minor gorge
Downloading safetensors index for unsloth/gemma-3-4b-it...
model.safetensors.index.json: 100%
90.6k/90.6k [00:00<00:00, 1.50MB/s]
Unsloth: Merging weights into 16bit: 0%| | 0/2 [00:00<?, ?it/s]
model-00001-of-00002.safetensors: 100%
4.96G/4.96G [06:21<00:00, 45.3MB/s]
Unsloth: Merging weights into 16bit: 0%| | 0/2 [06:39<?, ?it/s]
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
<ipython-input-21-0d65d3c1e00e> in <cell line: 0>()
1 if True: # Change to True to save finetune!
----> 2 model.save_pretrained_merged(MODEL_NAME, tokenizer)
7 frames
/usr/local/lib/python3.11/dist-packages/torch/serialization.py in load(f, map_location, pickle_module, weights_only, mmap, **pickle_load_args)
1478 if mmap:
1479 f_name = "" if not isinstance(f, str) else f"{f}, "
-> 1480 raise RuntimeError(
1481 "mmap can only be used with files saved with "
1482 f"`torch.save({f_name}_use_new_zipfile_serialization=True), "
RuntimeError: mmap can only be used with files saved with `torch.save(_use_new_zipfile_serialization=True), please torch.save your checkpoint with this option in order to use mmap.
the MODEL_NAME variable is "gemma-3-finetune"
the error occurred at this step here in the screenshot somehow. The training was fine and all, and almost nothing has changed (i opened and ran the free train colab again and it has the same error)
