|
|
@@ -386,8 +386,8 @@ int main(int argc, char ** argv) {
|
|
|
}
|
|
|
|
|
|
LOGLN(
|
|
|
- "recalculate the cached logits (check): embd_inp.empty() %s, n_matching_session_tokens %zu, embd_inp.size() %zu, session_tokens.size() %zu, embd_inp.size() %zu",
|
|
|
- log_tostr(embd_inp.empty()), n_matching_session_tokens, embd_inp.size(), session_tokens.size(), embd_inp.size());
|
|
|
+ "recalculate the cached logits (check): embd_inp.empty() %s, n_matching_session_tokens %zu, embd_inp.size() %zu, session_tokens.size() %zu",
|
|
|
+ log_tostr(embd_inp.empty()), n_matching_session_tokens, embd_inp.size(), session_tokens.size());
|
|
|
|
|
|
// if we will use the cache for the full prompt without reaching the end of the cache, force
|
|
|
// reevaluation of the last token to recalculate the cached logits
|