https://github.com/facebookresearch/LayerSkip

"LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", Accepted to ACL 2024
https://github.com/facebookresearch/LayerSkip

early-exit layer-drop llm optimization speculative-decoding transformers

Last synced: 2 months ago
JSON representation

"LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", Accepted to ACL 2024

• Host: GitHub
• URL: https://github.com/facebookresearch/LayerSkip
• Owner: facebookresearch
• License: other
• Created: 2024-02-26T19:49:36.000Z (10 months ago)
• Default Branch: main
• Last Pushed: 2024-10-21T15:47:14.000Z (2 months ago)
• Last Synced: 2024-10-23T04:50:35.066Z (2 months ago)
• Topics: early-exit, layer-drop, llm, optimization, speculative-decoding, transformers
• Language: Python
• Homepage:
• Size: 11 MB
• Stars: 107
• Watchers: 5
• Forks: 6
• Open Issues: 3

Awesome Lists containing this project