| { | |
| "best_metric": 11.399757966922145, | |
| "best_model_checkpoint": "/cluster/home/torstefl/Master/saved_model/Whisper/NB-RUND_02.06/checkpoint-3924", | |
| "epoch": 2.9977081741787623, | |
| "eval_steps": 327, | |
| "global_step": 3924, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 19.654556274414062, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.3443, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 3.5712509155273438, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.5902, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 3.6999974250793457, | |
| "learning_rate": 3e-06, | |
| "loss": 0.3244, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_rundkast_loss": 0.2645014524459839, | |
| "eval_rundkast_runtime": 288.2236, | |
| "eval_rundkast_samples_per_second": 4.646, | |
| "eval_rundkast_steps_per_second": 0.097, | |
| "eval_rundkast_wer": 13.709806430971044, | |
| "step": 327 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_nb_samtale_loss": 0.38987481594085693, | |
| "eval_nb_samtale_runtime": 137.2637, | |
| "eval_nb_samtale_samples_per_second": 3.883, | |
| "eval_nb_samtale_steps_per_second": 0.087, | |
| "eval_nb_samtale_wer": 14.713541666666666, | |
| "step": 327 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_bigbrother_loss": 0.772404134273529, | |
| "eval_bigbrother_runtime": 288.9747, | |
| "eval_bigbrother_samples_per_second": 4.741, | |
| "eval_bigbrother_steps_per_second": 0.1, | |
| "eval_bigbrother_wer": 36.54233464915814, | |
| "step": 327 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_NB_RUND_loss": 0.32218727469444275, | |
| "eval_NB_RUND_runtime": 433.4771, | |
| "eval_NB_RUND_samples_per_second": 4.319, | |
| "eval_NB_RUND_steps_per_second": 0.09, | |
| "eval_NB_RUND_wer": 14.207341670028237, | |
| "step": 327 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 3.632507085800171, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.2775, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 3.317390203475952, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2567, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 3.4814083576202393, | |
| "learning_rate": 6e-06, | |
| "loss": 0.2431, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_rundkast_loss": 0.216255784034729, | |
| "eval_rundkast_runtime": 280.6623, | |
| "eval_rundkast_samples_per_second": 4.771, | |
| "eval_rundkast_steps_per_second": 0.1, | |
| "eval_rundkast_wer": 12.158054711246201, | |
| "step": 654 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_nb_samtale_loss": 0.32306087017059326, | |
| "eval_nb_samtale_runtime": 134.9288, | |
| "eval_nb_samtale_samples_per_second": 3.95, | |
| "eval_nb_samtale_steps_per_second": 0.089, | |
| "eval_nb_samtale_wer": 13.802083333333334, | |
| "step": 654 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_bigbrother_loss": 0.6955679655075073, | |
| "eval_bigbrother_runtime": 288.247, | |
| "eval_bigbrother_samples_per_second": 4.753, | |
| "eval_bigbrother_steps_per_second": 0.101, | |
| "eval_bigbrother_wer": 37.13042777733022, | |
| "step": 654 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_NB_RUND_loss": 0.26444458961486816, | |
| "eval_NB_RUND_runtime": 428.1229, | |
| "eval_NB_RUND_samples_per_second": 4.373, | |
| "eval_NB_RUND_steps_per_second": 0.091, | |
| "eval_NB_RUND_wer": 12.972972972972974, | |
| "step": 654 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 3.4407951831817627, | |
| "learning_rate": 7e-06, | |
| "loss": 0.2299, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 3.4071555137634277, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.2244, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 5.109046459197998, | |
| "learning_rate": 9e-06, | |
| "loss": 0.2128, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_rundkast_loss": 0.19231900572776794, | |
| "eval_rundkast_runtime": 278.5019, | |
| "eval_rundkast_samples_per_second": 4.808, | |
| "eval_rundkast_steps_per_second": 0.101, | |
| "eval_rundkast_wer": 11.646136618141098, | |
| "step": 981 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_nb_samtale_loss": 0.30107787251472473, | |
| "eval_nb_samtale_runtime": 136.2934, | |
| "eval_nb_samtale_samples_per_second": 3.911, | |
| "eval_nb_samtale_steps_per_second": 0.088, | |
| "eval_nb_samtale_wer": 13.736979166666666, | |
| "step": 981 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_bigbrother_loss": 0.6737728714942932, | |
| "eval_bigbrother_runtime": 289.591, | |
| "eval_bigbrother_samples_per_second": 4.731, | |
| "eval_bigbrother_steps_per_second": 0.1, | |
| "eval_bigbrother_wer": 34.5524852976718, | |
| "step": 981 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_NB_RUND_loss": 0.2399020791053772, | |
| "eval_NB_RUND_runtime": 431.6962, | |
| "eval_NB_RUND_samples_per_second": 4.336, | |
| "eval_NB_RUND_steps_per_second": 0.09, | |
| "eval_NB_RUND_wer": 12.682533279548206, | |
| "step": 981 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.9982640743255615, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2085, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 4.208399772644043, | |
| "learning_rate": 9.973849372384938e-06, | |
| "loss": 0.2017, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 2.8742215633392334, | |
| "learning_rate": 9.947698744769874e-06, | |
| "loss": 0.1993, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 3.360987901687622, | |
| "learning_rate": 9.921548117154812e-06, | |
| "loss": 0.2002, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_rundkast_loss": 0.18137970566749573, | |
| "eval_rundkast_runtime": 279.393, | |
| "eval_rundkast_samples_per_second": 4.793, | |
| "eval_rundkast_steps_per_second": 0.1, | |
| "eval_rundkast_wer": 11.486162214045754, | |
| "step": 1308 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_nb_samtale_loss": 0.28179603815078735, | |
| "eval_nb_samtale_runtime": 136.0323, | |
| "eval_nb_samtale_samples_per_second": 3.918, | |
| "eval_nb_samtale_steps_per_second": 0.088, | |
| "eval_nb_samtale_wer": 13.264973958333334, | |
| "step": 1308 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bigbrother_loss": 0.6532533764839172, | |
| "eval_bigbrother_runtime": 290.948, | |
| "eval_bigbrother_samples_per_second": 4.709, | |
| "eval_bigbrother_steps_per_second": 0.1, | |
| "eval_bigbrother_wer": 36.59067107065174, | |
| "step": 1308 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_NB_RUND_loss": 0.22604148089885712, | |
| "eval_NB_RUND_runtime": 428.2146, | |
| "eval_NB_RUND_samples_per_second": 4.372, | |
| "eval_NB_RUND_steps_per_second": 0.091, | |
| "eval_NB_RUND_wer": 12.36789027833804, | |
| "step": 1308 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 2.6161742210388184, | |
| "learning_rate": 9.89539748953975e-06, | |
| "loss": 0.1459, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 3.147104501724243, | |
| "learning_rate": 9.869246861924687e-06, | |
| "loss": 0.1416, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 2.851698875427246, | |
| "learning_rate": 9.843096234309623e-06, | |
| "loss": 0.1375, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_rundkast_loss": 0.1737283319234848, | |
| "eval_rundkast_runtime": 280.5915, | |
| "eval_rundkast_samples_per_second": 4.772, | |
| "eval_rundkast_steps_per_second": 0.1, | |
| "eval_rundkast_wer": 11.158214685650297, | |
| "step": 1635 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_nb_samtale_loss": 0.27981099486351013, | |
| "eval_nb_samtale_runtime": 136.8976, | |
| "eval_nb_samtale_samples_per_second": 3.893, | |
| "eval_nb_samtale_steps_per_second": 0.088, | |
| "eval_nb_samtale_wer": 12.939453125, | |
| "step": 1635 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_bigbrother_loss": 0.6777428388595581, | |
| "eval_bigbrother_runtime": 287.7862, | |
| "eval_bigbrother_samples_per_second": 4.76, | |
| "eval_bigbrother_steps_per_second": 0.101, | |
| "eval_bigbrother_wer": 34.77805526464191, | |
| "step": 1635 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_NB_RUND_loss": 0.21971319615840912, | |
| "eval_NB_RUND_runtime": 432.7546, | |
| "eval_NB_RUND_samples_per_second": 4.326, | |
| "eval_NB_RUND_steps_per_second": 0.09, | |
| "eval_NB_RUND_wer": 12.041145623235176, | |
| "step": 1635 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 3.1212966442108154, | |
| "learning_rate": 9.81694560669456e-06, | |
| "loss": 0.1413, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 3.1864242553710938, | |
| "learning_rate": 9.790794979079498e-06, | |
| "loss": 0.1415, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 2.8078577518463135, | |
| "learning_rate": 9.764644351464436e-06, | |
| "loss": 0.1436, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_rundkast_loss": 0.1729215830564499, | |
| "eval_rundkast_runtime": 280.9351, | |
| "eval_rundkast_samples_per_second": 4.766, | |
| "eval_rundkast_steps_per_second": 0.1, | |
| "eval_rundkast_wer": 10.862262038073908, | |
| "step": 1962 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_nb_samtale_loss": 0.2810090184211731, | |
| "eval_nb_samtale_runtime": 134.891, | |
| "eval_nb_samtale_samples_per_second": 3.951, | |
| "eval_nb_samtale_steps_per_second": 0.089, | |
| "eval_nb_samtale_wer": 12.451171875, | |
| "step": 1962 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_bigbrother_loss": 0.6837819814682007, | |
| "eval_bigbrother_runtime": 285.9416, | |
| "eval_bigbrother_samples_per_second": 4.791, | |
| "eval_bigbrother_steps_per_second": 0.101, | |
| "eval_bigbrother_wer": 33.82743897526787, | |
| "step": 1962 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_NB_RUND_loss": 0.21930767595767975, | |
| "eval_NB_RUND_runtime": 436.0144, | |
| "eval_NB_RUND_samples_per_second": 4.293, | |
| "eval_NB_RUND_steps_per_second": 0.089, | |
| "eval_NB_RUND_wer": 11.649858814037918, | |
| "step": 1962 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 2.6239843368530273, | |
| "learning_rate": 9.738493723849372e-06, | |
| "loss": 0.1355, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 2.6334943771362305, | |
| "learning_rate": 9.71234309623431e-06, | |
| "loss": 0.1369, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 2.887988805770874, | |
| "learning_rate": 9.686192468619247e-06, | |
| "loss": 0.1334, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_rundkast_loss": 0.1666535586118698, | |
| "eval_rundkast_runtime": 280.7891, | |
| "eval_rundkast_samples_per_second": 4.769, | |
| "eval_rundkast_steps_per_second": 0.1, | |
| "eval_rundkast_wer": 11.022236442169254, | |
| "step": 2289 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_nb_samtale_loss": 0.2767205238342285, | |
| "eval_nb_samtale_runtime": 137.903, | |
| "eval_nb_samtale_samples_per_second": 3.865, | |
| "eval_nb_samtale_steps_per_second": 0.087, | |
| "eval_nb_samtale_wer": 12.59765625, | |
| "step": 2289 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_bigbrother_loss": 0.6614672541618347, | |
| "eval_bigbrother_runtime": 297.4947, | |
| "eval_bigbrother_samples_per_second": 4.605, | |
| "eval_bigbrother_steps_per_second": 0.097, | |
| "eval_bigbrother_wer": 38.96721179408684, | |
| "step": 2289 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_NB_RUND_loss": 0.21262812614440918, | |
| "eval_NB_RUND_runtime": 432.1545, | |
| "eval_NB_RUND_samples_per_second": 4.332, | |
| "eval_NB_RUND_steps_per_second": 0.09, | |
| "eval_NB_RUND_wer": 11.803146430012102, | |
| "step": 2289 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 2.951620578765869, | |
| "learning_rate": 9.660041841004185e-06, | |
| "loss": 0.1381, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 2.709320306777954, | |
| "learning_rate": 9.633891213389122e-06, | |
| "loss": 0.1371, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 2.9003231525421143, | |
| "learning_rate": 9.60774058577406e-06, | |
| "loss": 0.1336, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 2.6803975105285645, | |
| "learning_rate": 9.581589958158996e-06, | |
| "loss": 0.1419, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_rundkast_loss": 0.163014754652977, | |
| "eval_rundkast_runtime": 278.1861, | |
| "eval_rundkast_samples_per_second": 4.813, | |
| "eval_rundkast_steps_per_second": 0.101, | |
| "eval_rundkast_wer": 10.630299152135658, | |
| "step": 2616 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_nb_samtale_loss": 0.26299360394477844, | |
| "eval_nb_samtale_runtime": 135.7564, | |
| "eval_nb_samtale_samples_per_second": 3.926, | |
| "eval_nb_samtale_steps_per_second": 0.088, | |
| "eval_nb_samtale_wer": 12.744140625, | |
| "step": 2616 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bigbrother_loss": 0.6536648273468018, | |
| "eval_bigbrother_runtime": 293.9224, | |
| "eval_bigbrother_samples_per_second": 4.661, | |
| "eval_bigbrother_steps_per_second": 0.099, | |
| "eval_bigbrother_wer": 37.380165955047126, | |
| "step": 2616 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_NB_RUND_loss": 0.20609000325202942, | |
| "eval_NB_RUND_runtime": 431.4855, | |
| "eval_NB_RUND_samples_per_second": 4.339, | |
| "eval_NB_RUND_steps_per_second": 0.09, | |
| "eval_NB_RUND_wer": 11.678096006454215, | |
| "step": 2616 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 2.4410240650177, | |
| "learning_rate": 9.555439330543933e-06, | |
| "loss": 0.0925, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 2.712162733078003, | |
| "learning_rate": 9.529288702928871e-06, | |
| "loss": 0.0833, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 2.0603525638580322, | |
| "learning_rate": 9.503138075313809e-06, | |
| "loss": 0.0789, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_rundkast_loss": 0.17048095166683197, | |
| "eval_rundkast_runtime": 278.2837, | |
| "eval_rundkast_samples_per_second": 4.812, | |
| "eval_rundkast_steps_per_second": 0.101, | |
| "eval_rundkast_wer": 10.406334986402175, | |
| "step": 2943 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_nb_samtale_loss": 0.2874841094017029, | |
| "eval_nb_samtale_runtime": 136.4854, | |
| "eval_nb_samtale_samples_per_second": 3.905, | |
| "eval_nb_samtale_steps_per_second": 0.088, | |
| "eval_nb_samtale_wer": 12.679036458333334, | |
| "step": 2943 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_bigbrother_loss": 0.691508412361145, | |
| "eval_bigbrother_runtime": 287.4115, | |
| "eval_bigbrother_samples_per_second": 4.767, | |
| "eval_bigbrother_steps_per_second": 0.101, | |
| "eval_bigbrother_wer": 33.75493434302747, | |
| "step": 2943 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_NB_RUND_loss": 0.22030942142009735, | |
| "eval_NB_RUND_runtime": 428.4348, | |
| "eval_NB_RUND_samples_per_second": 4.369, | |
| "eval_NB_RUND_steps_per_second": 0.091, | |
| "eval_NB_RUND_wer": 11.53287615974183, | |
| "step": 2943 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 1.8269718885421753, | |
| "learning_rate": 9.476987447698746e-06, | |
| "loss": 0.0816, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 2.921381711959839, | |
| "learning_rate": 9.450836820083684e-06, | |
| "loss": 0.0822, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 2.1898179054260254, | |
| "learning_rate": 9.42468619246862e-06, | |
| "loss": 0.0816, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_rundkast_loss": 0.17146149277687073, | |
| "eval_rundkast_runtime": 276.8281, | |
| "eval_rundkast_samples_per_second": 4.837, | |
| "eval_rundkast_steps_per_second": 0.101, | |
| "eval_rundkast_wer": 10.574308110702288, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_nb_samtale_loss": 0.2887733280658722, | |
| "eval_nb_samtale_runtime": 135.3506, | |
| "eval_nb_samtale_samples_per_second": 3.938, | |
| "eval_nb_samtale_steps_per_second": 0.089, | |
| "eval_nb_samtale_wer": 12.744140625, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_bigbrother_loss": 0.7190759778022766, | |
| "eval_bigbrother_runtime": 287.0381, | |
| "eval_bigbrother_samples_per_second": 4.773, | |
| "eval_bigbrother_steps_per_second": 0.101, | |
| "eval_bigbrother_wer": 34.520261016676066, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_NB_RUND_loss": 0.2218938171863556, | |
| "eval_NB_RUND_runtime": 426.5073, | |
| "eval_NB_RUND_samples_per_second": 4.389, | |
| "eval_NB_RUND_steps_per_second": 0.091, | |
| "eval_NB_RUND_wer": 11.649858814037918, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 2.9775750637054443, | |
| "learning_rate": 9.398535564853557e-06, | |
| "loss": 0.0831, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.9110767841339111, | |
| "learning_rate": 9.372384937238495e-06, | |
| "loss": 0.0826, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 1.5339910984039307, | |
| "learning_rate": 9.346234309623433e-06, | |
| "loss": 0.0863, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_rundkast_loss": 0.1625809371471405, | |
| "eval_rundkast_runtime": 278.5757, | |
| "eval_rundkast_samples_per_second": 4.807, | |
| "eval_rundkast_steps_per_second": 0.101, | |
| "eval_rundkast_wer": 10.358342665173573, | |
| "step": 3597 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_nb_samtale_loss": 0.28124529123306274, | |
| "eval_nb_samtale_runtime": 136.7892, | |
| "eval_nb_samtale_samples_per_second": 3.897, | |
| "eval_nb_samtale_steps_per_second": 0.088, | |
| "eval_nb_samtale_wer": 12.483723958333332, | |
| "step": 3597 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_bigbrother_loss": 0.697869598865509, | |
| "eval_bigbrother_runtime": 286.0407, | |
| "eval_bigbrother_samples_per_second": 4.79, | |
| "eval_bigbrother_steps_per_second": 0.101, | |
| "eval_bigbrother_wer": 35.48698944654797, | |
| "step": 3597 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_NB_RUND_loss": 0.2132958024740219, | |
| "eval_NB_RUND_runtime": 426.3524, | |
| "eval_NB_RUND_samples_per_second": 4.391, | |
| "eval_NB_RUND_steps_per_second": 0.091, | |
| "eval_NB_RUND_wer": 11.411859620814845, | |
| "step": 3597 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 2.007660150527954, | |
| "learning_rate": 9.320083682008368e-06, | |
| "loss": 0.0852, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 2.9868288040161133, | |
| "learning_rate": 9.293933054393306e-06, | |
| "loss": 0.084, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 3.0625417232513428, | |
| "learning_rate": 9.267782426778244e-06, | |
| "loss": 0.0862, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 1.9115489721298218, | |
| "learning_rate": 9.241631799163181e-06, | |
| "loss": 0.0873, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_rundkast_loss": 0.16490745544433594, | |
| "eval_rundkast_runtime": 273.4131, | |
| "eval_rundkast_samples_per_second": 4.897, | |
| "eval_rundkast_steps_per_second": 0.102, | |
| "eval_rundkast_wer": 10.51031834906415, | |
| "step": 3924 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_nb_samtale_loss": 0.2840913236141205, | |
| "eval_nb_samtale_runtime": 134.8591, | |
| "eval_nb_samtale_samples_per_second": 3.952, | |
| "eval_nb_samtale_steps_per_second": 0.089, | |
| "eval_nb_samtale_wer": 12.3046875, | |
| "step": 3924 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bigbrother_loss": 0.7155491709709167, | |
| "eval_bigbrother_runtime": 294.2668, | |
| "eval_bigbrother_samples_per_second": 4.656, | |
| "eval_bigbrother_steps_per_second": 0.099, | |
| "eval_bigbrother_wer": 36.75984854587932, | |
| "step": 3924 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_NB_RUND_loss": 0.21525567770004272, | |
| "eval_NB_RUND_runtime": 427.008, | |
| "eval_NB_RUND_samples_per_second": 4.384, | |
| "eval_NB_RUND_steps_per_second": 0.091, | |
| "eval_NB_RUND_wer": 11.399757966922145, | |
| "step": 3924 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 39270, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 327, | |
| "total_flos": 5.435448297947136e+19, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |