← back to catalog · registered 2026-09-27 14:57

Johneeee/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-oQ6e_g128-fp16-text

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Johneeee%2FQwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-oQ6e_g128-fp16-text"
Response includes
  • classification m3
  • files 15
  • author_summary 82 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · 30-day
0
Likes
0
Model age
today
created 2026-09-27

Metadata

Tags
mlx safetensors qwen3_5 oq quantized 6-bit region:us

Related

Total size
20.0 GB
Files
15
Quantizations
1
Registered
2026-09-27 14:57
Last updated on HF
2026-09-27 15:03

Files by quantization

Auxiliary files 15 files 20.0 GB
model-00003-of-00005.safetensors 4.69 GB 5740d157 download
model-00001-of-00005.safetensors 4.68 GB d18c7b53 download
model-00004-of-00005.safetensors 4.68 GB db557730 download
model-00002-of-00005.safetensors 4.66 GB 185c21a4 download
model-00005-of-00005.safetensors 1.24 GB 61bbc5cc download
tokenizer.json 19.1 MB 6f32ce20 download
vocab.json 6.41 MB 0aa0ce06 download
model.safetensors.index.json 178 KB 179dbfa2 download
oq_imatrix_report.json 30.1 KB f4684b0e download
config.json 18.8 KB 6e2d0371 download
tokenizer_config.json 14.9 KB 5cc018ff download
chat_template.jinja 11.5 KB 82faea87 download
README.md 2.33 KB 5c7db3ea download
.gitattributes 1.53 KB 52373fe2 download
generation_config.json 213 B d04042de download

README current version from Hugging Face


library_name: mlx
tags:

  • mlx
  • oq
  • quantized

Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-oQ6e_g128-fp16-text

This model was quantized using oQ (oMLX v0.7.0.dev4) mixed-precision quantization.

Quantization details

  • Model type: qwen3_5
  • Bits: 6
  • Group size: 128
  • Format: MLX safetensors

"Done. Perplexity + KLD fidelity, all four on wikitext-2-raw (51,100 positions, top-512 support, fp32 logits) vs your oQ8e-mtp standard:

variant perplexity dPer% fwdKLD revKLD JSD top1
oQ8e-mtp (base) 7.9190 0 - - - 1.0000

oQ6e-mtp 7.9215 +0.032 0.0013 10.62 5.31 0.9839

oQ6e-g128-text 7.9233 +0.055 0.0017 10.88 5.44 0.9805

oQ63e-text 7.9401 +0.267 0.0031 11.78 5.89 0.9764

Read: lower Δperp, lower fwd KLD / JSD, higher top-1 = closer to your fp16-adjacent reference.

The gist:

  • oQ6e-mtp is the best of the three — nearly indistinguishable from base (Δperp 0.032%). MTP-headed, ~5.6s/window.
  • oQ6e-g128-text is a hair behind (Δperp 0.055%). Surprisingly, the wider g128 group holds up basically at parity — and it's the fastest (4.7s/window, no MTP head) — so it's the best size/speed tradeoff.
  • oQ63e-text is the worst: Δperp 0.27%, roughly 8-10× the divergence of the two oQ6 builds. Still usable, but it's distinctly the lowest-fidelity.

One thing to note: the -mtp tag is nominal here — all four configs report mtp_layers:[], so the model output is a single (1, seq, 248320) array and my comparison used the main-LM logits throughout. No MTP-head bias to correct.

The KLD asymmetry is the usual oQ signature: fwd KLD stays tiny (0.001–0.003 nat) because the reference covers what the quant agrees on; reverse KLD is large (10.6–11.8) because quantization concentrates probability on a few tokens the fp16 reference barely registers. Perplexity and top-1 divergences tell the practical story — all three are comfortably deployable, but if you want the closest-to-standard quality, take oQ6e-mtp; if you want speed at negligible cost, oQ6e-g128-text.

Artifacts: /Users/juha/.hermes/cache/scratch/kld_results/*.json (full per-variant numbers), kld_ref.npz + wikitext_windows.npy (reusable reference). Skill updated with the rebuilt harness entry points."

Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.