lyadalachanchu/gpt-oss-20b-aime26-generations

--- dataset_info: - config_name: default features: - name: run_id dtype: string - name: model dtype: string - name: tokenizer dtype: string - name: dataset dtype: string - name: dataset_config dtype: 'null' - name: dataset_split dtype: string - name: question_index dtype: int64 - name: k dtype: int64 - name: sample_index dtype: int64 - name: question dtype: string - name: ground_truth_answer dtype: string - name: ground_truth_normalized dtype: string - name: prediction_normalized dtype: string - name: is_correct dtype: bool - name: num_tokens dtype: int64 - name: text dtype: string - name: cot_text dtype: string - name: answer_with_working dtype: string - name: boxed_answer dtype: string splits: - name: generations num_bytes: 63581825 num_examples: 960 download_size: 30885959 dataset_size: 63581825 - config_name: default_question_metrics features: - name: run_id dtype: string - name: model dtype: string - name: tokenizer dtype: string - name: dataset dtype: string - name: dataset_config dtype: 'null' - name: dataset_split dtype: string - name: question_index dtype: int64 - name: k dtype: int64 - name: question dtype: string - name: ground_truth_answer dtype: string - name: ground_truth_normalized dtype: string - name: solve_count dtype: int64 - name: solve_rate dtype: float64 - name: pass_at_k dtype: bool - name: avg_generation_tokens dtype: float64 - name: correct_pattern dtype: string splits: - name: question_metrics num_bytes: 24848 num_examples: 30 download_size: 19008 dataset_size: 24848 - config_name: default_summary features: - name: run_id dtype: string - name: model dtype: string - name: tokenizer dtype: string - name: dataset dtype: string - name: dataset_config dtype: 'null' - name: dataset_split dtype: string - name: questions dtype: int64 - name: k dtype: int64 - name: avg_at_k dtype: float64 - name: pass_at_k dtype: float64 - name: solved_questions dtype: int64 - name: total_correct_generations dtype: int64 - name: total_generations dtype: int64 - name: avg_generation_tokens dtype: float64 - name: avg_question_generation_tokens dtype: float64 - name: timestamp dtype: string splits: - name: summary num_bytes: 184 num_examples: 1 download_size: 7661 dataset_size: 184 configs: - config_name: default data_files: - split: generations path: data/generations-* - config_name: default_question_metrics data_files: - split: question_metrics path: default_question_metrics/question_metrics-* - config_name: default_summary data_files: - split: summary path: default_summary/summary-* ---