training.log

nohup: ignoring input
LlamaConfig.validate patched.
Config: 9 heads / 3 kv_heads set.
Loaded 714 Sepedi sentences.

Map:   0%|          | 0/714 [00:00<?, ? examples/s]
Map: 100%|██████████| 714/714 [00:01<00:00, 656.00 examples/s]
Map: 100%|██████████| 714/714 [00:01<00:00, 654.73 examples/s]
Train: 642 | Eval: 72

Loading weights:   0%|          | 0/272 [00:00<?, ?it/s]
Loading weights: 100%|██████████| 272/272 [00:00<00:00, 5910.53it/s]
Model loaded. Params: 125,008,384
Starting NSO fine-tuning...

  0%|          | 0/105 [00:00<?, ?it/s]/home/sediba/.local/lib/python3.12/site-packages/torch/utils/data/dataloader.py:752: UserWarning: 'pin_memory' argument is set as true but no accelerator is found, then device pinned memory won't be used.
  super().__init__(loader)