haraldwolff/llama.cpp
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

Work around for recalculating logits in cached prompts (Fixes #1585) (#1609)

Browse source 
* Work around for recalculating logits in cached prompts
This commit is contained in:
DannyDaemonic 2023-05-29 05:13:40 -07:00 committed by GitHub
parent 0e730dd23b
commit 248367605e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 6 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
examples/main/main.cpp
View file

@ -360,6 +360,12 @@ int main(int argc, char ** argv) {
}
}
if (i > 0) {
// check if we've used up all the prompt but not all cached tokens
if (embd.size() == i && n_session_consumed < (int) session_tokens.size()) {
// force revaluation of the last token to recalculate logits
i--;
n_past--;
}
embd.erase(embd.begin(), embd.begin() + i);
}
}