Ashima/qwen3_0.6b-task738_augmented_many-hop_reasoning_Mar16-1639.json
收藏数据链接：
官方服务：
资源简介：
--- dataset_info: features: - name: problem dtype: string - name: solution dtype: string - name: win_rate dtype: float64 - name: completions list: string - name: rewards list: float64 - name: num_completions dtype: int64 splits: - name: train num_bytes: 19821814 num_examples: 1000 download_size: 5956341 dataset_size: 19821814 configs: - config_name: default data_files: - split: train path: data/train-* ---
提供机构：
Ashima
