harsha070/exp2-qwen-mbpp-s42-lambda-0p25
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: harsha070/exp2-qwen-mbpp-s42-lambda-0p25
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-16 06:15:17 +08:00
commit e9d6237854
47 changed files with 1417 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
.gitattributes vendored Normal file
View File

@@ -0,0 +1,36 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ckpt filter=lfs diff=lfs merge=lfs -text
*.ftz filter=lfs diff=lfs merge=lfs -text
*.gz filter=lfs diff=lfs merge=lfs -text
*.h5 filter=lfs diff=lfs merge=lfs -text
*.joblib filter=lfs diff=lfs merge=lfs -text
*.lfs.* filter=lfs diff=lfs merge=lfs -text
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz filter=lfs diff=lfs merge=lfs -text
*.onnx filter=lfs diff=lfs merge=lfs -text
*.ot filter=lfs diff=lfs merge=lfs -text
*.parquet filter=lfs diff=lfs merge=lfs -text
*.pb filter=lfs diff=lfs merge=lfs -text
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl filter=lfs diff=lfs merge=lfs -text
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
*.safetensors filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* filter=lfs diff=lfs merge=lfs -text
*.tar filter=lfs diff=lfs merge=lfs -text
*.tflite filter=lfs diff=lfs merge=lfs -text
*.tgz filter=lfs diff=lfs merge=lfs -text
*.wasm filter=lfs diff=lfs merge=lfs -text
*.xz 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
tokenizer.json filter=lfs diff=lfs merge=lfs -text

67
README.md Normal file
View File

@@ -0,0 +1,67 @@
---
base_model: harsha070/sft-warmup-qwen-v1
library_name: transformers
model_name: exp2-qwen-mbpp-s42-lambda-0p25
tags:
- generated_from_trainer
- trl
- grpo
licence: license
---
# Model Card for exp2-qwen-mbpp-s42-lambda-0p25
This model is a fine-tuned version of [harsha070/sft-warmup-qwen-v1](https://huggingface.co/harsha070/sft-warmup-qwen-v1).
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="harsha070/exp2-qwen-mbpp-s42-lambda-0p25", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/models-self5933/obfuscation-early-warning/runs/0pjebn91)
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 1.3.0
- Transformers: 5.7.0
- Pytorch: 2.11.0
- Datasets: 4.8.5
- Tokenizers: 0.22.2
## Citations
Cite GRPO as:
```bibtex
@article{shao2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@software{vonwerra2020trl,
title = {{TRL: Transformers Reinforcement Learning}},
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
license = {Apache-2.0},
url = {https://github.com/huggingface/trl},
year = {2020}
}
```

54
chat_template.jinja Normal file
View File

@@ -0,0 +1,54 @@
{%- if tools %}
{{- '<|im_start|>system\n' }}
{%- if messages[0]['role'] == 'system' %}
{{- messages[0]['content'] }}
{%- else %}
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
{%- endif %}
{{- "\n\n# 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' }}
{%- else %}
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- for message in messages %}
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
{%- elif message.role == "assistant" %}
{{- '<|im_start|>' + message.role }}
{%- if message.content %}
{{- '\n' + message.content }}
{%- endif %}
{%- for tool_call in message.tool_calls %}
{%- if tool_call.function is defined %}
{%- set tool_call = tool_call.function %}
{%- endif %}
{{- '\n<tool_call>\n{"name": "' }}
{{- tool_call.name }}
{{- '", "arguments": ' }}
{{- tool_call.arguments | tojson }}
{{- '}\n</tool_call>' }}
{%- endfor %}
{{- '<|im_end|>\n' }}
{%- elif message.role == "tool" %}
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
{{- '<|im_start|>user' }}
{%- endif %}
{{- '\n<tool_response>\n' }}
{{- message.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' }}
{%- endif %}

3
completions/completions_00010.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c0b0bbcd8288475d10ff51c0a3b2c94e4aba2f59bfb390f43371805becb83324
size 18571

3
completions/completions_00020.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b256bdb82db0190a21975217aa841c67d74931bf8c3ffa00453a8fefb78c106f
size 17589

3
completions/completions_00030.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:079488236c30234dc3a34d3888783dd18499b439b51d2e4dd954add2d614d2d2
size 18934

3
completions/completions_00040.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5cb0735f11a6886e2f64e59c95a51229ba0c381ce1423c80c3410ebfb8686abf
size 19177

3
completions/completions_00050.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:24aecf939a64b6a83235d7a696cb64ffdb984728095d6671c75d08a76ca9da82
size 18467

3
completions/completions_00060.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:43bbe94fe454fdaabe37623b5874f1db600a4af3441ef19372dd7fe53c99779a
size 17625

3
completions/completions_00070.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:fd384b2019d521ba630f03f2bee3e1e49c6eb06a42515a8e30035e4adfb6ccb3
size 18213

3
completions/completions_00080.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:10506d2efddf1d86bd715ebb49946dce52424cd65dc5aae299ad61175aedcd71
size 17656

3
completions/completions_00090.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:825bbcc5d4407626a7e0d9491d88dc3b6d295d97683fef685271fd13f5a6255e
size 17777

3
completions/completions_00100.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e7e78f4a11d283af6d2756606f8813105fa7c05b5ec742d7efdaaf6acdba0555
size 18693

3
completions/completions_00110.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2f42aa40f179a76d58bdf2347aea88c4912852e4008bc3702bf93f24137b633d
size 20772

3
completions/completions_00120.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d9b46d51775c0a293958904f37f85ffb891df9f911f1cd0fa9db9b1b6fbfd309
size 20236

3
completions/completions_00130.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:50a779da6108c97f9f08244f94d37e0fb5270026920acfc52e4e4ea70a63d37b
size 19154

3
completions/completions_00140.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e0119a211c3f20ca9ac836310bfce4df42d28220815582bb103b6e357269bc89
size 16770

3
completions/completions_00150.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:02798400369f039e2c191d03d4f1bf198bcbdc5ba50ec75a8669d4eb4e5be3b7
size 18198

3
completions/completions_00160.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:33654a88093883856bfe5471775b9c29561227ed6a6fc7d25ee767a86960c337
size 17214

3
completions/completions_00170.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9804d65dd7c72eb177925af42ed3f5068f323978a883c171576540d3316992a1
size 17010

3
completions/completions_00180.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:57d9b53e4f360f9ed44ac22efd2d492169ef67b00c34d90fa45df31d4062473d
size 18083

3
completions/completions_00190.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:cc35641cd8df86333d08a95f0c9da9477a986b375054cb9215cd1e3ee620be89
size 17747

3
completions/completions_00200.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6a7c27cf14176a14ef5c3578c1d617049f1821b8f4004d5f253038b5a5c107f1
size 15277

3
completions/completions_00210.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:61e1e3aa7227df76fd2d9d5b07e68cf591b725c201853e41478177d45ea32b79
size 16705

3
completions/completions_00220.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:dd2e12a808f38569b03d8e0773fb80d2bd4641cef554fd0aa913592b9e031c7d
size 17824

3
completions/completions_00230.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9a3d8f94d2b60ad86cb21cfbf9f5e41b81697accab9ab418c2a55d70d0de8a1c
size 17469

3
completions/completions_00240.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:711dc8506c6bd1cd8b245503e6346525655b1d6c3ba08d666d24b649dd72d22e
size 17930

3
completions/completions_00250.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:62b43606401ccc06c8fe10f06d51469a2ca203c6a2cef1e7259b91c71ec0ffab
size 17810

3
completions/completions_00260.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:88003fa5e5754092db19ca79aa5b767052c66736be63428b44a2695e1c6e0779
size 17296

3
completions/completions_00270.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:21798afe0325c99b58c2424791998d4e47c22f9e13bc0bee4a4a035d5d23ebe7
size 16380

3
completions/completions_00280.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7bd8c0cd1389dde99f201477a3e77ef19d6f5ce6d842cffd5e61ab68138d9db4
size 16891

3
completions/completions_00290.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:06905d2100233c07140141138813e668a3c268a669d3118531c04643ac705377
size 18454

3
completions/completions_00300.parquet Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:db26bb5dafb73b8989834bdfd5d72ddfaa3d75dec5d51e446de6ce03d3497163
size 17804

69
config.json Normal file
View File

@@ -0,0 +1,69 @@
{
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": null,
"dtype": "bfloat16",
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 11008,
"layer_types": [
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention"
],
"max_position_embeddings": 32768,
"max_window_layers": 70,
"model_type": "qwen2",
"num_attention_heads": 16,
"num_hidden_layers": 36,
"num_key_value_heads": 2,
"pad_token_id": 151643,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 1000000.0,
"rope_type": "default"
},
"sliding_window": null,
"tie_word_embeddings": true,
"transformers_version": "5.7.0",
"use_cache": false,
"use_sliding_window": false,
"vocab_size": 151936
}

13
generation_config.json Normal file
View File

@@ -0,0 +1,13 @@
{
"do_sample": true,
"eos_token_id": [
151645,
151643
],
"pad_token_id": 151643,
"repetition_penalty": 1.05,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8,
"transformers_version": "5.7.0"
}

54
last-checkpoint/chat_template.jinja Normal file
View File

@@ -0,0 +1,54 @@
{%- if tools %}
{{- '<|im_start|>system\n' }}
{%- if messages[0]['role'] == 'system' %}
{{- messages[0]['content'] }}
{%- else %}
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
{%- endif %}
{{- "\n\n# 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' }}
{%- else %}
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- for message in messages %}
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
{%- elif message.role == "assistant" %}
{{- '<|im_start|>' + message.role }}
{%- if message.content %}
{{- '\n' + message.content }}
{%- endif %}
{%- for tool_call in message.tool_calls %}
{%- if tool_call.function is defined %}
{%- set tool_call = tool_call.function %}
{%- endif %}
{{- '\n<tool_call>\n{"name": "' }}
{{- tool_call.name }}
{{- '", "arguments": ' }}
{{- tool_call.arguments | tojson }}
{{- '}\n</tool_call>' }}
{%- endfor %}
{{- '<|im_end|>\n' }}
{%- elif message.role == "tool" %}
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
{{- '<|im_start|>user' }}
{%- endif %}
{{- '\n<tool_response>\n' }}
{{- message.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' }}
{%- endif %}

69
last-checkpoint/config.json Normal file
View File

@@ -0,0 +1,69 @@
{
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": null,
"dtype": "bfloat16",
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 11008,
"layer_types": [
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention"
],
"max_position_embeddings": 32768,
"max_window_layers": 70,
"model_type": "qwen2",
"num_attention_heads": 16,
"num_hidden_layers": 36,
"num_key_value_heads": 2,
"pad_token_id": 151643,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 1000000.0,
"rope_type": "default"
},
"sliding_window": null,
"tie_word_embeddings": true,
"transformers_version": "5.7.0",
"use_cache": false,
"use_sliding_window": false,
"vocab_size": 151936
}

13
last-checkpoint/generation_config.json Normal file
View File

@@ -0,0 +1,13 @@
{
"do_sample": true,
"eos_token_id": [
151645,
151643
],
"pad_token_id": 151643,
"repetition_penalty": 1.05,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8,
"transformers_version": "5.7.0"
}

3
last-checkpoint/model.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:89a43560f87b842e22b16e9371ae166b13e2a7e87091cf48810e1cd6b62e9830
size 6171927112

3
last-checkpoint/tokenizer.json Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
size 11421892

30
last-checkpoint/tokenizer_config.json Normal file
View File

@@ -0,0 +1,30 @@
{
"add_prefix_space": false,
"backend": "tokenizers",
"bos_token": null,
"clean_up_tokenization_spaces": false,
"eos_token": "<|im_end|>",
"errors": "replace",
"extra_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|>"
],
"is_local": false,
"local_files_only": false,
"model_max_length": 131072,
"pad_token": "<|endoftext|>",
"split_special_tokens": false,
"tokenizer_class": "Qwen2Tokenizer",
"unk_token": null
}

874
last-checkpoint/trainer_state.json Normal file
View File

@@ -0,0 +1,874 @@
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 500,
"global_step": 300,
"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": 0.5875,
"completions/max_length": 512.0,
"completions/max_terminated_length": 440.4,
"completions/mean_length": 468.6125,
"completions/mean_terminated_length": 379.02000122070314,
"completions/min_length": 369.0,
"completions/min_terminated_length": 317.8,
"entropy": 0.2740801006555557,
"epoch": 0.06666666666666667,
"frac_reward_zero_std": 0.15,
"grad_norm": 1.3671875,
"kl": 0.04017456619621953,
"learning_rate": 9.7e-06,
"loss": -0.004630821943283081,
"num_tokens": 50357.0,
"reward": 0.803729248046875,
"reward_std": 0.41876387000083926,
"rewards/JointRewardFunction/mean": 0.803729248046875,
"rewards/JointRewardFunction/std": 0.41876387894153594,
"step": 10,
"step_time": 23.422285906800244
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completions/clipped_ratio": 0.75,
"completions/max_length": 512.0,
"completions/max_terminated_length": 328.2,
"completions/mean_length": 493.75,
"completions/mean_terminated_length": 310.2800018310547,
"completions/min_length": 446.2,
"completions/min_terminated_length": 292.6,
"entropy": 0.1757997965440154,
"epoch": 0.13333333333333333,
"frac_reward_zero_std": 0.65,
"grad_norm": 0.1865234375,
"kl": 0.064147645724006,
"learning_rate": 9.366666666666668e-06,
"loss": -0.00039904499426484107,
"num_tokens": 103617.0,
"reward": 0.85,
"reward_std": 0.3472102493047714,
"rewards/JointRewardFunction/mean": 0.85,
"rewards/JointRewardFunction/std": 0.3472102552652359,
"step": 20,
"step_time": 23.712279441399733
},
{
"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": 0.6625,
"completions/max_length": 512.0,
"completions/max_terminated_length": 376.5,
"completions/mean_length": 476.5875,
"completions/mean_terminated_length": 333.94667358398436,
"completions/min_length": 374.9,
"completions/min_terminated_length": 272.5,
"entropy": 0.21253416435793043,
"epoch": 0.2,
"frac_reward_zero_std": 0.3,
"grad_norm": 2.28125,
"kl": 0.07378085452364758,
"learning_rate": 9.033333333333334e-06,
"loss": -0.012095230817794799,
"num_tokens": 153160.0,
"reward": 0.95,
"reward_std": 0.43348987102508546,
"rewards/JointRewardFunction/mean": 0.95,
"rewards/JointRewardFunction/std": 0.43348987102508546,
"step": 30,
"step_time": 23.327648415401928
},
{
"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": 0.85,
"completions/max_length": 512.0,
"completions/max_terminated_length": 284.1,
"completions/mean_length": 503.7625,
"completions/mean_terminated_length": 278.8300018310547,
"completions/min_length": 477.8,
"completions/min_terminated_length": 273.0,
"entropy": 0.21520083369687198,
"epoch": 0.26666666666666666,
"frac_reward_zero_std": 0.35,
"grad_norm": 0.255859375,
"kl": 0.06080276914872229,
"learning_rate": 8.700000000000001e-06,
"loss": -0.0005816968623548746,
"num_tokens": 206773.0,
"reward": 0.7875,
"reward_std": 0.4616557478904724,
"rewards/JointRewardFunction/mean": 0.7875,
"rewards/JointRewardFunction/std": 0.46165576577186584,
"step": 40,
"step_time": 23.59478929130273
},
{
"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": 0.7625,
"completions/max_length": 512.0,
"completions/max_terminated_length": 325.7,
"completions/mean_length": 490.0375,
"completions/mean_terminated_length": 299.44667053222656,
"completions/min_length": 433.4,
"completions/min_terminated_length": 279.8,
"entropy": 0.25695387944579123,
"epoch": 0.3333333333333333,
"frac_reward_zero_std": 0.45,
"grad_norm": 1.6953125,
"kl": 0.08556670525576919,
"learning_rate": 8.366666666666667e-06,
"loss": -0.004827765375375747,
"num_tokens": 258692.0,
"reward": 0.9375,
"reward_std": 0.3981345325708389,
"rewards/JointRewardFunction/mean": 0.9375,
"rewards/JointRewardFunction/std": 0.3981345325708389,
"step": 50,
"step_time": 23.633761547702306
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completions/clipped_ratio": 0.75,
"completions/max_length": 512.0,
"completions/max_terminated_length": 332.6,
"completions/mean_length": 492.95,
"completions/mean_terminated_length": 307.1166687011719,
"completions/min_length": 437.8,
"completions/min_terminated_length": 284.2,
"entropy": 0.2603289651684463,
"epoch": 0.4,
"frac_reward_zero_std": 0.55,
"grad_norm": 1.671875,
"kl": 0.060612542228773235,
"learning_rate": 8.033333333333335e-06,
"loss": 0.0008684337139129638,
"num_tokens": 311528.0,
"reward": 0.925,
"reward_std": 0.2958350956439972,
"rewards/JointRewardFunction/mean": 0.925,
"rewards/JointRewardFunction/std": 0.29583510756492615,
"step": 60,
"step_time": 23.58587597120204
},
{
"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": 0.725,
"completions/max_length": 512.0,
"completions/max_terminated_length": 339.0,
"completions/mean_length": 492.8125,
"completions/mean_terminated_length": 313.85,
"completions/min_length": 433.5,
"completions/min_terminated_length": 279.9,
"entropy": 0.27547385785728695,
"epoch": 0.4666666666666667,
"frac_reward_zero_std": 0.6,
"grad_norm": 0.99609375,
"kl": 0.0518098235828802,
"learning_rate": 7.7e-06,
"loss": -0.0017207100987434386,
"num_tokens": 364749.0,
"reward": 0.9125,
"reward_std": 0.3383516758680344,
"rewards/JointRewardFunction/mean": 0.9125,
"rewards/JointRewardFunction/std": 0.3383516877889633,
"step": 70,
"step_time": 23.567738218297016
},
{
"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": 0.7375,
"completions/max_length": 512.0,
"completions/max_terminated_length": 449.5,
"completions/mean_length": 487.675,
"completions/mean_terminated_length": 422.1333343505859,
"completions/min_length": 393.1,
"completions/min_terminated_length": 393.1,
"entropy": 0.2849867718294263,
"epoch": 0.5333333333333333,
"frac_reward_zero_std": 0.4,
"grad_norm": 0.042724609375,
"kl": 0.054288532945793125,
"learning_rate": 7.3666666666666676e-06,
"loss": 0.01214314103126526,
"num_tokens": 417435.0,
"reward": 0.9625,
"reward_std": 0.44988160133361815,
"rewards/JointRewardFunction/mean": 0.9625,
"rewards/JointRewardFunction/std": 0.4498816192150116,
"step": 80,
"step_time": 23.797617995494512
},
{
"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": 0.7125,
"completions/max_length": 512.0,
"completions/max_terminated_length": 330.2,
"completions/mean_length": 487.5125,
"completions/mean_terminated_length": 302.90834045410156,
"completions/min_length": 421.4,
"completions/min_terminated_length": 267.8,
"entropy": 0.2951860463246703,
"epoch": 0.6,
"frac_reward_zero_std": 0.45,
"grad_norm": 1.265625,
"kl": 0.04278905827086419,
"learning_rate": 7.033333333333334e-06,
"loss": -0.0006712859496474266,
"num_tokens": 470584.0,
"reward": 0.9375,
"reward_std": 0.31479085683822633,
"rewards/JointRewardFunction/mean": 0.9375,
"rewards/JointRewardFunction/std": 0.3147908627986908,
"step": 90,
"step_time": 23.514957720592793
},
{
"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": 0.625,
"completions/max_length": 512.0,
"completions/max_terminated_length": 459.9,
"completions/mean_length": 472.3125,
"completions/mean_terminated_length": 418.8316680908203,
"completions/min_length": 365.9,
"completions/min_terminated_length": 365.9,
"entropy": 0.37076061628758905,
"epoch": 0.6666666666666666,
"frac_reward_zero_std": 0.4,
"grad_norm": 2.15625,
"kl": 0.05282424986362457,
"learning_rate": 6.700000000000001e-06,
"loss": -0.009071560204029083,
"num_tokens": 523253.0,
"reward": 0.9815673828125,
"reward_std": 0.4160596996545792,
"rewards/JointRewardFunction/mean": 0.9815673828125,
"rewards/JointRewardFunction/std": 0.41605971157550814,
"step": 100,
"step_time": 23.664498128004197
},
{
"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": 0.6125,
"completions/max_length": 509.7,
"completions/max_terminated_length": 484.8,
"completions/mean_length": 475.7,
"completions/mean_terminated_length": 437.1983367919922,
"completions/min_length": 383.8,
"completions/min_terminated_length": 383.8,
"entropy": 0.4262677112594247,
"epoch": 0.7333333333333333,
"frac_reward_zero_std": 0.4,
"grad_norm": 2.0625,
"kl": 0.03714689936023206,
"learning_rate": 6.366666666666668e-06,
"loss": 0.008247312903404237,
"num_tokens": 574257.0,
"reward": 1.037060546875,
"reward_std": 0.3255626171827316,
"rewards/JointRewardFunction/mean": 1.037060546875,
"rewards/JointRewardFunction/std": 0.3255626350641251,
"step": 110,
"step_time": 23.18678354880103
},
{
"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": 0.575,
"completions/max_length": 512.0,
"completions/max_terminated_length": 388.5,
"completions/mean_length": 472.7625,
"completions/mean_terminated_length": 342.3550079345703,
"completions/min_length": 396.3,
"completions/min_terminated_length": 293.9,
"entropy": 0.4861995566636324,
"epoch": 0.8,
"frac_reward_zero_std": 0.35,
"grad_norm": 1.40625,
"kl": 0.039038634288590404,
"learning_rate": 6.033333333333335e-06,
"loss": 0.012050890922546386,
"num_tokens": 624342.0,
"reward": 0.925,
"reward_std": 0.4410230278968811,
"rewards/JointRewardFunction/mean": 0.925,
"rewards/JointRewardFunction/std": 0.441023051738739,
"step": 120,
"step_time": 23.72601753709896
},
{
"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": 0.5375,
"completions/max_length": 511.6,
"completions/max_terminated_length": 476.8,
"completions/mean_length": 470.0375,
"completions/mean_terminated_length": 427.0658416748047,
"completions/min_length": 372.0,
"completions/min_terminated_length": 372.0,
"entropy": 0.41965660247951747,
"epoch": 0.8666666666666667,
"frac_reward_zero_std": 0.55,
"grad_norm": 1.4453125,
"kl": 0.03558391091646627,
"learning_rate": 5.7e-06,
"loss": 0.0031338028609752657,
"num_tokens": 676197.0,
"reward": 1.029736328125,
"reward_std": 0.3752464294433594,
"rewards/JointRewardFunction/mean": 1.029736328125,
"rewards/JointRewardFunction/std": 0.37524643540382385,
"step": 130,
"step_time": 23.592583017596915
},
{
"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": 0.5,
"completions/max_length": 512.0,
"completions/max_terminated_length": 470.4,
"completions/mean_length": 468.3375,
"completions/mean_terminated_length": 422.97833862304685,
"completions/min_length": 369.0,
"completions/min_terminated_length": 369.0,
"entropy": 0.4238532094284892,
"epoch": 0.9333333333333333,
"frac_reward_zero_std": 0.4,
"grad_norm": 0.048095703125,
"kl": 0.047195866727270185,
"learning_rate": 5.366666666666666e-06,
"loss": 0.011741240322589875,
"num_tokens": 726296.0,
"reward": 1.0271484375,
"reward_std": 0.39577038288116456,
"rewards/JointRewardFunction/mean": 1.0271484375,
"rewards/JointRewardFunction/std": 0.39577039480209353,
"step": 140,
"step_time": 23.827525900093313
},
{
"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": 0.475,
"completions/max_length": 507.6,
"completions/max_terminated_length": 417.3,
"completions/mean_length": 452.0375,
"completions/mean_terminated_length": 360.2283416748047,
"completions/min_length": 351.9,
"completions/min_terminated_length": 300.7,
"entropy": 0.405802302993834,
"epoch": 1.0,
"frac_reward_zero_std": 0.5,
"grad_norm": 1.3125,
"kl": 0.05754059529863298,
"learning_rate": 5.033333333333333e-06,
"loss": 0.020150861144065856,
"num_tokens": 774771.0,
"reward": 1.0125,
"reward_std": 0.37200968265533446,
"rewards/JointRewardFunction/mean": 1.0125,
"rewards/JointRewardFunction/std": 0.3720097005367279,
"step": 150,
"step_time": 23.477148605001275
},
{
"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": 0.35,
"completions/max_length": 504.7,
"completions/max_terminated_length": 460.6,
"completions/mean_length": 418.725,
"completions/mean_terminated_length": 376.1097686767578,
"completions/min_length": 290.2,
"completions/min_terminated_length": 290.2,
"entropy": 0.4137630261480808,
"epoch": 1.0666666666666667,
"frac_reward_zero_std": 0.55,
"grad_norm": 1.5234375,
"kl": 0.04866745978360996,
"learning_rate": 4.7e-06,
"loss": -0.012348555028438568,
"num_tokens": 820185.0,
"reward": 1.1374755859375,
"reward_std": 0.2693489044904709,
"rewards/JointRewardFunction/mean": 1.1374755859375,
"rewards/JointRewardFunction/std": 0.2693489044904709,
"step": 160,
"step_time": 23.104779590805993
},
{
"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": 0.45,
"completions/max_length": 512.0,
"completions/max_terminated_length": 459.3,
"completions/mean_length": 452.575,
"completions/mean_terminated_length": 405.1588165283203,
"completions/min_length": 342.3,
"completions/min_terminated_length": 342.3,
"entropy": 0.3708974776789546,
"epoch": 1.1333333333333333,
"frac_reward_zero_std": 0.8,
"grad_norm": 0.032470703125,
"kl": 0.05273645754205063,
"learning_rate": 4.366666666666667e-06,
"loss": 0.0020671430975198746,
"num_tokens": 868979.0,
"reward": 1.2064208984375,
"reward_std": 0.1077505886554718,
"rewards/JointRewardFunction/mean": 1.2064208984375,
"rewards/JointRewardFunction/std": 0.10775059163570404,
"step": 170,
"step_time": 23.706487477812335
},
{
"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": 0.3875,
"completions/max_length": 512.0,
"completions/max_terminated_length": 481.1,
"completions/mean_length": 445.025,
"completions/mean_terminated_length": 411.8959655761719,
"completions/min_length": 341.8,
"completions/min_terminated_length": 341.8,
"entropy": 0.3909618055447936,
"epoch": 1.2,
"frac_reward_zero_std": 0.55,
"grad_norm": 1.7109375,
"kl": 0.045815252687316385,
"learning_rate": 4.033333333333333e-06,
"loss": 0.004306042194366455,
"num_tokens": 917385.0,
"reward": 1.075,
"reward_std": 0.3337466180324554,
"rewards/JointRewardFunction/mean": 1.075,
"rewards/JointRewardFunction/std": 0.3337466180324554,
"step": 180,
"step_time": 23.554991897323635
},
{
"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": 0.6125,
"completions/max_length": 512.0,
"completions/max_terminated_length": 471.2,
"completions/mean_length": 468.075,
"completions/mean_terminated_length": 422.6735778808594,
"completions/min_length": 385.9,
"completions/min_terminated_length": 385.9,
"entropy": 0.36821637134999036,
"epoch": 1.2666666666666666,
"frac_reward_zero_std": 0.55,
"grad_norm": 1.8671875,
"kl": 0.04221574537805282,
"learning_rate": 3.7e-06,
"loss": 0.005308620631694794,
"num_tokens": 968751.0,
"reward": 0.95,
"reward_std": 0.3438155859708786,
"rewards/JointRewardFunction/mean": 0.95,
"rewards/JointRewardFunction/std": 0.34381560385227206,
"step": 190,
"step_time": 23.511522994117694
},
{
"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": 0.3125,
"completions/max_length": 497.9,
"completions/max_terminated_length": 470.5,
"completions/mean_length": 431.05,
"completions/mean_terminated_length": 400.3317901611328,
"completions/min_length": 318.5,
"completions/min_terminated_length": 318.5,
"entropy": 0.41096227150410414,
"epoch": 1.3333333333333333,
"frac_reward_zero_std": 0.65,
"grad_norm": 0.06640625,
"kl": 0.04462224093731493,
"learning_rate": 3.366666666666667e-06,
"loss": 0.0076727248728275296,
"num_tokens": 1016343.0,
"reward": 1.125,
"reward_std": 0.21674493551254273,
"rewards/JointRewardFunction/mean": 1.125,
"rewards/JointRewardFunction/std": 0.21674493551254273,
"step": 200,
"step_time": 23.060737183006133
},
{
"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": 0.3875,
"completions/max_length": 512.0,
"completions/max_terminated_length": 482.7,
"completions/mean_length": 458.65,
"completions/mean_terminated_length": 428.9585815429688,
"completions/min_length": 363.5,
"completions/min_terminated_length": 363.5,
"entropy": 0.3947587950155139,
"epoch": 1.4,
"frac_reward_zero_std": 0.65,
"grad_norm": 9.3125,
"kl": 0.10990666588768364,
"learning_rate": 3.0333333333333337e-06,
"loss": 0.010126692801713943,
"num_tokens": 1066311.0,
"reward": 1.111083984375,
"reward_std": 0.28541127145290374,
"rewards/JointRewardFunction/mean": 1.111083984375,
"rewards/JointRewardFunction/std": 0.2854112803936005,
"step": 210,
"step_time": 23.734341114398557
},
{
"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": 0.375,
"completions/max_length": 511.2,
"completions/max_terminated_length": 499.2,
"completions/mean_length": 448.1,
"completions/mean_terminated_length": 415.37428894042966,
"completions/min_length": 327.8,
"completions/min_terminated_length": 327.8,
"entropy": 0.4135129824280739,
"epoch": 1.4666666666666668,
"frac_reward_zero_std": 0.4,
"grad_norm": 2.0,
"kl": 0.04890955399023369,
"learning_rate": 2.7000000000000004e-06,
"loss": 0.005412362515926361,
"num_tokens": 1115131.0,
"reward": 1.0,
"reward_std": 0.407365021109581,
"rewards/JointRewardFunction/mean": 1.0,
"rewards/JointRewardFunction/std": 0.4073650389909744,
"step": 220,
"step_time": 23.674485975824062
},
{
"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": 0.3,
"completions/max_length": 493.0,
"completions/max_terminated_length": 468.3,
"completions/mean_length": 425.175,
"completions/mean_terminated_length": 400.7392883300781,
"completions/min_length": 313.4,
"completions/min_terminated_length": 313.4,
"entropy": 0.40025499686598776,
"epoch": 1.5333333333333332,
"frac_reward_zero_std": 0.6,
"grad_norm": 0.045166015625,
"kl": 0.044459241011645646,
"learning_rate": 2.3666666666666667e-06,
"loss": 0.0074857622385025024,
"num_tokens": 1162029.0,
"reward": 1.0625,
"reward_std": 0.332049286365509,
"rewards/JointRewardFunction/mean": 1.0625,
"rewards/JointRewardFunction/std": 0.3320492923259735,
"step": 230,
"step_time": 22.975970137718832
},
{
"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": 0.5125,
"completions/max_length": 507.5,
"completions/max_terminated_length": 417.0,
"completions/mean_length": 443.7875,
"completions/mean_terminated_length": 338.87250366210935,
"completions/min_length": 315.1,
"completions/min_terminated_length": 263.9,
"entropy": 0.3852341592311859,
"epoch": 1.6,
"frac_reward_zero_std": 0.5,
"grad_norm": 2.109375,
"kl": 0.039597276959102604,
"learning_rate": 2.0333333333333335e-06,
"loss": 0.015155516564846039,
"num_tokens": 1211952.0,
"reward": 1.03134765625,
"reward_std": 0.30796128809452056,
"rewards/JointRewardFunction/mean": 1.03134765625,
"rewards/JointRewardFunction/std": 0.30796128809452056,
"step": 240,
"step_time": 23.477432061507717
},
{
"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": 0.4125,
"completions/max_length": 499.5,
"completions/max_terminated_length": 454.8,
"completions/mean_length": 440.175,
"completions/mean_terminated_length": 402.6975067138672,
"completions/min_length": 342.6,
"completions/min_terminated_length": 342.6,
"entropy": 0.4049293929710984,
"epoch": 1.6666666666666665,
"frac_reward_zero_std": 0.55,
"grad_norm": 0.06689453125,
"kl": 0.057483326562214644,
"learning_rate": 1.7000000000000002e-06,
"loss": 0.005497528612613678,
"num_tokens": 1260194.0,
"reward": 1.100634765625,
"reward_std": 0.30466278195381163,
"rewards/JointRewardFunction/mean": 1.100634765625,
"rewards/JointRewardFunction/std": 0.30466278940439223,
"step": 250,
"step_time": 23.31596095281129
},
{
"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": 0.4125,
"completions/max_length": 512.0,
"completions/max_terminated_length": 486.0,
"completions/mean_length": 455.5875,
"completions/mean_terminated_length": 423.4404815673828,
"completions/min_length": 368.3,
"completions/min_terminated_length": 368.3,
"entropy": 0.37657185792922976,
"epoch": 1.7333333333333334,
"frac_reward_zero_std": 0.55,
"grad_norm": 1.78125,
"kl": 0.044227164250332865,
"learning_rate": 1.3666666666666668e-06,
"loss": 0.01007101833820343,
"num_tokens": 1312481.0,
"reward": 1.0563232421875,
"reward_std": 0.3230230301618576,
"rewards/JointRewardFunction/mean": 1.0563232421875,
"rewards/JointRewardFunction/std": 0.3230230316519737,
"step": 260,
"step_time": 23.83616074830352
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completions/clipped_ratio": 0.25,
"completions/max_length": 511.2,
"completions/max_terminated_length": 478.9,
"completions/mean_length": 419.7,
"completions/mean_terminated_length": 390.71238708496094,
"completions/min_length": 292.4,
"completions/min_terminated_length": 292.4,
"entropy": 0.3890227179042995,
"epoch": 1.8,
"frac_reward_zero_std": 0.85,
"grad_norm": 0.0517578125,
"kl": 0.04973521351348609,
"learning_rate": 1.0333333333333333e-06,
"loss": 0.00801372081041336,
"num_tokens": 1358297.0,
"reward": 1.1375,
"reward_std": 0.12793734967708587,
"rewards/JointRewardFunction/mean": 1.1375,
"rewards/JointRewardFunction/std": 0.12793734967708587,
"step": 270,
"step_time": 23.87891690217657
},
{
"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": 0.45,
"completions/max_length": 512.0,
"completions/max_terminated_length": 464.2,
"completions/mean_length": 440.9125,
"completions/mean_terminated_length": 394.0228607177734,
"completions/min_length": 329.2,
"completions/min_terminated_length": 329.2,
"entropy": 0.3685274325311184,
"epoch": 1.8666666666666667,
"frac_reward_zero_std": 0.6,
"grad_norm": 0.134765625,
"kl": 0.05201944473665208,
"learning_rate": 7.000000000000001e-07,
"loss": 0.013003082573413849,
"num_tokens": 1406762.0,
"reward": 1.123974609375,
"reward_std": 0.25532945692539216,
"rewards/JointRewardFunction/mean": 1.123974609375,
"rewards/JointRewardFunction/std": 0.25532945692539216,
"step": 280,
"step_time": 23.774091701293948
},
{
"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": 0.4125,
"completions/max_length": 508.9,
"completions/max_terminated_length": 447.5,
"completions/mean_length": 443.3125,
"completions/mean_terminated_length": 385.1429840087891,
"completions/min_length": 311.9,
"completions/min_terminated_length": 311.9,
"entropy": 0.3619745412841439,
"epoch": 1.9333333333333333,
"frac_reward_zero_std": 0.55,
"grad_norm": 1.296875,
"kl": 0.04537691879086196,
"learning_rate": 3.666666666666667e-07,
"loss": 0.015167883038520813,
"num_tokens": 1456063.0,
"reward": 1.1,
"reward_std": 0.2966939479112625,
"rewards/JointRewardFunction/mean": 1.1,
"rewards/JointRewardFunction/std": 0.296693953871727,
"step": 290,
"step_time": 23.711561490091846
},
{
"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": 0.3625,
"completions/max_length": 511.4,
"completions/max_terminated_length": 471.0,
"completions/mean_length": 445.95,
"completions/mean_terminated_length": 418.78345947265626,
"completions/min_length": 355.0,
"completions/min_terminated_length": 355.0,
"entropy": 0.3853150447830558,
"epoch": 2.0,
"frac_reward_zero_std": 0.6,
"grad_norm": 1.8515625,
"kl": 0.04785980319138616,
"learning_rate": 3.333333333333334e-08,
"loss": -0.0016140155494213104,
"num_tokens": 1504095.0,
"reward": 1.1375,
"reward_std": 0.2394672751426697,
"rewards/JointRewardFunction/mean": 1.1375,
"rewards/JointRewardFunction/std": 0.23946728110313414,
"step": 300,
"step_time": 23.609792864409975
}
],
"logging_steps": 10,
"max_steps": 300,
"num_input_tokens_seen": 1504095,
"num_train_epochs": 2,
"save_steps": 10,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}

3
last-checkpoint/training_args.bin Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3ce376f35c622be385d16371a211e21a08a548f37ed9b16106324741412ca461
size 7249

3
model.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:89a43560f87b842e22b16e9371ae166b13e2a7e87091cf48810e1cd6b62e9830
size 6171927112

3
tokenizer.json Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
size 11421892

30
tokenizer_config.json Normal file
View File

@@ -0,0 +1,30 @@
{
"add_prefix_space": false,
"backend": "tokenizers",
"bos_token": null,
"clean_up_tokenization_spaces": false,
"eos_token": "<|im_end|>",
"errors": "replace",
"extra_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|>"
],
"is_local": false,
"local_files_only": false,
"model_max_length": 131072,
"pad_token": "<|endoftext|>",
"split_special_tokens": false,
"tokenizer_class": "Qwen2Tokenizer",
"unk_token": null
}

3
training_args.bin Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3ce376f35c622be385d16371a211e21a08a548f37ed9b16106324741412ca461
size 7249