From 4c37bd0b546c421f0680839210f2e963d8d3f35a Mon Sep 17 00:00:00 2001 From: NanoCode012 Date: Mon, 28 Aug 2023 09:39:10 +0900 Subject: [PATCH] Fix(tokenizer): Make sure to add pad for CodeLlamaTokenizer (#489) --- src/axolotl/utils/models.py | 1 + 1 file changed, 1 insertion(+) diff --git a/src/axolotl/utils/models.py b/src/axolotl/utils/models.py index c2fbc19e3..6cd1cd7ea 100644 --- a/src/axolotl/utils/models.py +++ b/src/axolotl/utils/models.py @@ -59,6 +59,7 @@ def load_tokenizer(cfg): in [ "LlamaTokenizer", "LlamaTokenizerFast", + "CodeLlamaTokenizer", ] and hasattr(tokenizer, "pad_token") and not tokenizer.pad_token