README.md

metadata

license: cc-by-nc-4.0
language:
  - en
  - ko

Base Model:

We made a LLM model with yanolja/KoSOLAR-10.7B-v0.3

Model Description

We use QLoRA(128, 32) and NEFTune on LLM Fine-tuning lr=1e-5

Train Detail

Our Korean data used, and 1 epoch train

Others

We are making LLM model for Kolon !