fix config errors

config.json

@@ -44,6 +44,10 @@
     "torch_dtype": "bfloat16"
   },
   "audio_token_index": 262143,
+  "auto_map": {
+    "AutoConfig": "configuration_gemma3mm.Gemma3MMConfig",
+    "AutoModel": "modeling_gemma3mm.Gemma3MMForConditionalGeneration"
+  },
   "boa_token_index": 256001,
   "boi_token_index": 255999,
   "eoa_token_index": 256002,

configuration_gemma3mm.py

@@ -203,8 +203,4 @@ class Gemma3MMConfig(PretrainedConfig):
         self.audio_token_index = audio_token_index
         self.initializer_range = initializer_range
 
-        super().__init__(**kwargs)
-
-
-AutoConfig.register("gemma3mm", Gemma3MMConfig)
-Gemma3MMConfig.register_for_auto_class()
+        super().__init__(**kwargs)

modeling_gemma3mm.py

@@ -1,24 +1,3 @@
-#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
-#           This file was automatically generated from src/transformers/models/gemma3/modular_gemma3.py.
-#               Do NOT edit this file manually as any edits will be overwritten by the generation of
-#             the file from the modular. If any change should be done, please apply the change to the
-#                          modular_gemma3.py file directly. One of our CI enforces this.
-#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
-# coding=utf-8
-# Copyright 2025 Google Inc. HuggingFace Inc. team. All rights reserved.
-#
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
 import copy
 from collections.abc import Callable
 from dataclasses import dataclass
@@ -162,11 +141,18 @@ GEMMA3_INPUTS_DOCSTRING = r"""
             the complete sequence length.
 """
 
+@add_start_docstrings(
+    "The bare Gemma3 Model outputting raw hidden-states without any specific head on top.",
+    GEMMA3_START_DOCSTRING,
+)
+class Gemma3MMPreTrainedModel(Gemma3PreTrainedModel):
+    config_class = Gemma3MMConfig
+
 @add_start_docstrings(
     """The GEMMA3 model which consists of a vision backbone and a language model.""",
     GEMMA3_START_DOCSTRING,
 )
-class Gemma3MMForConditionalGeneration(Gemma3PreTrainedModel, GenerationMixin):
+class Gemma3MMForConditionalGeneration(Gemma3MMPreTrainedModel, GenerationMixin):
     def __init__(self, config: Gemma3MMConfig):
         super().__init__(config)
         self.vision_tower = AutoModel.from_config(config=config.vision_config)
@@ -631,10 +617,4 @@ class Gemma3MMForConditionalGeneration(Gemma3PreTrainedModel, GenerationMixin):
         return model_inputs
 
     def tie_weights(self):
-        return self.language_model.tie_weights()
-
-
-AutoConfig.register("gemma3mm", Gemma3MMConfig)
-AutoModel.register("gemma3mm", Gemma3MMForConditionalGeneration)
-Gemma3MMConfig.register_for_auto_class()
-Gemma3MMForConditionalGeneration.register_for_auto_class()
+        return self.language_model.tie_weights()

processor_config.json

@@ -1,4 +1,7 @@
 {
+  "auto_map": {
+    "AutoProcessor": "processing_gemma3mm.Gemma3MMProcessor"
+  },
   "image_seq_length": 256,
   "processor_class": "Gemma3MMProcessor"
 }