Honggy/llama_3_1_8B_Instruct_FFT_on_KMMLU

Llama-3.1-8B-Instruct-FFT-on-KMMLU

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the HAERAE-HUB/KMMLU dataset.

Model Description

• Model Type: Causal Language Model
• Base Model: meta-llama/Llama-3.1-8B-Instruct
• Language: Korean (한국어)
• License: llama3.1
• Fine-tuning Method: Full Fine-Tuning (FFT)

Dataset

This model was fine-tuned using the KMMLU (Korean Massive Multitask Language Understanding) dataset to improve its performance in Korean general knowledge and reasoning tasks.