PrototypeCommon
sail-sg/Attention-Sink
[ICLR 2025] When Attention Sink Emerges in Language Models: An Empirical View (Spotlight)
- Owner
- sail-sg (organization)
- Stars
- Stars: 164
- Forks
- Forks: 5
- Language
- Python
- License
- MIT
- Created
- 2024
- Topics
- attention-mechanismattention-sinklanguage-modellarge-language-models
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by sail-sg
All 31 repos by sail-sg- sail-sg/EditAnythingStars: 3,423PythonRC
- sail-sg/envpoolStars: 1,533C++Beta
- sail-sg/poolformerStars: 1,367PythonBeta
- sail-sg/understand-r1-zeroStars: 1,286PythonBeta
- sail-sg/voloStars: 947Jupyter NotebookBeta
- sail-sg/AdanStars: 825PythonBeta
Nearby in Python
All Python repos- thetahealth/mirobody-envStars: 164PythonPrototype
- VectifyAI/jev-doc-searchStars: 164PythonPrototype
- matt-pio00p9/YouTube-Views-Boost-Free-View-Booster-Growth-Tool-2026Stars: 164PythonPrototype
- tjweir/liftbookStars: 164PythonPrototype
- simonw/django-openidStars: 164PythonPrototype
- anirudhjoshi/lastpass2keepassStars: 164PythonPrototype