diff options
| author | Jared Van Bortel <jared@nomic.ai> | 2024-04-09 13:44:08 -0400 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-04-09 13:44:08 -0400 |
| commit | 1b67731e184e27a465b8c5476061294a4af668ea (patch) | |
| tree | 15a2d877029fb509a34e462c227475bc7d6dc31e /examples/tokenize/tokenize.cpp | |
| parent | c4a3a4ff47d62d2503ddf9bd91b58c21f04fe3c3 (diff) | |
BERT tokenizer fixes (#6498)
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
Diffstat (limited to 'examples/tokenize/tokenize.cpp')
| -rw-r--r-- | examples/tokenize/tokenize.cpp | 4 |
1 files changed, 1 insertions, 3 deletions
diff --git a/examples/tokenize/tokenize.cpp b/examples/tokenize/tokenize.cpp index d95a9247..8b1baea8 100644 --- a/examples/tokenize/tokenize.cpp +++ b/examples/tokenize/tokenize.cpp @@ -26,11 +26,9 @@ int main(int argc, char ** argv) { llama_context_params ctx_params = llama_context_default_params(); llama_context * ctx = llama_new_context_with_model(model, ctx_params); - const bool add_bos = llama_should_add_bos_token(model); - std::vector<llama_token> tokens; - tokens = ::llama_tokenize(model, prompt, add_bos, true); + tokens = ::llama_tokenize(model, prompt, true, true); for (int i = 0; i < (int) tokens.size(); i++) { if (printing_ids) { |