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

BERT tokenizer fixes (#6498)

Browse source 
Key changes:
* BERT conversion: fix abuse of LlamaHfVocab, do not set BOS or EOS
* Nomic Embed conversion: pad vocab instead of slicing embedding tensor
* llama_tokenize: handle added special tokens like HF does
This commit is contained in:
Jared Van Bortel 2024-04-09 13:44:08 -04:00 • committed by GitHub
parent c4a3a4ff47
commit 1b67731e18
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
20 changed files with 221 additions and 194 deletions
Download patch file Download diff file Expand all files Collapse all files

4
examples/lookup/lookup-create.cpp
View file

@ -28,10 +28,8 @@ int main(int argc, char ** argv){
GGML_ASSERT(model != nullptr);
// tokenize the prompt
const bool add_bos = llama_should_add_bos_token(model);
std::vector<llama_token> inp;
inp = ::llama_tokenize(ctx, params.prompt, add_bos, true);
inp = ::llama_tokenize(ctx, params.prompt, true, true);
fprintf(stderr, "%s: tokenization done\n", __func__);