2024-06-11 20:31:50,507 - INFO: Calling run.. 2024-06-11 20:31:50,508 - INFO: Problem Type: text_causal_classification_modeling 2024-06-11 20:31:50,508 - INFO: Global random seed: 622910 2024-06-11 20:31:50,508 - INFO: Preparing the data... 2024-06-11 20:31:50,508 - INFO: Setting up automatic validation split... 2024-06-11 20:31:50,549 - INFO: Preparing train and validation data 2024-06-11 20:31:50,549 - INFO: Loading train dataset... 2024-06-11 20:31:51,325 - INFO: Loading validation dataset... 2024-06-11 20:31:51,461 - INFO: Number of observations in train dataset: 9600 2024-06-11 20:31:51,461 - INFO: Number of observations in validation dataset: 2400 2024-06-11 20:31:51,625 - WARNING: EOS token id not matching between config and tokenizer. Overwriting with tokenizer id. 2024-06-11 20:31:51,625 - WARNING: PAD token id not matching between config and tokenizer. Overwriting with tokenizer id. 2024-06-11 20:31:51,632 - INFO: Using bfloat16 for backbone 2024-06-11 20:31:51,632 - INFO: Loading tiiuae/falcon-rw-1b. This may take a while. 2024-06-11 20:32:00,834 - INFO: Loaded tiiuae/falcon-rw-1b. 2024-06-11 20:32:00,836 - WARNING: EOS token id not matching between generation config and tokenizer. Overwriting with tokenizer id. 2024-06-11 20:32:00,837 - WARNING: PAD token id not matching between generation config and tokenizer. Overwriting with tokenizer id. 2024-06-11 20:32:00,837 - INFO: Lora module names: ['query_key_value', 'dense', 'dense_h_to_4h', 'dense_4h_to_h'] 2024-06-11 20:32:00,955 - INFO: Enough space available for saving model weights.Required space: 2591.11MB, Available space: 994803.55MB. 2024-06-11 20:32:00,959 - INFO: Optimizer AdamW has been provided with parameters {'eps': 1e-08, 'weight_decay': 0.0, 'betas': (0.8999999762, 0.9990000129), 'lr': 0.0001} 2024-06-11 20:32:01,103 - WARNING: No order set for keys: ['answer_column_label', 'num_classes']. 2024-06-11 20:32:01,123 - WARNING: No order set for keys: ['answer_column_label', 'num_classes']. 2024-06-11 20:32:02,308 - INFO: started process: 0, can_track: True, tracking_mode: TrackingMode.DURING_EPOCH 2024-06-11 20:32:02,309 - INFO: Training Epoch: 1 / 1 2024-06-11 20:32:02,310 - INFO: train loss: 0%| | 0/600 [00:00