Honest receipt — catcher-gated (shipped only because it passed EVERY gate)
Table with columns: gate, pristine, uncensored| gate | pristine | uncensored |
|---|
| refusal compliance | 0.060 | 1.000 |
| MMLU-val accuracy | 0.642 | 0.650 (Δ +0.007, held) |
| code-switch rate | 0.000 | 0.000 |
| degeneration rate | 0.000 | 0.000 |
| instruction-following | 1.000 | 1.000 |
Kept config: A:frac=0.5 (64 residual-write matrices edited). The gate rejects any
config that raises refusals but drops capability or degrades generation
(code-switching, empty/looping output, broken instruction-following). MMLU alone is
argmax-blind, so the generative gate is what keeps these coherent — a model that
code-switches or loops is not shipped.
Weights
- bf16 safetensors — full precision, intended as a fine-tuning base. Hidden
states stay readable (logit-lens compatible; a GGUF quant would not).
Method: zero training, zero fine-tuning — pure activation-derived weight excision,
gated on compliance and capability and generation quality.