ardauzunoglu/c4_lowq_200m2b_qwen3_8b_rewrites

--- dataset_info: features: - name: row_idx dtype: int64 - name: sample_idx dtype: int64 - name: text dtype: string - name: finish_reason dtype: string - name: stop_reason dtype: int64 - name: document dtype: string - name: prompt_token_count dtype: int64 splits: - name: train num_bytes: 2765163205 num_examples: 800385 download_size: 1566520638 dataset_size: 2765163205 configs: - config_name: default data_files: - split: train path: data/train-* ---