| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.906542056074766, | |
| "global_step": 3200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 9.7e-06, | |
| "loss": 9.9876, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 4.772322177886963, | |
| "eval_runtime": 43.7501, | |
| "eval_samples_per_second": 19.497, | |
| "eval_steps_per_second": 0.617, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.97e-05, | |
| "loss": 3.6526, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 3.8356621265411377, | |
| "eval_runtime": 43.2583, | |
| "eval_samples_per_second": 19.719, | |
| "eval_steps_per_second": 0.624, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.96e-05, | |
| "loss": 3.1832, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 3.42309832572937, | |
| "eval_runtime": 43.8662, | |
| "eval_samples_per_second": 19.446, | |
| "eval_steps_per_second": 0.616, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 3.0394, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "eval_loss": 3.35982608795166, | |
| "eval_runtime": 43.5204, | |
| "eval_samples_per_second": 19.6, | |
| "eval_steps_per_second": 0.62, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 4.96e-05, | |
| "loss": 2.9964, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_loss": 3.0924816131591797, | |
| "eval_runtime": 42.6866, | |
| "eval_samples_per_second": 19.983, | |
| "eval_steps_per_second": 0.633, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 5.96e-05, | |
| "loss": 2.9881, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "eval_loss": 3.0199828147888184, | |
| "eval_runtime": 44.2234, | |
| "eval_samples_per_second": 19.288, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 6.96e-05, | |
| "loss": 2.9678, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "eval_loss": 3.0672409534454346, | |
| "eval_runtime": 44.282, | |
| "eval_samples_per_second": 19.263, | |
| "eval_steps_per_second": 0.61, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 7.960000000000001e-05, | |
| "loss": 2.7541, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "eval_loss": 2.397754669189453, | |
| "eval_runtime": 43.7165, | |
| "eval_samples_per_second": 19.512, | |
| "eval_steps_per_second": 0.618, | |
| "eval_wer": 0.9788053949903661, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 8.960000000000001e-05, | |
| "loss": 1.876, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "eval_loss": 1.5307629108428955, | |
| "eval_runtime": 43.4064, | |
| "eval_samples_per_second": 19.651, | |
| "eval_steps_per_second": 0.622, | |
| "eval_wer": 0.7509098694069792, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 9.960000000000001e-05, | |
| "loss": 1.4334, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "eval_loss": 1.435613751411438, | |
| "eval_runtime": 43.9507, | |
| "eval_samples_per_second": 19.408, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.620316848640548, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 9.565610859728508e-05, | |
| "loss": 1.223, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "eval_loss": 1.0539987087249756, | |
| "eval_runtime": 43.8182, | |
| "eval_samples_per_second": 19.467, | |
| "eval_steps_per_second": 0.616, | |
| "eval_wer": 0.5885249411260972, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "learning_rate": 9.113122171945702e-05, | |
| "loss": 1.0139, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "eval_loss": 0.9824701547622681, | |
| "eval_runtime": 43.7265, | |
| "eval_samples_per_second": 19.508, | |
| "eval_steps_per_second": 0.617, | |
| "eval_wer": 0.4964675658317277, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 8.660633484162897e-05, | |
| "loss": 0.9349, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "eval_loss": 0.981073796749115, | |
| "eval_runtime": 43.357, | |
| "eval_samples_per_second": 19.674, | |
| "eval_steps_per_second": 0.623, | |
| "eval_wer": 0.4774138300149861, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "learning_rate": 8.20814479638009e-05, | |
| "loss": 0.8395, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "eval_loss": 0.9597522616386414, | |
| "eval_runtime": 44.326, | |
| "eval_samples_per_second": 19.244, | |
| "eval_steps_per_second": 0.609, | |
| "eval_wer": 0.44166131449368445, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 7.755656108597285e-05, | |
| "loss": 0.8417, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "eval_loss": 0.9240782856941223, | |
| "eval_runtime": 44.9894, | |
| "eval_samples_per_second": 18.96, | |
| "eval_steps_per_second": 0.6, | |
| "eval_wer": 0.43513166345536286, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 7.30316742081448e-05, | |
| "loss": 0.7091, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "eval_loss": 0.9342209100723267, | |
| "eval_runtime": 43.4211, | |
| "eval_samples_per_second": 19.645, | |
| "eval_steps_per_second": 0.622, | |
| "eval_wer": 0.4319203596660244, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.89, | |
| "learning_rate": 6.850678733031674e-05, | |
| "loss": 0.6622, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 15.89, | |
| "eval_loss": 0.895724892616272, | |
| "eval_runtime": 43.4542, | |
| "eval_samples_per_second": 19.63, | |
| "eval_steps_per_second": 0.621, | |
| "eval_wer": 0.41629201455791054, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 16.82, | |
| "learning_rate": 6.398190045248869e-05, | |
| "loss": 0.6762, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.82, | |
| "eval_loss": 0.9993765354156494, | |
| "eval_runtime": 44.1607, | |
| "eval_samples_per_second": 19.316, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.40023549561121813, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 17.76, | |
| "learning_rate": 5.945701357466064e-05, | |
| "loss": 0.6325, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 17.76, | |
| "eval_loss": 0.9341434240341187, | |
| "eval_runtime": 42.9623, | |
| "eval_samples_per_second": 19.855, | |
| "eval_steps_per_second": 0.628, | |
| "eval_wer": 0.41168914579319205, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 5.4932126696832586e-05, | |
| "loss": 0.5829, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "eval_loss": 1.0207712650299072, | |
| "eval_runtime": 42.8374, | |
| "eval_samples_per_second": 19.913, | |
| "eval_steps_per_second": 0.63, | |
| "eval_wer": 0.3980946264183258, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "learning_rate": 5.0407239819004526e-05, | |
| "loss": 0.5598, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "eval_loss": 1.0339125394821167, | |
| "eval_runtime": 42.5183, | |
| "eval_samples_per_second": 20.062, | |
| "eval_steps_per_second": 0.635, | |
| "eval_wer": 0.39359880111325196, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 20.56, | |
| "learning_rate": 4.588235294117647e-05, | |
| "loss": 0.5637, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 20.56, | |
| "eval_loss": 0.9660681486129761, | |
| "eval_runtime": 44.6662, | |
| "eval_samples_per_second": 19.097, | |
| "eval_steps_per_second": 0.604, | |
| "eval_wer": 0.399165061014772, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 21.5, | |
| "learning_rate": 4.1357466063348414e-05, | |
| "loss": 0.5396, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.5, | |
| "eval_loss": 1.05593740940094, | |
| "eval_runtime": 44.132, | |
| "eval_samples_per_second": 19.328, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.3923142795975166, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 22.43, | |
| "learning_rate": 3.683257918552037e-05, | |
| "loss": 0.5001, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 22.43, | |
| "eval_loss": 0.9710575938224792, | |
| "eval_runtime": 43.7084, | |
| "eval_samples_per_second": 19.516, | |
| "eval_steps_per_second": 0.618, | |
| "eval_wer": 0.38589167201883967, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 23.36, | |
| "learning_rate": 3.230769230769231e-05, | |
| "loss": 0.4797, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 23.36, | |
| "eval_loss": 0.9755498766899109, | |
| "eval_runtime": 43.0821, | |
| "eval_samples_per_second": 19.799, | |
| "eval_steps_per_second": 0.627, | |
| "eval_wer": 0.3775422821665596, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 24.3, | |
| "learning_rate": 2.7782805429864255e-05, | |
| "loss": 0.4993, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 24.3, | |
| "eval_loss": 0.9937364459037781, | |
| "eval_runtime": 43.0983, | |
| "eval_samples_per_second": 19.792, | |
| "eval_steps_per_second": 0.626, | |
| "eval_wer": 0.37315350032113037, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 25.23, | |
| "learning_rate": 2.3257918552036202e-05, | |
| "loss": 0.4728, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 25.23, | |
| "eval_loss": 1.028990387916565, | |
| "eval_runtime": 44.5142, | |
| "eval_samples_per_second": 19.162, | |
| "eval_steps_per_second": 0.607, | |
| "eval_wer": 0.37229715264397345, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 26.17, | |
| "learning_rate": 1.8733031674208146e-05, | |
| "loss": 0.4479, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 26.17, | |
| "eval_loss": 0.9995871186256409, | |
| "eval_runtime": 43.2227, | |
| "eval_samples_per_second": 19.735, | |
| "eval_steps_per_second": 0.625, | |
| "eval_wer": 0.36897880539499034, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 1.4208144796380091e-05, | |
| "loss": 0.4675, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "eval_loss": 1.0029648542404175, | |
| "eval_runtime": 42.9456, | |
| "eval_samples_per_second": 19.862, | |
| "eval_steps_per_second": 0.629, | |
| "eval_wer": 0.37101263112823807, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 28.04, | |
| "learning_rate": 9.683257918552037e-06, | |
| "loss": 0.4454, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 28.04, | |
| "eval_loss": 1.0169341564178467, | |
| "eval_runtime": 43.9547, | |
| "eval_samples_per_second": 19.406, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.37037037037037035, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 28.97, | |
| "learning_rate": 5.1583710407239815e-06, | |
| "loss": 0.4473, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 28.97, | |
| "eval_loss": 1.0175889730453491, | |
| "eval_runtime": 43.8487, | |
| "eval_samples_per_second": 19.453, | |
| "eval_steps_per_second": 0.616, | |
| "eval_wer": 0.3707985442089488, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 29.91, | |
| "learning_rate": 6.334841628959276e-07, | |
| "loss": 0.4005, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 29.91, | |
| "eval_loss": 1.0105210542678833, | |
| "eval_runtime": 42.6213, | |
| "eval_samples_per_second": 20.013, | |
| "eval_steps_per_second": 0.633, | |
| "eval_wer": 0.36951402269321343, | |
| "step": 3200 | |
| } | |
| ], | |
| "max_steps": 3210, | |
| "num_train_epochs": 30, | |
| "total_flos": 4.853367218401499e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |