reasoning-degeneration-dev/algorithmic-sft-training-configs-v1
收藏Hugging Face2026-03-23 更新2026-03-29 收录
下载链接：
https://hf-mirror.com/datasets/reasoning-degeneration-dev/algorithmic-sft-training-configs-v1
下载链接
链接失效反馈官方服务：
资源简介：
---
license: mit
tags:
- algorithmic-sft
- training-config
- llamafactory
- algorithmic-sft-vs-distillation
---
# algorithmic-sft-training-configs-v1
LlamaFactory training configs. All cutoff_len=32768. Countdown configs use new equation-answer format.
## Dataset Info
- **Rows**: 17
- **Columns**: 7
## Columns
| Column | Type | Description |
|--------|------|-------------|
| config_name | Value('string') | YAML filename |
| domain | Value('string') | *No description provided* |
| is_distillation | Value('bool') | *No description provided* |
| yaml_content | Value('string') | Full YAML config |
| model_name | Value('string') | *No description provided* |
| cutoff_len | Value('int64') | *No description provided* |
| dataset | Value('string') | *No description provided* |
## Generation Parameters
```json
{
"script_name": "upload after countdown redesign",
"description": "LlamaFactory training configs. All cutoff_len=32768. Countdown configs use new equation-answer format.",
"model": "Qwen/Qwen2.5-1.5B-Instruct",
"hyperparameters": {},
"input_datasets": []
}
```
## Experiment Documentation
For complete experiment details, see [https://github.com/Zayne-sprague/SC-Research-Notes/tree/main/experiments/algorithmic_sft_vs_distillation](https://github.com/Zayne-sprague/SC-Research-Notes/tree/main/experiments/algorithmic_sft_vs_distillation)
## Usage
```python
from datasets import load_dataset
dataset = load_dataset("reasoning-degeneration-dev/algorithmic-sft-training-configs-v1", split="train")
print(f"Loaded {len(dataset)} rows")
```
---
*This dataset is tracked in [reasoning-degeneration-dev/PROJECT-MANIFEST](https://huggingface.co/datasets/reasoning-degeneration-dev/PROJECT-MANIFEST)*
---
许可证：MIT许可证
标签：
- 算法监督微调（algorithmic SFT）
- 训练配置
- LlamaFactory
- 算法监督微调与知识蒸馏对比
---
# 算法监督微调训练配置v1（algorithmic-sft-training-configs-v1）
本数据集包含LlamaFactory训练配置，所有配置的截断长度（cutoff_len）均为32768，其中倒计时类配置采用新型的公式-答案格式。
## 数据集信息
- **样本条数**：17
- **字段数**：7
## 字段说明
| 字段名 | 数据类型 | 字段说明 |
|--------|----------|----------|
| config_name | 字符串 | YAML配置文件名 |
| domain | 字符串 | *暂无说明* |
| is_distillation | 布尔值 | *暂无说明* |
| yaml_content | 字符串 | 完整的YAML配置内容 |
| model_name | 字符串 | *暂无说明* |
| cutoff_len | 64位整数 | *暂无说明* |
| dataset | 字符串 | *暂无说明* |
## 生成参数
json
{
"script_name": "upload after countdown redesign",
"description": "LlamaFactory training configs. All cutoff_len=32768. Countdown configs use new equation-answer format.",
"model": "Qwen/Qwen2.5-1.5B-Instruct",
"hyperparameters": {},
"input_datasets": []
}
## 实验文档
完整实验细节请参阅：[https://github.com/Zayne-sprague/SC-Research-Notes/tree/main/experiments/algorithmic_sft_vs_distillation]
## 使用方法
python
from datasets import load_dataset
dataset = load_dataset("reasoning-degeneration-dev/algorithmic-sft-training-configs-v1", split="train")
print(f"Loaded {len(dataset)} rows")
*本数据集已在[reasoning-degeneration-dev/PROJECT-MANIFEST](https://huggingface.co/datasets/reasoning-degeneration-dev/PROJECT-MANIFEST)中进行追踪*
