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
844 Pulls Updated 3 weeks ago
Updated 3 weeks ago
3 weeks ago
dffc6569dea2 · 20GB
model
archqwen2
·
parameters32.8B
·
quantizationQ4_K_M
20GB
license
apache 2.0
10B
params
{
"stop": [
"<｜begin▁of▁sentence｜>",
"<｜end▁of▁sentence｜>",
148B
template
{{- if gt (len .Tools) 0 }}
<|im_start|>system
{{- if and (gt (len .Messages) 0) (eq (index .Messag
1.7kB
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
Donation
Your donation helps us continue our further development and improvement, a cup of coffee can do it.
- bitcoin:
bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge