| { | |
| "best_metric": 5.074450392391248, | |
| "best_model_checkpoint": "./checkpoint-1500", | |
| "epoch": 1.3533333333333333, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.87983978638184e-07, | |
| "loss": 0.676, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.712950600801067e-07, | |
| "loss": 0.3062, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.546061415220292e-07, | |
| "loss": 0.2604, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.379172229639519e-07, | |
| "loss": 0.2175, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.212283044058745e-07, | |
| "loss": 0.2147, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.04539385847797e-07, | |
| "loss": 0.1825, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.885180240320427e-07, | |
| "loss": 0.168, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.718291054739653e-07, | |
| "loss": 0.1771, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.551401869158879e-07, | |
| "loss": 0.1862, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.384512683578104e-07, | |
| "loss": 0.1867, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.21762349799733e-07, | |
| "loss": 0.1652, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.050734312416555e-07, | |
| "loss": 0.1583, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 7.883845126835781e-07, | |
| "loss": 0.1562, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.716955941255007e-07, | |
| "loss": 0.1607, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.550066755674232e-07, | |
| "loss": 0.1422, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 7.383177570093458e-07, | |
| "loss": 0.1508, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.216288384512683e-07, | |
| "loss": 0.1556, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.049399198931909e-07, | |
| "loss": 0.1581, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.882510013351135e-07, | |
| "loss": 0.1738, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.71562082777036e-07, | |
| "loss": 0.151, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.548731642189586e-07, | |
| "loss": 0.1471, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 6.381842456608812e-07, | |
| "loss": 0.1491, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.214953271028037e-07, | |
| "loss": 0.1388, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.048064085447263e-07, | |
| "loss": 0.1521, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 5.887850467289718e-07, | |
| "loss": 0.1572, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 5.720961281708944e-07, | |
| "loss": 0.1577, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.55407209612817e-07, | |
| "loss": 0.1416, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5.387182910547396e-07, | |
| "loss": 0.1375, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5.220293724966622e-07, | |
| "loss": 0.1228, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 5.053404539385847e-07, | |
| "loss": 0.1556, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.1683349609375, | |
| "eval_runtime": 4390.4106, | |
| "eval_samples_per_second": 3.535, | |
| "eval_steps_per_second": 0.221, | |
| "eval_wer": 5.095888987058885, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.886515353805072e-07, | |
| "loss": 0.1318, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.719626168224299e-07, | |
| "loss": 0.1586, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.5527369826435247e-07, | |
| "loss": 0.1403, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.38584779706275e-07, | |
| "loss": 0.144, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.225634178905207e-07, | |
| "loss": 0.1417, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.058744993324432e-07, | |
| "loss": 0.1544, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.891855807743658e-07, | |
| "loss": 0.1483, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 3.724966622162884e-07, | |
| "loss": 0.1469, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.558077436582109e-07, | |
| "loss": 0.1782, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 3.391188251001335e-07, | |
| "loss": 0.2161, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.22429906542056e-07, | |
| "loss": 0.1903, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.057409879839786e-07, | |
| "loss": 0.2003, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 2.890520694259012e-07, | |
| "loss": 0.2123, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.723631508678237e-07, | |
| "loss": 0.2347, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.5567423230974635e-07, | |
| "loss": 0.1982, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 2.389853137516689e-07, | |
| "loss": 0.1899, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 2.2229639519359144e-07, | |
| "loss": 0.1865, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 2.05607476635514e-07, | |
| "loss": 0.204, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.8891855807743656e-07, | |
| "loss": 0.2119, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.7222963951935914e-07, | |
| "loss": 0.2108, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.555407209612817e-07, | |
| "loss": 0.229, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.3885180240320425e-07, | |
| "loss": 0.2851, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.221628838451268e-07, | |
| "loss": 0.1927, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.054739652870494e-07, | |
| "loss": 0.1675, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 8.878504672897195e-08, | |
| "loss": 0.167, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 7.209612817089452e-08, | |
| "loss": 0.176, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 5.5407209612817086e-08, | |
| "loss": 0.1718, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.871829105473965e-08, | |
| "loss": 0.1831, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.2029372496662217e-08, | |
| "loss": 0.1793, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 5.3404539385847795e-09, | |
| "loss": 0.1732, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_loss": 0.164794921875, | |
| "eval_runtime": 4364.6715, | |
| "eval_samples_per_second": 3.556, | |
| "eval_steps_per_second": 0.222, | |
| "eval_wer": 5.074450392391248, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "step": 1500, | |
| "total_flos": 1.0191220995495297e+20, | |
| "train_loss": 0.1855107421875, | |
| "train_runtime": 20533.0269, | |
| "train_samples_per_second": 2.338, | |
| "train_steps_per_second": 0.073 | |
| } | |
| ], | |
| "max_steps": 1500, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.0191220995495297e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |