GPT-OS3-Chi-8B-A3B

  • Developed by: qingy2024
  • Base model: qingy2024/GPT-OS3-V2-8B-Base

GPT OSS Small (OS3) is a project to create usable and intelligent language models based on pruned GPT-OSS-20B variants by AmanPriyanshu. These are post trained with LoRA on the qingy2024/GPT-OS3-Dataset-v2 dataset to revert some of the "brain damage" due to the expert pruning.

(This is a Preview release, please don't use it or make GGUFs!)

Chi, Step 3000, V2 Dataset. Failed experiment with wrong chat template :/

Built with Axolotl

Downloads last month
11
Safetensors
Model size
8.37B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for qingy2024/GPT-OS3-Chi-8B-A3B

Finetuned
(1)
this model

Dataset used to train qingy2024/GPT-OS3-Chi-8B-A3B