Rashik24/wmt24pp-en-bn

该数据集是`google/wmt24pp`的一个过滤子集，专门用于英语-孟加拉语翻译任务。通过应用特定的过滤规则（如保留`is_bad_source`为false的行、确保`source`和`target`非空），从原始998行数据中筛选出960行有效数据。数据集保留了原始JSONL文件的列结构。This dataset is a filtered subset of `google/wmt24pp` specifically for English-Bengali translation tasks. By applying specific filtering rules (e.g., keeping rows where `is_bad_source` is false, ensuring `source` and `target` are non-empty), 960 valid rows were selected from the original 998 rows. The dataset preserves the column structure of the original JSONL file.