返回模型目录
josephmayo /

Fara-7B-Abliterated-v2-GGUF

HF 源站

josephmayo/Fara-7B-Abliterated-v2-GGUF is an uncensored open-source model hosted on Hugging Face.

GGUF7Babliterationrefusal-removalqwen2_5_vlfara-7bconversational
README.md说明 / 使用方法

base_model: josephmayo/Fara-7B-Abliterated-v2 library_name: gguf tags:

  • gguf
  • abliteration
  • refusal-removal
  • qwen2_5_vl
  • fara-7b quantized: true

Fara-7B-Abliterated-v2-GGUF

GGUF quantizations of josephmayo/Fara-7B-Abliterated-v2.

QuantFile
Q4_K_M ara-7b-abliterated-q4_k_m.gguf
Q6_K ara-7b-abliterated-q6_k.gguf
Q8_0 ara-7b-abliterated-q8_0.gguf

Base

  • Original: microsoft/Fara-7B
  • Abliterated: josephmayo/Fara-7B-Abliterated-v2
  • Method: Arditi-2024 refusal-direction orthogonalization on the Qwen2.5-VL language tower.

Research artifact only. Not for harmful use.