Hugging Face's logo

cloudyu
/
Truthful_DPO_TomGrc_FusionNet_34Bx2_MoE

Text Generation
Transformers
Safetensors
mixtral
Mixture of Experts
DPO
RL-TUNED
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
3

Can you make a 2.4bpw exl2 quantisation for this model?

#1
by xldistance - opened
Discussion
xldistance

@LoneStriker
It feels like this model is going to be strong

cloudyu
 Owner
•
edited Jan 25, 2024

please wait the test metrics.
I'm a little afraid of overfitting

cloudyu
 Owner

metrics ready
https://huggingface.co/cloudyu/Truthful_DPO_TomGrc_FusionNet_34Bx2_MoE/blob/main/34bx2-dpo.jpg

LoneStriker

quants were completed yesterday: https://huggingface.co/models?search=LoneStriker%20Truthful_DPO_TomGrc_FusionNet_34Bx2_MoE

cloudyu
 Owner

quants were completed yesterday: https://huggingface.co/models?search=LoneStriker%20Truthful_DPO_TomGrc_FusionNet_34Bx2_MoE

many thanks

Sign up or log in to comment