| { | |
| "best_metric": 0.35703112038112395, | |
| "best_model_checkpoint": "results/punjabi_model/20240829_200444/checkpoint-3000", | |
| "epoch": 9.999656616990592, | |
| "eval_steps": 3000, | |
| "global_step": 109200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.045784401254492596, | |
| "grad_norm": 2.4544355869293213, | |
| "learning_rate": 9.980000000000001e-06, | |
| "loss": 3.9787, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09156880250898519, | |
| "grad_norm": 5.300338268280029, | |
| "learning_rate": 1.9980000000000002e-05, | |
| "loss": 2.7973, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1373532037634778, | |
| "grad_norm": 5.187418460845947, | |
| "learning_rate": 2.9980000000000004e-05, | |
| "loss": 2.1081, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18313760501797038, | |
| "grad_norm": 5.451952934265137, | |
| "learning_rate": 3.998000000000001e-05, | |
| "loss": 1.7718, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22892200627246298, | |
| "grad_norm": 10.363802909851074, | |
| "learning_rate": 3.962053231939164e-05, | |
| "loss": 1.6058, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2747064075269556, | |
| "grad_norm": 3.4008052349090576, | |
| "learning_rate": 3.924030418250951e-05, | |
| "loss": 1.5005, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.2747064075269556, | |
| "eval_loss": 0.580813467502594, | |
| "eval_runtime": 645.4536, | |
| "eval_samples_per_second": 10.151, | |
| "eval_steps_per_second": 10.151, | |
| "eval_wer": 0.35703112038112395, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.32049080878144814, | |
| "grad_norm": 2.9593212604522705, | |
| "learning_rate": 3.886007604562738e-05, | |
| "loss": 1.4034, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.36627521003594077, | |
| "grad_norm": 4.573808193206787, | |
| "learning_rate": 3.847984790874525e-05, | |
| "loss": 1.3577, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.41205961129043334, | |
| "grad_norm": 2.684593915939331, | |
| "learning_rate": 3.809961977186312e-05, | |
| "loss": 1.3114, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.45784401254492596, | |
| "grad_norm": 2.6589484214782715, | |
| "learning_rate": 3.771939163498099e-05, | |
| "loss": 1.3163, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5036284137994186, | |
| "grad_norm": 3.144775390625, | |
| "learning_rate": 3.733992395437262e-05, | |
| "loss": 1.2375, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5494128150539112, | |
| "grad_norm": 3.1532704830169678, | |
| "learning_rate": 3.6959695817490496e-05, | |
| "loss": 1.2442, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5494128150539112, | |
| "eval_loss": 0.43034785985946655, | |
| "eval_runtime": 665.7368, | |
| "eval_samples_per_second": 9.842, | |
| "eval_steps_per_second": 9.842, | |
| "eval_wer": 0.2554395487189211, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5951972163084037, | |
| "grad_norm": 9.455696105957031, | |
| "learning_rate": 3.657946768060837e-05, | |
| "loss": 1.225, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.6409816175628963, | |
| "grad_norm": 3.3579087257385254, | |
| "learning_rate": 3.619923954372624e-05, | |
| "loss": 1.1457, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.686766018817389, | |
| "grad_norm": 4.421695709228516, | |
| "learning_rate": 3.581901140684411e-05, | |
| "loss": 1.1656, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7325504200718815, | |
| "grad_norm": 3.5701684951782227, | |
| "learning_rate": 3.5439543726235745e-05, | |
| "loss": 1.175, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.7783348213263741, | |
| "grad_norm": 4.08390474319458, | |
| "learning_rate": 3.505931558935361e-05, | |
| "loss": 1.131, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8241192225808667, | |
| "grad_norm": 3.4614651203155518, | |
| "learning_rate": 3.4679847908745255e-05, | |
| "loss": 1.1133, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8241192225808667, | |
| "eval_loss": 0.3901245594024658, | |
| "eval_runtime": 664.9166, | |
| "eval_samples_per_second": 9.854, | |
| "eval_steps_per_second": 9.854, | |
| "eval_wer": 0.2110935981607452, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8699036238353592, | |
| "grad_norm": 2.597609519958496, | |
| "learning_rate": 3.429961977186312e-05, | |
| "loss": 1.0977, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.9156880250898519, | |
| "grad_norm": 3.4706475734710693, | |
| "learning_rate": 3.3919391634980995e-05, | |
| "loss": 1.0986, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9614724263443445, | |
| "grad_norm": 2.1491544246673584, | |
| "learning_rate": 3.353916349809886e-05, | |
| "loss": 1.1154, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0072568275988372, | |
| "grad_norm": 1.495735764503479, | |
| "learning_rate": 3.3158935361216734e-05, | |
| "loss": 1.1157, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.0530412288533297, | |
| "grad_norm": 2.120736837387085, | |
| "learning_rate": 3.277870722433461e-05, | |
| "loss": 1.0495, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.0988256301078223, | |
| "grad_norm": 2.2971479892730713, | |
| "learning_rate": 3.239847908745247e-05, | |
| "loss": 1.0785, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0988256301078223, | |
| "eval_loss": 0.34569498896598816, | |
| "eval_runtime": 655.7186, | |
| "eval_samples_per_second": 9.992, | |
| "eval_steps_per_second": 9.992, | |
| "eval_wer": 0.19471569766063854, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.1446100313623149, | |
| "grad_norm": 1.080255150794983, | |
| "learning_rate": 3.2018250950570346e-05, | |
| "loss": 1.0311, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.1903944326168074, | |
| "grad_norm": 2.3899152278900146, | |
| "learning_rate": 3.163802281368821e-05, | |
| "loss": 1.107, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.2361788338713, | |
| "grad_norm": 2.9057199954986572, | |
| "learning_rate": 3.125855513307985e-05, | |
| "loss": 1.0711, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.2819632351257926, | |
| "grad_norm": 1.879520058631897, | |
| "learning_rate": 3.087832699619772e-05, | |
| "loss": 1.0348, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.3277476363802854, | |
| "grad_norm": 2.777270555496216, | |
| "learning_rate": 3.0498098859315592e-05, | |
| "loss": 1.0308, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.373532037634778, | |
| "grad_norm": 3.571882724761963, | |
| "learning_rate": 3.0117870722433462e-05, | |
| "loss": 1.0245, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.373532037634778, | |
| "eval_loss": 0.3307068943977356, | |
| "eval_runtime": 655.2961, | |
| "eval_samples_per_second": 9.999, | |
| "eval_steps_per_second": 9.999, | |
| "eval_wer": 0.1764061055675381, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.4193164388892705, | |
| "grad_norm": 2.154317855834961, | |
| "learning_rate": 2.973764258555133e-05, | |
| "loss": 1.0197, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.465100840143763, | |
| "grad_norm": 2.7176313400268555, | |
| "learning_rate": 2.9357414448669205e-05, | |
| "loss": 1.0042, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.5108852413982556, | |
| "grad_norm": 3.9277138710021973, | |
| "learning_rate": 2.8977946768060842e-05, | |
| "loss": 0.9801, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.5566696426527482, | |
| "grad_norm": 6.810076713562012, | |
| "learning_rate": 2.859771863117871e-05, | |
| "loss": 0.9256, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.6024540439072408, | |
| "grad_norm": 1.7569458484649658, | |
| "learning_rate": 2.821749049429658e-05, | |
| "loss": 0.981, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.6482384451617333, | |
| "grad_norm": 3.3207457065582275, | |
| "learning_rate": 2.783726235741445e-05, | |
| "loss": 0.9452, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.6482384451617333, | |
| "eval_loss": 0.3014201521873474, | |
| "eval_runtime": 651.2443, | |
| "eval_samples_per_second": 10.061, | |
| "eval_steps_per_second": 10.061, | |
| "eval_wer": 0.1733545068853547, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.694022846416226, | |
| "grad_norm": 1.4468517303466797, | |
| "learning_rate": 2.7457034220532324e-05, | |
| "loss": 0.9969, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.7398072476707185, | |
| "grad_norm": 1.2308686971664429, | |
| "learning_rate": 2.7077566539923958e-05, | |
| "loss": 0.9551, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.785591648925211, | |
| "grad_norm": 3.698532819747925, | |
| "learning_rate": 2.6697338403041827e-05, | |
| "loss": 0.9585, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.8313760501797036, | |
| "grad_norm": 1.4992170333862305, | |
| "learning_rate": 2.6317110266159697e-05, | |
| "loss": 0.951, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.8771604514341964, | |
| "grad_norm": 3.0142011642456055, | |
| "learning_rate": 2.5936882129277566e-05, | |
| "loss": 0.9677, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.922944852688689, | |
| "grad_norm": 1.9342625141143799, | |
| "learning_rate": 2.5556653992395443e-05, | |
| "loss": 0.9656, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.922944852688689, | |
| "eval_loss": 0.3123963475227356, | |
| "eval_runtime": 653.0667, | |
| "eval_samples_per_second": 10.033, | |
| "eval_steps_per_second": 10.033, | |
| "eval_wer": 0.15938233272498875, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.9687292539431815, | |
| "grad_norm": 5.628562927246094, | |
| "learning_rate": 2.5177186311787077e-05, | |
| "loss": 0.9511, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.0145136551976743, | |
| "grad_norm": 2.930298089981079, | |
| "learning_rate": 2.4796958174904946e-05, | |
| "loss": 0.918, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.060298056452167, | |
| "grad_norm": 1.2405227422714233, | |
| "learning_rate": 2.4416730038022816e-05, | |
| "loss": 0.9372, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.1060824577066595, | |
| "grad_norm": 2.1558220386505127, | |
| "learning_rate": 2.4036501901140686e-05, | |
| "loss": 0.9673, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.151866858961152, | |
| "grad_norm": 1.4096624851226807, | |
| "learning_rate": 2.365703422053232e-05, | |
| "loss": 0.9826, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.1976512602156446, | |
| "grad_norm": 0.9921212196350098, | |
| "learning_rate": 2.3276806083650192e-05, | |
| "loss": 0.9261, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.1976512602156446, | |
| "eval_loss": 0.30950331687927246, | |
| "eval_runtime": 651.0611, | |
| "eval_samples_per_second": 10.064, | |
| "eval_steps_per_second": 10.064, | |
| "eval_wer": 0.15439904244981156, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.243435661470137, | |
| "grad_norm": 3.321803331375122, | |
| "learning_rate": 2.2896577946768065e-05, | |
| "loss": 0.8988, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.2892200627246297, | |
| "grad_norm": 1.4012964963912964, | |
| "learning_rate": 2.2516349809885935e-05, | |
| "loss": 0.8717, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.3350044639791223, | |
| "grad_norm": 1.9718306064605713, | |
| "learning_rate": 2.2136121673003805e-05, | |
| "loss": 0.9545, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.380788865233615, | |
| "grad_norm": 2.5009524822235107, | |
| "learning_rate": 2.1755893536121674e-05, | |
| "loss": 0.9311, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.4265732664881074, | |
| "grad_norm": 1.2410893440246582, | |
| "learning_rate": 2.137642585551331e-05, | |
| "loss": 0.8701, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.4723576677426, | |
| "grad_norm": 1.5738617181777954, | |
| "learning_rate": 2.099619771863118e-05, | |
| "loss": 0.8938, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.4723576677426, | |
| "eval_loss": 0.26886311173439026, | |
| "eval_runtime": 654.1765, | |
| "eval_samples_per_second": 10.016, | |
| "eval_steps_per_second": 10.016, | |
| "eval_wer": 0.15074304946552583, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.5181420689970926, | |
| "grad_norm": 2.2281384468078613, | |
| "learning_rate": 2.061596958174905e-05, | |
| "loss": 0.8609, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.563926470251585, | |
| "grad_norm": 2.347487449645996, | |
| "learning_rate": 2.023574144486692e-05, | |
| "loss": 0.9237, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.6097108715060777, | |
| "grad_norm": 1.710715413093567, | |
| "learning_rate": 1.9855513307984794e-05, | |
| "loss": 0.8778, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.6554952727605707, | |
| "grad_norm": 9.36032772064209, | |
| "learning_rate": 1.947680608365019e-05, | |
| "loss": 0.9092, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.701279674015063, | |
| "grad_norm": 1.8656507730484009, | |
| "learning_rate": 1.909657794676806e-05, | |
| "loss": 0.8829, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.747064075269556, | |
| "grad_norm": 0.8339200019836426, | |
| "learning_rate": 1.8716349809885934e-05, | |
| "loss": 0.8979, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.747064075269556, | |
| "eval_loss": 0.2853592336177826, | |
| "eval_runtime": 652.9254, | |
| "eval_samples_per_second": 10.035, | |
| "eval_steps_per_second": 10.035, | |
| "eval_wer": 0.14718778886492379, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.792848476524048, | |
| "grad_norm": 1.8189595937728882, | |
| "learning_rate": 1.8336121673003804e-05, | |
| "loss": 0.9216, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.838632877778541, | |
| "grad_norm": 1.2487553358078003, | |
| "learning_rate": 1.7955893536121673e-05, | |
| "loss": 0.8714, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.8844172790330336, | |
| "grad_norm": 2.8581480979919434, | |
| "learning_rate": 1.757642585551331e-05, | |
| "loss": 0.8675, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.930201680287526, | |
| "grad_norm": 1.5087124109268188, | |
| "learning_rate": 1.719619771863118e-05, | |
| "loss": 0.9003, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.9759860815420187, | |
| "grad_norm": 2.1860768795013428, | |
| "learning_rate": 1.6816730038022814e-05, | |
| "loss": 0.906, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.0217704827965113, | |
| "grad_norm": 2.0595364570617676, | |
| "learning_rate": 1.6436501901140687e-05, | |
| "loss": 0.8624, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.0217704827965113, | |
| "eval_loss": 0.282227486371994, | |
| "eval_runtime": 652.2893, | |
| "eval_samples_per_second": 10.045, | |
| "eval_steps_per_second": 10.045, | |
| "eval_wer": 0.14017800004740347, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.067554884051004, | |
| "grad_norm": 2.31689453125, | |
| "learning_rate": 1.6056273764258557e-05, | |
| "loss": 0.8498, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.1133392853054964, | |
| "grad_norm": 1.297116756439209, | |
| "learning_rate": 1.5676045627376426e-05, | |
| "loss": 0.8301, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.159123686559989, | |
| "grad_norm": 1.9447718858718872, | |
| "learning_rate": 1.52958174904943e-05, | |
| "loss": 0.8445, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.2049080878144816, | |
| "grad_norm": 1.8142869472503662, | |
| "learning_rate": 1.4915589353612167e-05, | |
| "loss": 0.8988, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.250692489068974, | |
| "grad_norm": 26.81559944152832, | |
| "learning_rate": 1.4535361216730039e-05, | |
| "loss": 0.8358, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.2964768903234667, | |
| "grad_norm": 1.4768437147140503, | |
| "learning_rate": 1.4155133079847908e-05, | |
| "loss": 0.8269, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.2964768903234667, | |
| "eval_loss": 0.26364651322364807, | |
| "eval_runtime": 653.8923, | |
| "eval_samples_per_second": 10.02, | |
| "eval_steps_per_second": 10.02, | |
| "eval_wer": 0.13765376502097604, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.3422612915779593, | |
| "grad_norm": 3.271960496902466, | |
| "learning_rate": 1.3774904942965781e-05, | |
| "loss": 0.8389, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.388045692832452, | |
| "grad_norm": 1.8337602615356445, | |
| "learning_rate": 1.3395437262357415e-05, | |
| "loss": 0.8295, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.4338300940869444, | |
| "grad_norm": 2.3357372283935547, | |
| "learning_rate": 1.301596958174905e-05, | |
| "loss": 0.8646, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.479614495341437, | |
| "grad_norm": 3.577268600463867, | |
| "learning_rate": 1.2636501901140685e-05, | |
| "loss": 0.844, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.52539889659593, | |
| "grad_norm": 3.0567266941070557, | |
| "learning_rate": 1.2256273764258558e-05, | |
| "loss": 0.8212, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.571183297850422, | |
| "grad_norm": 5.759117126464844, | |
| "learning_rate": 1.1876045627376427e-05, | |
| "loss": 0.85, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.571183297850422, | |
| "eval_loss": 0.24618284404277802, | |
| "eval_runtime": 648.8718, | |
| "eval_samples_per_second": 10.098, | |
| "eval_steps_per_second": 10.098, | |
| "eval_wer": 0.13248678628143443, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.616967699104915, | |
| "grad_norm": 1.7661995887756348, | |
| "learning_rate": 1.1495817490494297e-05, | |
| "loss": 0.8213, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.6627521003594077, | |
| "grad_norm": 2.657759428024292, | |
| "learning_rate": 1.1115589353612168e-05, | |
| "loss": 0.8379, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.7085365016139002, | |
| "grad_norm": 2.6153953075408936, | |
| "learning_rate": 1.0735361216730038e-05, | |
| "loss": 0.807, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.754320902868393, | |
| "grad_norm": 4.843667507171631, | |
| "learning_rate": 1.035513307984791e-05, | |
| "loss": 0.8265, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.8001053041228854, | |
| "grad_norm": 3.4964911937713623, | |
| "learning_rate": 9.97490494296578e-06, | |
| "loss": 0.85, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.845889705377378, | |
| "grad_norm": 2.8312175273895264, | |
| "learning_rate": 9.594676806083652e-06, | |
| "loss": 0.8557, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.845889705377378, | |
| "eval_loss": 0.24823951721191406, | |
| "eval_runtime": 641.5128, | |
| "eval_samples_per_second": 10.213, | |
| "eval_steps_per_second": 10.213, | |
| "eval_wer": 0.12771088620795903, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.8916741066318705, | |
| "grad_norm": 2.185664653778076, | |
| "learning_rate": 9.214448669201521e-06, | |
| "loss": 0.8157, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 3.937458507886363, | |
| "grad_norm": 3.490175724029541, | |
| "learning_rate": 8.834220532319391e-06, | |
| "loss": 0.8202, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.9832429091408557, | |
| "grad_norm": 1.9320560693740845, | |
| "learning_rate": 8.453992395437262e-06, | |
| "loss": 0.8344, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.029027310395349, | |
| "grad_norm": 1.6495819091796875, | |
| "learning_rate": 8.073764258555134e-06, | |
| "loss": 0.8167, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.074811711649841, | |
| "grad_norm": 5.037049770355225, | |
| "learning_rate": 7.694296577946768e-06, | |
| "loss": 0.8033, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 4.120596112904334, | |
| "grad_norm": 2.1924211978912354, | |
| "learning_rate": 7.314828897338404e-06, | |
| "loss": 0.8177, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.120596112904334, | |
| "eval_loss": 0.2399299442768097, | |
| "eval_runtime": 647.8262, | |
| "eval_samples_per_second": 10.114, | |
| "eval_steps_per_second": 10.114, | |
| "eval_wer": 0.12562513332227251, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.166380514158826, | |
| "grad_norm": 0.6167405843734741, | |
| "learning_rate": 6.9346007604562745e-06, | |
| "loss": 0.8205, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.212164915413319, | |
| "grad_norm": 1.5424410104751587, | |
| "learning_rate": 6.554372623574144e-06, | |
| "loss": 0.8096, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.257949316667811, | |
| "grad_norm": 3.1014537811279297, | |
| "learning_rate": 6.1741444866920154e-06, | |
| "loss": 0.7904, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.303733717922304, | |
| "grad_norm": 1.1060818433761597, | |
| "learning_rate": 5.794676806083651e-06, | |
| "loss": 0.7818, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.349518119176796, | |
| "grad_norm": 1.345553994178772, | |
| "learning_rate": 5.414448669201522e-06, | |
| "loss": 0.8144, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 4.395302520431289, | |
| "grad_norm": 2.172055721282959, | |
| "learning_rate": 5.034220532319392e-06, | |
| "loss": 0.7933, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 4.395302520431289, | |
| "eval_loss": 0.2388318032026291, | |
| "eval_runtime": 640.8224, | |
| "eval_samples_per_second": 10.224, | |
| "eval_steps_per_second": 10.224, | |
| "eval_wer": 0.12215282880235122, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 4.441086921685781, | |
| "grad_norm": 3.54559063911438, | |
| "learning_rate": 4.653992395437263e-06, | |
| "loss": 0.814, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 4.486871322940274, | |
| "grad_norm": 3.8439674377441406, | |
| "learning_rate": 4.273764258555134e-06, | |
| "loss": 0.8641, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 4.5326557241947665, | |
| "grad_norm": 1.967161774635315, | |
| "learning_rate": 3.893536121673004e-06, | |
| "loss": 0.7962, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 4.5784401254492595, | |
| "grad_norm": 1.7992150783538818, | |
| "learning_rate": 3.5133079847908747e-06, | |
| "loss": 0.7871, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 4.6242245267037525, | |
| "grad_norm": 1.4352868795394897, | |
| "learning_rate": 3.1330798479087456e-06, | |
| "loss": 0.839, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 4.670008927958245, | |
| "grad_norm": 1.4436633586883545, | |
| "learning_rate": 2.7536121673003807e-06, | |
| "loss": 0.7864, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.670008927958245, | |
| "eval_loss": 0.237684428691864, | |
| "eval_runtime": 642.8597, | |
| "eval_samples_per_second": 10.192, | |
| "eval_steps_per_second": 10.192, | |
| "eval_wer": 0.12045815458273092, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.715793329212737, | |
| "grad_norm": 6.0789055824279785, | |
| "learning_rate": 2.373384030418251e-06, | |
| "loss": 0.7787, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 4.76157773046723, | |
| "grad_norm": 3.2274794578552246, | |
| "learning_rate": 1.993916349809886e-06, | |
| "loss": 0.8234, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.807362131721723, | |
| "grad_norm": 2.8651950359344482, | |
| "learning_rate": 1.6136882129277568e-06, | |
| "loss": 0.8, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 4.853146532976215, | |
| "grad_norm": 1.4273629188537598, | |
| "learning_rate": 1.2334600760456275e-06, | |
| "loss": 0.7861, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 4.898930934230708, | |
| "grad_norm": 1.4547092914581299, | |
| "learning_rate": 8.532319391634982e-07, | |
| "loss": 0.7801, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 4.9447153354852, | |
| "grad_norm": 2.2551023960113525, | |
| "learning_rate": 4.737642585551331e-07, | |
| "loss": 0.8371, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.9447153354852, | |
| "eval_loss": 0.23816268146038055, | |
| "eval_runtime": 641.5849, | |
| "eval_samples_per_second": 10.212, | |
| "eval_steps_per_second": 10.212, | |
| "eval_wer": 0.11996634353298097, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.990499736739693, | |
| "grad_norm": 1.8520026206970215, | |
| "learning_rate": 9.353612167300382e-08, | |
| "loss": 0.7373, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 5.036627521003594, | |
| "grad_norm": 5.179400444030762, | |
| "learning_rate": 2.0232462686567166e-05, | |
| "loss": 0.836, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 5.082411922258086, | |
| "grad_norm": 7.242797374725342, | |
| "learning_rate": 2.004589552238806e-05, | |
| "loss": 0.8493, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 5.128196323512579, | |
| "grad_norm": 3.6222658157348633, | |
| "learning_rate": 1.9859328358208957e-05, | |
| "loss": 0.8105, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 5.1739807247670715, | |
| "grad_norm": 2.839559316635132, | |
| "learning_rate": 1.9672761194029853e-05, | |
| "loss": 0.8075, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 5.2197651260215645, | |
| "grad_norm": 2.1728529930114746, | |
| "learning_rate": 1.9486940298507463e-05, | |
| "loss": 0.7846, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 5.2197651260215645, | |
| "eval_loss": 0.24261941015720367, | |
| "eval_runtime": 650.6489, | |
| "eval_samples_per_second": 10.07, | |
| "eval_steps_per_second": 10.07, | |
| "eval_wer": 0.13190609371666942, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 5.2655495272760575, | |
| "grad_norm": 1.945101022720337, | |
| "learning_rate": 1.9300373134328362e-05, | |
| "loss": 0.8194, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 5.31133392853055, | |
| "grad_norm": 1.497315764427185, | |
| "learning_rate": 1.9113805970149254e-05, | |
| "loss": 0.8367, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 5.357118329785042, | |
| "grad_norm": 3.0750606060028076, | |
| "learning_rate": 1.8927238805970153e-05, | |
| "loss": 0.8442, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 5.402902731039535, | |
| "grad_norm": 2.4695701599121094, | |
| "learning_rate": 1.8740671641791046e-05, | |
| "loss": 0.8124, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 5.448687132294028, | |
| "grad_norm": 1.8706340789794922, | |
| "learning_rate": 1.85544776119403e-05, | |
| "loss": 0.775, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 5.49447153354852, | |
| "grad_norm": 5.028220176696777, | |
| "learning_rate": 1.8367910447761194e-05, | |
| "loss": 0.808, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 5.49447153354852, | |
| "eval_loss": 0.24327826499938965, | |
| "eval_runtime": 650.6547, | |
| "eval_samples_per_second": 10.07, | |
| "eval_steps_per_second": 10.07, | |
| "eval_wer": 0.13021734493138348, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 5.540255934803013, | |
| "grad_norm": 1.9498703479766846, | |
| "learning_rate": 1.818134328358209e-05, | |
| "loss": 0.7813, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 5.586040336057505, | |
| "grad_norm": 3.7338151931762695, | |
| "learning_rate": 1.7994776119402986e-05, | |
| "loss": 0.7698, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 5.631824737311998, | |
| "grad_norm": 3.364290475845337, | |
| "learning_rate": 1.780820895522388e-05, | |
| "loss": 0.7783, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 5.67760913856649, | |
| "grad_norm": 3.4009501934051514, | |
| "learning_rate": 1.7622014925373137e-05, | |
| "loss": 0.7652, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 5.723393539820983, | |
| "grad_norm": 2.9030606746673584, | |
| "learning_rate": 1.743544776119403e-05, | |
| "loss": 0.7946, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 5.769177941075475, | |
| "grad_norm": 1.3028395175933838, | |
| "learning_rate": 1.724888059701493e-05, | |
| "loss": 0.7456, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 5.769177941075475, | |
| "eval_loss": 0.2477836161851883, | |
| "eval_runtime": 648.6941, | |
| "eval_samples_per_second": 10.1, | |
| "eval_steps_per_second": 10.1, | |
| "eval_wer": 0.1254769974639141, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 5.814962342329968, | |
| "grad_norm": 1.8734581470489502, | |
| "learning_rate": 1.706231343283582e-05, | |
| "loss": 0.7893, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 5.8607467435844605, | |
| "grad_norm": 2.084261178970337, | |
| "learning_rate": 1.6876119402985077e-05, | |
| "loss": 0.7331, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 5.9065311448389535, | |
| "grad_norm": 2.031810760498047, | |
| "learning_rate": 1.668955223880597e-05, | |
| "loss": 0.747, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 5.952315546093446, | |
| "grad_norm": 4.709017276763916, | |
| "learning_rate": 1.650298507462687e-05, | |
| "loss": 0.7473, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 5.998099947347939, | |
| "grad_norm": 3.2480199337005615, | |
| "learning_rate": 1.631641791044776e-05, | |
| "loss": 0.7195, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 6.043884348602431, | |
| "grad_norm": 2.1403627395629883, | |
| "learning_rate": 1.6129850746268657e-05, | |
| "loss": 0.7217, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 6.043884348602431, | |
| "eval_loss": 0.24559348821640015, | |
| "eval_runtime": 650.412, | |
| "eval_samples_per_second": 10.074, | |
| "eval_steps_per_second": 10.074, | |
| "eval_wer": 0.12621175132137186, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 6.089668749856924, | |
| "grad_norm": 2.90395188331604, | |
| "learning_rate": 1.5943283582089553e-05, | |
| "loss": 0.7285, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 6.135453151111417, | |
| "grad_norm": 2.9291510581970215, | |
| "learning_rate": 1.5757089552238805e-05, | |
| "loss": 0.7324, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 6.181237552365909, | |
| "grad_norm": 12.322357177734375, | |
| "learning_rate": 1.5570522388059705e-05, | |
| "loss": 0.7198, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 6.227021953620402, | |
| "grad_norm": 4.177962303161621, | |
| "learning_rate": 1.5383955223880597e-05, | |
| "loss": 0.7206, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 6.272806354874894, | |
| "grad_norm": 1.2690945863723755, | |
| "learning_rate": 1.5197388059701494e-05, | |
| "loss": 0.729, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 6.318590756129387, | |
| "grad_norm": 1.8948352336883545, | |
| "learning_rate": 1.5010820895522389e-05, | |
| "loss": 0.7115, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 6.318590756129387, | |
| "eval_loss": 0.2540062367916107, | |
| "eval_runtime": 648.0874, | |
| "eval_samples_per_second": 10.11, | |
| "eval_steps_per_second": 10.11, | |
| "eval_wer": 0.12306534568983907, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 6.364375157383879, | |
| "grad_norm": 2.9900403022766113, | |
| "learning_rate": 1.4824253731343286e-05, | |
| "loss": 0.7681, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 6.410159558638372, | |
| "grad_norm": 3.5702733993530273, | |
| "learning_rate": 1.4638059701492537e-05, | |
| "loss": 0.7142, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 6.455943959892864, | |
| "grad_norm": 2.3290789127349854, | |
| "learning_rate": 1.4451492537313434e-05, | |
| "loss": 0.71, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 6.501728361147357, | |
| "grad_norm": 6.307437896728516, | |
| "learning_rate": 1.4264925373134328e-05, | |
| "loss": 0.6944, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 6.547512762401849, | |
| "grad_norm": 5.269557952880859, | |
| "learning_rate": 1.4078358208955226e-05, | |
| "loss": 0.7638, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 6.593297163656342, | |
| "grad_norm": 1.6152143478393555, | |
| "learning_rate": 1.389179104477612e-05, | |
| "loss": 0.6701, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 6.593297163656342, | |
| "eval_loss": 0.2436157912015915, | |
| "eval_runtime": 649.7342, | |
| "eval_samples_per_second": 10.084, | |
| "eval_steps_per_second": 10.084, | |
| "eval_wer": 0.12396008627432391, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 6.639081564910835, | |
| "grad_norm": 2.4076077938079834, | |
| "learning_rate": 1.3705223880597016e-05, | |
| "loss": 0.7058, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 6.684865966165328, | |
| "grad_norm": 3.9522953033447266, | |
| "learning_rate": 1.351865671641791e-05, | |
| "loss": 0.6958, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 6.73065036741982, | |
| "grad_norm": 1.8242387771606445, | |
| "learning_rate": 1.3332089552238807e-05, | |
| "loss": 0.7166, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 6.776434768674313, | |
| "grad_norm": 4.560197353363037, | |
| "learning_rate": 1.3145522388059701e-05, | |
| "loss": 0.6946, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 6.822219169928806, | |
| "grad_norm": 2.9841842651367188, | |
| "learning_rate": 1.2958955223880599e-05, | |
| "loss": 0.7179, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 6.868003571183298, | |
| "grad_norm": 1.7246359586715698, | |
| "learning_rate": 1.2772761194029852e-05, | |
| "loss": 0.6966, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 6.868003571183298, | |
| "eval_loss": 0.23714399337768555, | |
| "eval_runtime": 653.4649, | |
| "eval_samples_per_second": 10.027, | |
| "eval_steps_per_second": 10.027, | |
| "eval_wer": 0.12096774193548387, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 6.91378797243779, | |
| "grad_norm": 3.9530229568481445, | |
| "learning_rate": 1.2586194029850747e-05, | |
| "loss": 0.6542, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 6.959572373692283, | |
| "grad_norm": 4.2303690910339355, | |
| "learning_rate": 1.2399626865671643e-05, | |
| "loss": 0.683, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 7.005356774946776, | |
| "grad_norm": 1.5437530279159546, | |
| "learning_rate": 1.2213059701492537e-05, | |
| "loss": 0.6415, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 7.051141176201268, | |
| "grad_norm": 2.246727466583252, | |
| "learning_rate": 1.2026492537313435e-05, | |
| "loss": 0.6831, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 7.096925577455761, | |
| "grad_norm": 5.2752580642700195, | |
| "learning_rate": 1.1839925373134329e-05, | |
| "loss": 0.6666, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 7.142709978710253, | |
| "grad_norm": 1.453157663345337, | |
| "learning_rate": 1.1653358208955226e-05, | |
| "loss": 0.7056, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 7.142709978710253, | |
| "eval_loss": 0.23014602065086365, | |
| "eval_runtime": 648.8331, | |
| "eval_samples_per_second": 10.098, | |
| "eval_steps_per_second": 10.098, | |
| "eval_wer": 0.11875755492877628, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 7.188494379964746, | |
| "grad_norm": 1.6704216003417969, | |
| "learning_rate": 1.1467164179104479e-05, | |
| "loss": 0.7214, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 7.234278781219238, | |
| "grad_norm": 2.2454934120178223, | |
| "learning_rate": 1.1280597014925375e-05, | |
| "loss": 0.6921, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 7.280063182473731, | |
| "grad_norm": 0.8851762413978577, | |
| "learning_rate": 1.1094029850746269e-05, | |
| "loss": 0.6605, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 7.3258475837282235, | |
| "grad_norm": 2.482797861099243, | |
| "learning_rate": 1.0907462686567165e-05, | |
| "loss": 0.6383, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 7.3716319849827165, | |
| "grad_norm": 2.4076988697052, | |
| "learning_rate": 1.072089552238806e-05, | |
| "loss": 0.6638, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 7.417416386237209, | |
| "grad_norm": 2.0605874061584473, | |
| "learning_rate": 1.0534328358208956e-05, | |
| "loss": 0.6654, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 7.417416386237209, | |
| "eval_loss": 0.24034035205841064, | |
| "eval_runtime": 650.4954, | |
| "eval_samples_per_second": 10.072, | |
| "eval_steps_per_second": 10.072, | |
| "eval_wer": 0.11891754165580337, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 7.463200787491702, | |
| "grad_norm": 1.975950002670288, | |
| "learning_rate": 1.034776119402985e-05, | |
| "loss": 0.6744, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 7.508985188746194, | |
| "grad_norm": 1.52113676071167, | |
| "learning_rate": 1.0161940298507465e-05, | |
| "loss": 0.6832, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 7.554769590000687, | |
| "grad_norm": 2.152534246444702, | |
| "learning_rate": 9.975373134328359e-06, | |
| "loss": 0.6719, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 7.600553991255179, | |
| "grad_norm": 1.5796542167663574, | |
| "learning_rate": 9.788805970149254e-06, | |
| "loss": 0.6515, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 7.646338392509672, | |
| "grad_norm": 1.5929287672042847, | |
| "learning_rate": 9.60223880597015e-06, | |
| "loss": 0.666, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 7.692122793764165, | |
| "grad_norm": 1.7882949113845825, | |
| "learning_rate": 9.415671641791046e-06, | |
| "loss": 0.6339, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 7.692122793764165, | |
| "eval_loss": 0.23006904125213623, | |
| "eval_runtime": 649.7568, | |
| "eval_samples_per_second": 10.084, | |
| "eval_steps_per_second": 10.084, | |
| "eval_wer": 0.11488232087412008, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 7.737907195018657, | |
| "grad_norm": 1.4368743896484375, | |
| "learning_rate": 9.22910447761194e-06, | |
| "loss": 0.6861, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 7.78369159627315, | |
| "grad_norm": 2.0316922664642334, | |
| "learning_rate": 9.042537313432836e-06, | |
| "loss": 0.6513, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 7.829475997527642, | |
| "grad_norm": 1.2255696058273315, | |
| "learning_rate": 8.855970149253732e-06, | |
| "loss": 0.6336, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 7.875260398782135, | |
| "grad_norm": 2.4979262351989746, | |
| "learning_rate": 8.669776119402986e-06, | |
| "loss": 0.639, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 7.921044800036627, | |
| "grad_norm": 2.6587982177734375, | |
| "learning_rate": 8.483208955223882e-06, | |
| "loss": 0.6506, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 7.96682920129112, | |
| "grad_norm": 2.314595937728882, | |
| "learning_rate": 8.296641791044778e-06, | |
| "loss": 0.6649, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 7.96682920129112, | |
| "eval_loss": 0.24233344197273254, | |
| "eval_runtime": 654.1461, | |
| "eval_samples_per_second": 10.016, | |
| "eval_steps_per_second": 10.016, | |
| "eval_wer": 0.11500675499514114, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 8.012613602545613, | |
| "grad_norm": 2.341566562652588, | |
| "learning_rate": 8.110074626865673e-06, | |
| "loss": 0.6722, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 8.058398003800106, | |
| "grad_norm": 3.6063222885131836, | |
| "learning_rate": 7.923507462686567e-06, | |
| "loss": 0.6567, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 8.104182405054598, | |
| "grad_norm": 3.134598970413208, | |
| "learning_rate": 7.736940298507463e-06, | |
| "loss": 0.6455, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 8.14996680630909, | |
| "grad_norm": 2.014883041381836, | |
| "learning_rate": 7.550373134328359e-06, | |
| "loss": 0.6618, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 8.195751207563584, | |
| "grad_norm": 2.0325255393981934, | |
| "learning_rate": 7.363805970149255e-06, | |
| "loss": 0.609, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 8.241535608818076, | |
| "grad_norm": 2.343564748764038, | |
| "learning_rate": 7.177238805970151e-06, | |
| "loss": 0.5974, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 8.241535608818076, | |
| "eval_loss": 0.24035315215587616, | |
| "eval_runtime": 656.8993, | |
| "eval_samples_per_second": 9.974, | |
| "eval_steps_per_second": 9.974, | |
| "eval_wer": 0.11389869877462018, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 8.287320010072568, | |
| "grad_norm": 3.118222951889038, | |
| "learning_rate": 6.9906716417910455e-06, | |
| "loss": 0.6189, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 8.33310441132706, | |
| "grad_norm": 1.9298070669174194, | |
| "learning_rate": 6.804104477611941e-06, | |
| "loss": 0.6321, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 8.378888812581554, | |
| "grad_norm": 2.3452179431915283, | |
| "learning_rate": 6.617537313432837e-06, | |
| "loss": 0.6131, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 8.424673213836046, | |
| "grad_norm": 1.260360836982727, | |
| "learning_rate": 6.43134328358209e-06, | |
| "loss": 0.6485, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 8.470457615090538, | |
| "grad_norm": 2.775022506713867, | |
| "learning_rate": 6.2447761194029854e-06, | |
| "loss": 0.6397, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 8.516242016345032, | |
| "grad_norm": 2.084218740463257, | |
| "learning_rate": 6.058208955223881e-06, | |
| "loss": 0.5994, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 8.516242016345032, | |
| "eval_loss": 0.2302207350730896, | |
| "eval_runtime": 661.0282, | |
| "eval_samples_per_second": 9.912, | |
| "eval_steps_per_second": 9.912, | |
| "eval_wer": 0.11068118793107534, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 8.562026417599524, | |
| "grad_norm": 3.495532751083374, | |
| "learning_rate": 5.871641791044776e-06, | |
| "loss": 0.6213, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 8.607810818854016, | |
| "grad_norm": 1.2543615102767944, | |
| "learning_rate": 5.68544776119403e-06, | |
| "loss": 0.6118, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 8.653595220108508, | |
| "grad_norm": 2.4980568885803223, | |
| "learning_rate": 5.498880597014926e-06, | |
| "loss": 0.5992, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 8.699379621363002, | |
| "grad_norm": 3.1688904762268066, | |
| "learning_rate": 5.312313432835822e-06, | |
| "loss": 0.629, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 8.745164022617494, | |
| "grad_norm": 1.777855396270752, | |
| "learning_rate": 5.125746268656717e-06, | |
| "loss": 0.5818, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 8.790948423871987, | |
| "grad_norm": 4.508159637451172, | |
| "learning_rate": 4.939179104477612e-06, | |
| "loss": 0.6323, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 8.790948423871987, | |
| "eval_loss": 0.23431606590747833, | |
| "eval_runtime": 665.4475, | |
| "eval_samples_per_second": 9.846, | |
| "eval_steps_per_second": 9.846, | |
| "eval_wer": 0.11001753928562964, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 8.836732825126479, | |
| "grad_norm": 2.9219791889190674, | |
| "learning_rate": 4.752611940298508e-06, | |
| "loss": 0.5897, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 8.882517226380973, | |
| "grad_norm": 1.1439307928085327, | |
| "learning_rate": 4.5660447761194035e-06, | |
| "loss": 0.611, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 8.928301627635465, | |
| "grad_norm": 2.547057628631592, | |
| "learning_rate": 4.379477611940298e-06, | |
| "loss": 0.6163, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 8.974086028889957, | |
| "grad_norm": 1.996872901916504, | |
| "learning_rate": 4.193283582089553e-06, | |
| "loss": 0.6063, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 9.019870430144449, | |
| "grad_norm": 1.6827927827835083, | |
| "learning_rate": 4.006716417910448e-06, | |
| "loss": 0.594, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 9.065654831398943, | |
| "grad_norm": 1.5130544900894165, | |
| "learning_rate": 3.820149253731343e-06, | |
| "loss": 0.5862, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 9.065654831398943, | |
| "eval_loss": 0.22885525226593018, | |
| "eval_runtime": 661.9197, | |
| "eval_samples_per_second": 9.898, | |
| "eval_steps_per_second": 9.898, | |
| "eval_wer": 0.10877912350975326, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 9.111439232653435, | |
| "grad_norm": 2.690355062484741, | |
| "learning_rate": 3.6335820895522388e-06, | |
| "loss": 0.5986, | |
| "step": 99500 | |
| }, | |
| { | |
| "epoch": 9.157223633907927, | |
| "grad_norm": 1.3742425441741943, | |
| "learning_rate": 3.4470149253731346e-06, | |
| "loss": 0.5974, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 9.20300803516242, | |
| "grad_norm": 2.864790678024292, | |
| "learning_rate": 3.2608208955223884e-06, | |
| "loss": 0.5916, | |
| "step": 100500 | |
| }, | |
| { | |
| "epoch": 9.248792436416913, | |
| "grad_norm": 1.65906822681427, | |
| "learning_rate": 3.074253731343284e-06, | |
| "loss": 0.5724, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 9.294576837671405, | |
| "grad_norm": 1.9599213600158691, | |
| "learning_rate": 2.8876865671641795e-06, | |
| "loss": 0.5857, | |
| "step": 101500 | |
| }, | |
| { | |
| "epoch": 9.340361238925897, | |
| "grad_norm": 3.1552865505218506, | |
| "learning_rate": 2.7011194029850745e-06, | |
| "loss": 0.6196, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 9.340361238925897, | |
| "eval_loss": 0.23824112117290497, | |
| "eval_runtime": 661.8149, | |
| "eval_samples_per_second": 9.9, | |
| "eval_steps_per_second": 9.9, | |
| "eval_wer": 0.10700149320945225, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 9.386145640180391, | |
| "grad_norm": 2.2402780055999756, | |
| "learning_rate": 2.5145522388059703e-06, | |
| "loss": 0.6165, | |
| "step": 102500 | |
| }, | |
| { | |
| "epoch": 9.431930041434883, | |
| "grad_norm": 3.58760929107666, | |
| "learning_rate": 2.327985074626866e-06, | |
| "loss": 0.584, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 9.477714442689376, | |
| "grad_norm": 4.692273139953613, | |
| "learning_rate": 2.1417910447761194e-06, | |
| "loss": 0.5801, | |
| "step": 103500 | |
| }, | |
| { | |
| "epoch": 9.523498843943868, | |
| "grad_norm": 1.1370134353637695, | |
| "learning_rate": 1.955223880597015e-06, | |
| "loss": 0.5876, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 9.569283245198362, | |
| "grad_norm": 1.0059013366699219, | |
| "learning_rate": 1.7686567164179106e-06, | |
| "loss": 0.6329, | |
| "step": 104500 | |
| }, | |
| { | |
| "epoch": 9.615067646452854, | |
| "grad_norm": 2.666593551635742, | |
| "learning_rate": 1.582089552238806e-06, | |
| "loss": 0.6077, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 9.615067646452854, | |
| "eval_loss": 0.22313112020492554, | |
| "eval_runtime": 661.0901, | |
| "eval_samples_per_second": 9.911, | |
| "eval_steps_per_second": 9.911, | |
| "eval_wer": 0.10743997535019317, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 9.660852047707346, | |
| "grad_norm": 3.046264410018921, | |
| "learning_rate": 1.3955223880597016e-06, | |
| "loss": 0.5714, | |
| "step": 105500 | |
| }, | |
| { | |
| "epoch": 9.706636448961838, | |
| "grad_norm": 1.997536301612854, | |
| "learning_rate": 1.2089552238805971e-06, | |
| "loss": 0.6201, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 9.752420850216332, | |
| "grad_norm": 3.185781478881836, | |
| "learning_rate": 1.0223880597014927e-06, | |
| "loss": 0.5782, | |
| "step": 106500 | |
| }, | |
| { | |
| "epoch": 9.798205251470824, | |
| "grad_norm": 4.050471782684326, | |
| "learning_rate": 8.361940298507463e-07, | |
| "loss": 0.5841, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 9.843989652725316, | |
| "grad_norm": 2.3069846630096436, | |
| "learning_rate": 6.496268656716419e-07, | |
| "loss": 0.5601, | |
| "step": 107500 | |
| }, | |
| { | |
| "epoch": 9.88977405397981, | |
| "grad_norm": 3.8783347606658936, | |
| "learning_rate": 4.6343283582089555e-07, | |
| "loss": 0.5949, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 9.88977405397981, | |
| "eval_loss": 0.22683313488960266, | |
| "eval_runtime": 661.0107, | |
| "eval_samples_per_second": 9.912, | |
| "eval_steps_per_second": 9.912, | |
| "eval_wer": 0.1066222654120547, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 9.935558455234302, | |
| "grad_norm": 1.1555778980255127, | |
| "learning_rate": 2.768656716417911e-07, | |
| "loss": 0.5902, | |
| "step": 108500 | |
| }, | |
| { | |
| "epoch": 9.981342856488794, | |
| "grad_norm": 0.8959473371505737, | |
| "learning_rate": 9.029850746268658e-08, | |
| "loss": 0.5785, | |
| "step": 109000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 109200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 3000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0560519550521973e+20, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |