README.md

metadata

base_model: failspy/Meta-Llama-3-8B-Instruct-abliterated-v3
library_name: transformers
license: llama3
tags:
  - mlx

0xBreath/Meta-Llama-3-8B-Instruct-abliterated-v3-q8-mlx

The Model 0xBreath/Meta-Llama-3-8B-Instruct-abliterated-v3-q8-mlx was converted to MLX format from failspy/Meta-Llama-3-8B-Instruct-abliterated-v3 using mlx-lm version 0.19.0.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("0xBreath/Meta-Llama-3-8B-Instruct-abliterated-v3-q8-mlx")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)