huihui_ai/
tinyr1-abliterated:32b-preview

1,281 7 months ago

Tiny-R1-32B-Preview, which outperforms the 70B model Deepseek-R1-Distill-Llama-70B and nearly matches the full R1 model in math.

tools 32b

7 months ago

dffc6569dea2 · 20GB ·

model
qwen2
·
32.8B
·
Q4_K_M
template
{{- if gt (len .Tools) 0 }} <|im_start|>system {{- if and (gt (len .Messages) 0) (eq (index .Message
license
apache 2.0
params
{ "stop": [ "<｜begin▁of▁sentence｜>", "<｜end▁of▁sentence｜>",

Readme

This is an uncensored version of qihoo360/TinyR1-32B-Preview 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.

References

Huggingface

x.com/support_huihui

Donation

Your donation helps us continue our further development and improvement, a cup of coffee can do it.
  • bitcoin:
  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge