reward_humanlike/mean": 0.044519555568695066,
+      "rewards/reward_humanlike/std": 0.051969482377171514,
+      "rewards/reward_nonslop/mean": 0.5143645763397217,
+      "rewards/reward_nonslop/std": 0.024226398020982743,
+      "rewards/reward_paragraph_structure/mean": 0.9907094240188599,
+      "rewards/reward_paragraph_structure/std": 0.023819129168987273,
+      "rewards/reward_readability_flesch/mean": 0.7974974989891053,
+      "rewards/reward_readability_flesch/std": 0.12351922243833542,
+      "rewards/reward_variance_humanlike/mean": 0.38650373220443723,
+      "rewards/reward_variance_humanlike/std": 0.05399425402283668,
+      "step": 35
+    },
+    {
+      "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": 846.8,
+      "completions/max_terminated_length": 846.8,
+      "completions/mean_length": 555.05,
+      "completions/mean_terminated_length": 555.05,
+      "completions/min_length": 432.2,
+      "completions/min_terminated_length": 432.2,
+      "epoch": 0.0007297938332421091,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.089969202876091,
+      "learning_rate": 2e-05,
+      "loss": 0.0135,
+      "num_tokens": 372868.0,
+      "reward": 2.6368720293045045,
+      "reward_std": 0.17332236915826799,
+      "rewards/reward_humanlike/mean": 0.05930730104446411,
+      "rewards/reward_humanlike/std": 0.04687353018671274,
+      "rewards/reward_nonslop/mean": 0.4885416626930237,
+      "rewards/reward_nonslop/std": 0.03745536170899868,
+      "rewards/reward_paragraph_structure/mean": 0.890105378627777,
+      "rewards/reward_paragraph_structure/std": 0.09538187831640244,
+      "rewards/reward_readability_flesch/mean": 0.7819342494010926,
+      "rewards/reward_readability_flesch/std": 0.11071507334709167,
+      "rewards/reward_variance_humanlike/mean": 0.4169835090637207,
+      "rewards/reward_variance_humanlike/std": 0.0439291026443243,
+      "step": 40
+    },
+    {
+      "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": 613.2,
+      "completions/max_terminated_length": 613.2,
+      "completions/mean_length": 505.075,
+      "completions/mean_terminated_length": 505.075,
+      "completions/min_length": 374.0,
+      "completions/min_terminated_length": 374.0,
+      "epoch": 0.0008210180623973727,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0890246331691742,
+      "learning_rate": 2e-05,
+      "loss": 0.0102,
+      "num_tokens": 418490.0,
+      "reward": 2.6527084350585937,
+      "reward_std": 0.16761653274297714,
+      "rewards/reward_humanlike/mean": 0.08072967529296875,
+      "rewards/reward_humanlike/std": 0.06607948099263013,
+      "rewards/reward_nonslop/mean": 0.5106609463691711,
+      "rewards/reward_nonslop/std": 0.0351794958114624,
+      "rewards/reward_paragraph_structure/mean": 0.9813281297683716,
+      "rewards/reward_paragraph_structure/std": 0.05892824307084084,
+      "rewards/reward_readability_flesch/mean": 0.6913772821426392,
+      "rewards/reward_readability_flesch/std": 0.13855507373809814,
+      "rewards/reward_variance_humanlike/mean": 0.38861238956451416,
+      "rewards/reward_variance_humanlike/std": 0.039696966856718065,
+      "step": 45
+    },
+    {
+      "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": 607.2,
+      "completions/max_terminated_length": 607.2,
+      "completions/mean_length": 500.2125,
+      "completions/mean_terminated_length": 500.2125,
+      "completions/min_length": 407.0,
+      "completions/min_terminated_length": 407.0,
+      "epoch": 0.0009122422915526364,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.07912778109312057,
+      "learning_rate": 2e-05,
+      "loss": 0.005,
+      "num_tokens": 463659.0,
+      "reward": 2.753040885925293,
+      "reward_std": 0.136872561275959,
+      "rewards/reward_humanlike/mean": 0.03859691172838211,
+      "rewards/reward_humanlike/std": 0.035087807662785055,
+      "rewards/reward_nonslop/mean": 0.5123575925827026,
+      "rewards/reward_nonslop/std": 0.02690323181450367,
+      "rewards/reward_paragraph_structure/mean": 0.992300295829773,
+      "rewards/reward_paragraph_structure/std": 0.030798836052417754,
+      "rewards/reward_readability_flesch/mean": 0.8399829745292664,
+      "rewards/reward_readability_flesch/std": 0.11433027684688568,
+      "rewards/reward_variance_humanlike/mean": 0.3698031187057495,
+      "rewards/reward_variance_humanlike/std": 0.04352555349469185,
+      "step": 50
+    },
+    {
+      "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": 562.0,
+      "completions/max_terminated_length": 562.0,
+      "completions/mean_length": 444.2875,
+      "completions/mean_terminated_length": 444.2875,
+      "completions/min_length": 313.0,
+      "completions/min_terminated_length": 313.0,
+      "epoch": 0.0010034665207079,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0981728583574295,
+      "learning_rate": 2e-05,
+      "loss": -0.0035,
+      "num_tokens": 504402.0,
+      "reward": 2.7707664489746096,
+      "reward_std": 0.15653531849384308,
+      "rewards/reward_humanlike/mean": 0.06527388915419578,
+      "rewards/reward_humanlike/std": 0.038607833255082366,
+      "rewards/reward_nonslop/mean": 0.5381535172462464,
+      "rewards/reward_nonslop/std": 0.030977290123701096,
+      "rewards/reward_paragraph_structure/mean": 0.9606678128242493,
+      "rewards/reward_paragraph_structure/std": 0.11089053153991699,
+      "rewards/reward_readability_flesch/mean": 0.7880517601966858,
+      "rewards/reward_readability_flesch/std": 0.14046915918588637,
+      "rewards/reward_variance_humanlike/mean": 0.4186194181442261,
+      "rewards/reward_variance_humanlike/std": 0.0556840606033802,
+      "step": 55
+    },
+    {
+      "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": 576.2,
+      "completions/max_terminated_length": 576.2,
+      "completions/mean_length": 473.1625,
+      "completions/mean_terminated_length": 473.1625,
+      "completions/min_length": 373.8,
+      "completions/min_terminated_length": 373.8,
+      "epoch": 0.0010946907498631637,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0779431164264679,
+      "learning_rate": 2e-05,
+      "loss": 0.0033,
+      "num_tokens": 547375.0,
+      "reward": 2.721493101119995,
+      "reward_std": 0.16468897759914397,
+      "rewards/reward_humanlike/mean": 0.06399692744016647,
+      "rewards/reward_humanlike/std": 0.06561810113489627,
+      "rewards/reward_nonslop/mean": 0.526066243648529,
+      "rewards/reward_nonslop/std": 0.03075612857937813,
+      "rewards/reward_paragraph_structure/mean": 0.9890687108039856,
+      "rewards/reward_paragraph_structure/std": 0.043725161254405974,
+      "rewards/reward_readability_flesch/mean": 0.7518861651420593,
+      "rewards/reward_readability_flesch/std": 0.13989782482385635,
+      "rewards/reward_variance_humanlike/mean": 0.39047507643699647,
+      "rewards/reward_variance_humanlike/std": 0.04762849882245064,
+      "step": 60
+    },
+    {
+      "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": 578.8,
+      "completions/max_terminated_length": 578.8,
+      "completions/mean_length": 483.25,
+      "completions/mean_terminated_length": 483.25,
+      "completions/min_length": 392.0,
+      "completions/min_terminated_length": 392.0,
+      "epoch": 0.0011859149790184272,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08613153547048569,
+      "learning_rate": 2e-05,
+      "loss": -0.0079,
+      "num_tokens": 591107.0,
+      "reward": 2.756886434555054,
+      "reward_std": 0.15308991372585296,
+      "rewards/reward_humanlike/mean": 0.03657452017068863,
+      "rewards/reward_humanlike/std": 0.030589387845247985,
+      "rewards/reward_nonslop/mean": 0.5334658443927764,
+      "rewards/reward_nonslop/std": 0.02915770635008812,
+      "rewards/reward_paragraph_structure/mean": 0.9779379844665528,
+      "rewards/reward_paragraph_structure/std": 0.07469315528869629,
+      "rewards/reward_readability_flesch/mean": 0.8177438259124756,
+      "rewards/reward_readability_flesch/std": 0.12392935305833816,
+      "rewards/reward_variance_humanlike/mean": 0.3911642551422119,
+      "rewards/reward_variance_humanlike/std": 0.04321202002465725,
+      "step": 65
+    },
+    {
+      "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": 607.0,
+      "completions/max_terminated_length": 607.0,
+      "completions/mean_length": 494.2,
+      "completions/mean_terminated_length": 494.2,
+      "completions/min_length": 396.0,
+      "completions/min_terminated_length": 396.0,
+      "epoch": 0.001277139208173691,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08903255313634872,
+      "learning_rate": 2e-05,
+      "loss": 0.0097,
+      "num_tokens": 635795.0,
+      "reward": 2.869403123855591,
+      "reward_std": 0.12632807344198227,
+      "rewards/reward_humanlike/mean": 0.04444831982254982,
+      "rewards/reward_humanlike/std": 0.036120910011231896,
+      "rewards/reward_nonslop/mean": 0.5243099689483642,
+      "rewards/reward_nonslop/std": 0.027720437198877335,
+      "rewards/reward_paragraph_structure/mean": 0.9873847007751465,
+      "rewards/reward_paragraph_structure/std": 0.04481239095330238,
+      "rewards/reward_readability_flesch/mean": 0.9202077388763428,
+      "rewards/reward_readability_flesch/std": 0.09855079799890518,
+      "rewards/reward_variance_humanlike/mean": 0.39305235743522643,
+      "rewards/reward_variance_humanlike/std": 0.04738411866128445,
+      "step": 70
+    },
+    {
+      "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": 554.2,
+      "completions/max_terminated_length": 554.2,
+      "completions/mean_length": 441.2,
+      "completions/mean_terminated_length": 441.2,
+      "completions/min_length": 343.0,
+      "completions/min_terminated_length": 343.0,
+      "epoch": 0.0013683634373289546,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.08839516341686249,
+      "learning_rate": 2e-05,
+      "loss": 0.006,
+      "num_tokens": 676227.0,
+      "reward": 2.7607101917266847,
+      "reward_std": 0.14567108005285262,
+      "rewards/reward_humanlike/mean": 0.0049436971545219425,
+      "rewards/reward_humanlike/std": 0.009033319354057313,
+      "rewards/reward_nonslop/mean": 0.5163205862045288,
+      "rewards/reward_nonslop/std": 0.034804318100214005,
+      "rewards/reward_paragraph_structure/mean": 0.9944571018218994,
+      "rewards/reward_paragraph_structure/std": 0.022171604633331298,
+      "rewards/reward_readability_flesch/mean": 0.8619415044784546,
+      "rewards/reward_readability_flesch/std": 0.12174343764781952,
+      "rewards/reward_variance_humanlike/mean": 0.38304726481437684,
+      "rewards/reward_variance_humanlike/std": 0.05802188515663147,
+      "step": 75
+    },
+    {
+      "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": 717.2,
+      "completions/max_terminated_length": 717.2,
+      "completions/mean_length": 521.9,
+      "completions/mean_terminated_length": 521.9,
+      "completions/min_length": 412.2,
+      "completions/min_terminated_length": 412.2,
+      "epoch": 0.0014595876664842181,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.0736362412571907,
+      "learning_rate": 2e-05,
+      "loss": 0.0207,
+      "num_tokens": 723019.0,
+      "reward": 2.7467071056365966,
+      "reward_std": 0.13217758685350417,
+      "rewards/reward_humanlike/mean": 0.04404948055744171,
+      "rewards/reward_humanlike/std": 0.03210974643006921,
+      "rewards/reward_nonslop/mean": 0.5157673060894012,
+      "rewards/reward_nonslop/std": 0.03307870253920555,
+      "rewards/reward_paragraph_structure/mean": 0.9820503711700439,
+      "rewards/reward_paragraph_structure/std": 0.06460869461297988,
+      "rewards/reward_readability_flesch/mean": 0.8017061710357666,
+      "rewards/reward_readability_flesch/std": 0.09596328362822533,
+      "rewards/reward_variance_humanlike/mean": 0.4031337320804596,
+      "rewards/reward_variance_humanlike/std": 0.038825053721666336,
+      "step": 80
+    },
+    {
+      "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": 568.2,
+      "completions/max_terminated_length": 568.2,
+      "completions/mean_length": 492.875,
+      "completions/mean_terminated_length": 492.875,
+      "completions/min_length": 405.2,
+      "completions/min_terminated_length": 405.2,
+      "epoch": 0.0015508118956394819,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09337656199932098,
+      "learning_rate": 2e-05,
+      "loss": 0.0205,
+      "num_tokens": 767585.0,
+      "reward": 2.8245923042297365,
+      "reward_std": 0.12125235199928283,
+      "rewards/reward_humanlike/mean": 0.04374306052923203,
+      "rewards/reward_humanlike/std": 0.05065076029859483,
+      "rewards/reward_nonslop/mean": 0.5121960043907166,
+      "rewards/reward_nonslop/std": 0.024268346652388573,
+      "rewards/reward_paragraph_structure/mean": 0.9952531695365906,
+      "rewards/reward_paragraph_structure/std": 0.01898730993270874,
+      "rewards/reward_readability_flesch/mean": 0.8540295481681823,
+      "rewards/reward_readability_flesch/std": 0.11251321136951446,
+      "rewards/reward_variance_humanlike/mean": 0.4193705081939697,
+      "rewards/reward_variance_humanlike/std": 0.0530940480530262,
+      "step": 85
+    },
+    {
+      "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": 529.2,
+      "completions/max_terminated_length": 529.2,
+      "completions/mean_length": 429.45,
+      "completions/mean_terminated_length": 429.45,
+      "completions/min_length": 340.6,
+      "completions/min_terminated_length": 340.6,
+      "epoch": 0.0016420361247947454,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10793354362249374,
+      "learning_rate": 2e-05,
+      "loss": 0.0084,
+      "num_tokens": 807109.0,
+      "reward": 2.8659895420074464,
+      "reward_std": 0.22321247607469558,
+      "rewards/reward_humanlike/mean": 0.11830640286207199,
+      "rewards/reward_humanlike/std": 0.10621838448569179,
+      "rewards/reward_nonslop/mean": 0.5382362484931946,
+      "rewards/reward_nonslop/std": 0.02679550349712372,
+      "rewards/reward_paragraph_structure/mean": 0.9388323545455932,
+      "rewards/reward_paragraph_structure/std": 0.08712138235569,
+      "rewards/reward_readability_flesch/mean": 0.8695886135101318,
+      "rewards/reward_readability_flesch/std": 0.11793182790279388,
+      "rewards/reward_variance_humanlike/mean": 0.4010259807109833,
+      "rewards/reward_variance_humanlike/std": 0.05023952871561051,
+      "step": 90
+    },
+    {
+      "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": 516.2,
+      "completions/max_terminated_length": 516.2,
+      "completions/mean_length": 427.1625,
+      "completions/mean_terminated_length": 427.1625,
+      "completions/min_length": 327.4,
+      "completions/min_terminated_length": 327.4,
+      "epoch": 0.0017332603539500092,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10153933614492416,
+      "learning_rate": 2e-05,
+      "loss": 0.0076,
+      "num_tokens": 846354.0,
+      "reward": 2.9194432735443114,
+      "reward_std": 0.1527949333190918,
+      "rewards/reward_humanlike/mean": 0.13011178895831108,
+      "rewards/reward_humanlike/std": 0.1040344807319343,
+      "rewards/reward_nonslop/mean": 0.539959442615509,
+      "rewards/reward_nonslop/std": 0.027887169644236563,
+      "rewards/reward_paragraph_structure/mean": 0.97805335521698,
+      "rewards/reward_paragraph_structure/std": 0.0572780579328537,
+      "rewards/reward_readability_flesch/mean": 0.8915346980094909,
+      "rewards/reward_readability_flesch/std": 0.10210664793848992,
+      "rewards/reward_variance_humanlike/mean": 0.37978398203849795,
+      "rewards/reward_variance_humanlike/std": 0.04532681703567505,
+      "step": 95
+    },
+    {
+      "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": 507.2,
+      "completions/max_terminated_length": 507.2,
+      "completions/mean_length": 416.6125,
+      "completions/mean_terminated_length": 416.6125,
+      "completions/min_length": 341.0,
+      "completions/min_terminated_length": 341.0,
+      "epoch": 0.0018244845831052727,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11056620627641678,
+      "learning_rate": 2e-05,
+      "loss": -0.0238,
+      "num_tokens": 884675.0,
+      "reward": 2.8297982215881348,
+      "reward_std": 0.1837340384721756,
+      "rewards/reward_humanlike/mean": 0.029251274466514588,
+      "rewards/reward_humanlike/std": 0.030034957686439157,
+      "rewards/reward_nonslop/mean": 0.5330106735229492,
+      "rewards/reward_nonslop/std": 0.02519891820847988,
+      "rewards/reward_paragraph_structure/mean": 0.9299438714981079,
+      "rewards/reward_paragraph_structure/std": 0.11956685781478882,
+      "rewards/reward_readability_flesch/mean": 0.9390206336975098,
+      "rewards/reward_readability_flesch/std": 0.1079161748290062,
+      "rewards/reward_variance_humanlike/mean": 0.3985717177391052,
+      "rewards/reward_variance_humanlike/std": 0.046227993071079256,
+      "step": 100
+    },
+    {
+      "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": 484.8,
+      "completions/max_terminated_length": 484.8,
+      "completions/mean_length": 397.0,
+      "completions/mean_terminated_length": 397.0,
+      "completions/min_length": 311.6,
+      "completions/min_terminated_length": 311.6,
+      "epoch": 0.0019157088122605363,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12234635651111603,
+      "learning_rate": 2e-05,
+      "loss": -0.0084,
+      "num_tokens": 921523.0,
+      "reward": 2.7906761169433594,
+      "reward_std": 0.18639333844184874,
+      "rewards/reward_humanlike/mean": 0.03967294096946716,
+      "rewards/reward_humanlike/std": 0.047859178856015205,
+      "rewards/reward_nonslop/mean": 0.5455231070518494,
+      "rewards/reward_nonslop/std": 0.02572299391031265,
+      "rewards/reward_paragraph_structure/mean": 0.9380416989326477,
+      "rewards/reward_paragraph_structure/std": 0.14332405030727385,
+      "rewards/reward_readability_flesch/mean": 0.8651458501815796,
+      "rewards/reward_readability_flesch/std": 0.10486775860190392,
+      "rewards/reward_variance_humanlike/mean": 0.40229247212409974,
+      "rewards/reward_variance_humanlike/std": 0.05056553930044174,
+      "step": 105
+    },
+    {
+      "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": 535.0,
+      "completions/max_terminated_length": 535.0,
+      "completions/mean_length": 407.4,
+      "completions/mean_terminated_length": 407.4,
+      "completions/min_length": 323.6,
+      "completions/min_terminated_length": 323.6,
+      "epoch": 0.0020069330414158,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12049026042222977,
+      "learning_rate": 2e-05,
+      "loss": -0.0015,
+      "num_tokens": 959235.0,
+      "reward": 2.8142509937286375,
+      "reward_std": 0.20959972888231276,
+      "rewards/reward_humanlike/mean": 0.10315801277756691,
+      "rewards/reward_humanlike/std": 0.06646046359091998,
+      "rewards/reward_nonslop/mean": 0.549621844291687,
+      "rewards/reward_nonslop/std": 0.02784626930952072,
+      "rewards/reward_paragraph_structure/mean": 0.9427387475967407,
+      "rewards/reward_paragraph_structure/std": 0.0946300521492958,
+      "rewards/reward_readability_flesch/mean": 0.8157400369644165,
+      "rewards/reward_readability_flesch/std": 0.12322266846895218,
+      "rewards/reward_variance_humanlike/mean": 0.40299235582351683,
+      "rewards/reward_variance_humanlike/std": 0.03882267326116562,
+      "step": 110
+    },
+    {
+      "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": 546.2,
+      "completions/max_terminated_length": 546.2,
+      "completions/mean_length": 453.1875,
+      "completions/mean_terminated_length": 453.1875,
+      "completions/min_length": 365.0,
+      "completions/min_terminated_length": 365.0,
+      "epoch": 0.0020981572705710636,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10685638338327408,
+      "learning_rate": 2e-05,
+      "loss": -0.0023,
+      "num_tokens": 1000562.0,
+      "reward": 2.7050733089447023,
+      "reward_std": 0.17174948751926422,
+      "rewards/reward_humanlike/mean": 0.03010150045156479,
+      "rewards/reward_humanlike/std": 0.02183089666068554,
+      "rewards/reward_nonslop/mean": 0.537644910812378,
+      "rewards/reward_nonslop/std": 0.022509072348475455,
+      "rewards/reward_paragraph_structure/mean": 0.9359247803688049,
+      "rewards/reward_paragraph_structure/std": 0.13663459122180938,
+      "rewards/reward_readability_flesch/mean": 0.7878659129142761,
+      "rewards/reward_readability_flesch/std": 0.1104049876332283,
+      "rewards/reward_variance_humanlike/mean": 0.4135362088680267,
+      "rewards/reward_variance_humanlike/std": 0.04251410141587257,
+      "step": 115
+    },
+    {
+      "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": 563.4,
+      "completions/max_terminated_length": 563.4,
+      "completions/mean_length": 454.7875,
+      "completions/mean_terminated_length": 454.7875,
+      "completions/min_length": 365.0,
+      "completions/min_terminated_length": 365.0,
+      "epoch": 0.0021893814997263274,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10802900046110153,
+      "learning_rate": 2e-05,
+      "loss": 0.0097,
+      "num_tokens": 1042177.0,
+      "reward": 2.8397992134094237,
+      "reward_std": 0.15860502868890763,
+      "rewards/reward_humanlike/mean": 0.04295992851257324,
+      "rewards/reward_humanlike/std": 0.05235542356967926,
+      "rewards/reward_nonslop/mean": 0.5383237481117249,
+      "rewards/reward_nonslop/std": 0.029689770936965943,
+      "rewards/reward_paragraph_structure/mean": 1.0,
+      "rewards/reward_paragraph_structure/std": 0.0,
+      "rewards/reward_readability_flesch/mean": 0.8279854476451873,
+      "rewards/reward_readability_flesch/std": 0.11809881925582885,
+      "rewards/reward_variance_humanlike/mean": 0.43053008913993834,
+      "rewards/reward_variance_humanlike/std": 0.04717860147356987,
+      "step": 120
+    },
+    {
+      "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": 586.2,
+      "completions/max_terminated_length": 586.2,
+      "completions/mean_length": 476.3125,
+      "completions/mean_terminated_length": 476.3125,
+      "completions/min_length": 379.4,
+      "completions/min_terminated_length": 379.4,
+      "epoch": 0.002280605728881591,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11722852289676666,
+      "learning_rate": 2e-05,
+      "loss": 0.0084,
+      "num_tokens": 1085338.0,
+      "reward": 2.8460022449493407,
+      "reward_std": 0.12902745008468627,
+      "rewards/reward_humanlike/mean": 0.08100048899650573,
+      "rewards/reward_humanlike/std": 0.0518460288643837,
+      "rewards/reward_nonslop/mean": 0.5345206022262573,
+      "rewards/reward_nonslop/std": 0.028127148374915124,
+      "rewards/reward_paragraph_structure/mean": 0.9582128286361694,
+      "rewards/reward_paragraph_structure/std": 0.052062156796455386,
+      "rewards/reward_readability_flesch/mean": 0.8341620564460754,
+      "rewards/reward_readability_flesch/std": 0.11833389699459076,
+      "rewards/reward_variance_humanlike/mean": 0.43810625076293946,
+      "rewards/reward_variance_humanlike/std": 0.06058142632246018,
+      "step": 125
+    },
+    {
+      "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": 637.2,
+      "completions/max_terminated_length": 637.2,
+      "completions/mean_length": 525.2625,
+      "completions/mean_terminated_length": 525.2625,
+      "completions/min_length": 415.6,
+      "completions/min_terminated_length": 415.6,
+      "epoch": 0.0023718299580368545,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.1034693568944931,
+      "learning_rate": 2e-05,
+      "loss": -0.0015,
+      "num_tokens": 1132447.0,
+      "reward": 2.9677632331848143,
+      "reward_std": 0.13178672641515732,
+      "rewards/reward_humanlike/mean": 0.11690731346607208,
+      "rewards/reward_humanlike/std": 0.07614295929670334,
+      "rewards/reward_nonslop/mean": 0.5198043942451477,
+      "rewards/reward_nonslop/std": 0.027982910722494127,
+      "rewards/reward_paragraph_structure/mean": 0.9756827473640441,
+      "rewards/reward_paragraph_structure/std": 0.044346729014068845,
+      "rewards/reward_readability_flesch/mean": 0.8949468612670899,
+      "rewards/reward_readability_flesch/std": 0.08357911184430122,
+      "rewards/reward_variance_humanlike/mean": 0.4604219257831573,
+      "rewards/reward_variance_humanlike/std": 0.05337681397795677,
+      "step": 130
+    },
+    {
+      "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": 588.0,
+      "completions/max_terminated_length": 588.0,
+      "completions/mean_length": 498.6875,
+      "completions/mean_terminated_length": 498.6875,
+      "completions/min_length": 399.0,
+      "completions/min_terminated_length": 399.0,
+      "epoch": 0.0024630541871921183,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.1009395495057106,
+      "learning_rate": 2e-05,
+      "loss": -0.0043,
+      "num_tokens": 1177382.0,
+      "reward": 2.822851276397705,
+      "reward_std": 0.13398206382989883,
+      "rewards/reward_humanlike/mean": 0.06381046175956726,
+      "rewards/reward_humanlike/std": 0.0503330685198307,
+      "rewards/reward_nonslop/mean": 0.5348796844482422,
+      "rewards/reward_nonslop/std": 0.025143631175160407,
+      "rewards/reward_paragraph_structure/mean": 0.9924662351608277,
+      "rewards/reward_paragraph_structure/std": 0.03013508329167962,
+      "rewards/reward_readability_flesch/mean": 0.7840398788452149,
+      "rewards/reward_readability_flesch/std": 0.11849776804447174,
+      "rewards/reward_variance_humanlike/mean": 0.44765504002571105,
+      "rewards/reward_variance_humanlike/std": 0.045124633610248564,
+      "step": 135
+    },
+    {
+      "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": 667.8,
+      "completions/max_terminated_length": 667.8,
+      "completions/mean_length": 552.2125,
+      "completions/mean_terminated_length": 552.2125,
+      "completions/min_length": 431.8,
+      "completions/min_terminated_length": 431.8,
+      "epoch": 0.002554278416347382,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.10111426562070847,
+      "learning_rate": 2e-05,
+      "loss": -0.017,
+      "num_tokens": 1226679.0,
+      "reward": 2.9173266887664795,
+      "reward_std": 0.1144576370716095,
+      "rewards/reward_humanlike/mean": 0.11132953837513923,
+      "rewards/reward_humanlike/std": 0.05890989601612091,
+      "rewards/reward_nonslop/mean": 0.4942618072032928,
+      "rewards/reward_nonslop/std": 0.02765493802726269,
+      "rewards/reward_paragraph_structure/mean": 0.9944768905639648,
+      "rewards/reward_paragraph_structure/std": 0.02089656200259924,
+      "rewards/reward_readability_flesch/mean": 0.8778629779815674,
+      "rewards/reward_readability_flesch/std": 0.08600130677223206,
+      "rewards/reward_variance_humanlike/mean": 0.4393955051898956,
+      "rewards/reward_variance_humanlike/std": 0.04750688448548317,
+      "step": 140
+    },
+    {
+      "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": 640.6,
+      "completions/max_terminated_length": 640.6,
+      "completions/mean_length": 534.1125,
+      "completions/mean_terminated_length": 534.1125,
+      "completions/min_length": 438.2,
+      "completions/min_terminated_length": 438.2,
+      "epoch": 0.0026455026455026454,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11239214986562729,
+      "learning_rate": 2e-05,
+      "loss": 0.0035,
+      "num_tokens": 1274592.0,
+      "reward": 3.053114986419678,
+      "reward_std": 0.11381979882717133,
+      "rewards/reward_humanlike/mean": 0.12257216051220894,
+      "rewards/reward_humanlike/std": 0.08092293813824654,
+      "rewards/reward_nonslop/mean": 0.5075180411338807,
+      "rewards/reward_nonslop/std": 0.026819777116179465,
+      "rewards/reward_paragraph_structure/mean": 0.9958648920059204,
+      "rewards/reward_paragraph_structure/std": 0.015135037153959275,
+      "rewards/reward_readability_flesch/mean": 0.9384488105773926,
+      "rewards/reward_readability_flesch/std": 0.07422738522291183,
+      "rewards/reward_variance_humanlike/mean": 0.4887110412120819,
+      "rewards/reward_variance_humanlike/std": 0.044287893921136856,
+      "step": 145
+    },
+    {
+      "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": 625.8,
+      "completions/max_terminated_length": 625.8,
+      "completions/mean_length": 526.2,
+      "completions/mean_terminated_length": 526.2,
+      "completions/min_length": 432.6,
+      "completions/min_terminated_length": 432.6,
+      "epoch": 0.002736726874657909,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09958786517381668,
+      "learning_rate": 2e-05,
+      "loss": 0.0085,
+      "num_tokens": 1321840.0,
+      "reward": 2.912756013870239,
+      "reward_std": 0.12361426800489425,
+      "rewards/reward_humanlike/mean": 0.05477507412433624,
+      "rewards/reward_humanlike/std": 0.04108944982290268,
+      "rewards/reward_nonslop/mean": 0.5123514175415039,
+      "rewards/reward_nonslop/std": 0.027717596665024758,
+      "rewards/reward_paragraph_structure/mean": 0.9881778597831726,
+      "rewards/reward_paragraph_structure/std": 0.04215195477008819,
+      "rewards/reward_readability_flesch/mean": 0.8884697198867798,
+      "rewards/reward_readability_flesch/std": 0.09501284211874009,
+      "rewards/reward_variance_humanlike/mean": 0.4689819157123566,
+      "rewards/reward_variance_humanlike/std": 0.04468975365161896,
+      "step": 150
+    },
+    {
+      "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": 709.8,
+      "completions/max_terminated_length": 709.8,
+      "completions/mean_length": 540.75,
+      "completions/mean_terminated_length": 540.75,
+      "completions/min_length": 391.2,
+      "completions/min_terminated_length": 391.2,
+      "epoch": 0.002827951103813173,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.09662537276744843,
+      "learning_rate": 2e-05,
+      "loss": 0.0434,
+      "num_tokens": 1370236.0,
+      "reward": 2.9029190063476564,
+      "reward_std": 0.21406158804893494,
+      "rewards/reward_humanlike/mean": 0.16697487533092498,
+      "rewards/reward_humanlike/std": 0.10503775626420975,
+      "rewards/reward_nonslop/mean": 0.516099750995636,
+      "rewards/reward_nonslop/std": 0.0328385628759861,
+      "rewards/reward_paragraph_structure/mean": 0.9108669638633728,
+      "rewards/reward_paragraph_structure/std": 0.046271538734436034,
+      "rewards/reward_readability_flesch/mean": 0.8546391785144806,
+      "rewards/reward_readability_flesch/std": 0.08806488886475564,
+      "rewards/reward_variance_humanlike/mean": 0.45433822870254514,
+      "rewards/reward_variance_humanlike/std": 0.056633827090263364,
+      "step": 155
+    },
+    {
+      "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": 593.2,
+      "completions/max_terminated_length": 593.2,
+      "completions/mean_length": 503.625,
+      "completions/mean_terminated_length": 503.625,
+      "completions/min_length": 417.6,
+      "completions/min_terminated_length": 417.6,
+      "epoch": 0.0029191753329684362,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.14527224004268646,
+      "learning_rate": 2e-05,
+      "loss": 0.0157,
+      "num_tokens": 1415710.0,
+      "reward": 3.0420565605163574,
+      "reward_std": 0.18344633877277375,
+      "rewards/reward_humanlike/mean": 0.15477735549211502,
+      "rewards/reward_humanlike/std": 0.10218072310090065,
+      "rewards/reward_nonslop/mean": 0.5330479383468628,
+      "rewards/reward_nonslop/std": 0.027725811675190927,
+      "rewards/reward_paragraph_structure/mean": 0.9918029308319092,
+      "rewards/reward_paragraph_structure/std": 0.03271809332072735,
+      "rewards/reward_readability_flesch/mean": 0.8464983224868774,
+      "rewards/reward_readability_flesch/std": 0.11713154464960099,
+      "rewards/reward_variance_humanlike/mean": 0.515929889678955,
+      "rewards/reward_variance_humanlike/std": 0.05413371622562409,
+      "step": 160
+    },
+    {
+      "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": 584.4,
+      "completions/max_terminated_length": 584.4,
+      "completions/mean_length": 489.9875,
+      "completions/mean_terminated_length": 489.9875,
+      "completions/min_length": 375.6,
+      "completions/min_terminated_length": 375.6,
+      "epoch": 0.0030103995621237,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11605068296194077,
+      "learning_rate": 2e-05,
+      "loss": -0.0128,
+      "num_tokens": 1460061.0,
+      "reward": 3.143190383911133,
+      "reward_std": 0.11709334850311279,
+      "rewards/reward_humanlike/mean": 0.16331919580698012,
+      "rewards/reward_humanlike/std": 0.08971775695681572,
+      "rewards/reward_nonslop/mean": 0.5315526962280274,
+      "rewards/reward_nonslop/std": 0.02810417078435421,
+      "rewards/reward_paragraph_structure/mean": 0.9975423097610474,
+      "rewards/reward_paragraph_structure/std": 0.00911269262433052,
+      "rewards/reward_readability_flesch/mean": 0.9527804493904114,
+      "rewards/reward_readability_flesch/std": 0.07519614920020104,
+      "rewards/reward_variance_humanlike/mean": 0.4979958117008209,
+      "rewards/reward_variance_humanlike/std": 0.04819941446185112,
+      "step": 165
+    },
+    {
+      "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": 593.0,
+      "completions/max_terminated_length": 593.0,
+      "completions/mean_length": 476.4625,
+      "completions/mean_terminated_length": 476.4625,
+      "completions/min_length": 369.2,
+      "completions/min_terminated_length": 369.2,
+      "epoch": 0.0031016237912789638,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.1157560646533966,
+      "learning_rate": 2e-05,
+      "loss": -0.0103,
+      "num_tokens": 1503314.0,
+      "reward": 3.1130752086639406,
+      "reward_std": 0.1477765291929245,
+      "rewards/reward_humanlike/mean": 0.1538683906197548,
+      "rewards/reward_humanlike/std": 0.0856837049126625,
+      "rewards/reward_nonslop/mean": 0.5342989683151245,
+      "rewards/reward_nonslop/std": 0.027773698419332506,
+      "rewards/reward_paragraph_structure/mean": 0.9942754626274108,
+      "rewards/reward_paragraph_structure/std": 0.022898165881633757,
+      "rewards/reward_readability_flesch/mean": 0.9252384662628174,
+      "rewards/reward_readability_flesch/std": 0.09302579760551452,
+      "rewards/reward_variance_humanlike/mean": 0.5053938448429107,
+      "rewards/reward_variance_humanlike/std": 0.052392928302288054,
+      "step": 170
+    },
+    {
+      "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": 580.0,
+      "completions/max_terminated_length": 580.0,
+      "completions/mean_length": 488.4125,
+      "completions/mean_terminated_length": 488.4125,
+      "completions/min_length": 401.8,
+      "completions/min_terminated_length": 401.8,
+      "epoch": 0.0031928480204342275,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.11335834860801697,
+      "learning_rate": 2e-05,
+      "loss": 0.0026,
+      "num_tokens": 1547459.0,
+      "reward": 3.2409387588500977,
+      "reward_std": 0.15198988765478133,
+      "rewards/reward_humanlike/mean": 0.26733590960502623,
+      "rewards/reward_humanlike/std": 0.1318629041314125,
+      "rewards/reward_nonslop/mean": 0.53455730676651,
+      "rewards/reward_nonslop/std": 0.023813217505812646,
+      "rewards/reward_paragraph_structure/mean": 0.9959634304046631,
+      "rewards/reward_paragraph_structure/std": 0.010735219717025757,
+      "rewards/reward_readability_flesch/mean": 0.9626064658164978,
+      "rewards/reward_readability_flesch/std": 0.05338000953197479,
+      "rewards/reward_variance_humanlike/mean": 0.4804756224155426,
+      "rewards/reward_variance_humanlike/std": 0.05087298080325127,
+      "step": 175
+    },
+    {
+      "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": 509.0,
+      "completions/max_terminated_length": 509.0,
+      "completions/mean_length": 419.4375,
+      "completions/mean_terminated_length": 419.4375,
+      "completions/min_length": 339.6,
+      "completions/min_terminated_length": 339.6,
+      "epoch": 0.003284072249589491,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12220724672079086,
+      "learning_rate": 2e-05,
+      "loss": 0.0048,
+      "num_tokens": 1586054.0,
+      "reward": 3.0313026905059814,
+      "reward_std": 0.19189467430114746,
+      "rewards/reward_humanlike/mean": 0.1291750580072403,
+      "rewards/reward_humanlike/std": 0.08153657093644143,
+      "rewards/reward_nonslop/mean": 0.5568328022956848,
+      "rewards/reward_nonslop/std": 0.02491689659655094,
+      "rewards/reward_paragraph_structure/mean": 0.9697872281074524,
+      "rewards/reward_paragraph_structure/std": 0.07481733858585357,
+      "rewards/reward_readability_flesch/mean": 0.863669466972351,
+      "rewards/reward_readability_flesch/std": 0.11045316830277443,
+      "rewards/reward_variance_humanlike/mean": 0.5118380665779114,
+      "rewards/reward_variance_humanlike/std": 0.0632518894970417,
+      "step": 180
+    },
+    {
+      "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": 573.2,
+      "completions/max_terminated_length": 573.2,
+      "completions/mean_length": 443.35,
+      "completions/mean_terminated_length": 443.35,
+      "completions/min_length": 346.0,
+      "completions/min_terminated_length": 346.0,
+      "epoch": 0.0033752964787447546,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.12768340110778809,
+      "learning_rate": 2e-05,
+      "loss": 0.0094,
+      "num_tokens": 1626690.0,
+      "reward": 3.0511626243591308,
+      "reward_std": 0.2748347967863083,
+      "rewards/reward_humanlike/mean": 0.226414555311203,
+      "rewards/reward_humanlike/std": 0.14178435355424882,
+      "rewards/reward_nonslop/mean": 0.5576748013496399,
+      "rewards/reward_nonslop/std": 0.027306027337908746,
+      "rewards/reward_paragraph_structure/mean": 0.9287306308746338,
+      "rewards/reward_paragraph_structure/std": 0.08972582221031189,
+      "rewards/reward_readability_flesch/mean": 0.8113695025444031,
+      "rewards/reward_readability_flesch/std": 0.16540859043598174,
+      "rewards/reward_variance_humanlike/mean": 0.5269731521606446,
+      "rewards/reward_variance_humanlike/std": 0.05631731376051903,
+      "step": 185
+    },
+    {
+      "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": 464.4,
+      "completions/max_terminated_length": 464.4,
+      "completions/mean_length": 386.95,
+      "completions/mean_terminated_length": 386.95,
+      "completions/min_length": 311.6,
+      "completions/min_terminated_length": 311.6,
+      "epoch": 0.0034665207079000184,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.17250695824623108,
+      "learning_rate": 2e-05,
+      "loss": -0.0028,
+      "num_tokens": 1662670.0,
+      "reward": 3.181674861907959,
+      "reward_std": 0.24182533621788024,
+      "rewards/reward_humanlike/mean": 0.2005528837442398,
+      "rewards/reward_humanlike/std": 0.1301863819360733,
+      "rewards/reward_nonslop/mean": 0.5789134621620178,
+      "rewards/reward_nonslop/std": 0.02567046955227852,
+      "rewards/reward_paragraph_structure/mean": 0.9434401750564575,
+      "rewards/reward_paragraph_structure/std": 0.11112795919179916,
+      "rewards/reward_readability_flesch/mean": 0.9058632373809814,
+      "rewards/reward_readability_flesch/std": 0.1350622519850731,
+      "rewards/reward_variance_humanlike/mean": 0.5529051184654236,
+      "rewards/reward_variance_humanlike/std": 0.05817737057805061,
+      "step": 190
+    },
+    {
+      "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": 492.0,
+      "completions/max_terminated_length": 492.0,
+      "completions/mean_length": 398.65,
+      "completions/mean_terminated_length": 398.65,
+      "completions/min_length": 294.0,
+      "completions/min_terminated_length": 294.0,
+      "epoch": 0.0035577449370552817,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.16075296700000763,
+      "learning_rate": 2e-05,
+      "loss": 0.0197,
+      "num_tokens": 1699698.0,
+      "reward": 3.3926229000091555,
+      "reward_std": 0.22744941860437393,
+      "rewards/reward_humanlike/mean": 0.39020901918411255,
+      "rewards/reward_humanlike/std": 0.13960812389850616,
+      "rewards/reward_nonslop/mean": 0.5689973950386047,
+      "rewards/reward_nonslop/std": 0.0317101176828146,
+      "rewards/reward_paragraph_structure/mean": 0.9568790316581726,
+      "rewards/reward_paragraph_structure/std": 0.09259039759635926,
+      "rewards/reward_readability_flesch/mean": 0.8731052994728088,
+      "rewards/reward_readability_flesch/std": 0.1623595878481865,
+      "rewards/reward_variance_humanlike/mean": 0.6034322381019592,
+      "rewards/reward_variance_humanlike/std": 0.04017769172787666,
+      "step": 195
+    },
+    {
+      "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": 541.8,
+      "completions/max_terminated_length": 541.8,
+      "completions/mean_length": 429.8125,
+      "completions/mean_terminated_length": 429.8125,
+      "completions/min_length": 336.2,
+      "completions/min_terminated_length": 336.2,
+      "epoch": 0.0036489691662105455,
+      "frac_reward_zero_std": 0.0,
+      "grad_norm": 0.19189226627349854,
+      "learning_rate": 2e-05,
+      "loss": 0.0243,
+      "num_tokens": 1739219.0,
+      "reward": 3.2115593910217286,
+      "reward_std": 0.2621428996324539,
+      "rewards/reward_humanlike/mean": 0.39398524165153503,
+      "rewards/reward_humanlike/std": 0.12724200785160064,
+      "rewards/reward_nonslop/mean": 0.559182322025299,
+      "rewards/reward_nonslop/std": 0.02537613846361637,
+      "rewards/reward_paragraph_structure/mean": 0.9473167061805725,
+      "rewards/reward_paragraph_structure/std": 0.10079236477613449,
+      "rewards/reward_readability_flesch/mean": 0.7008966207504272,
+      "rewards/reward_readability_flesch/std": 0.22132455557584763,
+      "rewards/reward_variance_humanlike/mean": 0.6101784110069275,
+      "rewards/reward_variance_humanlike/std": 0.03768599070608616,
+      "step": 200
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 250,
+  "num_input_tokens_seen": 1739219,
+  "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
+}

checkpoint-200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d6d1dbe50a50058c947771a7fd7c03a443069be77d949e0bd75335cc07d1af9
+size 6865

checkpoint-250/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen3-14B
+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

checkpoint-250/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-250/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50bef24e37209e74e948a6fa60c599f0456e385a86d29115aa8539658e8dd66d
+size 513877864