{ "vocab_size": 259, "bos_id": 256, "eos_id": 257, "counts": { "train": 45362173, "val": 1622371, "test": 3074698 }, "bytes_per_word": 9.865595756933395, "stats": { "raw": 279, "kept": 279, "dropped_short": 0, "dropped_ksr": 0, "dropped_dup": 0, "content_bytes": 50058684, "words": 5074066 }, "normalization": { "zwnj_policy": "keep", "digit_policy": "keep_native", "remove_diacritics": false }, "source": "Omarrran/KS-PRET-5M_5_million_kashmiri_Pretrainning_LLM_dataset_12M_tokens_2026", "dtype": "uint16" }