johnbean393/chiboard-1-s1

Chiboard-1 S1

Full fine-tune of LiquidAI/LFM2.5-350M-Base for Chinese pinyin-to-Hanzi conversion and revision. This is the 350M student initialization (S1) for the later Chiboard teacher-distillation stage; it is not the final shipped model.

Training exposure: all plain SFT rows once plus all revision SFT rows twice, concatenated and globally shuffled before one training pass.

Prompt format:

<|startoftext|>{committed_context}<|reserved_6|>{raw_pinyin}<|reserved_7|>{display}<|reserved_8|>{target}<|im_end|>

The tokenizer adds exactly one BOS. Loss is computed on the entire target plus <|im_end|> and nowhere in the prompt.

Exposure summary

{
  "eval": {
    "plain_eval_rows": 25000,
    "plain_split": "dev",
    "revision_eval_rows": 25000,
    "revision_split": "validation",
    "total_eval_rows": 50000
  },
  "revisions": {
    "model": "9960764e30892e01f29a6dc23df2533fcd8bd5ae",
    "plain": "54f9a912fce5f60df4918f83b3d9f0b5595f47b4",
    "revision": "62ffeead7a381d4d88d640a44531e3f3cbb9996e"
  },
  "train": {
    "effective_train_rows": 78233116,
    "logical_mixture_rows": 78233116,
    "max_train_samples": 0,
    "plain_row_exposure_share": 0.6970899382302502,
    "plain_source_rows": 54535518,
    "revision_repeats": 2,
    "revision_row_exposure_share": 0.30291006176974977,
    "revision_source_rows": 11848799,
    "shuffle_seed": 20260710
  }
}