Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Login
Top
New
Best
Ask
Show
Jobs
Post-transformer inference: 224× compression of Llama-70B with improved accuracy | Heykuki News
Post-transformer inference: 224× compression of Llama-70B with improved accuracy
zenodo.org
72 points
anima-core
6 months ago
56 comments
Loading...