tokenized-v2-cpt-structured-v2/rag/state.json

{
  "dataset": "mkd-chanwoo/keural-rag-chatml-ko",
  "data_files": [
    "keural_synthetic_sft_shard_0000.jsonl"
  ],
  "split": "train",
  "revision": "aa9023d231a3451d25135d5aa84b12b48ca06450",
  "max_output_tokens": 50000000,
  "tokens_per_chunk": 25000000,
  "shuffle_buffer": 10000,
  "seed": 4242,
  "allowed_licenses": [
    "cc-by-4.0"
  ],
  "exclude_source_regex": null,
  "tokenizer_vocab_size": 20000,
  "tokenizer": "/workspace/KAWK50M/artifacts/tokenizer-20k-v3",
  "documents_seen": 38912,
  "tokens_written": 50000000,
  "completed_chunks": 2,
  "current_chunk_tokens": 0,
  "counts": {
    "accepted": 35723,
    "license_cc-by-4.0": 38912,
    "rejected_boilerplate_heavy": 73,
    "rejected_code_heavy": 18,
    "rejected_code_or_math_heavy": 17,
    "rejected_digit_heavy": 1650,
    "rejected_explicit_code_or_math": 55,
    "rejected_list_heavy": 322,
    "rejected_low_hangul_letter_ratio": 319,
    "rejected_low_hangul_visible_ratio": 10,
    "rejected_repeated_character_run": 9,
    "rejected_short_line_heavy": 2,
    "rejected_symbol_heavy": 560,
    "rejected_table_heavy": 154,
    "seen": 38912
  },
  "complete": true
}