PrototypeCommon
lucidrains/Mega-pytorch
Implementation of Mega, the Single-head Attention with Multi-headed EMA architecture that currently holds SOTA on Long Range Arena
- Owner
- lucidrains (personal account)
- Stars
- Stars: 206
- Forks
- Forks: 10
- Language
- Python
- License
- MIT
- Created
- 2022
- Topics
- artificial-intelligenceattention-mechanismsdeep-learningexponential-moving-averagelong-range-arena
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by lucidrains
All 175 repos by lucidrains- lucidrains/vit-pytorchStars: 25,555PythonLTS
- lucidrains/DALLE2-pytorchStars: 11,306PythonStable
- lucidrains/denoising-diffusion-pytorchStars: 10,723PythonStable
- lucidrains/imagen-pytorchStars: 8,430PythonStable
- lucidrains/PaLM-rlhf-pytorchStars: 7,862PythonStable
- lucidrains/x-transformersStars: 5,962PythonStable
Nearby in Python
All Python repos- iamyoki/qwen-image-2.1-skillStars: 206PythonPrototype
- ashleydarosa/CC-Checker-Validator-GeneratorStars: 206PythonPrototype
- sepehr071/iran-mcpStars: 206PythonPrototype
- peppermintos/iceStars: 206PythonPrototype
- PirateBox-Dev/PirateBoxScripts_WebserverStars: 206PythonPrototype
- kennethreitz-archive/argsStars: 206PythonPrototype