Keep getting this error message when starting training.

[PhilFW2]

File "C:\ProgramData\miniconda3\envs\lora\lib\site-packages\torch\nn\modules\conv.py", line 309, in _conv_forward
return F.conv1d(input, weight, bias, self.stride,
RuntimeError: Given groups=1, weight of size [320, 1024, 7], expected input[4, 384, 40] to have 1024 channels, but got 384 channels instead

All other steps work - the raw data processing etc..
I am using a game laptop so only have like 4 GB of GPU

Perhaps this is the issue

[MaxMax2016]

do you use whisper large v2?