huihui_ai/aya-expanse-abliterated

huihui_ai/ aya-expanse-abliterated

2,113 Downloads Updated 1 year ago

Cohere For AI's language models trained to perform well across 23 different languages.

tools 8b 32b

ollama run huihui_ai/aya-expanse-abliterated

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

from ollama import chat

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

import ollama from 'ollama'

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

Applications

Claude Code ollama launch claude --model huihui_ai/aya-expanse-abliterated

Codex ollama launch codex --model huihui_ai/aya-expanse-abliterated

OpenCode ollama launch opencode --model huihui_ai/aya-expanse-abliterated

OpenClaw ollama launch openclaw --model huihui_ai/aya-expanse-abliterated

Models

View all →

Name

9 models

Size

Context

Input

aya-expanse-abliterated:latest

5.1GB · 8K context window · Text · 1 year ago

aya-expanse-abliterated:latest

5.1GB

Text

aya-expanse-abliterated:8b

latest

5.1GB · 8K context window · Text · 1 year ago

aya-expanse-abliterated:8b latest

5.1GB

Text

aya-expanse-abliterated:32b

20GB · 8K context window · Text · 1 year ago

aya-expanse-abliterated:32b

20GB

Text

Readme

This is an uncensored version of CohereForAI/aya-expanse 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