aminediroHF/async-grpo-ckpt-smoke-r1d-1.5b
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: aminediroHF/async-grpo-ckpt-smoke-r1d-1.5b
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-09-06 08:19:14 +08:00
commit 09a35f6b7b
21 changed files with 604 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

70
README.md Normal file
View File

@@ -0,0 +1,70 @@
---
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
library_name: transformers
model_name: async-grpo-ckpt-smoke-r1d-1.5b
tags:
- generated_from_trainer
- trackio:https://huggingface.co/spaces/aminediroHF/async-grpo-ckpt-smoke-static-053862
- trackio
- hf_jobs
- async-grpo
- trl
licence: license
---
# Model Card for async-grpo-ckpt-smoke-r1d-1.5b
This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B).
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="aminediroHF/async-grpo-ckpt-smoke-r1d-1.5b", 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 AsyncGRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 1.11.0.dev0
- Transformers: 5.15.0
- Pytorch: 2.13.0+cu130
- Datasets: 5.0.1
- Tokenizers: 0.22.2
## Citations
Cite AsyncGRPO 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}
}
```

1
chat_template.jinja Normal file
View File

@@ -0,0 +1 @@
{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{bos_token}}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is none %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls']%}{%- if not ns.is_first %}{{'<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<｜tool▁call▁end｜>'}}{%- set ns.is_first = true -%}{%- else %}{{'\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<｜tool▁call▁end｜>'}}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- endfor %}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{%- if ns.is_tool %}{{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<｜Assistant｜>' + content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- set ns.is_output_first = false %}{%- else %}{{'\n<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<｜tool▁outputs▁end｜>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<｜Assistant｜><think>\n'}}{% endif %}

62
config.json Normal file
View File

@@ -0,0 +1,62 @@
{
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 151646,
"dtype": "float32",
"eos_token_id": 151643,
"hidden_act": "silu",
"hidden_size": 1536,
"initializer_range": 0.02,
"intermediate_size": 8960,
"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"
],
"max_position_embeddings": 131072,
"max_window_layers": 21,
"model_type": "qwen2",
"num_attention_heads": 12,
"num_hidden_layers": 28,
"num_key_value_heads": 2,
"pad_token_id": 151643,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 10000,
"rope_type": "default"
},
"sliding_window": null,
"tie_word_embeddings": false,
"transformers_version": "5.15.0",
"use_cache": false,
"use_mrope": false,
"use_sliding_window": false,
"vocab_size": 151936
}

12
generation_config.json Normal file
View File

@@ -0,0 +1,12 @@
{
"_from_model_config": true,
"bos_token_id": 151646,
"do_sample": true,
"eos_token_id": [
151643
],
"pad_token_id": 151643,
"temperature": 0.6,
"top_p": 0.95,
"transformers_version": "5.15.0"
}

1
last-checkpoint/chat_template.jinja Normal file
View File

@@ -0,0 +1 @@
{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{bos_token}}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is none %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls']%}{%- if not ns.is_first %}{{'<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<｜tool▁call▁end｜>'}}{%- set ns.is_first = true -%}{%- else %}{{'\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<｜tool▁call▁end｜>'}}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- endfor %}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{%- if ns.is_tool %}{{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<｜Assistant｜>' + content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- set ns.is_output_first = false %}{%- else %}{{'\n<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<｜tool▁outputs▁end｜>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<｜Assistant｜><think>\n'}}{% endif %}

62
last-checkpoint/config.json Normal file
View File

@@ -0,0 +1,62 @@
{
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 151646,
"dtype": "float32",
"eos_token_id": 151643,
"hidden_act": "silu",
"hidden_size": 1536,
"initializer_range": 0.02,
"intermediate_size": 8960,
"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"
],
"max_position_embeddings": 131072,
"max_window_layers": 21,
"model_type": "qwen2",
"num_attention_heads": 12,
"num_hidden_layers": 28,
"num_key_value_heads": 2,
"pad_token_id": 151643,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 10000,
"rope_type": "default"
},
"sliding_window": null,
"tie_word_embeddings": false,
"transformers_version": "5.15.0",
"use_cache": false,
"use_mrope": false,
"use_sliding_window": false,
"vocab_size": 151936
}

12
last-checkpoint/generation_config.json Normal file
View File

@@ -0,0 +1,12 @@
{
"_from_model_config": true,
"bos_token_id": 151646,
"do_sample": true,
"eos_token_id": [
151643
],
"pad_token_id": 151643,
"temperature": 0.6,
"top_p": 0.95,
"transformers_version": "5.15.0"
}

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

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b9c47b492488c981054bd27ecc224d98bf429a7b0e0c1f09826d988b43ae4a8b
size 7108390424

3
last-checkpoint/optimizer.pt Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ed8a5af882a4788a95cebc2d8051545a8aa7c63d14fad89365e01f5a1e623db8
size 14217004240

3
last-checkpoint/rng_state.pth Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
size 14645

1
last-checkpoint/rollout_state.json Normal file
View File

@@ -0,0 +1 @@
{"rows_consumed": 56, "groups_trained": 18}

3
last-checkpoint/scheduler.pt Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8ab327a1048744a1568a3c4ad6453c9b156bc3dcfcd06c5a301e6c619985da55
size 1465

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

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:322664cdc3082b6eba003af5228a77ca1d7936d402e584ecde8f15d3d98bdb72
size 11421911

26
last-checkpoint/tokenizer_config.json Normal file
View File

@@ -0,0 +1,26 @@
{
"backend": "tokenizers",
"bos_token": "<｜begin▁of▁sentence｜>",
"clean_up_tokenization_spaces": false,
"eos_token": "<｜end▁of▁sentence｜>",
"is_local": false,
"legacy": true,
"local_files_only": false,
"model_max_length": 16384,
"pad_token": "<｜end▁of▁sentence｜>",
"response_template": {
"defaults": {
"role": "assistant"
},
"fields": {
"content": {
"close_pattern": "<｜end▁of▁sentence｜>\\s*",
"content": "text"
}
},
"start_anchor": "<｜Assistant｜><think>\n"
},
"sp_model_kwargs": {},
"tokenizer_class": "TokenizersBackend",
"unk_token": null
}

268
last-checkpoint/trainer_state.json Normal file
View File

@@ -0,0 +1,268 @@
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5,
"eval_steps": 500,
"global_step": 8,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"buffer_qsize": 0.0,
"clip_ratio/high_max": 3.222041777917184e-05,
"clip_ratio/high_mean": 3.222041777917184e-05,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 3.222041777917184e-05,
"completions/mean_length": 3876.25,
"entropy": 0.9389231279492378,
"epoch": 0.25,
"forward_time_s": 0.30089330673217773,
"generation_tok_per_s": 6915.577392578125,
"grad_norm": 0.36549293994903564,
"kl": 0.0003301510150777176,
"learning_rate": 1e-06,
"loss": 0.02711719088256359,
"queue_wait_time_s": 2.0328194051980972,
"ratio": 1.0000957623124123,
"reward": 0.4375,
"reward_std": 0.49206146597862244,
"rewards/r1_distill_math_reward": 0.4375,
"scoring_time_ms": 18.674388885498047,
"step": 1,
"step_time": 6.262407066999003,
"train_seq_len": 4371.125,
"training_tok/s": 12972.816542826044,
"wait_scoring_ms": 0.20329749584197998,
"weight_sync_time_s": 0.14123249053955078
},
{
"buffer_qsize": 3.5,
"clip_ratio/high_max": 1.6543144738534465e-05,
"clip_ratio/high_mean": 1.6543144738534465e-05,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 1.6543144738534465e-05,
"completions/mean_length": 6120.375,
"entropy": 0.7296525910496712,
"epoch": 0.5,
"forward_time_s": 0.25667285919189453,
"generation_tok_per_s": 13383.16943359375,
"grad_norm": 0.2867599427700043,
"iteration_time_s": 7.6814478730084375,
"kl": 0.00025582036323612556,
"learning_rate": 1e-06,
"loss": 3.546476364135742e-05,
"queue_wait_time_s": 0.3530673235654831,
"ratio": 1.0000218078494072,
"reward": 0.75,
"reward_std": 0.4330126941204071,
"rewards/r1_distill_math_reward": 0.75,
"scoring_time_ms": 5.158496618270874,
"step": 2,
"step_time": 7.5191129069717135,
"train_seq_len": 6374.125,
"training_tok/s": 13318.513123161385,
"wait_scoring_ms": 5.808052182197571
},
{
"buffer_qsize": 18.0,
"clip_ratio/high_max": 7.113443098205607e-05,
"clip_ratio/high_mean": 7.113443098205607e-05,
"clip_ratio/low_mean": 1.5646326573914848e-05,
"clip_ratio/low_min": 1.5646326573914848e-05,
"clip_ratio/region_mean": 8.678075755597092e-05,
"completions/mean_length": 6053.8125,
"entropy": 0.8014494031667709,
"epoch": 0.75,
"forward_time_s": 0.2557235360145569,
"generation_tok_per_s": 13382.43359375,
"grad_norm": 0.3024653494358063,
"iteration_time_s": 7.571961241017561,
"kl": 0.00028099300106987357,
"learning_rate": 1e-06,
"loss": 0.00955304503440857,
"queue_wait_time_s": 0.00046597421169281006,
"ratio": 1.0000539049506187,
"reward": 0.5625,
"reward_std": 0.38186579942703247,
"rewards/r1_distill_math_reward": 0.5625,
"scoring_time_ms": 0.7072950005531311,
"step": 3,
"step_time": 7.4889817090006545,
"train_seq_len": 6390.375,
"training_tok/s": 13393.731171849211,
"wait_scoring_ms": 12.160233974456787
},
{
"buffer_qsize": 34.0,
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 4.13285370086669e-05,
"clip_ratio/low_min": 4.13285370086669e-05,
"clip_ratio/region_mean": 4.13285370086669e-05,
"completions/mean_length": 6813.625,
"entropy": 0.811307743191719,
"epoch": 1.0,
"forward_time_s": 0.2839837670326233,
"generation_tok_per_s": 13381.1396484375,
"grad_norm": 0.2735491394996643,
"iteration_time_s": 8.428669046988944,
"kl": 0.00027251767460256815,
"learning_rate": 1e-06,
"loss": 0.038605645298957825,
"queue_wait_time_s": 0.0004654228687286377,
"ratio": 0.9999778047204018,
"reward": 0.25,
"reward_std": 0.4330126941204071,
"rewards/r1_distill_math_reward": 0.25,
"scoring_time_ms": 1.8631829619407654,
"step": 4,
"step_time": 8.344294181006262,
"train_seq_len": 7421.5,
"training_tok/s": 13408.494449990198,
"wait_scoring_ms": 14.634688377380371,
"weight_sync_time_s": 0.15913748741149902
},
{
"buffer_qsize": 0.0,
"clip_ratio/high_max": 1.8846135390049312e-05,
"clip_ratio/high_mean": 1.8846135390049312e-05,
"clip_ratio/low_mean": 1.011736094369553e-05,
"clip_ratio/low_min": 1.011736094369553e-05,
"clip_ratio/region_mean": 2.8963496333744843e-05,
"completions/mean_length": 5484.0,
"entropy": 0.7321677580475807,
"epoch": 0.125,
"forward_time_s": 0.3462470471858978,
"generation_tok_per_s": 9634.451171875,
"grad_norm": 0.28255149722099304,
"kl": 0.00025452028421568684,
"learning_rate": 1e-06,
"loss": -0.004137326031923294,
"queue_wait_time_s": 1.8160016685724258,
"ratio": 1.0000781491398811,
"reward": 0.5625,
"reward_std": 0.38186579942703247,
"rewards/r1_distill_math_reward": 0.5625,
"scoring_time_ms": 13.104030728340149,
"step": 5,
"step_time": 7.895209455047734,
"train_seq_len": 5787.375,
"training_tok/s": 13242.969307036337,
"wait_scoring_ms": 0.1778009943664074,
"weight_sync_time_s": 0.13536810874938965
},
{
"buffer_qsize": 3.0,
"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/mean_length": 6134.3125,
"entropy": 0.6220929026603699,
"epoch": 0.25,
"forward_time_s": 0.26168057322502136,
"generation_tok_per_s": 11878.65380859375,
"grad_norm": 0.17205984890460968,
"iteration_time_s": 7.687154909974197,
"kl": 0.00023902823340904433,
"learning_rate": 1e-06,
"loss": 0.0026915371417999268,
"queue_wait_time_s": 0.3177753835916519,
"ratio": 0.9999898672103882,
"reward": 0.375,
"reward_std": 0.21650634706020355,
"rewards/r1_distill_math_reward": 0.375,
"scoring_time_ms": 8.408703327178955,
"step": 6,
"step_time": 7.5453005689778365,
"train_seq_len": 6619.25,
"training_tok/s": 13203.215699232549,
"wait_scoring_ms": 0.5662760138511658
},
{
"buffer_qsize": 18.0,
"clip_ratio/high_max": 5.6621894145791885e-05,
"clip_ratio/high_mean": 5.6621894145791885e-05,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 5.6621894145791885e-05,
"completions/mean_length": 5713.375,
"entropy": 0.8666651099920273,
"epoch": 0.375,
"forward_time_s": 0.2426280975341797,
"generation_tok_per_s": 13567.3173828125,
"grad_norm": 0.30514439940452576,
"iteration_time_s": 7.157755050022388,
"kl": 0.0002972106722154422,
"learning_rate": 1e-06,
"loss": 0.015170708298683167,
"queue_wait_time_s": 0.0004561096429824829,
"ratio": 1.0000762566924095,
"reward": 0.75,
"reward_std": 0.40742091834545135,
"rewards/r1_distill_math_reward": 0.75,
"scoring_time_ms": 2.2001885175704956,
"step": 7,
"step_time": 7.077003714046441,
"train_seq_len": 5968.5,
"training_tok/s": 13234.408437249378,
"wait_scoring_ms": 14.30916166305542
},
{
"buffer_qsize": 34.0,
"clip_ratio/high_max": 9.804690307646524e-06,
"clip_ratio/high_mean": 9.804690307646524e-06,
"clip_ratio/low_mean": 3.2692873901396524e-05,
"clip_ratio/low_min": 3.2692873901396524e-05,
"clip_ratio/region_mean": 4.249756420904305e-05,
"completions/mean_length": 5408.0625,
"entropy": 0.9108624383807182,
"epoch": 0.5,
"forward_time_s": 0.23351335525512695,
"generation_tok_per_s": 13553.60693359375,
"grad_norm": 0.3738342225551605,
"iteration_time_s": 6.897600525000598,
"kl": 0.0003173556542606093,
"learning_rate": 1e-06,
"loss": 0.0034758001565933228,
"queue_wait_time_s": 0.0004056990146636963,
"ratio": 0.9998656511306763,
"reward": 0.5,
"reward_std": 0.4330126941204071,
"rewards/r1_distill_math_reward": 0.5,
"scoring_time_ms": 18.20103907585144,
"step": 8,
"step_time": 6.822746234043734,
"train_seq_len": 5848.75,
"training_tok/s": 13207.699007613326,
"wait_scoring_ms": 35.41268348693848,
"weight_sync_time_s": 0.23273587226867676
}
],
"logging_steps": 1,
"max_steps": 8,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"save_steps": 4,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 6909415142986752.0,
"train_batch_size": 2,
"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 sha256:5ab9aeab1b506ada6e8f21d2b4d576c5b644e131107ebbd7055c5587655acf4b
size 6033

3
model.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:52f925bd55fdae95b0027ece929c8659de0e81f79d22d76ff9535ba7a2acda90
size 7108390424

3
tokenizer.json Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:322664cdc3082b6eba003af5228a77ca1d7936d402e584ecde8f15d3d98bdb72
size 11421911

26
tokenizer_config.json Normal file
View File

@@ -0,0 +1,26 @@
{
"backend": "tokenizers",
"bos_token": "<｜begin▁of▁sentence｜>",
"clean_up_tokenization_spaces": false,
"eos_token": "<｜end▁of▁sentence｜>",
"is_local": false,
"legacy": true,
"local_files_only": false,
"model_max_length": 16384,
"pad_token": "<｜end▁of▁sentence｜>",
"response_template": {
"defaults": {
"role": "assistant"
},
"fields": {
"content": {
"close_pattern": "<｜end▁of▁sentence｜>\\s*",
"content": "text"
}
},
"start_anchor": "<｜Assistant｜><think>\n"
},
"sp_model_kwargs": {},
"tokenizer_class": "TokenizersBackend",
"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:d7dae1dc2eb45fb4b5a0a33e34e6ad39d4eab678a33f05e303250ca5e217bfbe
size 6033