PrototypeCommon
alainnothere/llm-circuit-finder
I replicated Ng's RYS method and found that duplicating 3 specific layers in Qwen2.5-32B boosts reasoning by 17% and duplicating layers 12-14 in Devstral-24B improves logical deduction from 0.22→0.76 on BBH — no training, no weight changes, just routing hidden states through the same circuit twice.…
- Owner
- alainnothere (personal account)
- Stars
- Stars: 244
- Forks
- Forks: 20
- Language
- Python
- License
- None
- Created
- 2026
- Tier
- Prototype (100 to 249 ★) What this tier means
Nearby in Python
All Python repos- Amal-David/awesome-jevStars: 244PythonPrototype
- Mr-Destroyer/ZimZillaStars: 244PythonPrototype
- yzg1226/mcd-party-officerStars: 244PythonPrototype
- herumi/fmathStars: 244PythonPrototype
- andrewf/pcap2harStars: 244PythonPrototype
- simplistix/testfixturesStars: 244PythonPrototype