Insert single space in front of the prompt

- this is to match original llama tokenizer behavior
ggml-org · Mar 17, 2023 · 7566d1a · 7566d1a
1 parent 7e1041a
commit 7566d1a
Showing 1 changed file with 2 additions and 0 deletions.
diff --git a/main.cpp b/main.cpp
@@ -845,6 +845,8 @@ int main(int argc, char ** argv) {
 
     std::vector<float> logits;
 
+    // Add a space in front of the first character to match OG llama tokenizer behavior
+    params.prompt.insert(0, 1, ' ');
     // tokenize the prompt
     std::vector<gpt_vocab::id> embd_inp = ::llama_tokenize(vocab, params.prompt, true);