mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-25 02:44:36 +00:00
server : fix EOS token detection with disabled cache (#5938)
This commit is contained in:
parent
581ed5c4fe
commit
af37fd8b30
@ -1123,7 +1123,7 @@ struct server_context {
|
|||||||
});
|
});
|
||||||
}
|
}
|
||||||
|
|
||||||
if (!slot.cache_tokens.empty() && result.tok == llama_token_eos(model)) {
|
if (result.tok == llama_token_eos(model)) {
|
||||||
slot.stopped_eos = true;
|
slot.stopped_eos = true;
|
||||||
slot.has_next_token = false;
|
slot.has_next_token = false;
|
||||||
|
|
||||||
|
Loading…
Reference in New Issue
Block a user