huihui_ai/smallthinker-abliterated:3b-fp16

huihui_ai/ smallthinker-abliterated:3b-fp16

904 Downloads Updated 1 year ago

A new small reasoning model fine-tuned from the Qwen 2.5 3B Instruct model.

ollama run huihui_ai/smallthinker-abliterated:3b-fp16

curl http://localhost:11434/api/chat \
  -d '{
    "model": "huihui_ai/smallthinker-abliterated:3b-fp16",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='huihui_ai/smallthinker-abliterated:3b-fp16',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'huihui_ai/smallthinker-abliterated:3b-fp16',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 year ago

1 year ago

edeedd824468 · 6.2GB ·

model

archqwen2

parameters3.09B

quantizationF16

6.2GB

template

{{- range $i, $_ := .Messages }} {{- $last := eq (len (slice $.Messages $i)) 1 -}} <|im_start|>{{ .R

255B

license

Qwen RESEARCH LICENSE AGREEMENT Qwen RESEARCH LICENSE AGREEMENT Release Date: September 19, 2024 By

7.4kB

system

You are a helpful assistant.

28B

Readme

This is an uncensored version of PowerInfer/SmallThinker-3B-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