| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02857142857142857, | |
| "grad_norm": 320.9814758300781, | |
| "learning_rate": 0.0004952380952380952, | |
| "loss": 5.144, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05714285714285714, | |
| "grad_norm": 8.983612060546875, | |
| "learning_rate": 0.0004904761904761905, | |
| "loss": 3.5072, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08571428571428572, | |
| "grad_norm": 5.370372772216797, | |
| "learning_rate": 0.0004857142857142857, | |
| "loss": 2.2891, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11428571428571428, | |
| "grad_norm": 2.666455030441284, | |
| "learning_rate": 0.00048095238095238095, | |
| "loss": 2.0865, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 4.363830089569092, | |
| "learning_rate": 0.0004761904761904762, | |
| "loss": 2.061, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17142857142857143, | |
| "grad_norm": 5.7685394287109375, | |
| "learning_rate": 0.0004714285714285714, | |
| "loss": 1.8436, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 4.38090181350708, | |
| "learning_rate": 0.00046666666666666666, | |
| "loss": 1.8594, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "grad_norm": 6.654545783996582, | |
| "learning_rate": 0.00046190476190476195, | |
| "loss": 1.8804, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2571428571428571, | |
| "grad_norm": 2.162074565887451, | |
| "learning_rate": 0.00045714285714285713, | |
| "loss": 1.6293, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 4.620967388153076, | |
| "learning_rate": 0.00045238095238095237, | |
| "loss": 1.6454, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3142857142857143, | |
| "grad_norm": 5.860279083251953, | |
| "learning_rate": 0.00044761904761904766, | |
| "loss": 1.6508, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.34285714285714286, | |
| "grad_norm": 2.8689663410186768, | |
| "learning_rate": 0.00044285714285714284, | |
| "loss": 1.3364, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.37142857142857144, | |
| "grad_norm": 2.600432872772217, | |
| "learning_rate": 0.0004380952380952381, | |
| "loss": 1.4913, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.5309250354766846, | |
| "learning_rate": 0.00043333333333333337, | |
| "loss": 1.3719, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 2.634753704071045, | |
| "learning_rate": 0.00042857142857142855, | |
| "loss": 1.5274, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.45714285714285713, | |
| "grad_norm": 3.284735918045044, | |
| "learning_rate": 0.0004238095238095238, | |
| "loss": 1.5985, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4857142857142857, | |
| "grad_norm": 3.2241106033325195, | |
| "learning_rate": 0.0004190476190476191, | |
| "loss": 1.2673, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5142857142857142, | |
| "grad_norm": 2.514350175857544, | |
| "learning_rate": 0.0004142857142857143, | |
| "loss": 1.4211, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5428571428571428, | |
| "grad_norm": 3.476823091506958, | |
| "learning_rate": 0.00040952380952380955, | |
| "loss": 1.4254, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 2.4598169326782227, | |
| "learning_rate": 0.0004047619047619048, | |
| "loss": 1.2857, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.915236234664917, | |
| "learning_rate": 0.0004, | |
| "loss": 1.2144, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6285714285714286, | |
| "grad_norm": 3.5647623538970947, | |
| "learning_rate": 0.00039523809523809526, | |
| "loss": 1.2258, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6571428571428571, | |
| "grad_norm": 4.019155025482178, | |
| "learning_rate": 0.0003904761904761905, | |
| "loss": 1.4103, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6857142857142857, | |
| "grad_norm": 3.1268930435180664, | |
| "learning_rate": 0.0003857142857142857, | |
| "loss": 1.2308, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 2.708326578140259, | |
| "learning_rate": 0.00038095238095238096, | |
| "loss": 1.3197, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7428571428571429, | |
| "grad_norm": 2.4128565788269043, | |
| "learning_rate": 0.0003761904761904762, | |
| "loss": 1.1922, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.7714285714285715, | |
| "grad_norm": 4.487890243530273, | |
| "learning_rate": 0.00037142857142857143, | |
| "loss": 1.2505, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.1995606422424316, | |
| "learning_rate": 0.00036666666666666667, | |
| "loss": 1.2047, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8285714285714286, | |
| "grad_norm": 3.0884156227111816, | |
| "learning_rate": 0.0003619047619047619, | |
| "loss": 1.1397, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 3.9434640407562256, | |
| "learning_rate": 0.00035714285714285714, | |
| "loss": 1.0844, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8857142857142857, | |
| "grad_norm": 3.6582884788513184, | |
| "learning_rate": 0.00035238095238095243, | |
| "loss": 1.1461, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9142857142857143, | |
| "grad_norm": 2.6335928440093994, | |
| "learning_rate": 0.0003476190476190476, | |
| "loss": 1.4124, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9428571428571428, | |
| "grad_norm": 2.8153300285339355, | |
| "learning_rate": 0.00034285714285714285, | |
| "loss": 1.1619, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9714285714285714, | |
| "grad_norm": 5.384872913360596, | |
| "learning_rate": 0.00033809523809523814, | |
| "loss": 1.1091, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.4129958152771, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 1.1665, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0720068216323853, | |
| "eval_runtime": 16.9829, | |
| "eval_samples_per_second": 11.777, | |
| "eval_steps_per_second": 1.472, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0285714285714285, | |
| "grad_norm": 7.848823547363281, | |
| "learning_rate": 0.00032857142857142856, | |
| "loss": 1.073, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.0571428571428572, | |
| "grad_norm": 3.559180498123169, | |
| "learning_rate": 0.00032380952380952385, | |
| "loss": 1.1773, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.0857142857142856, | |
| "grad_norm": 2.477151870727539, | |
| "learning_rate": 0.00031904761904761903, | |
| "loss": 1.0647, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.1142857142857143, | |
| "grad_norm": 3.4121389389038086, | |
| "learning_rate": 0.00031428571428571427, | |
| "loss": 1.0428, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 6.0573015213012695, | |
| "learning_rate": 0.00030952380952380956, | |
| "loss": 1.0667, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.1714285714285715, | |
| "grad_norm": 3.7485954761505127, | |
| "learning_rate": 0.0003047619047619048, | |
| "loss": 1.0497, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 4.239710330963135, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1227, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2285714285714286, | |
| "grad_norm": 7.383611679077148, | |
| "learning_rate": 0.00029523809523809526, | |
| "loss": 1.0, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.2571428571428571, | |
| "grad_norm": 3.1444430351257324, | |
| "learning_rate": 0.0002904761904761905, | |
| "loss": 0.9194, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.2857142857142856, | |
| "grad_norm": 8.754597663879395, | |
| "learning_rate": 0.0002857142857142857, | |
| "loss": 1.2234, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3142857142857143, | |
| "grad_norm": 5.383545875549316, | |
| "learning_rate": 0.00028095238095238097, | |
| "loss": 1.0219, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.342857142857143, | |
| "grad_norm": 3.977374792098999, | |
| "learning_rate": 0.0002761904761904762, | |
| "loss": 1.061, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.3714285714285714, | |
| "grad_norm": 3.6732325553894043, | |
| "learning_rate": 0.0002714285714285714, | |
| "loss": 1.1131, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 4.493168830871582, | |
| "learning_rate": 0.0002666666666666667, | |
| "loss": 1.0033, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 2.9993937015533447, | |
| "learning_rate": 0.0002619047619047619, | |
| "loss": 1.0023, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.457142857142857, | |
| "grad_norm": 3.984879732131958, | |
| "learning_rate": 0.0002571428571428571, | |
| "loss": 0.9796, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.4857142857142858, | |
| "grad_norm": 5.652857303619385, | |
| "learning_rate": 0.0002523809523809524, | |
| "loss": 0.912, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5142857142857142, | |
| "grad_norm": 4.275585174560547, | |
| "learning_rate": 0.0002476190476190476, | |
| "loss": 0.9618, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.5428571428571427, | |
| "grad_norm": 4.000895977020264, | |
| "learning_rate": 0.00024285714285714286, | |
| "loss": 1.0019, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 3.681532144546509, | |
| "learning_rate": 0.0002380952380952381, | |
| "loss": 0.9769, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 6.31843376159668, | |
| "learning_rate": 0.00023333333333333333, | |
| "loss": 1.0508, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.6285714285714286, | |
| "grad_norm": 3.165881633758545, | |
| "learning_rate": 0.00022857142857142857, | |
| "loss": 1.1192, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.657142857142857, | |
| "grad_norm": 3.3488543033599854, | |
| "learning_rate": 0.00022380952380952383, | |
| "loss": 0.945, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.6857142857142857, | |
| "grad_norm": 3.479367256164551, | |
| "learning_rate": 0.00021904761904761904, | |
| "loss": 1.1191, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 3.9680163860321045, | |
| "learning_rate": 0.00021428571428571427, | |
| "loss": 1.0116, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.7428571428571429, | |
| "grad_norm": 3.4398531913757324, | |
| "learning_rate": 0.00020952380952380954, | |
| "loss": 1.0162, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.7714285714285714, | |
| "grad_norm": 4.210978031158447, | |
| "learning_rate": 0.00020476190476190477, | |
| "loss": 0.9214, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 3.9029576778411865, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0135, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.8285714285714287, | |
| "grad_norm": 3.1066203117370605, | |
| "learning_rate": 0.00019523809523809525, | |
| "loss": 1.0136, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.8571428571428572, | |
| "grad_norm": 4.775870323181152, | |
| "learning_rate": 0.00019047619047619048, | |
| "loss": 0.8558, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.8857142857142857, | |
| "grad_norm": 4.7907490730285645, | |
| "learning_rate": 0.00018571428571428572, | |
| "loss": 0.9436, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.9142857142857141, | |
| "grad_norm": 4.235854148864746, | |
| "learning_rate": 0.00018095238095238095, | |
| "loss": 0.9884, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.9428571428571428, | |
| "grad_norm": 3.239032030105591, | |
| "learning_rate": 0.00017619047619047622, | |
| "loss": 0.9304, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.9714285714285715, | |
| "grad_norm": 3.5313665866851807, | |
| "learning_rate": 0.00017142857142857143, | |
| "loss": 0.983, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.0609097480773926, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 0.9815, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8912971615791321, | |
| "eval_runtime": 17.0248, | |
| "eval_samples_per_second": 11.748, | |
| "eval_steps_per_second": 1.468, | |
| "step": 700 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1050, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4188970942464000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |