yousan

← Back to Works

Japanese 0.3B Mistral pretrained model

A 0.3B Japanese base model pretrained with the Mistral architecture.

Model card image on Hugging Face
Date
Tech
LLM, Mistral, Pretraining, Hugging Face

Overview

I pretrained a 0.3B Japanese base model with the Mistral architecture and published it on Hugging Face. The work was done in April 2024.

What I did

I ran the pretraining and published the model.

Links