Muse-Glimmer-30B-abliterated

Muse-Glimmer-30B-abliterated

Muse-Glimmer-30B with its refusal direction removed, produced with Heretic. BF16, 13 shards, 59.55 GB.

Abliteration identifies the direction in the model's residual stream that corresponds to refusing a request, and subtracts that component from the weights. It is a direct edit to the checkpoint, not fine-tuning: no gradient steps, no training data.

Result

KL divergence measures how far the abliterated model's next-token distribution has moved from the original on prompts it would never have refused — the collateral cost of the edit. Heretic warns that values above ~0.5 usually indicate significant damage to the original model's capabilities; this is well under half that.

How this configuration was chosen

Heretic's abliteration is parameterised by how hard to cut, where in the layer stack the cut peaks, how far it spreads, and which layer the refusal direction is read from. Those interact, so the configuration was selected by search rather than by hand, in two stages:

1. 120 trials were evaluated against a fixed 1,000-prompt subsample, three workers sharing one Optuna study. Cheap enough to explore the space.
2. The 15 best candidates were then re-measured against the full 7,011 harmful / 12,000 harmless sets.

The second stage was not a formality. Refusal count is a rare-event count, so a 1,000-prompt estimate is noisy: candidates the first stage ranked 23 against 34 came back 261 against 259 when measured properly — a genuine inversion. KL, being a smooth statistic, reproduced almost exactly at both sample sizes.

The winning configuration removed 3× more refusals than the runner-up while doing less damage to the model (KL 0.224 against 0.334), so it is not simply the most aggressive setting available.

Parameters

Direction scope global, direction index 36.589359.

Note that min_weight is nearly equal to max_weight in both components: the winning cut is close to uniform across the layers it touches, rather than a narrow peak.

Integrity

Verified before publication: 13 shards in the index, 13 on disk, no missing shards, no orphans, declared size matching the bytes on disk exactly, and every shard's safetensors header parsing with precisely the tensor names its index entry claims.

Use with vLLM

vllm serve dudeman2512/Muse-Glimmer-30B-abliterated

Caveat

This model has had its refusal behaviour removed. It will attempt requests the original declines. Whatever guardrails you need belong at the application layer. I'm not responsible for what or how you use it for and it is not designed for malicious use cases.