metadata

license: apache-2.0
library_name: transformers
inference: true
tags:
  - unsloth
  - trl
  - sft

Uploaded model

Developed by: Sakalti
License: apache2.0
Finetuned from model : Sakalti/Saba1-1.8B This qwen model was trained 2x faster with Unsloth and Huggingface's TRL library.

paramaters(no embedding layer) 1.31B paramaters(yes embedding layer) 1.54B layers: 28layers

概要

Saba1.5はSaba1をファインチューニングしたモデルです。性能はまだわかりませんが上昇してるでしょう。