Datasets:
| { | |
| "train_count": 10000000, | |
| "val_count": 500000, | |
| "test_count": 500000, | |
| "split_format": "hf_parquet_uint16_target_plus_padded_meta_v1", | |
| "subset_policy": { | |
| "train_count": 10000000, | |
| "val_count": 500000, | |
| "test_count": 500000 | |
| }, | |
| "rows_per_file": 32768, | |
| "compression": "zstd", | |
| "source_format": "compact_jsonl_gz_plus_target_uint16_x1000", | |
| "local_path_fields_removed": true | |
| } |