Base Model:

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

Model Description

We use QLoRA(64, 16) and NEFTune on LLM Fine-tuning lr=1e-5

Our Korean data used, and 1 epoch train

We are making LLM model for Kolon !

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Quantizations