huihui_ai/tess-r1-abliterated:70b-q8

huihui_ai/ tess-r1-abliterated:70b-q8_0

1,648 Downloads Updated 11 months ago

Tess-R1 is designed with test-time compute in mind, and has the capabilities to produce a Chain-of-Thought (CoT) reasoning before producing the final output.

70b

ollama run huihui_ai/tess-r1-abliterated:70b-q8_0

curl http://localhost:11434/api/chat \
  -d '{
    "model": "huihui_ai/tess-r1-abliterated:70b-q8_0",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='huihui_ai/tess-r1-abliterated:70b-q8_0',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'huihui_ai/tess-r1-abliterated:70b-q8_0',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 11 months ago

11 months ago

951d115398ba · 75GB ·

model

archllama

parameters70.6B

quantizationQ8_0

75GB

template

{{- if .System }}<|start_header_id|>system<|end_header_id|> {{ .System }}<|eot_id|> {{- end }} {{- r

249B

system

You are Tess-R1, an advanced AI that was created for complex reasoning. Given a user query, you are

431B

params

{ "stop": [ "<|end_of_text|>", "<|eot_id|>" ], "temperature": 0.75,

102B

Readme

This is an uncensored version of migtissera/Tess-R1-Limerick-Llama-3.1-70B 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

Donation

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

bitcoin:

  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge