Reading metadata...: 2165it [00:00, 13419.24it/s] | 0/60000 [00:00> The following columns in the training set don't have a corresponding argument in `WhisperForConditionalGeneration.forward` and have been ignored: input_length. If input_length are not expected by `WhisperForConditionalGeneration.forward`, you can safely ignore this message. [WARNING|logging.py:329] 2023-11-18 12:45:05,113 >> `use_cache = True` is incompatible with gradient checkpointing. Setting `use_cache = False`... 0%| | 20/60000 [05:10<155:19:13, 9.32s/it]