| { | |
| "best_metric": 1.6565031400746337, | |
| "best_model_checkpoint": "./checkpoint-1000", | |
| "epoch": 66.66666666666667, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2.1000000000000002e-06, | |
| "loss": 1.5608, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 1.2470210790634155, | |
| "eval_runtime": 87.1381, | |
| "eval_samples_per_second": 2.743, | |
| "eval_steps_per_second": 0.344, | |
| "eval_wer": 9.884408846818967, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 0.7986, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 0.46967005729675293, | |
| "eval_runtime": 113.6552, | |
| "eval_samples_per_second": 2.103, | |
| "eval_steps_per_second": 0.264, | |
| "eval_wer": 17.629926276508602, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.100000000000001e-06, | |
| "loss": 0.2654, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.23039382696151733, | |
| "eval_runtime": 129.519, | |
| "eval_samples_per_second": 1.845, | |
| "eval_steps_per_second": 0.232, | |
| "eval_wer": 19.495767725493764, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.1537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "eval_loss": 0.17667056620121002, | |
| "eval_runtime": 163.2833, | |
| "eval_samples_per_second": 1.464, | |
| "eval_steps_per_second": 0.184, | |
| "eval_wer": 80.85009556748885, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 9.766666666666667e-06, | |
| "loss": 0.0873, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_loss": 0.1339290887117386, | |
| "eval_runtime": 144.8884, | |
| "eval_samples_per_second": 1.65, | |
| "eval_steps_per_second": 0.207, | |
| "eval_wer": 12.332756894511695, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 9.48888888888889e-06, | |
| "loss": 0.0314, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.09715640544891357, | |
| "eval_runtime": 92.9931, | |
| "eval_samples_per_second": 2.57, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 2.0478747610812778, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 9.211111111111111e-06, | |
| "loss": 0.0063, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "eval_loss": 0.11041156202554703, | |
| "eval_runtime": 92.8999, | |
| "eval_samples_per_second": 2.573, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6838081368890507, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 8.933333333333333e-06, | |
| "loss": 0.0037, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "eval_loss": 0.11353909224271774, | |
| "eval_runtime": 93.2005, | |
| "eval_samples_per_second": 2.564, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.9386547738236097, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 8.655555555555557e-06, | |
| "loss": 0.0027, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.11416648328304291, | |
| "eval_runtime": 93.2482, | |
| "eval_samples_per_second": 2.563, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.6656048056794392, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 8.377777777777779e-06, | |
| "loss": 0.0016, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "eval_loss": 0.1213652566075325, | |
| "eval_runtime": 93.3293, | |
| "eval_samples_per_second": 2.561, | |
| "eval_steps_per_second": 0.321, | |
| "eval_wer": 1.847638117775553, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "learning_rate": 8.1e-06, | |
| "loss": 0.0013, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "eval_loss": 0.11827071011066437, | |
| "eval_runtime": 93.0929, | |
| "eval_samples_per_second": 2.567, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.6018931464457997, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 7.822222222222224e-06, | |
| "loss": 0.001, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.1215524896979332, | |
| "eval_runtime": 93.0458, | |
| "eval_samples_per_second": 2.569, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.7384181305178845, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 21.67, | |
| "learning_rate": 7.544444444444445e-06, | |
| "loss": 0.0009, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 21.67, | |
| "eval_loss": 0.12276798486709595, | |
| "eval_runtime": 92.9939, | |
| "eval_samples_per_second": 2.57, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6291981432602167, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 23.33, | |
| "learning_rate": 7.266666666666668e-06, | |
| "loss": 0.0008, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 23.33, | |
| "eval_loss": 0.12465143948793411, | |
| "eval_runtime": 92.9329, | |
| "eval_samples_per_second": 2.572, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6656048056794392, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 6.9888888888888895e-06, | |
| "loss": 0.0008, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.12543533742427826, | |
| "eval_runtime": 93.1244, | |
| "eval_samples_per_second": 2.566, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.647401474469828, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 6.711111111111111e-06, | |
| "loss": 0.0007, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "eval_loss": 0.12520712614059448, | |
| "eval_runtime": 93.0139, | |
| "eval_samples_per_second": 2.57, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6291981432602167, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 28.33, | |
| "learning_rate": 6.433333333333333e-06, | |
| "loss": 0.0007, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 28.33, | |
| "eval_loss": 0.1271507740020752, | |
| "eval_runtime": 92.9162, | |
| "eval_samples_per_second": 2.572, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6291981432602167, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 6.155555555555556e-06, | |
| "loss": 0.0007, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.12834006547927856, | |
| "eval_runtime": 93.0797, | |
| "eval_samples_per_second": 2.568, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.702011468098662, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 31.67, | |
| "learning_rate": 5.877777777777778e-06, | |
| "loss": 0.0006, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 31.67, | |
| "eval_loss": 0.12884125113487244, | |
| "eval_runtime": 92.7701, | |
| "eval_samples_per_second": 2.576, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6656048056794392, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.0006, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 0.12947146594524384, | |
| "eval_runtime": 92.971, | |
| "eval_samples_per_second": 2.571, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.702011468098662, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 5.322222222222223e-06, | |
| "loss": 0.0006, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 0.12980106472969055, | |
| "eval_runtime": 92.8564, | |
| "eval_samples_per_second": 2.574, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6656048056794392, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 36.67, | |
| "learning_rate": 5.044444444444445e-06, | |
| "loss": 0.0006, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 36.67, | |
| "eval_loss": 0.13186444342136383, | |
| "eval_runtime": 93.0759, | |
| "eval_samples_per_second": 2.568, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.702011468098662, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 38.33, | |
| "learning_rate": 4.766666666666667e-06, | |
| "loss": 0.0006, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 38.33, | |
| "eval_loss": 0.13074839115142822, | |
| "eval_runtime": 93.1217, | |
| "eval_samples_per_second": 2.567, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.729316464913079, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 4.488888888888889e-06, | |
| "loss": 0.0005, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.1326516568660736, | |
| "eval_runtime": 92.8713, | |
| "eval_samples_per_second": 2.573, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.7475197961226905, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "learning_rate": 4.211111111111112e-06, | |
| "loss": 0.0005, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "eval_loss": 0.13262800872325897, | |
| "eval_runtime": 92.9841, | |
| "eval_samples_per_second": 2.57, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.702011468098662, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 43.33, | |
| "learning_rate": 3.9333333333333335e-06, | |
| "loss": 0.0005, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 43.33, | |
| "eval_loss": 0.13290846347808838, | |
| "eval_runtime": 92.8115, | |
| "eval_samples_per_second": 2.575, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.647401474469828, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 3.6555555555555562e-06, | |
| "loss": 0.0005, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 0.1338340789079666, | |
| "eval_runtime": 93.0235, | |
| "eval_samples_per_second": 2.569, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 1.702011468098662, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 46.67, | |
| "learning_rate": 3.377777777777778e-06, | |
| "loss": 0.0005, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 46.67, | |
| "eval_loss": 0.1339310258626938, | |
| "eval_runtime": 92.9221, | |
| "eval_samples_per_second": 2.572, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.674706471284245, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 48.33, | |
| "learning_rate": 3.1000000000000004e-06, | |
| "loss": 0.0005, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 48.33, | |
| "eval_loss": 0.13417784869670868, | |
| "eval_runtime": 92.8791, | |
| "eval_samples_per_second": 2.573, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.7111131337034675, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 2.8222222222222223e-06, | |
| "loss": 0.0005, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.13492102921009064, | |
| "eval_runtime": 92.778, | |
| "eval_samples_per_second": 2.576, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.7111131337034675, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 51.67, | |
| "learning_rate": 2.5444444444444446e-06, | |
| "loss": 0.0004, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 51.67, | |
| "eval_loss": 0.13491509854793549, | |
| "eval_runtime": 92.9969, | |
| "eval_samples_per_second": 2.57, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.7111131337034675, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 53.33, | |
| "learning_rate": 2.266666666666667e-06, | |
| "loss": 0.0005, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 53.33, | |
| "eval_loss": 0.13460427522659302, | |
| "eval_runtime": 92.9646, | |
| "eval_samples_per_second": 2.571, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6565031400746337, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "learning_rate": 1.988888888888889e-06, | |
| "loss": 0.0005, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_loss": 0.13541632890701294, | |
| "eval_runtime": 92.8779, | |
| "eval_samples_per_second": 2.573, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.7111131337034675, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 56.67, | |
| "learning_rate": 1.7111111111111112e-06, | |
| "loss": 0.0004, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 56.67, | |
| "eval_loss": 0.13564594089984894, | |
| "eval_runtime": 92.7968, | |
| "eval_samples_per_second": 2.576, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.7111131337034675, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 58.33, | |
| "learning_rate": 1.4333333333333335e-06, | |
| "loss": 0.0004, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 58.33, | |
| "eval_loss": 0.13617311418056488, | |
| "eval_runtime": 93.0052, | |
| "eval_samples_per_second": 2.57, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6565031400746337, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 1.1555555555555556e-06, | |
| "loss": 0.0004, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 0.13621629774570465, | |
| "eval_runtime": 92.7177, | |
| "eval_samples_per_second": 2.578, | |
| "eval_steps_per_second": 0.324, | |
| "eval_wer": 1.7111131337034675, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 61.67, | |
| "learning_rate": 8.777777777777778e-07, | |
| "loss": 0.0004, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 61.67, | |
| "eval_loss": 0.1365342140197754, | |
| "eval_runtime": 92.7764, | |
| "eval_samples_per_second": 2.576, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6565031400746337, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 63.33, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.0004, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 63.33, | |
| "eval_loss": 0.13664031028747559, | |
| "eval_runtime": 92.9018, | |
| "eval_samples_per_second": 2.573, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6565031400746337, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 3.2222222222222227e-07, | |
| "loss": 0.0004, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_loss": 0.13669337332248688, | |
| "eval_runtime": 93.0078, | |
| "eval_samples_per_second": 2.57, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6565031400746337, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "learning_rate": 4.444444444444445e-08, | |
| "loss": 0.0004, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "eval_loss": 0.1367739737033844, | |
| "eval_runtime": 92.9503, | |
| "eval_samples_per_second": 2.571, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 1.6565031400746337, | |
| "step": 1000 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 67, | |
| "total_flos": 1.839327907479552e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |