Verified Patch Catalog (P0 Config & P1 Weights)
Every patch is pre-registered, evaluated across 10,000 paired bootstrap and permutation tests, checked against the secret perturbed WSP-Holdout split and StrongREJECT/XSTest safety gates, and signed with Ed25519.
Qwen2.5-Coder-1.5B Repetition-Penalty Identifier & JSON Brace Fix
Neutralizes repetition_penalty=1.1 in Qwen/Qwen2.5-Coder-1.5B-Instruct generation_config.json that corrupts repeated variable names and nested closing braces in structured code/JSON output.
$ pip install wsphub && wsp apply Qwen/Qwen2.5-Coder-1.5B-Instruct --patch wsp-qwen2.5-coder-1.5b-rep-penalty-syntax-fix --out ./patched
Qwen3.5-0.8B 32k Thinking-Loop Surgical Cure (MIT CSAIL Exp-8)
Eliminates infinite 32,768-token chain-of-thought looping in Qwen/Qwen3.5-0.8B by modifying 64 output columns of mlp.down_proj in layers 15 and 19 (131,072 bf16 weights, 0.016% of parameters). Lifts independent MATH-500 accuracy from 5.0% (7.9% paper) to 51.0% (48.1% paper).
$ pip install wsphub && wsp apply Qwen/Qwen3.5-0.8B --patch wsp-qwen3.5-0.8b-math-loop-exp8 --out ./patched
Qwen3.5-0.8B Missing generation_config & Dual Stop-Token Fix
Adds the missing generation_config.json to Qwen/Qwen3.5-0.8B with dual EOS stop tokens [248046 (<|im_end|>), 248044 (<|endoftext|>)] and verified sampling parameters, curing post-turn hallucination bleed.
$ pip install wsphub && wsp apply Qwen/Qwen3.5-0.8B --patch wsp-qwen3.5-0.8b-missing-genconfig-eos-fix --out ./patched