Taywon/dolma3_pool_stratified_1k

--- dataset_info: features: - name: doc_id dtype: string - name: bin_id dtype: int64 - name: weborganizer_topic dtype: string - name: weborganizer_format dtype: string - name: estimated_token_count dtype: int64 - name: shard_path dtype: string - name: year dtype: float64 - name: url_domain dtype: string - name: bin_classification dtype: string - name: sample dtype: string - name: text dtype: string splits: - name: train num_bytes: 4498221 num_examples: 1000 download_size: 2524947 dataset_size: 4498221 configs: - config_name: default data_files: - split: train path: data/train-* ---