---
tags: ["manticore"]
---
openaccess-ai-collective Manticore-13b-Chat-Pyg with the Guanaco 13b qLoRa from TimDettmers applied, uncensored
---

Quantized by mindrage to 4bit GPTQ, groupsize 128, no-act-order.

Command used to quantize: 
python3 llama.py Manticore-13B-Chat-Pyg-Guanaco-GPTQ-4bit-128g.no-act-order.safetensors c4 --wbits 4 --true-sequential --groupsize 128 --save_safetensors



The model seems to have benefited from further augmentation by openaccess-ai-collective with the Guanaco qLora by TimDettmer, pending further testing.
It is very diverse and can generate output that seems eerily smart at times.


Promptimg style:

user: "USER:"
bot: "ASSISTANT:"
context: "This is a conversation between an advanced AI and a human user."