yhyu13
Upload model
2b5cd1b
|
raw
history blame
614 Bytes
metadata
license: apache-2.0

GPTQ 4-bit no actor version for compatibility that works in textgen-webui

Generated by using scripts from https://gitee.com/yhyu13/llama_-tools

Original weights: https://huggingface.co./openaccess-ai-collective/manticore-13b


Manticore is by far the most satisfying LLaMA model I've used due to its fine-tuning datasets are generated from the best of Vicuna, Wizard, and ShareGPT/GPT4.

Here is a conversation generated in textgen-webui, it shows step by step Chain of Thoughts comments along side generating complex code sample in an engaging tone!

sample