---

license: cc-by-sa-4.0

base_model: johnbean393/chiboard-1-t1

tags:

• gguf
• chinese
• lfm2
• chiboard
• pinyin
• ime
• sft
• teacher

language:

• zh

---

chiboard-1-t1-GGUF

GGUF export of johnbean393/chiboard-1-t1, a 1.2B full-fine-tuned teacher initialization for Chiboard Chinese pinyin-to-Hanzi conversion and revision.

Prompt format:

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

The source model is trained with completion-only loss on {target}<|im_end|>.

Serve with exactly one <|startoftext|> token. Most runtimes add it automatically, so do not also embed it in the prompt string.

This is T1, the 1.2B teacher initialization for Chiboard.

Files

• Chiboard-T1-Q8_0.gguf: Q8_0 GGUF quantization of johnbean393/chiboard-1-t1.

Training

• Base model: LiquidAI/LFM2.5-1.2B-Base
• Datasets: johnbean393/chiboard-1-sft, johnbean393/chiboard-1-revision-sft
• Source model: johnbean393/chiboard-1-t1
• Hub target: johnbean393/chiboard-1-t1-GGUF
• Mixture: all plain SFT rows once plus all revision SFT rows twice
• Mixture order: concatenated and globally shuffled before one training pass
• Peak learning rate: 3e-5
• Cosine learning-rate floor: 3e-6

Conversion

• Source revision: 4086865d8813a01909a420579ee7b15821bf80b0
• GGUF architecture: lfm2
• GGUF version: 3
• Quantization: Q8_0
• Context length: 128000
• SHA-256: a0e3e866a6015c5d29179655cf2459537a87ed40c24c6a497ce953811564b2e7