lIlBrother
/

Llama-3.1-8B-Instruct-KoEn-FFT-Merge

Model card Files Files and versions Community

Llama-3.1-8B-Instruct-KoEn-FFT-Merge / README.md

lIlBrother's picture

Update README.md

4880cd4 verified about 1 month ago

|

665 Bytes

	---
	base_model:
	- unsloth/Meta-Llama-3.1-8B
	datasets:
	- lcw99/wikipedia-korean-20240501-1million-qna
	- MarkrAI/KOpen-HQ-Hermes-2.5-60K
	- garage-bAInd/Open-Platypus
	- rwkv-x-dev/openorca-gpt4
	- gbharti/finance-alpaca
	language:
	- ko
	- en
	license: llama3.1
	---

	https://arca.live/b/alpaca/118261066?p=2
	의 방법을 활용하여 만들어졌음.

	- lcw99/wikipedia-korean-20240501-1million-qna
	- MarkrAI/KOpen-HQ-Hermes-2.5-60K
	- garage-bAInd/Open-Platypus
	- rwkv-x-dev/openorca-gpt4
	- gbharti/finance-alpaca
	- 내가 직접 만든 데이터

	를 적당히 샘플링하여 만들었음.

	LogicKor 점수는 가장 높음.

	해당 모델은 DPO 학습되지 않았음.