---
title: devbox_32768
canonical_url: "https://www.modelscope.cn/models/shiw1141/devbox_32768"
md_url: "https://www.modelscope.cn/models/shiw1141/devbox_32768.md"
repository: shiw1141/devbox_32768
last_updated: 2026-08-03
license: other
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - qwen3
architectures:
  - Qwen3ForCausalLM
base_model:
  - /root/models/OneReason-8B-pretrain-competition
base_model_relation: finetune
parameters: 8.4B
tensor_type:
  - BF16
library_name:
  - safetensors
downloads: 27
stars: 0
tags:
  - llama-factory
  - full
  - generated_from_trainer
---

# devbox_32768

> devbox_32768 - shiw1141 在 ModelScope 开源的模型。This model is a fine-tuned version of /root/models/OneReason-8B-pretrain-competition on the datafinal dataset.

shiw1141/devbox_32768 是 ModelScope 魔搭社区上的 8.4B 参数text-generation模型，采用 other 许可，基于 /root/models/OneReason-8B-pretrain-competition 构建。

- **Repository**: shiw1141/devbox_32768
- **License**: other
- **Tasks**: text-generation
- **Parameters**: 8.4B
- **Base model**: /root/models/OneReason-8B-pretrain-competition
- **Tags**: llama-factory, full, generated_from_trainer
- **Downloads**: 27
- **Stars**: 0
- **Last updated**: 2026-08-03

Source: https://www.modelscope.cn/models/shiw1141/devbox_32768

---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# 03-full-32768

This model is a fine-tuned version of [/root/models/OneReason-8B-pretrain-competition](https://huggingface.co//root/models/OneReason-8B-pretrain-competition) on the data_final dataset.

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 20260701
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- total_eval_batch_size: 4
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 0.03
- num_epochs: 1

### Training results



### Framework versions

- Transformers 5.6.0
- Pytorch 2.5.1+cu124
- Datasets 4.0.0
- Tokenizers 0.22.2