README.md

---
license: other
language:
- ko
- en
tags:
- lfm
- korean
- sft
- lfm-chat
- jsonl
---

# LFM2.5-KO-Dataset-Index-and-Sources

Snapshot of gyunggyung/LLM-Ko-Datasets README/LICENSE used as a dataset index while building the SFT mix.

This dataset is part of the `LFM2.5-8B-A1B-KO-SFT` / Agentic SFT workflow.

- Main SFT model: https://huggingface.co/LLM-OS-Models/LFM2.5-8B-A1B-KO-SFT
- CPT base model: https://huggingface.co/LLM-OS-Models/LFM2.5-8B-A1B-KO-CPT-FULL
- Agentic follow-up model: https://huggingface.co/LLM-OS-Models/LFM2.5-8B-A1B-KO-Agentic-SFT
- SFT GitHub: https://github.com/gyunggyung/LFM25-KO-SFT
- CPT GitHub: https://github.com/gyunggyung/LFM25-KO-CPT

## Source Attribution

- Snapshot of https://github.com/gyunggyung/LLM-Ko-Datasets README/LICENSE.

Additional public references:

- Liquid LFM base model: https://huggingface.co/LiquidAI/LFM2.5-8B-A1B
- Liquid chat template docs: https://docs.liquid.ai/lfm/key-concepts/chat-template
- Liquid tool-use docs: https://docs.liquid.ai/lfm/key-concepts/tool-use
- Legalize-KR organization: https://github.com/legalize-kr
- KoTSQA v2.0: https://huggingface.co/datasets/etri-lirs/KoTSQA-v.2.0
- Korean dataset index reviewed for candidates: https://github.com/gyunggyung/LLM-Ko-Datasets

## Notes

- This is an index/reference artifact, not a training corpus by itself.
- It documents Korean dataset candidates reviewed while building the KO SFT mix.

## Summary

| field | value |
|---|---:|
| kind | `source_index` |
| sample count | n/a |
| token count | n/a |
| max sequence / sample length | n/a |
| uploaded size bytes | 106659 |

## Format

- `raw_lfm_chat_jsonl`: JSONL rows with a `text` field containing LFM ChatML-like conversation text.
- `prepared_tokenized`: NumPy response-only SFT arrays built with the LFM tokenizer:
  - `tokens.npy`
  - `epoch_0/inst_start.npy`
  - `epoch_0/inst_len.npy`
  - `epoch_0/resp_start.npy`
  - `epoch_0/resp_len.npy`
  - `tokenizer.json`

## Local Source Path

```text
/home/work/.data/lfm2_ko_sft/downloads/LLM-Ko-Datasets
```

## License And Usage Notes

This release republishes preprocessing artifacts used for the LFM2.5 Korean CPT/SFT
workflow. Source components come from multiple public or locally prepared
datasets, so downstream users should verify each upstream source license before
redistribution or commercial use. Legal and finance examples are for model
training/evaluation only and are not legal, financial, or investment advice.

## Stats

```json
{
  "path": "/home/work/.data/lfm2_ko_sft/downloads/LLM-Ko-Datasets",
  "size_bytes": 106659,
  "file_count": 31
}
```