huihui_ai/tinyr1-abliterated:32b-preview-q8

huihui_ai/ tinyr1-abliterated:32b-preview-q8_0

1,540 Downloads Updated 1 year 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

ollama run huihui_ai/tinyr1-abliterated:32b-preview-q8_0

curl http://localhost:11434/api/chat \
  -d '{
    "model": "huihui_ai/tinyr1-abliterated:32b-preview-q8_0",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='huihui_ai/tinyr1-abliterated:32b-preview-q8_0',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'huihui_ai/tinyr1-abliterated:32b-preview-q8_0',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 year ago

1 year ago

3d2fadf6c7f6 · 35GB ·

model

archqwen2

parameters32.8B

quantizationQ8_0

35GB

template

{{- if gt (len .Tools) 0 }} <|im_start|>system {{- if and (gt (len .Messages) 0) (eq (index .Message

1.7kB

license

apache 2.0

10B

params

{ "stop": [ "<｜begin▁of▁sentence｜>", "<｜end▁of▁sentence｜>",

148B

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