OntologerFounder's picture
Upload folder using huggingface_hub
2aed5c9 verified
|
Raw
History Blame Contribute Delete
2.4 kB
metadata
license: apache-2.0
language:
  - en
tags:
  - qwen3
  - heretic
  - abliterated
  - uncensored
  - iq1_s
  - 2-bit
  - importance-matrix
pipeline_tag: text-generation
model-index:
  - name: FaceNet-Qwen3-8B-2Bit-Heretic
    results: []

FaceNet-Qwen3-8B-2Bit-Heretic

Qwen3-8B abliterated with Heretic, quantized to 2-bit (IQ1_S) using a 30MB importance matrix for coherence preservation.

Heretic Abliteration

  • Base model: Qwen/Qwen3-8B
  • Method: Heretic (p-e-w/heretic), 20 Optuna trials, auto-selected best
  • Refusal rate: 18/100 (down from ~99/100 baseline, 82% reduction)
  • KL divergence: 0.112 (well within safe range)
  • Abliteration applied to full-precision weights, then quantized

Quantization

  • Format: IQ1_S (importance-weighted 2-bit)
  • Bits per weight: 2.06 bpw
  • Size: 2.0 GB (from 16.4 GB F16)
  • Importance matrix: 30MB diverse corpus (Frankenstein + froggeric/imatrix + eaddario code/math/general English)
  • Speed: ~95 tok/s on NVIDIA GB10 (Blackwell)

Variants

Variant Quant Size Speed BPW
2-bit (this) IQ1_S 2.0 GB 95 t/s 2.06
3-bit Q2_K 3.1 GB 58 t/s 3.20
5-bit Q4_K_M 4.8 GB 42 t/s 4.90

Usage

from llama_cpp import Llama

llm = Llama.from_pretrained(
    repo_id="sidmishra/FaceNet-Qwen3-8B-2Bit-Heretic",
    filename="qwen3-8b-heretic-iq1_s.gguf",
    n_ctx=32768,
    n_gpu_layers=-1,
    verbose=False
)

response = llm.create_chat_completion(
    messages=[{"role": "user", "content": "Your prompt here"}],
    max_tokens=200
)

Or with llama-server:

llama-server -m qwen3-8b-heretic-iq1_s.gguf --host 0.0.0.0 --port 8080 -ngl 99

Limitations

  • Output uses reasoning_content field (Qwen3 thinking format)
  • Occasional mild repetition artifacts at 2-bit (much rarer with the 30MB imatrix)
  • For maximum quality, use the 3-bit or 5-bit variants

Acknowledgments