Downloads · 30 days
0
Gander-heretic-ARA
justsumguy/Gander-heretic-ARA
Gander-heretic-ARA is a machine learning model from justsumguy. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.
Gander-Omni/Gander, release Gander-v63-Unit8 on MiniCPM-o 4.5, with refusal behaviour reduced by abliteration of the Thinker's language tower. It is BF16 and keeps upstream's layout, thinker/ and talker/, so it drops…
Access
Public
Updated Sep 23, 2026
Repo size
20 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors18.7 GB · 94%
At a glance
- Library
- transformers
- License
- apache-2.0
- Access
- Public
- Created
- Sep 23, 2026
- Updated
- Sep 23, 2026
- SHA
- 2be955d6
Base models
Model card
From the Hugging Face model README
Gander-heretic-ARA
Gander-Omni/Gander, release Gander-v63-Unit8 on
MiniCPM-o 4.5, with refusal behaviour reduced by abliteration of the Thinker's language tower.
It is BF16 and keeps upstream's layout, thinker/ and talker/, so it drops into the
Gander runtime the same
way as the upstream release. An NVFP4 build is at
justsumguy/Gander-heretic-ARA-NVFP4.
What changed
Only the Thinker's language tower (llm.*, Qwen3-8B shape, 36 layers). The Thinker's vision
tower, audio tower, resampler and audio projection are upstream's, unmodified. So are the
Talker (talker/) and its token2wav assets.
- 72 matrices were changed:
self_attn.o_projandmlp.down_projin all 36 layers. - The largest relative change is ‖ΔW‖/‖W‖ = 0.135.
Abliteration
- Tool: heretic, the unmerged
arabranch atedc3b12(Arbitrary-Rank Ablation). - Mode: ARA-LoRA with rank-128 adapters. After selection, the adapter delta was merged into the BF16 Thinker weights.
- Search setup:
- The language tower was extracted as a standalone Qwen3 model and loaded in bitsandbytes NF4 during the search.
- Hardware: RTX 5060 Ti 16 GB.
- Thinking was disabled in the chat template. System prompt:
You are a helpful assistant.
- Optimiser: 150 trials, 45 of them startup trials. KL target 0.01, row normalisation off, direction not orthogonalised, maximum response length 100.
- Prompts:
- Direction:
mlabonne/harmful_behaviorstrain[:400] andmlabonne/harmless_alpacatrain[:400]. - Evaluation: test[:100] of each.
- Direction:
Selected trial: 4.
| Parameter | Value |
|---|---|
| layers | 6–31 |
| preserve_good_behavior_weight | 0.5598 |
| steer_bad_behavior_weight | 0.0008 |
| overcorrect_relative_weight | 0.4266 |
| neighbor_count | 3 |
| Metric | Refusals (of 100) | KL divergence |
|---|---|---|
| Base model | 98 | – |
| Study (trial 4) | 3 | 0.0325 |
| Recheck (adapter re-fit for export) | 7 | 0.1276 |
The ARA-LoRA re-fit is not bit-reproducible, so the recheck numbers are the ones that describe these weights.
The refusal evaluation is text-only, through the extracted language tower with a chat template. It does not exercise Gander's duplex unit format.
No healing step (fine-tuning after ablation) was applied. Post-abliteration stability needs further testing, including in full-duplex sessions, which have not been evaluated on these weights.
Usage
As upstream: point the Gander runtime's duplex.checkpoint at thinker/ and
duplex.talker_checkpoint at talker/. The unit contract is unchanged: 8 Thinker text tokens
and 50 S3 speech tokens per 1 s unit.
License
Apache-2.0, as upstream. See thinker/LICENSE and thinker/NOTICE. The modification is the abliteration of the Thinker's language tower described above.
- Library
- transformers
- License
- apache-2.0
- Created
- Sep 23, 2026
- Updated
- Sep 23, 2026