| { | |
| "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": 7.102932453155518, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.1951, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 7.465461254119873, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.2188, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 6.309134006500244, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1466, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 3.431251049041748, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.0925, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 5.682745456695557, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 1.0697, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 5.679488182067871, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 1.0761, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.901688098907471, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 1.0359, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 3.6228830814361572, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.9778, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.557098388671875, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.9634, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.9451111555099487, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.9079, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 2.730046033859253, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 1.3387, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.5574017763137817, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.9057, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.219175100326538, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.8104, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.109845519065857, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.8269, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.4855905771255493, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.9437, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 1.108618140220642, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.8208, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.993137776851654, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.8353, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 1.164468765258789, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.8716, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.8304758071899414, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.7681, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.831884503364563, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.8329, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.251667857170105, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 1.2754, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.770209550857544, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.8055, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.6954714059829712, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.8038, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.7610450387001038, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.7889, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.6107993721961975, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.8067, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.6758484840393066, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.8279, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.591647744178772, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.7699, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.599900484085083, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.7895, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.560782790184021, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.8391, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.5509406924247742, | |
| "learning_rate": 0.0, | |
| "loss": 0.767, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "step": 30, | |
| "total_flos": 20031294111744.0, | |
| "train_loss": 0.9370571772257487, | |
| "train_runtime": 785.5201, | |
| "train_samples_per_second": 3.819, | |
| "train_steps_per_second": 0.038 | |
| } | |
| ], | |
| "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": 20031294111744.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |