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

Add support for BERT embedding models (#5423)

Browse source 
* BERT model graph construction (build_bert)
* WordPiece tokenizer (llm_tokenize_wpm)
* Add flag for non-causal attention models
* Allow for models that only output embeddings
* Support conversion of BERT models to GGUF
* Based on prior work by @xyzhang626 and @skeskinen

---------

Co-authored-by: Jared Van Bortel <jared@nomic.ai>
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Douglas Hanley 2024-02-11 10:21:38 -06:00 • committed by GitHub
parent 97a336507e
commit 2891c8aa9a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
8 changed files with 616 additions and 52 deletions
Download patch file Download diff file Expand all files Collapse all files

12
examples/embedding/embedding.cpp
View file

@ -87,7 +87,17 @@ int main(int argc, char ** argv) {
}
const int n_embd = llama_n_embd(model);
const auto * embeddings = llama_get_embeddings(ctx);
auto * embeddings = llama_get_embeddings(ctx);
// l2-normalize embeddings
float norm = 0;
for (int i = 0; i < n_embd; i++) {
norm += embeddings[i] * embeddings[i];
}
norm = sqrt(norm);
for (int i = 0; i < n_embd; i++) {
embeddings[i] /= norm;
}
for (int i = 0; i < n_embd; i++) {
printf("%f ", embeddings[i]);