root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-09-22 21:16:20 +00:00
Code Issues Actions 16 Packages Projects Releases Wiki Activity

llama : check all graph nodes when searching for result_embd_pooled (#8956)

Browse Source 
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
This commit is contained in:
fairydreaming 2024-08-11 10:35:26 +02:00 committed by GitHub
parent 7c5bfd57f8
commit 33309f661a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 8 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
src/llama.cpp
View File

@ -14722,12 +14722,15 @@ static int llama_decode_internal(
res = nullptr;
embd = nullptr;
} else if (cparams.embeddings) {
res = nullptr; // do not extract logits for embedding case
embd = gf->nodes[gf->n_nodes - 1];
if (strcmp(embd->name, "result_embd_pooled") != 0) {
embd = gf->nodes[gf->n_nodes - 2];
res = nullptr; // do not extract logits for embedding case
embd = nullptr;
for (int i = gf->n_nodes - 1; i >= 0; --i) {
if (strcmp(gf->nodes[i]->name, "result_embd_pooled") == 0) {
embd = gf->nodes[i];
break;
}
}
GGML_ASSERT(strcmp(embd->name, "result_embd_pooled") == 0 && "missing embeddings tensor");
GGML_ASSERT(embd != nullptr && "missing embeddings tensor");
} else {
embd = nullptr; // do not extract embeddings when not needed
GGML_ASSERT(strcmp(res->name, "result_output") == 0 && "missing result_output tensor");