2024-03-19 19:13:56,684 - INFO: Problem Type: text_causal_language_modeling 2024-03-19 19:13:56,684 - INFO: Global random seed: 187907 2024-03-19 19:13:56,684 - INFO: Preparing the data... 2024-03-19 19:13:56,684 - INFO: Setting up automatic validation split... 2024-03-19 19:13:56,710 - INFO: Preparing train and validation data 2024-03-19 19:13:56,710 - INFO: Loading train dataset... 2024-03-19 19:13:57,105 - INFO: Stop token ids: [tensor([ 529, 29989, 5205, 29989, 29958])] 2024-03-19 19:13:57,113 - INFO: Loading validation dataset... 2024-03-19 19:13:57,196 - INFO: Stop token ids: [tensor([ 529, 29989, 5205, 29989, 29958])] 2024-03-19 19:13:57,201 - INFO: Number of observations in train dataset: 567 2024-03-19 19:13:57,201 - INFO: Number of observations in validation dataset: 6 2024-03-19 19:13:57,463 - INFO: Stop token ids: [tensor([ 529, 29989, 5205, 29989, 29958], device='cuda:0')] 2024-03-19 19:13:57,466 - WARNING: PAD token id not matching between config and tokenizer. Overwriting with tokenizer id. 2024-03-19 19:13:57,466 - INFO: Setting pretraining_tp of model config to 1. 2024-03-19 19:13:57,470 - INFO: Using float16 for backbone 2024-03-19 19:13:57,470 - INFO: Loading TinyLlama/TinyLlama-1.1B-Chat-v0.6. This may take a while. 2024-03-19 19:14:04,419 - INFO: Loaded TinyLlama/TinyLlama-1.1B-Chat-v0.6. 2024-03-19 19:14:04,420 - WARNING: PAD token id not matching between generation config and tokenizer. Overwriting with tokenizer id. 2024-03-19 19:14:04,421 - INFO: Lora module names: ['q_proj', 'k_proj', 'v_proj', 'o_proj', 'gate_proj', 'up_proj', 'down_proj'] 2024-03-19 19:14:04,593 - INFO: Enough space available for saving model weights.Required space: 2173.51MB, Available space: 995215.66MB. 2024-03-19 19:14:05,761 - INFO: Training Epoch: 1 / 15 2024-03-19 19:14:05,762 - INFO: train loss: 0%| | 0/283 [00:00