| { | |
| "best_metric": 15.188899501220417, | |
| "best_model_checkpoint": "all_lang_models/punjabi_models/whisper-medium-pa_alldata_multigpu/checkpoint-2000", | |
| "epoch": 3.9525691699604746, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3.5436764027111585e-06, | |
| "loss": 2.3388, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 6.7460168970208566e-06, | |
| "loss": 1.2063, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 8.123811710560552e-06, | |
| "loss": 0.6582, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.014161010104347e-06, | |
| "loss": 0.402, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.673229499590088e-06, | |
| "loss": 0.2813, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.994059405940595e-06, | |
| "loss": 0.2054, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.974257425742576e-06, | |
| "loss": 0.1687, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.954455445544556e-06, | |
| "loss": 0.1521, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.934653465346535e-06, | |
| "loss": 0.1404, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.914851485148515e-06, | |
| "loss": 0.1322, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.12457275390625, | |
| "eval_runtime": 339.7255, | |
| "eval_samples_per_second": 9.625, | |
| "eval_steps_per_second": 0.077, | |
| "eval_wer": 29.207789451342457, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.895049504950496e-06, | |
| "loss": 0.1206, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.875247524752476e-06, | |
| "loss": 0.1193, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 9.855445544554456e-06, | |
| "loss": 0.1126, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 9.835643564356437e-06, | |
| "loss": 0.1105, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.815841584158416e-06, | |
| "loss": 0.1103, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 9.796039603960396e-06, | |
| "loss": 0.1041, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 9.776237623762377e-06, | |
| "loss": 0.1011, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.756435643564357e-06, | |
| "loss": 0.0997, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.736633663366337e-06, | |
| "loss": 0.0928, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.716831683168318e-06, | |
| "loss": 0.0929, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.09466552734375, | |
| "eval_runtime": 371.39, | |
| "eval_samples_per_second": 8.805, | |
| "eval_steps_per_second": 0.07, | |
| "eval_wer": 22.495489759100074, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 9.697029702970298e-06, | |
| "loss": 0.0939, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 9.677227722772277e-06, | |
| "loss": 0.089, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 9.657425742574257e-06, | |
| "loss": 0.0889, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 9.637623762376238e-06, | |
| "loss": 0.086, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 9.617821782178218e-06, | |
| "loss": 0.0876, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 9.598019801980199e-06, | |
| "loss": 0.0802, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 9.57821782178218e-06, | |
| "loss": 0.0809, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 9.558415841584158e-06, | |
| "loss": 0.0803, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 9.538613861386138e-06, | |
| "loss": 0.0787, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 9.518811881188119e-06, | |
| "loss": 0.0798, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_loss": 0.08154296875, | |
| "eval_runtime": 353.4734, | |
| "eval_samples_per_second": 9.251, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 19.680568820969967, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 9.4990099009901e-06, | |
| "loss": 0.0785, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 9.47920792079208e-06, | |
| "loss": 0.0785, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 9.45940594059406e-06, | |
| "loss": 0.0736, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.43960396039604e-06, | |
| "loss": 0.0737, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 9.41980198019802e-06, | |
| "loss": 0.0741, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 9.4e-06, | |
| "loss": 0.0739, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 9.38019801980198e-06, | |
| "loss": 0.0729, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 9.360396039603962e-06, | |
| "loss": 0.0718, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 9.340594059405941e-06, | |
| "loss": 0.0701, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 9.320792079207922e-06, | |
| "loss": 0.0705, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_loss": 0.074951171875, | |
| "eval_runtime": 358.951, | |
| "eval_samples_per_second": 9.11, | |
| "eval_steps_per_second": 0.072, | |
| "eval_wer": 18.505253104106973, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.300990099009902e-06, | |
| "loss": 0.0689, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 9.281188118811882e-06, | |
| "loss": 0.0698, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 9.261386138613863e-06, | |
| "loss": 0.0683, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.241584158415843e-06, | |
| "loss": 0.0687, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 9.221782178217824e-06, | |
| "loss": 0.0685, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 9.201980198019802e-06, | |
| "loss": 0.0662, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 9.182178217821783e-06, | |
| "loss": 0.0664, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 9.162376237623763e-06, | |
| "loss": 0.0679, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 9.142574257425744e-06, | |
| "loss": 0.0658, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 9.122772277227724e-06, | |
| "loss": 0.0636, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 0.07171630859375, | |
| "eval_runtime": 702.6659, | |
| "eval_samples_per_second": 4.654, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 18.462803777990025, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.102970297029705e-06, | |
| "loss": 0.0604, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.083168316831683e-06, | |
| "loss": 0.055, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.063366336633664e-06, | |
| "loss": 0.0547, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.043564356435644e-06, | |
| "loss": 0.0562, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 9.023762376237625e-06, | |
| "loss": 0.0526, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 9.003960396039605e-06, | |
| "loss": 0.0536, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 8.984158415841586e-06, | |
| "loss": 0.0558, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 8.964356435643566e-06, | |
| "loss": 0.0553, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.944554455445545e-06, | |
| "loss": 0.055, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 8.924752475247525e-06, | |
| "loss": 0.0544, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_loss": 0.070068359375, | |
| "eval_runtime": 329.5551, | |
| "eval_samples_per_second": 9.922, | |
| "eval_steps_per_second": 0.079, | |
| "eval_wer": 17.107078425130002, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 8.904950495049506e-06, | |
| "loss": 0.054, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 8.885148514851486e-06, | |
| "loss": 0.0522, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 8.865346534653467e-06, | |
| "loss": 0.0551, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 8.845544554455447e-06, | |
| "loss": 0.053, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 8.825742574257426e-06, | |
| "loss": 0.0543, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 8.805940594059406e-06, | |
| "loss": 0.055, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 8.786138613861387e-06, | |
| "loss": 0.0519, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 8.766336633663367e-06, | |
| "loss": 0.0524, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 8.746534653465347e-06, | |
| "loss": 0.0545, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 8.726732673267328e-06, | |
| "loss": 0.0552, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 0.06689453125, | |
| "eval_runtime": 525.7194, | |
| "eval_samples_per_second": 6.22, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 16.605645760373555, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 8.706930693069308e-06, | |
| "loss": 0.0545, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 8.687128712871287e-06, | |
| "loss": 0.054, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 8.667326732673268e-06, | |
| "loss": 0.0515, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 8.647524752475248e-06, | |
| "loss": 0.0522, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 8.627722772277228e-06, | |
| "loss": 0.0513, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 8.607920792079209e-06, | |
| "loss": 0.0526, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 8.58811881188119e-06, | |
| "loss": 0.0522, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 8.568316831683168e-06, | |
| "loss": 0.0531, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 8.548514851485148e-06, | |
| "loss": 0.0529, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 8.528712871287129e-06, | |
| "loss": 0.052, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_loss": 0.065673828125, | |
| "eval_runtime": 350.8659, | |
| "eval_samples_per_second": 9.32, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 16.40666454420036, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 8.50891089108911e-06, | |
| "loss": 0.0507, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 8.48910891089109e-06, | |
| "loss": 0.0508, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 8.46930693069307e-06, | |
| "loss": 0.0504, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.44950495049505e-06, | |
| "loss": 0.0526, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.42970297029703e-06, | |
| "loss": 0.0498, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 8.40990099009901e-06, | |
| "loss": 0.0507, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 8.39009900990099e-06, | |
| "loss": 0.052, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 8.37029702970297e-06, | |
| "loss": 0.0513, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 8.350495049504951e-06, | |
| "loss": 0.0513, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 8.330693069306932e-06, | |
| "loss": 0.0532, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 0.064453125, | |
| "eval_runtime": 587.0112, | |
| "eval_samples_per_second": 5.571, | |
| "eval_steps_per_second": 0.044, | |
| "eval_wer": 16.22890799108564, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 8.310891089108912e-06, | |
| "loss": 0.0506, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 8.29108910891089e-06, | |
| "loss": 0.0516, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 8.271287128712871e-06, | |
| "loss": 0.0493, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 8.251485148514852e-06, | |
| "loss": 0.0528, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 8.231683168316832e-06, | |
| "loss": 0.0489, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 8.211881188118813e-06, | |
| "loss": 0.0487, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 8.192079207920793e-06, | |
| "loss": 0.0475, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 8.172277227722772e-06, | |
| "loss": 0.0509, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 8.152475247524752e-06, | |
| "loss": 0.0496, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 8.132673267326733e-06, | |
| "loss": 0.0492, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 0.06268310546875, | |
| "eval_runtime": 347.8208, | |
| "eval_samples_per_second": 9.401, | |
| "eval_steps_per_second": 0.075, | |
| "eval_wer": 15.915844210973152, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8.112871287128715e-06, | |
| "loss": 0.0514, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.093069306930693e-06, | |
| "loss": 0.0426, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.073267326732674e-06, | |
| "loss": 0.0417, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.053465346534654e-06, | |
| "loss": 0.0384, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 8.033663366336635e-06, | |
| "loss": 0.0385, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 8.013861386138615e-06, | |
| "loss": 0.0421, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.994059405940596e-06, | |
| "loss": 0.0398, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 7.974257425742576e-06, | |
| "loss": 0.0394, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 7.954455445544555e-06, | |
| "loss": 0.0395, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 7.934653465346535e-06, | |
| "loss": 0.0411, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_loss": 0.06378173828125, | |
| "eval_runtime": 349.9241, | |
| "eval_samples_per_second": 9.345, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 15.804414729916164, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 7.914851485148516e-06, | |
| "loss": 0.0395, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 7.895049504950496e-06, | |
| "loss": 0.042, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 7.875247524752477e-06, | |
| "loss": 0.0389, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 7.855445544554457e-06, | |
| "loss": 0.041, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 7.835643564356436e-06, | |
| "loss": 0.0377, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 7.815841584158416e-06, | |
| "loss": 0.0391, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 7.796039603960397e-06, | |
| "loss": 0.0396, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 7.776237623762377e-06, | |
| "loss": 0.0383, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 7.756435643564357e-06, | |
| "loss": 0.0408, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 7.736633663366338e-06, | |
| "loss": 0.0409, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_loss": 0.0638427734375, | |
| "eval_runtime": 342.8981, | |
| "eval_samples_per_second": 9.536, | |
| "eval_steps_per_second": 0.076, | |
| "eval_wer": 15.671760585800701, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 7.716831683168318e-06, | |
| "loss": 0.0407, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 7.697029702970297e-06, | |
| "loss": 0.0402, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 7.677227722772278e-06, | |
| "loss": 0.0395, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 7.657425742574258e-06, | |
| "loss": 0.0414, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 7.637623762376238e-06, | |
| "loss": 0.0407, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 7.617821782178219e-06, | |
| "loss": 0.0397, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 7.5980198019801985e-06, | |
| "loss": 0.0402, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 7.578217821782179e-06, | |
| "loss": 0.0388, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 7.558415841584159e-06, | |
| "loss": 0.0395, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 7.538613861386139e-06, | |
| "loss": 0.0401, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_loss": 0.061920166015625, | |
| "eval_runtime": 386.8864, | |
| "eval_samples_per_second": 8.452, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 15.536453358802929, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 7.518811881188119e-06, | |
| "loss": 0.0392, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 7.4990099009901e-06, | |
| "loss": 0.0406, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 7.479207920792079e-06, | |
| "loss": 0.0405, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 7.45940594059406e-06, | |
| "loss": 0.0392, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 7.43960396039604e-06, | |
| "loss": 0.0395, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 7.419801980198021e-06, | |
| "loss": 0.0397, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 7.4e-06, | |
| "loss": 0.0393, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 7.380198019801981e-06, | |
| "loss": 0.0402, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 7.360396039603961e-06, | |
| "loss": 0.0405, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 7.340594059405941e-06, | |
| "loss": 0.0386, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 0.061187744140625, | |
| "eval_runtime": 322.9988, | |
| "eval_samples_per_second": 10.124, | |
| "eval_steps_per_second": 0.08, | |
| "eval_wer": 15.549718773214474, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 7.320792079207921e-06, | |
| "loss": 0.0401, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 7.300990099009902e-06, | |
| "loss": 0.0386, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 7.281188118811881e-06, | |
| "loss": 0.0374, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 7.261386138613862e-06, | |
| "loss": 0.0398, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 7.241584158415842e-06, | |
| "loss": 0.0418, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.2217821782178225e-06, | |
| "loss": 0.0393, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 7.201980198019802e-06, | |
| "loss": 0.039, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 7.182178217821783e-06, | |
| "loss": 0.0386, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 7.162376237623763e-06, | |
| "loss": 0.0396, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 7.142574257425743e-06, | |
| "loss": 0.039, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.0611572265625, | |
| "eval_runtime": 340.1349, | |
| "eval_samples_per_second": 9.614, | |
| "eval_steps_per_second": 0.076, | |
| "eval_wer": 15.539106441685238, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 7.122772277227723e-06, | |
| "loss": 0.0393, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.1029702970297035e-06, | |
| "loss": 0.0382, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 7.083168316831683e-06, | |
| "loss": 0.0308, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 7.0633663366336635e-06, | |
| "loss": 0.0307, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.043564356435644e-06, | |
| "loss": 0.0309, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 7.0237623762376235e-06, | |
| "loss": 0.0304, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 7.003960396039604e-06, | |
| "loss": 0.0306, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 6.984158415841584e-06, | |
| "loss": 0.0306, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 6.964356435643565e-06, | |
| "loss": 0.0315, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 6.9445544554455444e-06, | |
| "loss": 0.0312, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_loss": 0.06396484375, | |
| "eval_runtime": 350.8877, | |
| "eval_samples_per_second": 9.319, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 15.523187944391385, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 6.924752475247525e-06, | |
| "loss": 0.0302, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 6.904950495049505e-06, | |
| "loss": 0.0318, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 6.885148514851485e-06, | |
| "loss": 0.0308, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 6.865346534653466e-06, | |
| "loss": 0.0307, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 6.845544554455447e-06, | |
| "loss": 0.0307, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 6.825742574257427e-06, | |
| "loss": 0.03, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 6.805940594059407e-06, | |
| "loss": 0.0334, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 6.786138613861387e-06, | |
| "loss": 0.0307, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 6.7663366336633675e-06, | |
| "loss": 0.0304, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 6.746534653465347e-06, | |
| "loss": 0.0306, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_loss": 0.0655517578125, | |
| "eval_runtime": 328.5852, | |
| "eval_samples_per_second": 9.952, | |
| "eval_steps_per_second": 0.079, | |
| "eval_wer": 15.350737557041283, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 6.7267326732673276e-06, | |
| "loss": 0.0302, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 6.706930693069308e-06, | |
| "loss": 0.0314, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 6.6871287128712884e-06, | |
| "loss": 0.0321, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 6.667326732673268e-06, | |
| "loss": 0.0327, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 6.6475247524752485e-06, | |
| "loss": 0.0306, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 6.627722772277229e-06, | |
| "loss": 0.031, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 6.6079207920792085e-06, | |
| "loss": 0.032, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 6.588118811881189e-06, | |
| "loss": 0.0303, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 6.568316831683169e-06, | |
| "loss": 0.0318, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 6.548514851485149e-06, | |
| "loss": 0.0293, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_loss": 0.06573486328125, | |
| "eval_runtime": 340.4912, | |
| "eval_samples_per_second": 9.604, | |
| "eval_steps_per_second": 0.076, | |
| "eval_wer": 15.456860872333653, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 6.528712871287129e-06, | |
| "loss": 0.0317, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 6.50891089108911e-06, | |
| "loss": 0.0316, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 6.4891089108910894e-06, | |
| "loss": 0.0297, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 6.46930693069307e-06, | |
| "loss": 0.0324, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 6.44950495049505e-06, | |
| "loss": 0.032, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 6.429702970297031e-06, | |
| "loss": 0.0315, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 6.40990099009901e-06, | |
| "loss": 0.032, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 6.390099009900991e-06, | |
| "loss": 0.0329, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 6.370297029702971e-06, | |
| "loss": 0.0308, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 6.350495049504951e-06, | |
| "loss": 0.0318, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_loss": 0.0635986328125, | |
| "eval_runtime": 327.2093, | |
| "eval_samples_per_second": 9.994, | |
| "eval_steps_per_second": 0.079, | |
| "eval_wer": 15.313594396688954, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 6.330693069306931e-06, | |
| "loss": 0.0314, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 6.310891089108912e-06, | |
| "loss": 0.0321, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 6.291089108910891e-06, | |
| "loss": 0.032, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 6.271287128712872e-06, | |
| "loss": 0.0314, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 6.251485148514852e-06, | |
| "loss": 0.0322, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 6.231683168316833e-06, | |
| "loss": 0.0328, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 6.211881188118812e-06, | |
| "loss": 0.0311, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 6.192079207920793e-06, | |
| "loss": 0.0321, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 6.172277227722773e-06, | |
| "loss": 0.0312, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 6.152475247524753e-06, | |
| "loss": 0.0314, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "eval_loss": 0.06402587890625, | |
| "eval_runtime": 327.0528, | |
| "eval_samples_per_second": 9.998, | |
| "eval_steps_per_second": 0.079, | |
| "eval_wer": 15.188899501220417, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 5100, | |
| "num_train_epochs": 11, | |
| "total_flos": 5.2250072905718274e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |