diff --git a/gguf-py/gguf/tensor_mapping.py b/gguf-py/gguf/tensor_mapping.py
index b114df56b9..ca4172a4f7 100644
--- a/gguf-py/gguf/tensor_mapping.py
+++ b/gguf-py/gguf/tensor_mapping.py
@@ -148,7 +148,7 @@ class TensorNameMap:
             "model.layers.{bid}.input_layernorm",                   # llama4
             "layers.{bid}.input_layernorm",                         # embeddinggemma
             "transformer_encoder.{bid}.attention_norm",             # neobert
-            "layers.{bid}.attn_norm",                              # bert
+            "layers.{bid}.attn_norm",                               # modern-bert
             "model.layers.{bid}.operator_norm",                     # lfm2
             "model.transformer.blocks.{bid}.attn_norm",             # llada
             "layers.{bid}.input_layernorm",                         # qwen3-embedding