| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.857142857142857, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 6.764435291290283, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.2099, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 7.017722129821777, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.225, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 6.019818305969238, | |
| "learning_rate": 1e-05, | |
| "loss": 1.171, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.9952032566070557, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.1149, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 5.4799628257751465, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 1.0653, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 5.356297969818115, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 1.0985, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.822650909423828, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 1.0429, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 3.220806121826172, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 1.0131, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.365323781967163, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 1.0406, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.9435462951660156, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.9629, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 2.59014892578125, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 1.5017, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.4885938167572021, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.9285, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.1829850673675537, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.9084, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.2689722776412964, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.9337, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.215389370918274, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.9306, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 0.8309641480445862, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.819, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.9525875449180603, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.8622, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 1.046868920326233, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.9215, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 1.0977541208267212, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.8397, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.820371150970459, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.7826, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.2211568355560303, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 1.3281, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.7234557867050171, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.8169, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.7530570030212402, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.8442, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.6466705203056335, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.8461, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.6406083106994629, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.8771, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.7005373239517212, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.8336, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.5983966588973999, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.8448, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.5818594098091125, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.8554, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.5654487013816833, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.87, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.5420190691947937, | |
| "learning_rate": 0.0, | |
| "loss": 0.8261, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "step": 30, | |
| "total_flos": 27056444309504.0, | |
| "train_loss": 0.9771380583445232, | |
| "train_runtime": 1000.7763, | |
| "train_samples_per_second": 2.998, | |
| "train_steps_per_second": 0.03 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 27056444309504.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |