| { | |
| "best_global_step": 10000, | |
| "best_metric": 0.2586507557925852, | |
| "best_model_checkpoint": "/home/cluster-dgxa100/slp01/bagas-fine-tune-whisper/whisper-tiny-javanese-openslr-v3/checkpoint-10000", | |
| "epoch": 1.0807262117036798, | |
| "eval_steps": 500, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01080672178094775, | |
| "grad_norm": 25.346445083618164, | |
| "learning_rate": 1.94e-06, | |
| "loss": 3.5433, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0216134435618955, | |
| "grad_norm": 21.656307220458984, | |
| "learning_rate": 3.94e-06, | |
| "loss": 2.0264, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.03242016534284325, | |
| "grad_norm": 18.657211303710938, | |
| "learning_rate": 5.94e-06, | |
| "loss": 1.5688, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.043226887123791, | |
| "grad_norm": 16.42237663269043, | |
| "learning_rate": 7.94e-06, | |
| "loss": 1.3214, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.054033608904738746, | |
| "grad_norm": 18.631206512451172, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 1.1788, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.054033608904738746, | |
| "eval_loss": 0.967095136642456, | |
| "eval_runtime": 5770.5819, | |
| "eval_samples_per_second": 3.207, | |
| "eval_steps_per_second": 0.802, | |
| "eval_wer": 0.6590292385770924, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.0648403306856865, | |
| "grad_norm": 15.337555885314941, | |
| "learning_rate": 1.1940000000000001e-05, | |
| "loss": 1.0627, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.07564705246663425, | |
| "grad_norm": 14.623177528381348, | |
| "learning_rate": 1.394e-05, | |
| "loss": 0.9632, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.086453774247582, | |
| "grad_norm": 17.126712799072266, | |
| "learning_rate": 1.5940000000000003e-05, | |
| "loss": 0.906, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.09726049602852975, | |
| "grad_norm": 16.75067710876465, | |
| "learning_rate": 1.794e-05, | |
| "loss": 0.8503, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.10806721780947749, | |
| "grad_norm": 14.265076637268066, | |
| "learning_rate": 1.9940000000000002e-05, | |
| "loss": 0.8015, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10806721780947749, | |
| "eval_loss": 0.6976613402366638, | |
| "eval_runtime": 5463.0331, | |
| "eval_samples_per_second": 3.387, | |
| "eval_steps_per_second": 0.847, | |
| "eval_wer": 0.5304858499049883, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11887393959042525, | |
| "grad_norm": 13.737130165100098, | |
| "learning_rate": 1.9784444444444446e-05, | |
| "loss": 0.7589, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.129680661371373, | |
| "grad_norm": 18.01378631591797, | |
| "learning_rate": 1.9562222222222225e-05, | |
| "loss": 0.7589, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.14048738315232073, | |
| "grad_norm": 11.696120262145996, | |
| "learning_rate": 1.934e-05, | |
| "loss": 0.7087, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.1512941049332685, | |
| "grad_norm": 13.419560432434082, | |
| "learning_rate": 1.911777777777778e-05, | |
| "loss": 0.683, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.16210082671421625, | |
| "grad_norm": 12.753211975097656, | |
| "learning_rate": 1.8895555555555557e-05, | |
| "loss": 0.6498, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.16210082671421625, | |
| "eval_loss": 0.5724753737449646, | |
| "eval_runtime": 4564.4621, | |
| "eval_samples_per_second": 4.054, | |
| "eval_steps_per_second": 1.013, | |
| "eval_wer": 0.6670133485560569, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.172907548495164, | |
| "grad_norm": 11.64907455444336, | |
| "learning_rate": 1.8673333333333333e-05, | |
| "loss": 0.6216, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.18371427027611173, | |
| "grad_norm": 13.781865119934082, | |
| "learning_rate": 1.8451111111111113e-05, | |
| "loss": 0.6138, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.1945209920570595, | |
| "grad_norm": 12.58388900756836, | |
| "learning_rate": 1.822888888888889e-05, | |
| "loss": 0.595, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.20532771383800724, | |
| "grad_norm": 14.661055564880371, | |
| "learning_rate": 1.8006666666666668e-05, | |
| "loss": 0.5938, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.21613443561895498, | |
| "grad_norm": 11.948161125183105, | |
| "learning_rate": 1.7784444444444448e-05, | |
| "loss": 0.5828, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21613443561895498, | |
| "eval_loss": 0.5093731880187988, | |
| "eval_runtime": 5328.6402, | |
| "eval_samples_per_second": 3.473, | |
| "eval_steps_per_second": 0.868, | |
| "eval_wer": 0.4828939857208768, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22694115739990273, | |
| "grad_norm": 12.322188377380371, | |
| "learning_rate": 1.7562222222222224e-05, | |
| "loss": 0.5752, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.2377478791808505, | |
| "grad_norm": 17.046159744262695, | |
| "learning_rate": 1.734e-05, | |
| "loss": 0.5663, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.24855460096179824, | |
| "grad_norm": 10.154263496398926, | |
| "learning_rate": 1.711777777777778e-05, | |
| "loss": 0.537, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.259361322742746, | |
| "grad_norm": 11.958285331726074, | |
| "learning_rate": 1.6895555555555556e-05, | |
| "loss": 0.5246, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.27016804452369375, | |
| "grad_norm": 10.264266014099121, | |
| "learning_rate": 1.6673333333333335e-05, | |
| "loss": 0.5226, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27016804452369375, | |
| "eval_loss": 0.46415480971336365, | |
| "eval_runtime": 4645.745, | |
| "eval_samples_per_second": 3.983, | |
| "eval_steps_per_second": 0.996, | |
| "eval_wer": 0.38602898052064843, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.28097476630464147, | |
| "grad_norm": 12.049257278442383, | |
| "learning_rate": 1.6451111111111115e-05, | |
| "loss": 0.493, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.29178148808558924, | |
| "grad_norm": 9.821508407592773, | |
| "learning_rate": 1.622888888888889e-05, | |
| "loss": 0.5153, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.302588209866537, | |
| "grad_norm": 10.481095314025879, | |
| "learning_rate": 1.6006666666666667e-05, | |
| "loss": 0.5, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.3133949316474847, | |
| "grad_norm": 10.193309783935547, | |
| "learning_rate": 1.5784444444444447e-05, | |
| "loss": 0.5248, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.3242016534284325, | |
| "grad_norm": 12.328668594360352, | |
| "learning_rate": 1.5562222222222223e-05, | |
| "loss": 0.4955, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3242016534284325, | |
| "eval_loss": 0.4340818226337433, | |
| "eval_runtime": 4456.9484, | |
| "eval_samples_per_second": 4.152, | |
| "eval_steps_per_second": 1.038, | |
| "eval_wer": 0.39154200455117727, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.33500837520938026, | |
| "grad_norm": 12.583343505859375, | |
| "learning_rate": 1.5340000000000002e-05, | |
| "loss": 0.5082, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.345815096990328, | |
| "grad_norm": 8.40932846069336, | |
| "learning_rate": 1.511777777777778e-05, | |
| "loss": 0.4905, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.35662181877127574, | |
| "grad_norm": 14.150980949401855, | |
| "learning_rate": 1.4895555555555556e-05, | |
| "loss": 0.466, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.36742854055222346, | |
| "grad_norm": 13.014771461486816, | |
| "learning_rate": 1.4673333333333336e-05, | |
| "loss": 0.4788, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.37823526233317123, | |
| "grad_norm": 11.843710899353027, | |
| "learning_rate": 1.4451111111111112e-05, | |
| "loss": 0.4616, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.37823526233317123, | |
| "eval_loss": 0.4127795398235321, | |
| "eval_runtime": 4528.1925, | |
| "eval_samples_per_second": 4.086, | |
| "eval_steps_per_second": 1.022, | |
| "eval_wer": 0.35399088200564593, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.389041984114119, | |
| "grad_norm": 11.520469665527344, | |
| "learning_rate": 1.422888888888889e-05, | |
| "loss": 0.4695, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.3998487058950667, | |
| "grad_norm": 10.21032428741455, | |
| "learning_rate": 1.400666666666667e-05, | |
| "loss": 0.47, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.4106554276760145, | |
| "grad_norm": 9.393896102905273, | |
| "learning_rate": 1.3784444444444445e-05, | |
| "loss": 0.4656, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.42146214945696225, | |
| "grad_norm": 10.503016471862793, | |
| "learning_rate": 1.3562222222222223e-05, | |
| "loss": 0.4446, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.43226887123790997, | |
| "grad_norm": 10.747596740722656, | |
| "learning_rate": 1.3340000000000001e-05, | |
| "loss": 0.4474, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.43226887123790997, | |
| "eval_loss": 0.3900074064731598, | |
| "eval_runtime": 4858.8536, | |
| "eval_samples_per_second": 3.808, | |
| "eval_steps_per_second": 0.952, | |
| "eval_wer": 0.36136504038974343, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.44307559301885774, | |
| "grad_norm": 13.275285720825195, | |
| "learning_rate": 1.3117777777777779e-05, | |
| "loss": 0.4488, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.45388231479980545, | |
| "grad_norm": 11.318832397460938, | |
| "learning_rate": 1.2897777777777778e-05, | |
| "loss": 0.4292, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.4646890365807532, | |
| "grad_norm": 10.3064546585083, | |
| "learning_rate": 1.2675555555555557e-05, | |
| "loss": 0.4302, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.475495758361701, | |
| "grad_norm": 11.634562492370605, | |
| "learning_rate": 1.2453333333333335e-05, | |
| "loss": 0.426, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.4863024801426487, | |
| "grad_norm": 10.647918701171875, | |
| "learning_rate": 1.2231111111111111e-05, | |
| "loss": 0.4387, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.4863024801426487, | |
| "eval_loss": 0.37359631061553955, | |
| "eval_runtime": 4990.4878, | |
| "eval_samples_per_second": 3.708, | |
| "eval_steps_per_second": 0.927, | |
| "eval_wer": 0.35633684967821144, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.4971092019235965, | |
| "grad_norm": 9.396610260009766, | |
| "learning_rate": 1.200888888888889e-05, | |
| "loss": 0.4195, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.5079159237045442, | |
| "grad_norm": 10.845105171203613, | |
| "learning_rate": 1.1786666666666668e-05, | |
| "loss": 0.4056, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.518722645485492, | |
| "grad_norm": 9.404190063476562, | |
| "learning_rate": 1.1564444444444445e-05, | |
| "loss": 0.4306, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.5295293672664397, | |
| "grad_norm": 9.176289558410645, | |
| "learning_rate": 1.1342222222222224e-05, | |
| "loss": 0.4239, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.5403360890473875, | |
| "grad_norm": 10.088706016540527, | |
| "learning_rate": 1.1120000000000002e-05, | |
| "loss": 0.4154, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5403360890473875, | |
| "eval_loss": 0.36057594418525696, | |
| "eval_runtime": 5945.658, | |
| "eval_samples_per_second": 3.112, | |
| "eval_steps_per_second": 0.778, | |
| "eval_wer": 0.32743452795220485, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5511428108283353, | |
| "grad_norm": 9.688194274902344, | |
| "learning_rate": 1.0897777777777778e-05, | |
| "loss": 0.4115, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.5619495326092829, | |
| "grad_norm": 9.752260208129883, | |
| "learning_rate": 1.0675555555555558e-05, | |
| "loss": 0.3854, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.5727562543902307, | |
| "grad_norm": 10.447392463684082, | |
| "learning_rate": 1.0453333333333334e-05, | |
| "loss": 0.4141, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.5835629761711785, | |
| "grad_norm": 11.185776710510254, | |
| "learning_rate": 1.0231111111111112e-05, | |
| "loss": 0.3924, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.5943696979521262, | |
| "grad_norm": 10.3914794921875, | |
| "learning_rate": 1.000888888888889e-05, | |
| "loss": 0.419, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5943696979521262, | |
| "eval_loss": 0.3494803309440613, | |
| "eval_runtime": 6902.9208, | |
| "eval_samples_per_second": 2.681, | |
| "eval_steps_per_second": 0.67, | |
| "eval_wer": 0.314375307908257, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.605176419733074, | |
| "grad_norm": 11.420536041259766, | |
| "learning_rate": 9.786666666666667e-06, | |
| "loss": 0.4096, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.6159831415140217, | |
| "grad_norm": 9.05328369140625, | |
| "learning_rate": 9.564444444444445e-06, | |
| "loss": 0.3917, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.6267898632949694, | |
| "grad_norm": 10.281911849975586, | |
| "learning_rate": 9.342222222222223e-06, | |
| "loss": 0.3965, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6375965850759172, | |
| "grad_norm": 10.587265014648438, | |
| "learning_rate": 9.12e-06, | |
| "loss": 0.374, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.648403306856865, | |
| "grad_norm": 7.721372127532959, | |
| "learning_rate": 8.897777777777779e-06, | |
| "loss": 0.3799, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.648403306856865, | |
| "eval_loss": 0.3397567868232727, | |
| "eval_runtime": 7002.8513, | |
| "eval_samples_per_second": 2.642, | |
| "eval_steps_per_second": 0.661, | |
| "eval_wer": 0.2921668139413039, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6592100286378128, | |
| "grad_norm": 6.785597324371338, | |
| "learning_rate": 8.675555555555556e-06, | |
| "loss": 0.3953, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.6700167504187605, | |
| "grad_norm": 9.53781509399414, | |
| "learning_rate": 8.453333333333334e-06, | |
| "loss": 0.3786, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.6808234721997082, | |
| "grad_norm": 8.857239723205566, | |
| "learning_rate": 8.231111111111112e-06, | |
| "loss": 0.3744, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.691630193980656, | |
| "grad_norm": 9.638261795043945, | |
| "learning_rate": 8.00888888888889e-06, | |
| "loss": 0.3809, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.7024369157616037, | |
| "grad_norm": 8.304004669189453, | |
| "learning_rate": 7.786666666666666e-06, | |
| "loss": 0.3802, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7024369157616037, | |
| "eval_loss": 0.3289755880832672, | |
| "eval_runtime": 5885.8991, | |
| "eval_samples_per_second": 3.144, | |
| "eval_steps_per_second": 0.786, | |
| "eval_wer": 0.3044049452998538, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7132436375425515, | |
| "grad_norm": 9.978581428527832, | |
| "learning_rate": 7.564444444444446e-06, | |
| "loss": 0.3537, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.7240503593234993, | |
| "grad_norm": 10.849929809570312, | |
| "learning_rate": 7.342222222222223e-06, | |
| "loss": 0.3762, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.7348570811044469, | |
| "grad_norm": 11.856138229370117, | |
| "learning_rate": 7.1200000000000004e-06, | |
| "loss": 0.3477, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.7456638028853947, | |
| "grad_norm": 10.761491775512695, | |
| "learning_rate": 6.897777777777779e-06, | |
| "loss": 0.361, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.7564705246663425, | |
| "grad_norm": 9.24421501159668, | |
| "learning_rate": 6.675555555555556e-06, | |
| "loss": 0.3611, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7564705246663425, | |
| "eval_loss": 0.3224972188472748, | |
| "eval_runtime": 5632.6127, | |
| "eval_samples_per_second": 3.285, | |
| "eval_steps_per_second": 0.821, | |
| "eval_wer": 0.2823372093932546, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7672772464472902, | |
| "grad_norm": 10.52470874786377, | |
| "learning_rate": 6.453333333333334e-06, | |
| "loss": 0.3638, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.778083968228238, | |
| "grad_norm": 9.080463409423828, | |
| "learning_rate": 6.231111111111111e-06, | |
| "loss": 0.3532, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.7888906900091858, | |
| "grad_norm": 8.789374351501465, | |
| "learning_rate": 6.00888888888889e-06, | |
| "loss": 0.3592, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.7996974117901334, | |
| "grad_norm": 8.97732162475586, | |
| "learning_rate": 5.7866666666666674e-06, | |
| "loss": 0.3611, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.8105041335710812, | |
| "grad_norm": 10.455592155456543, | |
| "learning_rate": 5.5644444444444444e-06, | |
| "loss": 0.3548, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.8105041335710812, | |
| "eval_loss": 0.31678903102874756, | |
| "eval_runtime": 3060.9871, | |
| "eval_samples_per_second": 6.045, | |
| "eval_steps_per_second": 1.511, | |
| "eval_wer": 0.27332869353060313, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.821310855352029, | |
| "grad_norm": 8.56920051574707, | |
| "learning_rate": 5.342222222222223e-06, | |
| "loss": 0.3628, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.8321175771329767, | |
| "grad_norm": 11.37761402130127, | |
| "learning_rate": 5.12e-06, | |
| "loss": 0.3353, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.8429242989139245, | |
| "grad_norm": 9.396086692810059, | |
| "learning_rate": 4.897777777777778e-06, | |
| "loss": 0.3704, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.8537310206948722, | |
| "grad_norm": 10.0977144241333, | |
| "learning_rate": 4.677777777777778e-06, | |
| "loss": 0.364, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.8645377424758199, | |
| "grad_norm": 8.653088569641113, | |
| "learning_rate": 4.455555555555555e-06, | |
| "loss": 0.346, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8645377424758199, | |
| "eval_loss": 0.3104597330093384, | |
| "eval_runtime": 3053.8514, | |
| "eval_samples_per_second": 6.059, | |
| "eval_steps_per_second": 1.515, | |
| "eval_wer": 0.26601709428444076, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8753444642567677, | |
| "grad_norm": 9.058122634887695, | |
| "learning_rate": 4.233333333333334e-06, | |
| "loss": 0.3382, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.8861511860377155, | |
| "grad_norm": 12.135452270507812, | |
| "learning_rate": 4.011111111111111e-06, | |
| "loss": 0.3456, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.8969579078186632, | |
| "grad_norm": 6.601293563842773, | |
| "learning_rate": 3.7888888888888893e-06, | |
| "loss": 0.3404, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.9077646295996109, | |
| "grad_norm": 9.51930046081543, | |
| "learning_rate": 3.566666666666667e-06, | |
| "loss": 0.3479, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.9185713513805587, | |
| "grad_norm": 7.031350135803223, | |
| "learning_rate": 3.3444444444444445e-06, | |
| "loss": 0.3547, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9185713513805587, | |
| "eval_loss": 0.3063461184501648, | |
| "eval_runtime": 3070.7291, | |
| "eval_samples_per_second": 6.026, | |
| "eval_steps_per_second": 1.506, | |
| "eval_wer": 0.27081068822871623, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9293780731615064, | |
| "grad_norm": 11.10822868347168, | |
| "learning_rate": 3.1222222222222228e-06, | |
| "loss": 0.3454, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.9401847949424542, | |
| "grad_norm": 9.607211112976074, | |
| "learning_rate": 2.9e-06, | |
| "loss": 0.3319, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.950991516723402, | |
| "grad_norm": 10.614663124084473, | |
| "learning_rate": 2.677777777777778e-06, | |
| "loss": 0.3441, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.9617982385043498, | |
| "grad_norm": 8.344138145446777, | |
| "learning_rate": 2.455555555555556e-06, | |
| "loss": 0.3466, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 0.9726049602852974, | |
| "grad_norm": 11.955930709838867, | |
| "learning_rate": 2.2333333333333333e-06, | |
| "loss": 0.3211, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9726049602852974, | |
| "eval_loss": 0.30189329385757446, | |
| "eval_runtime": 3095.8164, | |
| "eval_samples_per_second": 5.977, | |
| "eval_steps_per_second": 1.494, | |
| "eval_wer": 0.28268910454413937, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9834116820662452, | |
| "grad_norm": 9.438616752624512, | |
| "learning_rate": 2.011111111111111e-06, | |
| "loss": 0.343, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 0.994218403847193, | |
| "grad_norm": 10.029309272766113, | |
| "learning_rate": 1.788888888888889e-06, | |
| "loss": 0.3582, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.0050791592370454, | |
| "grad_norm": 9.47360610961914, | |
| "learning_rate": 1.566666666666667e-06, | |
| "loss": 0.3024, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 1.0158858810179932, | |
| "grad_norm": 9.3403959274292, | |
| "learning_rate": 1.3444444444444446e-06, | |
| "loss": 0.2811, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.026692602798941, | |
| "grad_norm": 9.723664283752441, | |
| "learning_rate": 1.1222222222222222e-06, | |
| "loss": 0.2718, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.026692602798941, | |
| "eval_loss": 0.2989746034145355, | |
| "eval_runtime": 3189.5179, | |
| "eval_samples_per_second": 5.802, | |
| "eval_steps_per_second": 1.45, | |
| "eval_wer": 0.2659936346077151, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0374993245798887, | |
| "grad_norm": 7.739469051361084, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 0.2765, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.0483060463608365, | |
| "grad_norm": 8.379693984985352, | |
| "learning_rate": 6.777777777777779e-07, | |
| "loss": 0.2872, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 1.0591127681417842, | |
| "grad_norm": 8.849838256835938, | |
| "learning_rate": 4.5555555555555563e-07, | |
| "loss": 0.2782, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.069919489922732, | |
| "grad_norm": 8.006597518920898, | |
| "learning_rate": 2.3333333333333336e-07, | |
| "loss": 0.2673, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 1.0807262117036798, | |
| "grad_norm": 10.859480857849121, | |
| "learning_rate": 1.1111111111111112e-08, | |
| "loss": 0.2859, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0807262117036798, | |
| "eval_loss": 0.2979792058467865, | |
| "eval_runtime": 3174.7879, | |
| "eval_samples_per_second": 5.828, | |
| "eval_steps_per_second": 1.457, | |
| "eval_wer": 0.2586507557925852, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0807262117036798, | |
| "step": 10000, | |
| "total_flos": 3.93912009474048e+18, | |
| "train_loss": 0.09938106536865235, | |
| "train_runtime": 25030.863, | |
| "train_samples_per_second": 6.392, | |
| "train_steps_per_second": 0.4 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.93912009474048e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |