llama.cpp
3e12fbde
- llama: avoid copying logits during prompt decode in MTP (#23198)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
85 days ago
llama: avoid copying logits during prompt decode in MTP (#23198) * llama: avoid copying logits during prompt decode in MTP * review: update comment * llama-graph: call set_output for t_h_pre_norm
References
#23198 - llama: avoid copying logits during prompt decode in MTP
Author
am17an
Parents
39cf5d61
Loading