--- dataset_info: features: - name: trace_id dtype: string - name: example_id dtype: int64 - name: rollout_number dtype: int64 - name: task dtype: 'null' - name: prompt list: - name: role dtype: string - name: content dtype: string - name: tool_calls list: 'null' - name: tool_call_id dtype: string - name: completion list: - name: role dtype: string - name: content dtype: string - name: reasoning_content dtype: string - name: tool_calls list: 'null' - name: tool_call_id dtype: string - name: answer dtype: 'null' - name: reward dtype: float64 - name: score dtype: 'null' - name: correct dtype: 'null' - name: num_steps dtype: 'null' - name: total_time dtype: float64 - name: latency_ms dtype: int64 - name: created_at dtype: string - name: info struct: - name: id dtype: string - name: difficulty dtype: string - name: category dtype: string - name: timing struct: - name: start_time dtype: float64 - name: setup struct: - name: start dtype: float64 - name: end dtype: float64 - name: duration dtype: float64 - name: generation struct: - name: start dtype: float64 - name: end dtype: float64 - name: duration dtype: float64 - name: scoring struct: - name: start dtype: float64 - name: end dtype: float64 - name: duration dtype: float64 - name: model struct: - name: spans list: - name: start dtype: float64 - name: end dtype: float64 - name: duration dtype: float64 - name: duration dtype: float64 - name: env struct: - name: spans list: 'null' - name: duration dtype: float64 - name: total dtype: float64 - name: overhead dtype: float64 - name: token_usage struct: - name: input_tokens dtype: float64 - name: output_tokens dtype: float64 - name: final_output_tokens dtype: float64 - name: final_input_tokens dtype: float64 - name: is_completed dtype: bool - name: is_truncated dtype: bool - name: stop_condition dtype: string - name: metrics struct: - name: exact_match_reward dtype: float64 - name: num_turns dtype: float64 - name: exact_match_reward dtype: float64 splits: - name: train num_bytes: 38423 num_examples: 15 download_size: 67262 dataset_size: 38423 configs: - config_name: default data_files: - split: train path: data/train-* ---