Joshua Ainslie
1 paper · 1 atlas page
Papers (1)
Source
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Ainslie, Lee-Thorp, Jong, Zemlyanskiy et al. · EMNLP 2023 2023
arXiv ↗In the Atlas (1)
Co-authors (5)
- Michiel de Jong1 shared paper
- Federico Lebrón1 shared paper
- James Lee-Thorp1 shared paper
- Sumit Sanghai1 shared paper
- Yury Zemlyanskiy1 shared paper