| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.523809523809524, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 1.0693126916885376, | |
| "learning_rate": 0.000196, | |
| "loss": 9.5582, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 1.2914847135543823, | |
| "learning_rate": 0.000192, | |
| "loss": 9.2227, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 1.9945707321166992, | |
| "learning_rate": 0.000188, | |
| "loss": 8.7759, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.011664628982544, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 8.2142, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.3491569757461548, | |
| "learning_rate": 0.00018, | |
| "loss": 7.9009, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.9671052098274231, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 7.6909, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.116225004196167, | |
| "learning_rate": 0.000172, | |
| "loss": 7.5216, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 1.0181453227996826, | |
| "learning_rate": 0.000168, | |
| "loss": 7.4308, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.9790288209915161, | |
| "learning_rate": 0.000164, | |
| "loss": 7.4444, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.9529135823249817, | |
| "learning_rate": 0.00016, | |
| "loss": 7.3866, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.7934174537658691, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 7.3822, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.7163369059562683, | |
| "learning_rate": 0.000152, | |
| "loss": 7.3191, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 1.347898244857788, | |
| "learning_rate": 0.000148, | |
| "loss": 7.2142, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.7439594864845276, | |
| "learning_rate": 0.000144, | |
| "loss": 7.2718, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 1.4523087739944458, | |
| "learning_rate": 0.00014, | |
| "loss": 7.2901, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.7053799033164978, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 7.2125, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.6156577467918396, | |
| "learning_rate": 0.000132, | |
| "loss": 7.229, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.40743571519851685, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 7.2999, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 1.0756566524505615, | |
| "learning_rate": 0.000124, | |
| "loss": 7.212, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.7201813459396362, | |
| "learning_rate": 0.00012, | |
| "loss": 7.1826, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5773327946662903, | |
| "learning_rate": 0.000116, | |
| "loss": 7.1456, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.6004664301872253, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 7.0349, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 1.4157112836837769, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 7.2767, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.852541446685791, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 7.1745, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.7835370898246765, | |
| "learning_rate": 0.0001, | |
| "loss": 7.2477, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.414756178855896, | |
| "learning_rate": 9.6e-05, | |
| "loss": 7.1781, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.7038145661354065, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 7.1103, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.8368222713470459, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 7.1595, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.6943209171295166, | |
| "learning_rate": 8.4e-05, | |
| "loss": 7.1062, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.4186341464519501, | |
| "learning_rate": 8e-05, | |
| "loss": 7.2348, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 1.0224595069885254, | |
| "learning_rate": 7.6e-05, | |
| "loss": 7.1154, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.428688645362854, | |
| "learning_rate": 7.2e-05, | |
| "loss": 7.1194, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.913233757019043, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 7.1919, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.5481642484664917, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 7.126, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.49522772431373596, | |
| "learning_rate": 6e-05, | |
| "loss": 7.1564, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "grad_norm": 0.40602990984916687, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 7.081, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 7.0476190476190474, | |
| "grad_norm": 0.4593268036842346, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 7.0596, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 7.238095238095238, | |
| "grad_norm": 0.44626158475875854, | |
| "learning_rate": 4.8e-05, | |
| "loss": 7.0674, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.428571428571429, | |
| "grad_norm": 0.6573432087898254, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 7.0745, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.619047619047619, | |
| "grad_norm": 0.3820817172527313, | |
| "learning_rate": 4e-05, | |
| "loss": 7.0785, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.809523809523809, | |
| "grad_norm": 0.8610634803771973, | |
| "learning_rate": 3.6e-05, | |
| "loss": 7.1974, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.44188380241394043, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 7.0847, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 8.19047619047619, | |
| "grad_norm": 0.6792606711387634, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 7.1191, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.380952380952381, | |
| "grad_norm": 0.4903930723667145, | |
| "learning_rate": 2.4e-05, | |
| "loss": 7.1288, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.5853165984153748, | |
| "learning_rate": 2e-05, | |
| "loss": 7.0479, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.761904761904763, | |
| "grad_norm": 0.6836739182472229, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 7.0448, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.952380952380953, | |
| "grad_norm": 0.5737291574478149, | |
| "learning_rate": 1.2e-05, | |
| "loss": 7.0717, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 9.142857142857142, | |
| "grad_norm": 1.709892988204956, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 6.9482, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 9.333333333333334, | |
| "grad_norm": 0.61203932762146, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 7.1598, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "grad_norm": 0.3827505111694336, | |
| "learning_rate": 0.0, | |
| "loss": 7.1112, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "step": 100, | |
| "total_flos": 500328301455504.0, | |
| "train_loss": 7.34823148727417, | |
| "train_runtime": 426.3789, | |
| "train_samples_per_second": 3.94, | |
| "train_steps_per_second": 0.235 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 500328301455504.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |