| { | |
| "best_metric": 17.39402275853445, | |
| "best_model_checkpoint": "whisper-small-german-retrain/checkpoint-3000", | |
| "epoch": 130.43478260869566, | |
| "eval_steps": 1000, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.1739130434782608, | |
| "grad_norm": 2.7223403453826904, | |
| "learning_rate": 1.75e-06, | |
| "loss": 0.6229, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.3478260869565215, | |
| "grad_norm": 2.2247281074523926, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.3885, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.521739130434782, | |
| "grad_norm": 1.4519989490509033, | |
| "learning_rate": 5.25e-06, | |
| "loss": 0.2281, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 1.1439716815948486, | |
| "learning_rate": 7e-06, | |
| "loss": 0.1268, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.869565217391305, | |
| "grad_norm": 0.7530131340026855, | |
| "learning_rate": 8.75e-06, | |
| "loss": 0.0546, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 13.043478260869565, | |
| "grad_norm": 1.6841254234313965, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.0209, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.217391304347826, | |
| "grad_norm": 0.25525057315826416, | |
| "learning_rate": 1.2249999999999998e-05, | |
| "loss": 0.0098, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.227842316031456, | |
| "learning_rate": 1.4e-05, | |
| "loss": 0.0065, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.565217391304348, | |
| "grad_norm": 0.9949818253517151, | |
| "learning_rate": 1.575e-05, | |
| "loss": 0.0078, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 21.73913043478261, | |
| "grad_norm": 0.39849549531936646, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.0078, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 23.91304347826087, | |
| "grad_norm": 0.8715730905532837, | |
| "learning_rate": 1.7305555555555553e-05, | |
| "loss": 0.0071, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "grad_norm": 0.20837093889713287, | |
| "learning_rate": 1.711111111111111e-05, | |
| "loss": 0.0049, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 28.26086956521739, | |
| "grad_norm": 0.10564578324556351, | |
| "learning_rate": 1.6916666666666667e-05, | |
| "loss": 0.0036, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 30.434782608695652, | |
| "grad_norm": 0.2511199712753296, | |
| "learning_rate": 1.6722222222222222e-05, | |
| "loss": 0.0029, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 32.608695652173914, | |
| "grad_norm": 0.05474904552102089, | |
| "learning_rate": 1.6527777777777777e-05, | |
| "loss": 0.0016, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "grad_norm": 0.0346197709441185, | |
| "learning_rate": 1.633333333333333e-05, | |
| "loss": 0.0014, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 36.95652173913044, | |
| "grad_norm": 0.022180059924721718, | |
| "learning_rate": 1.613888888888889e-05, | |
| "loss": 0.0017, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 39.130434782608695, | |
| "grad_norm": 0.011027761735022068, | |
| "learning_rate": 1.594444444444444e-05, | |
| "loss": 0.0009, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 41.30434782608695, | |
| "grad_norm": 0.010779116302728653, | |
| "learning_rate": 1.575e-05, | |
| "loss": 0.0007, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 43.47826086956522, | |
| "grad_norm": 0.008251098915934563, | |
| "learning_rate": 1.5555555555555555e-05, | |
| "loss": 0.0006, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 43.47826086956522, | |
| "eval_loss": 0.5149791240692139, | |
| "eval_runtime": 40.8401, | |
| "eval_samples_per_second": 7.027, | |
| "eval_steps_per_second": 0.22, | |
| "eval_wer": 17.994247842941103, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 45.65217391304348, | |
| "grad_norm": 0.007410943973809481, | |
| "learning_rate": 1.536111111111111e-05, | |
| "loss": 0.0006, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 47.82608695652174, | |
| "grad_norm": 0.006867844145745039, | |
| "learning_rate": 1.5166666666666667e-05, | |
| "loss": 0.0005, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.00803290493786335, | |
| "learning_rate": 1.497222222222222e-05, | |
| "loss": 0.0005, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 52.17391304347826, | |
| "grad_norm": 0.005789933726191521, | |
| "learning_rate": 1.4777777777777777e-05, | |
| "loss": 0.0005, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 54.34782608695652, | |
| "grad_norm": 0.005295329727232456, | |
| "learning_rate": 1.4583333333333333e-05, | |
| "loss": 0.0004, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 56.52173913043478, | |
| "grad_norm": 0.004821299575269222, | |
| "learning_rate": 1.4388888888888886e-05, | |
| "loss": 0.0004, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 58.69565217391305, | |
| "grad_norm": 0.004488619975745678, | |
| "learning_rate": 1.4194444444444443e-05, | |
| "loss": 0.0004, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 60.869565217391305, | |
| "grad_norm": 0.0040580639615654945, | |
| "learning_rate": 1.4e-05, | |
| "loss": 0.0004, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 63.04347826086956, | |
| "grad_norm": 0.004379584453999996, | |
| "learning_rate": 1.3805555555555553e-05, | |
| "loss": 0.0003, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 65.21739130434783, | |
| "grad_norm": 0.003948231227695942, | |
| "learning_rate": 1.361111111111111e-05, | |
| "loss": 0.0003, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 67.3913043478261, | |
| "grad_norm": 0.0037093530409038067, | |
| "learning_rate": 1.3416666666666666e-05, | |
| "loss": 0.0003, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 69.56521739130434, | |
| "grad_norm": 0.0037487975787371397, | |
| "learning_rate": 1.3222222222222221e-05, | |
| "loss": 0.0003, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 71.73913043478261, | |
| "grad_norm": 0.003608859609812498, | |
| "learning_rate": 1.3027777777777776e-05, | |
| "loss": 0.0003, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 73.91304347826087, | |
| "grad_norm": 0.0034225275740027428, | |
| "learning_rate": 1.2833333333333331e-05, | |
| "loss": 0.0003, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 76.08695652173913, | |
| "grad_norm": 0.003319317242130637, | |
| "learning_rate": 1.2638888888888888e-05, | |
| "loss": 0.0003, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 78.26086956521739, | |
| "grad_norm": 0.003253102768212557, | |
| "learning_rate": 1.2444444444444445e-05, | |
| "loss": 0.0002, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 80.43478260869566, | |
| "grad_norm": 0.003029271960258484, | |
| "learning_rate": 1.2249999999999998e-05, | |
| "loss": 0.0002, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 82.6086956521739, | |
| "grad_norm": 0.002931032096967101, | |
| "learning_rate": 1.2055555555555555e-05, | |
| "loss": 0.0002, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 84.78260869565217, | |
| "grad_norm": 0.002701626857742667, | |
| "learning_rate": 1.1861111111111111e-05, | |
| "loss": 0.0002, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "grad_norm": 0.002894147066399455, | |
| "learning_rate": 1.1666666666666665e-05, | |
| "loss": 0.0002, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "eval_loss": 0.5610591173171997, | |
| "eval_runtime": 40.2761, | |
| "eval_samples_per_second": 7.126, | |
| "eval_steps_per_second": 0.223, | |
| "eval_wer": 17.456546204826807, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 89.1304347826087, | |
| "grad_norm": 0.0025357184931635857, | |
| "learning_rate": 1.1472222222222221e-05, | |
| "loss": 0.0002, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 91.30434782608695, | |
| "grad_norm": 0.0023813284933567047, | |
| "learning_rate": 1.1277777777777778e-05, | |
| "loss": 0.0002, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 93.47826086956522, | |
| "grad_norm": 0.002499406924471259, | |
| "learning_rate": 1.1083333333333331e-05, | |
| "loss": 0.0002, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 95.65217391304348, | |
| "grad_norm": 0.002111413050442934, | |
| "learning_rate": 1.0888888888888888e-05, | |
| "loss": 0.0002, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 97.82608695652173, | |
| "grad_norm": 0.0020701128523796797, | |
| "learning_rate": 1.0694444444444444e-05, | |
| "loss": 0.0002, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 0.0028157017659395933, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.0002, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 102.17391304347827, | |
| "grad_norm": 0.0020910503808408976, | |
| "learning_rate": 1.0305555555555554e-05, | |
| "loss": 0.0002, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 104.34782608695652, | |
| "grad_norm": 0.0019400820601731539, | |
| "learning_rate": 1.011111111111111e-05, | |
| "loss": 0.0002, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 106.52173913043478, | |
| "grad_norm": 0.0016938770422711968, | |
| "learning_rate": 9.916666666666666e-06, | |
| "loss": 0.0002, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 108.69565217391305, | |
| "grad_norm": 0.0018923500319942832, | |
| "learning_rate": 9.722222222222221e-06, | |
| "loss": 0.0002, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 110.8695652173913, | |
| "grad_norm": 0.0017991146305575967, | |
| "learning_rate": 9.527777777777776e-06, | |
| "loss": 0.0001, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 113.04347826086956, | |
| "grad_norm": 0.0018069138750433922, | |
| "learning_rate": 9.333333333333333e-06, | |
| "loss": 0.0001, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 115.21739130434783, | |
| "grad_norm": 0.0017327632522210479, | |
| "learning_rate": 9.13888888888889e-06, | |
| "loss": 0.0001, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 117.3913043478261, | |
| "grad_norm": 0.0015717543428763747, | |
| "learning_rate": 8.944444444444443e-06, | |
| "loss": 0.0001, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 119.56521739130434, | |
| "grad_norm": 0.001588036073371768, | |
| "learning_rate": 8.75e-06, | |
| "loss": 0.0001, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 121.73913043478261, | |
| "grad_norm": 0.0015617014141753316, | |
| "learning_rate": 8.555555555555554e-06, | |
| "loss": 0.0001, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 123.91304347826087, | |
| "grad_norm": 0.0015488613862544298, | |
| "learning_rate": 8.361111111111111e-06, | |
| "loss": 0.0001, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 126.08695652173913, | |
| "grad_norm": 0.0016123673412948847, | |
| "learning_rate": 8.166666666666666e-06, | |
| "loss": 0.0001, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 128.2608695652174, | |
| "grad_norm": 0.0014842275995761156, | |
| "learning_rate": 7.97222222222222e-06, | |
| "loss": 0.0001, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 130.43478260869566, | |
| "grad_norm": 0.001505482941865921, | |
| "learning_rate": 7.777777777777777e-06, | |
| "loss": 0.0001, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 130.43478260869566, | |
| "eval_loss": 0.5873442888259888, | |
| "eval_runtime": 40.5248, | |
| "eval_samples_per_second": 7.082, | |
| "eval_steps_per_second": 0.222, | |
| "eval_wer": 17.39402275853445, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 218, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.43204299022336e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |