From 637ed095a0db0cc8195a2cacf32d603a28edb6a2 Mon Sep 17 00:00:00 2001
From: NanoCode012 <kevinvong@rocketmail.com>
Date: Sun, 29 Oct 2023 21:32:37 +0900
Subject: [PATCH] fix(config): Set eos/bos to tokenizer if different (#801)

* fix(config): Set eos/bos to tokenizer if different

* chore: fix lint
---
 src/axolotl/utils/models.py | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/src/axolotl/utils/models.py b/src/axolotl/utils/models.py
index d0042abc49..5ce7de5e51 100644
--- a/src/axolotl/utils/models.py
+++ b/src/axolotl/utils/models.py
@@ -386,6 +386,20 @@ def load_model(
         )
         model.config.max_position_embeddings = cfg.sequence_len
 
+    if (
+        hasattr(model.config, "bos_token_id")
+        and model.config.bos_token_id
+        and model.config.bos_token_id != tokenizer.bos_token_id
+    ):
+        model.config.bos_token_id = tokenizer.bos_token_id
+
+    if (
+        hasattr(model.config, "eos_token_id")
+        and model.config.eos_token_id
+        and model.config.eos_token_id != tokenizer.eos_token_id
+    ):
+        model.config.eos_token_id = tokenizer.eos_token_id
+
     if model.device.type == "cuda":
         log_gpu_memory_usage(LOG, "after model load", model.device)