From d1a1b614cdd690c79579c5628af1998dcd01e502 Mon Sep 17 00:00:00 2001
From: Jared Van Bortel <jared@nomic.ai>
Date: Thu, 4 Apr 2024 17:54:46 -0400
Subject: [PATCH] spm : fix special_add_bos default

When this is not set in HF `tokenizer_config.json`, it should default to
true.
---
 llama.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/llama.cpp b/llama.cpp
index 9af370658..850cfb91b 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -11562,7 +11562,7 @@ static std::vector<llama_vocab::id> llama_tokenize_internal(const llama_vocab &
                 // tokenizer.encode('', add_special_tokens=True)  returns [1]
                 // tokenizer.encode('', add_special_tokens=False) returns []
 
-                if (add_special && vocab.special_add_bos == 1) {
+                if (add_special && vocab.special_add_bos != 0) {
                     GGML_ASSERT(vocab.special_bos_id != -1);
                     output.push_back(vocab.special_bos_id);
                 }