haoranli-ml/sanity_check_subset_single_policy_run

--- dataset_info: features: - name: problem dtype: string - name: answer dtype: string - name: prefix dtype: string - name: prefix_end_index dtype: int64 - name: response dtype: string - name: original_mean_reward_of_question dtype: float64 - name: original_mean_reward_source dtype: string - name: sources list: string - name: correct dtype: bool - name: difficulty dtype: string - name: prefix_tokens dtype: int64 - name: branch_rollouts list: string - name: branch_rewards list: int64 - name: branch_mean_reward dtype: float64 - name: gemini_summary_of_future dtype: string - name: index dtype: int64 - name: row_id dtype: int64 - name: prefix_len dtype: int64 splits: - name: train num_bytes: 30858863 num_examples: 42 download_size: 29962741 dataset_size: 30858863 configs: - config_name: default data_files: - split: train path: data/train-* ---