llama.cpp
65cdf34b
- llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
2 years ago
llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)
References
#2433 - Use n_embd_gqa instead of n_embd to handle llama-2 70B
Author
randxie
Parents
edcc7ae7
Loading