Mistral Small 3 sets a new benchmark in the “small” Large Language Models category below 70B.
tools
24b
1,782 Pulls Updated 5 weeks ago
Updated 5 weeks ago
5 weeks ago
a52475f69130 · 25GB
model
archllama
·
parameters23.6B
·
quantizationQ8_0
25GB
params
{
"temperature": 0.15
}
21B
system
You are Mistral Small 3, a Large Language Model (LLM) created by Mistral AI, a French startup headqu
644B
template
{{- range $index, $_ := .Messages }}
{{- if eq .Role "system" }}[SYSTEM_PROMPT]{{ .Content }}[/SYSTE
695B
license
Apache License
Version 2.0, January 2004
11kB
Readme
This is an uncensored version of mistralai/Mistral-Small-24B-Instruct-2501 created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.