Hugging Face's logo

OpenAssistant
/
oasst-rlhf-2-llama-30b-7k-steps-xor

Model card Files Community
8

Converted , Merged weights to share with everyone here

#7
by Yhyu13 - opened
Discussion
Yhyu13

https://huggingface.co./Yhyu13/llama-30B-hf-openassitant
https://huggingface.co./Yhyu13/oasst-rlhf-2-llama-30b-7k-steps-hf

gptq : https://huggingface.co./Yhyu13/oasst-rlhf-2-llama-30b-7k-steps-gptq-4bit

Yhyu13

The response is pretty good, OA is able to think step by step without explicitly prompting, we see similar behavior in chatgpt as well

Yhyu13 changed discussion status to closed
Yhyu13 changed discussion status to open

Sign up or log in to comment