[Bug] can't export gemma4 e2b after fine-tuning.

[Tejas1Koli]

Note: Please do not remove the questions. Answer beside them.

1. Did you update? pip install --upgrade unsloth unsloth_zoo -- yes
2. Colab or Kaggle or local / cloud -- local Unsloth studio
3. Number GPUs used, use nvidia-smi -- Mac m2 air 16gb

can't export gemma4 e2b (q6_k) as gguf after fine-tuning , this is the error it shows up 'Can not map tensor'

GGUF export failed: Unsloth: Failed to convert vision projector to GGUF with command /Users/tejaskoli/.unsloth/studio/unsloth_studio/bin/python /Users/tejaskoli/.unsloth/llama.cpp/unsloth_convert_hf_to_gguf.py --outfile /Users/tejaskoli/.unsloth/studio/exports/gemma-4-E2B-it-GGUF/_tmp_model_f23c556f/gemma-4-E2B-it.BF16-mmproj.gguf --outtype bf16 --mmproj --split-max-size 50G /var/folders/sb/t0nhqnps7tz6xt6bnhhtgz700000gn/T/tmpty_0kjg8/merged: Command '['/Users/tejaskoli/.unsloth/studio/unsloth_studio/bin/python', '/Users/tejaskoli/.unsloth/llama.cpp/unsloth_convert_hf_to_gguf.py', '--outfile', '/Users/tejaskoli/.unsloth/studio/exports/gemma-4-E2B-it-GGUF/_tmp_model_f23c556f/gemma-4-E2B-it.BF16-mmproj.gguf', '--outtype', 'bf16', '--mmproj', '--split-max-size', '50G', '/var/folders/sb/t0nhqnps7tz6xt6bnhhtgz700000gn/T/tmpty_0kjg8/merged']' returned non-zero exit status 1. --- converter stdout --- INFO:hf-to-gguf:Loading model: merged INFO:hf-to-gguf:Model architecture: Gemma4ForConditionalGeneration INFO:hf-to-gguf:gguf: loading model weight map from 'model.safetensors.index.json' INFO:hf-to-gguf:gguf: indexing model part 'model-00001-of-00003.safetensors' INFO:hf-to-gguf:gguf: indexing model part 'model-00002-of-00003.safetensors' INFO:hf-to-gguf:gguf: indexing model part 'model-00003-of-00003.safetensors' INFO:gguf.gguf_writer:gguf: This GGUF file is for Little Endian only INFO:hf-to-gguf:Exporting model... Traceback (most recent call last): File "/Users/tejaskoli/.unsloth/llama.cpp/unsloth_convert_hf_to_gguf.py", line 298, in main() ~~~~^^ File "/Users/tejaskoli/.unsloth/llama.cpp/unsloth_convert_hf_to_gguf.py", line 292, in main model_instance.write() ~~~~~~~~~~~~~~~~~~~~^^ File "/Users/tejaskoli/.unsloth/llama.cpp/conversion/base.py", line 1027, in write self.prepare_tensors() ~~~~~~~~~~~~~~~~~~~~^^ File "/Users/tejaskoli/.unsloth/llama.cpp/conversion/base.py", line 881, in prepare_tensors for new_name, data_torch in (self.modify_tensors(data_torch, name, bid)): ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/tejaskoli/.unsloth/llama.cpp/conversion/gemma.py", line 869, in modify_tensors mapped_name = self.map_tensor_name(name, (".weight", ".bias", ".input_max", ".input_min", ".output_max", ".output_min")) File "/Users/tejaskoli/.unsloth/llama.cpp/conversion/base.py", line 609, in map_tensor_name raise ValueError(f"Can not map tensor {name!r}") ValueError: Can not map tensor 'audio_tower.layers.0.feed_forward1.ffw_layer_1.input_max'