llama.cpp
f07cd35d
- speculative : fix off-by-one for n_drafted
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 year ago
speculative : fix off-by-one for n_drafted
References
#3624 - speculative : add tree-based sampling example
Author
ggerganov
Committer
ggerganov
Parents
373d782d
Loading