| { |
| "best_global_step": 13560, |
| "best_metric": 0.38738118946320266, |
| "best_model_checkpoint": "/content/drive/MyDrive/asr_Doracheckpoint/checkpoint-13560", |
| "epoch": 33.0, |
| "eval_steps": 500, |
| "global_step": 14916, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.11074197120708748, |
| "grad_norm": 22.475807189941406, |
| "learning_rate": 6.5044247787610615e-06, |
| "loss": 24.7683, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.22148394241417496, |
| "grad_norm": 16.99262237548828, |
| "learning_rate": 1.3141592920353982e-05, |
| "loss": 24.4852, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.33222591362126247, |
| "grad_norm": 24.19233512878418, |
| "learning_rate": 1.9778761061946902e-05, |
| "loss": 24.1312, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.4429678848283499, |
| "grad_norm": 21.80573081970215, |
| "learning_rate": 2.641592920353982e-05, |
| "loss": 23.2677, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.5537098560354374, |
| "grad_norm": 36.373104095458984, |
| "learning_rate": 3.305309734513274e-05, |
| "loss": 20.9703, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.6644518272425249, |
| "grad_norm": 17.80034828186035, |
| "learning_rate": 3.9690265486725656e-05, |
| "loss": 9.2538, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.7751937984496124, |
| "grad_norm": 3.4296886920928955, |
| "learning_rate": 4.632743362831858e-05, |
| "loss": 3.853, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.8859357696566998, |
| "grad_norm": 1.1884547472000122, |
| "learning_rate": 5.29646017699115e-05, |
| "loss": 3.2597, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.9966777408637874, |
| "grad_norm": 0.490660160779953, |
| "learning_rate": 5.960176991150442e-05, |
| "loss": 3.1597, |
| "step": 450 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_cer": 1.0, |
| "eval_loss": 3.1571714878082275, |
| "eval_runtime": 526.4938, |
| "eval_samples_per_second": 19.337, |
| "eval_steps_per_second": 2.418, |
| "eval_wer": 1.0, |
| "step": 452 |
| }, |
| { |
| "epoch": 1.106312292358804, |
| "grad_norm": 0.9700332283973694, |
| "learning_rate": 6.623893805309734e-05, |
| "loss": 3.1331, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.2170542635658914, |
| "grad_norm": 0.4179767370223999, |
| "learning_rate": 7.287610619469026e-05, |
| "loss": 3.1371, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.327796234772979, |
| "grad_norm": 1.74432373046875, |
| "learning_rate": 7.951327433628318e-05, |
| "loss": 3.1282, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.4385382059800664, |
| "grad_norm": 0.5509875416755676, |
| "learning_rate": 8.61504424778761e-05, |
| "loss": 3.1373, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.5492801771871538, |
| "grad_norm": 0.9913957715034485, |
| "learning_rate": 9.278761061946901e-05, |
| "loss": 3.12, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.6600221483942414, |
| "grad_norm": 0.4770459234714508, |
| "learning_rate": 9.942477876106195e-05, |
| "loss": 3.1318, |
| "step": 750 |
| }, |
| { |
| "epoch": 1.770764119601329, |
| "grad_norm": 0.5994167923927307, |
| "learning_rate": 0.00010606194690265486, |
| "loss": 3.1267, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.8815060908084162, |
| "grad_norm": 0.8976415395736694, |
| "learning_rate": 0.00011269911504424777, |
| "loss": 3.1187, |
| "step": 850 |
| }, |
| { |
| "epoch": 1.9922480620155039, |
| "grad_norm": 0.9691668152809143, |
| "learning_rate": 0.0001193362831858407, |
| "loss": 3.107, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_cer": 1.0, |
| "eval_loss": 3.1111810207366943, |
| "eval_runtime": 526.3967, |
| "eval_samples_per_second": 19.341, |
| "eval_steps_per_second": 2.418, |
| "eval_wer": 1.0, |
| "step": 904 |
| }, |
| { |
| "epoch": 2.1018826135105204, |
| "grad_norm": 1.487389326095581, |
| "learning_rate": 0.0001259734513274336, |
| "loss": 3.0986, |
| "step": 950 |
| }, |
| { |
| "epoch": 2.212624584717608, |
| "grad_norm": 2.4860403537750244, |
| "learning_rate": 0.00013261061946902652, |
| "loss": 3.1104, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.3233665559246957, |
| "grad_norm": 0.33336666226387024, |
| "learning_rate": 0.00013924778761061947, |
| "loss": 3.0526, |
| "step": 1050 |
| }, |
| { |
| "epoch": 2.434108527131783, |
| "grad_norm": 0.37972861528396606, |
| "learning_rate": 0.00014588495575221239, |
| "loss": 3.048, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.5448504983388704, |
| "grad_norm": 0.7906190156936646, |
| "learning_rate": 0.0001525221238938053, |
| "loss": 3.0403, |
| "step": 1150 |
| }, |
| { |
| "epoch": 2.655592469545958, |
| "grad_norm": 0.23263435065746307, |
| "learning_rate": 0.0001591592920353982, |
| "loss": 3.0586, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.7663344407530452, |
| "grad_norm": 0.5591597557067871, |
| "learning_rate": 0.00016579646017699114, |
| "loss": 3.0339, |
| "step": 1250 |
| }, |
| { |
| "epoch": 2.877076411960133, |
| "grad_norm": 0.3939036726951599, |
| "learning_rate": 0.00017243362831858406, |
| "loss": 3.0192, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.9878183831672205, |
| "grad_norm": 0.914394199848175, |
| "learning_rate": 0.000179070796460177, |
| "loss": 2.9634, |
| "step": 1350 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_cer": 1.0, |
| "eval_loss": 2.8615641593933105, |
| "eval_runtime": 514.5892, |
| "eval_samples_per_second": 19.785, |
| "eval_steps_per_second": 2.474, |
| "eval_wer": 1.0, |
| "step": 1356 |
| }, |
| { |
| "epoch": 3.097452934662237, |
| "grad_norm": 0.3776165246963501, |
| "learning_rate": 0.0001857079646017699, |
| "loss": 2.8303, |
| "step": 1400 |
| }, |
| { |
| "epoch": 3.2081949058693247, |
| "grad_norm": 1.6732350587844849, |
| "learning_rate": 0.0001923451327433628, |
| "loss": 2.6318, |
| "step": 1450 |
| }, |
| { |
| "epoch": 3.318936877076412, |
| "grad_norm": 1.371052622795105, |
| "learning_rate": 0.00019898230088495576, |
| "loss": 2.2794, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.4296788482834994, |
| "grad_norm": 1.3353525400161743, |
| "learning_rate": 0.00020561946902654865, |
| "loss": 1.7858, |
| "step": 1550 |
| }, |
| { |
| "epoch": 3.540420819490587, |
| "grad_norm": 1.5970370769500732, |
| "learning_rate": 0.00021225663716814157, |
| "loss": 1.5807, |
| "step": 1600 |
| }, |
| { |
| "epoch": 3.6511627906976747, |
| "grad_norm": 0.7207068204879761, |
| "learning_rate": 0.0002188938053097345, |
| "loss": 1.3541, |
| "step": 1650 |
| }, |
| { |
| "epoch": 3.761904761904762, |
| "grad_norm": 0.6358190774917603, |
| "learning_rate": 0.0002255309734513274, |
| "loss": 1.2085, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.8726467331118495, |
| "grad_norm": 0.7399587035179138, |
| "learning_rate": 0.00023216814159292032, |
| "loss": 1.1028, |
| "step": 1750 |
| }, |
| { |
| "epoch": 3.983388704318937, |
| "grad_norm": 0.690819263458252, |
| "learning_rate": 0.00023880530973451327, |
| "loss": 1.0782, |
| "step": 1800 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_cer": 0.18455519214451996, |
| "eval_loss": 0.6612873673439026, |
| "eval_runtime": 514.2625, |
| "eval_samples_per_second": 19.797, |
| "eval_steps_per_second": 2.475, |
| "eval_wer": 0.6691590477052594, |
| "step": 1808 |
| }, |
| { |
| "epoch": 4.093023255813954, |
| "grad_norm": 1.2990792989730835, |
| "learning_rate": 0.00024544247787610616, |
| "loss": 1.0172, |
| "step": 1850 |
| }, |
| { |
| "epoch": 4.203765227021041, |
| "grad_norm": 1.0645387172698975, |
| "learning_rate": 0.0002520796460176991, |
| "loss": 0.9697, |
| "step": 1900 |
| }, |
| { |
| "epoch": 4.314507198228128, |
| "grad_norm": 0.7005025148391724, |
| "learning_rate": 0.000258716814159292, |
| "loss": 0.8897, |
| "step": 1950 |
| }, |
| { |
| "epoch": 4.425249169435216, |
| "grad_norm": 0.742680013179779, |
| "learning_rate": 0.00026535398230088497, |
| "loss": 0.863, |
| "step": 2000 |
| }, |
| { |
| "epoch": 4.535991140642303, |
| "grad_norm": 1.0513403415679932, |
| "learning_rate": 0.00027199115044247783, |
| "loss": 0.8685, |
| "step": 2050 |
| }, |
| { |
| "epoch": 4.646733111849391, |
| "grad_norm": 0.652015209197998, |
| "learning_rate": 0.0002786283185840708, |
| "loss": 0.8753, |
| "step": 2100 |
| }, |
| { |
| "epoch": 4.7574750830564785, |
| "grad_norm": 0.8807318806648254, |
| "learning_rate": 0.0002852654867256637, |
| "loss": 0.8124, |
| "step": 2150 |
| }, |
| { |
| "epoch": 4.868217054263566, |
| "grad_norm": 0.7368378043174744, |
| "learning_rate": 0.0002919026548672566, |
| "loss": 0.7985, |
| "step": 2200 |
| }, |
| { |
| "epoch": 4.978959025470654, |
| "grad_norm": 0.7042475938796997, |
| "learning_rate": 0.00029853982300884956, |
| "loss": 0.7799, |
| "step": 2250 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_cer": 0.12444249443400947, |
| "eval_loss": 0.4389006197452545, |
| "eval_runtime": 533.481, |
| "eval_samples_per_second": 19.084, |
| "eval_steps_per_second": 2.386, |
| "eval_wer": 0.4757671766090341, |
| "step": 2260 |
| }, |
| { |
| "epoch": 5.08859357696567, |
| "grad_norm": 0.6251569986343384, |
| "learning_rate": 0.00029942477876106194, |
| "loss": 0.7953, |
| "step": 2300 |
| }, |
| { |
| "epoch": 5.1993355481727574, |
| "grad_norm": 0.6194866895675659, |
| "learning_rate": 0.00029868731563421826, |
| "loss": 0.7399, |
| "step": 2350 |
| }, |
| { |
| "epoch": 5.310077519379845, |
| "grad_norm": 0.8238500952720642, |
| "learning_rate": 0.00029794985250737463, |
| "loss": 0.7196, |
| "step": 2400 |
| }, |
| { |
| "epoch": 5.420819490586933, |
| "grad_norm": 0.5640735030174255, |
| "learning_rate": 0.00029721238938053095, |
| "loss": 0.7039, |
| "step": 2450 |
| }, |
| { |
| "epoch": 5.53156146179402, |
| "grad_norm": 1.1068897247314453, |
| "learning_rate": 0.0002964749262536873, |
| "loss": 0.7047, |
| "step": 2500 |
| }, |
| { |
| "epoch": 5.642303433001107, |
| "grad_norm": 0.7027403712272644, |
| "learning_rate": 0.00029573746312684364, |
| "loss": 0.7088, |
| "step": 2550 |
| }, |
| { |
| "epoch": 5.753045404208195, |
| "grad_norm": 0.8116627335548401, |
| "learning_rate": 0.00029499999999999996, |
| "loss": 0.6937, |
| "step": 2600 |
| }, |
| { |
| "epoch": 5.863787375415282, |
| "grad_norm": 0.6587276458740234, |
| "learning_rate": 0.00029426253687315633, |
| "loss": 0.6556, |
| "step": 2650 |
| }, |
| { |
| "epoch": 5.97452934662237, |
| "grad_norm": 0.6965680122375488, |
| "learning_rate": 0.00029352507374631265, |
| "loss": 0.6621, |
| "step": 2700 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_cer": 0.10366158777585194, |
| "eval_loss": 0.36891257762908936, |
| "eval_runtime": 516.652, |
| "eval_samples_per_second": 19.706, |
| "eval_steps_per_second": 2.464, |
| "eval_wer": 0.4011405811532543, |
| "step": 2712 |
| }, |
| { |
| "epoch": 6.084163898117387, |
| "grad_norm": 11.249834060668945, |
| "learning_rate": 0.000292787610619469, |
| "loss": 12.4669, |
| "step": 2750 |
| }, |
| { |
| "epoch": 6.194905869324474, |
| "grad_norm": 0.5191205739974976, |
| "learning_rate": 0.00029205014749262534, |
| "loss": 2.6523, |
| "step": 2800 |
| }, |
| { |
| "epoch": 6.305647840531561, |
| "grad_norm": 0.20015376806259155, |
| "learning_rate": 0.0002913126843657817, |
| "loss": 0.9416, |
| "step": 2850 |
| }, |
| { |
| "epoch": 6.416389811738649, |
| "grad_norm": 0.12533999979496002, |
| "learning_rate": 0.00029057522123893803, |
| "loss": 0.748, |
| "step": 2900 |
| }, |
| { |
| "epoch": 6.5271317829457365, |
| "grad_norm": 0.12881335616111755, |
| "learning_rate": 0.00028983775811209435, |
| "loss": 0.7512, |
| "step": 2950 |
| }, |
| { |
| "epoch": 6.637873754152824, |
| "grad_norm": 0.1284707486629486, |
| "learning_rate": 0.0002891002949852507, |
| "loss": 0.6778, |
| "step": 3000 |
| }, |
| { |
| "epoch": 6.748615725359912, |
| "grad_norm": 0.19739222526550293, |
| "learning_rate": 0.00028836283185840704, |
| "loss": 0.6532, |
| "step": 3050 |
| }, |
| { |
| "epoch": 6.859357696566999, |
| "grad_norm": 0.22524283826351166, |
| "learning_rate": 0.0002876253687315634, |
| "loss": 0.654, |
| "step": 3100 |
| }, |
| { |
| "epoch": 6.970099667774086, |
| "grad_norm": 0.12503944337368011, |
| "learning_rate": 0.00028688790560471973, |
| "loss": 0.6669, |
| "step": 3150 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_cer": 0.10294362938842999, |
| "eval_loss": 0.3620103597640991, |
| "eval_runtime": 549.5105, |
| "eval_samples_per_second": 18.527, |
| "eval_steps_per_second": 2.317, |
| "eval_wer": 0.3994387616547479, |
| "step": 3164 |
| }, |
| { |
| "epoch": 7.079734219269103, |
| "grad_norm": 0.11869937181472778, |
| "learning_rate": 0.00028615044247787605, |
| "loss": 0.6446, |
| "step": 3200 |
| }, |
| { |
| "epoch": 7.190476190476191, |
| "grad_norm": 0.18956558406352997, |
| "learning_rate": 0.0002854129793510324, |
| "loss": 0.654, |
| "step": 3250 |
| }, |
| { |
| "epoch": 7.301218161683278, |
| "grad_norm": 0.13704822957515717, |
| "learning_rate": 0.00028467551622418874, |
| "loss": 0.6491, |
| "step": 3300 |
| }, |
| { |
| "epoch": 7.411960132890365, |
| "grad_norm": 0.18376004695892334, |
| "learning_rate": 0.0002839380530973451, |
| "loss": 0.6623, |
| "step": 3350 |
| }, |
| { |
| "epoch": 7.522702104097453, |
| "grad_norm": 0.1584952026605606, |
| "learning_rate": 0.00028320058997050143, |
| "loss": 0.6523, |
| "step": 3400 |
| }, |
| { |
| "epoch": 7.63344407530454, |
| "grad_norm": 0.11089819669723511, |
| "learning_rate": 0.0002824631268436578, |
| "loss": 0.6266, |
| "step": 3450 |
| }, |
| { |
| "epoch": 7.7441860465116275, |
| "grad_norm": 0.08805970847606659, |
| "learning_rate": 0.0002817256637168141, |
| "loss": 0.6284, |
| "step": 3500 |
| }, |
| { |
| "epoch": 7.8549280177187155, |
| "grad_norm": 0.10621275007724762, |
| "learning_rate": 0.0002809882005899705, |
| "loss": 0.6337, |
| "step": 3550 |
| }, |
| { |
| "epoch": 7.965669988925803, |
| "grad_norm": 0.10317539423704147, |
| "learning_rate": 0.0002802507374631268, |
| "loss": 0.695, |
| "step": 3600 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_cer": 0.10233082652239812, |
| "eval_loss": 0.360572874546051, |
| "eval_runtime": 551.9954, |
| "eval_samples_per_second": 18.444, |
| "eval_steps_per_second": 2.306, |
| "eval_wer": 0.3974291662894904, |
| "step": 3616 |
| }, |
| { |
| "epoch": 8.07530454042082, |
| "grad_norm": 0.10587169229984283, |
| "learning_rate": 0.0002795132743362832, |
| "loss": 0.6347, |
| "step": 3650 |
| }, |
| { |
| "epoch": 8.186046511627907, |
| "grad_norm": 0.11036044359207153, |
| "learning_rate": 0.0002787758112094395, |
| "loss": 0.6384, |
| "step": 3700 |
| }, |
| { |
| "epoch": 8.296788482834994, |
| "grad_norm": 0.08095814287662506, |
| "learning_rate": 0.0002780383480825958, |
| "loss": 0.7067, |
| "step": 3750 |
| }, |
| { |
| "epoch": 8.407530454042082, |
| "grad_norm": 0.12639254331588745, |
| "learning_rate": 0.0002773008849557522, |
| "loss": 0.6403, |
| "step": 3800 |
| }, |
| { |
| "epoch": 8.518272425249169, |
| "grad_norm": 0.1153155043721199, |
| "learning_rate": 0.0002765634218289085, |
| "loss": 0.6301, |
| "step": 3850 |
| }, |
| { |
| "epoch": 8.629014396456256, |
| "grad_norm": 0.17786628007888794, |
| "learning_rate": 0.0002758259587020649, |
| "loss": 0.6628, |
| "step": 3900 |
| }, |
| { |
| "epoch": 8.739756367663345, |
| "grad_norm": 0.14786475896835327, |
| "learning_rate": 0.0002750884955752212, |
| "loss": 0.6585, |
| "step": 3950 |
| }, |
| { |
| "epoch": 8.850498338870432, |
| "grad_norm": 0.11910561472177505, |
| "learning_rate": 0.0002743510324483776, |
| "loss": 0.6215, |
| "step": 4000 |
| }, |
| { |
| "epoch": 8.96124031007752, |
| "grad_norm": 0.0890498086810112, |
| "learning_rate": 0.0002736135693215339, |
| "loss": 0.6378, |
| "step": 4050 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_cer": 0.10174340602359835, |
| "eval_loss": 0.3602811396121979, |
| "eval_runtime": 561.7285, |
| "eval_samples_per_second": 18.124, |
| "eval_steps_per_second": 2.266, |
| "eval_wer": 0.394930750429981, |
| "step": 4068 |
| }, |
| { |
| "epoch": 9.070874861572536, |
| "grad_norm": 0.13651859760284424, |
| "learning_rate": 0.00027287610619469027, |
| "loss": 0.6548, |
| "step": 4100 |
| }, |
| { |
| "epoch": 9.181616832779623, |
| "grad_norm": 0.0844583809375763, |
| "learning_rate": 0.0002721386430678466, |
| "loss": 0.6158, |
| "step": 4150 |
| }, |
| { |
| "epoch": 9.29235880398671, |
| "grad_norm": 0.13365715742111206, |
| "learning_rate": 0.0002714011799410029, |
| "loss": 0.6464, |
| "step": 4200 |
| }, |
| { |
| "epoch": 9.4031007751938, |
| "grad_norm": 0.1272304654121399, |
| "learning_rate": 0.0002706637168141593, |
| "loss": 0.7067, |
| "step": 4250 |
| }, |
| { |
| "epoch": 9.513842746400886, |
| "grad_norm": 0.09687255322933197, |
| "learning_rate": 0.0002699262536873156, |
| "loss": 0.6279, |
| "step": 4300 |
| }, |
| { |
| "epoch": 9.624584717607974, |
| "grad_norm": 0.10860361158847809, |
| "learning_rate": 0.00026918879056047197, |
| "loss": 0.633, |
| "step": 4350 |
| }, |
| { |
| "epoch": 9.73532668881506, |
| "grad_norm": 0.0992758646607399, |
| "learning_rate": 0.0002684513274336283, |
| "loss": 0.642, |
| "step": 4400 |
| }, |
| { |
| "epoch": 9.846068660022148, |
| "grad_norm": 0.2094111442565918, |
| "learning_rate": 0.0002677138643067846, |
| "loss": 0.6265, |
| "step": 4450 |
| }, |
| { |
| "epoch": 9.956810631229235, |
| "grad_norm": 0.12189333885908127, |
| "learning_rate": 0.000266976401179941, |
| "loss": 0.6484, |
| "step": 4500 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_cer": 0.10201535995822787, |
| "eval_loss": 0.3589155077934265, |
| "eval_runtime": 549.3437, |
| "eval_samples_per_second": 18.533, |
| "eval_steps_per_second": 2.317, |
| "eval_wer": 0.3956730334027338, |
| "step": 4520 |
| }, |
| { |
| "epoch": 10.066445182724253, |
| "grad_norm": 0.09829001873731613, |
| "learning_rate": 0.0002662389380530973, |
| "loss": 0.6291, |
| "step": 4550 |
| }, |
| { |
| "epoch": 10.17718715393134, |
| "grad_norm": 0.0978526845574379, |
| "learning_rate": 0.00026550147492625367, |
| "loss": 0.6322, |
| "step": 4600 |
| }, |
| { |
| "epoch": 10.287929125138428, |
| "grad_norm": 0.12002046406269073, |
| "learning_rate": 0.00026476401179941, |
| "loss": 0.6412, |
| "step": 4650 |
| }, |
| { |
| "epoch": 10.398671096345515, |
| "grad_norm": 0.09279460459947586, |
| "learning_rate": 0.00026402654867256636, |
| "loss": 0.6204, |
| "step": 4700 |
| }, |
| { |
| "epoch": 10.509413067552602, |
| "grad_norm": 0.14314547181129456, |
| "learning_rate": 0.0002632890855457227, |
| "loss": 0.6272, |
| "step": 4750 |
| }, |
| { |
| "epoch": 10.62015503875969, |
| "grad_norm": 0.14566180109977722, |
| "learning_rate": 0.00026255162241887905, |
| "loss": 0.645, |
| "step": 4800 |
| }, |
| { |
| "epoch": 10.730897009966778, |
| "grad_norm": 0.11837723851203918, |
| "learning_rate": 0.00026181415929203537, |
| "loss": 0.6399, |
| "step": 4850 |
| }, |
| { |
| "epoch": 10.841638981173865, |
| "grad_norm": 0.09135855734348297, |
| "learning_rate": 0.0002610766961651917, |
| "loss": 0.6787, |
| "step": 4900 |
| }, |
| { |
| "epoch": 10.952380952380953, |
| "grad_norm": 0.1067446619272232, |
| "learning_rate": 0.00026033923303834806, |
| "loss": 0.637, |
| "step": 4950 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_cer": 0.10139893103973428, |
| "eval_loss": 0.3589625060558319, |
| "eval_runtime": 549.1857, |
| "eval_samples_per_second": 18.538, |
| "eval_steps_per_second": 2.318, |
| "eval_wer": 0.3934642889472255, |
| "step": 4972 |
| }, |
| { |
| "epoch": 11.062015503875969, |
| "grad_norm": 0.11917246133089066, |
| "learning_rate": 0.0002596017699115044, |
| "loss": 0.6844, |
| "step": 5000 |
| }, |
| { |
| "epoch": 11.172757475083056, |
| "grad_norm": 0.07237949967384338, |
| "learning_rate": 0.00025886430678466075, |
| "loss": 0.6233, |
| "step": 5050 |
| }, |
| { |
| "epoch": 11.283499446290143, |
| "grad_norm": 0.10749184340238571, |
| "learning_rate": 0.00025812684365781707, |
| "loss": 0.6101, |
| "step": 5100 |
| }, |
| { |
| "epoch": 11.39424141749723, |
| "grad_norm": 0.15197442471981049, |
| "learning_rate": 0.00025738938053097344, |
| "loss": 0.6294, |
| "step": 5150 |
| }, |
| { |
| "epoch": 11.50498338870432, |
| "grad_norm": 0.10893998295068741, |
| "learning_rate": 0.00025665191740412976, |
| "loss": 0.6493, |
| "step": 5200 |
| }, |
| { |
| "epoch": 11.615725359911407, |
| "grad_norm": 0.14979778230190277, |
| "learning_rate": 0.00025591445427728613, |
| "loss": 0.6168, |
| "step": 5250 |
| }, |
| { |
| "epoch": 11.726467331118494, |
| "grad_norm": 0.1401524394750595, |
| "learning_rate": 0.00025517699115044245, |
| "loss": 0.606, |
| "step": 5300 |
| }, |
| { |
| "epoch": 11.837209302325581, |
| "grad_norm": 0.08199647068977356, |
| "learning_rate": 0.0002544395280235988, |
| "loss": 0.6264, |
| "step": 5350 |
| }, |
| { |
| "epoch": 11.947951273532668, |
| "grad_norm": 0.16091564297676086, |
| "learning_rate": 0.00025370206489675514, |
| "loss": 0.6251, |
| "step": 5400 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_cer": 0.10111972500018131, |
| "eval_loss": 0.3570472300052643, |
| "eval_runtime": 552.621, |
| "eval_samples_per_second": 18.423, |
| "eval_steps_per_second": 2.304, |
| "eval_wer": 0.39235991671947135, |
| "step": 5424 |
| }, |
| { |
| "epoch": 12.057585825027685, |
| "grad_norm": 0.13361454010009766, |
| "learning_rate": 0.0002529646017699115, |
| "loss": 0.6816, |
| "step": 5450 |
| }, |
| { |
| "epoch": 12.168327796234774, |
| "grad_norm": 0.10576008260250092, |
| "learning_rate": 0.00025222713864306783, |
| "loss": 0.6356, |
| "step": 5500 |
| }, |
| { |
| "epoch": 12.279069767441861, |
| "grad_norm": 0.07862374186515808, |
| "learning_rate": 0.00025148967551622415, |
| "loss": 0.6357, |
| "step": 5550 |
| }, |
| { |
| "epoch": 12.389811738648948, |
| "grad_norm": 0.1243850514292717, |
| "learning_rate": 0.0002507522123893805, |
| "loss": 0.6165, |
| "step": 5600 |
| }, |
| { |
| "epoch": 12.500553709856035, |
| "grad_norm": 0.08273769915103912, |
| "learning_rate": 0.00025001474926253684, |
| "loss": 0.6517, |
| "step": 5650 |
| }, |
| { |
| "epoch": 12.611295681063122, |
| "grad_norm": 0.0845748633146286, |
| "learning_rate": 0.00024927728613569316, |
| "loss": 0.6227, |
| "step": 5700 |
| }, |
| { |
| "epoch": 12.72203765227021, |
| "grad_norm": 0.1590346395969391, |
| "learning_rate": 0.00024853982300884953, |
| "loss": 0.6199, |
| "step": 5750 |
| }, |
| { |
| "epoch": 12.832779623477299, |
| "grad_norm": 0.08831535279750824, |
| "learning_rate": 0.00024780235988200585, |
| "loss": 0.6347, |
| "step": 5800 |
| }, |
| { |
| "epoch": 12.943521594684386, |
| "grad_norm": 0.07994233816862106, |
| "learning_rate": 0.0002470648967551622, |
| "loss": 0.6312, |
| "step": 5850 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_cer": 0.10085502317047523, |
| "eval_loss": 0.3568708002567291, |
| "eval_runtime": 550.4174, |
| "eval_samples_per_second": 18.497, |
| "eval_steps_per_second": 2.313, |
| "eval_wer": 0.39111070878971665, |
| "step": 5876 |
| }, |
| { |
| "epoch": 13.053156146179402, |
| "grad_norm": 0.15768156945705414, |
| "learning_rate": 0.00024632743362831854, |
| "loss": 0.6196, |
| "step": 5900 |
| }, |
| { |
| "epoch": 13.16389811738649, |
| "grad_norm": 0.1647799015045166, |
| "learning_rate": 0.0002455899705014749, |
| "loss": 0.6328, |
| "step": 5950 |
| }, |
| { |
| "epoch": 13.274640088593577, |
| "grad_norm": 0.10064481943845749, |
| "learning_rate": 0.00024485250737463123, |
| "loss": 0.6426, |
| "step": 6000 |
| }, |
| { |
| "epoch": 13.385382059800664, |
| "grad_norm": 0.1014849841594696, |
| "learning_rate": 0.0002441150442477876, |
| "loss": 0.6521, |
| "step": 6050 |
| }, |
| { |
| "epoch": 13.496124031007753, |
| "grad_norm": 0.09592822939157486, |
| "learning_rate": 0.00024337758112094392, |
| "loss": 0.6411, |
| "step": 6100 |
| }, |
| { |
| "epoch": 13.60686600221484, |
| "grad_norm": 0.08239690959453583, |
| "learning_rate": 0.00024264011799410027, |
| "loss": 0.6931, |
| "step": 6150 |
| }, |
| { |
| "epoch": 13.717607973421927, |
| "grad_norm": 0.0996900349855423, |
| "learning_rate": 0.00024190265486725662, |
| "loss": 0.6447, |
| "step": 6200 |
| }, |
| { |
| "epoch": 13.828349944629014, |
| "grad_norm": 0.10746508836746216, |
| "learning_rate": 0.00024116519174041293, |
| "loss": 0.6146, |
| "step": 6250 |
| }, |
| { |
| "epoch": 13.939091915836102, |
| "grad_norm": 0.09893930703401566, |
| "learning_rate": 0.0002404277286135693, |
| "loss": 0.6226, |
| "step": 6300 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_cer": 0.10101094342632949, |
| "eval_loss": 0.3556014895439148, |
| "eval_runtime": 552.96, |
| "eval_samples_per_second": 18.412, |
| "eval_steps_per_second": 2.302, |
| "eval_wer": 0.3921245587037205, |
| "step": 6328 |
| }, |
| { |
| "epoch": 14.048726467331118, |
| "grad_norm": 0.08149947971105576, |
| "learning_rate": 0.00023969026548672563, |
| "loss": 0.6498, |
| "step": 6350 |
| }, |
| { |
| "epoch": 14.159468438538205, |
| "grad_norm": 0.09183058887720108, |
| "learning_rate": 0.000238952802359882, |
| "loss": 0.6332, |
| "step": 6400 |
| }, |
| { |
| "epoch": 14.270210409745294, |
| "grad_norm": 0.10758010298013687, |
| "learning_rate": 0.00023821533923303832, |
| "loss": 0.6713, |
| "step": 6450 |
| }, |
| { |
| "epoch": 14.380952380952381, |
| "grad_norm": 0.08294181525707245, |
| "learning_rate": 0.0002374778761061947, |
| "loss": 0.6163, |
| "step": 6500 |
| }, |
| { |
| "epoch": 14.491694352159469, |
| "grad_norm": 0.11392789334058762, |
| "learning_rate": 0.000236740412979351, |
| "loss": 0.6247, |
| "step": 6550 |
| }, |
| { |
| "epoch": 14.602436323366556, |
| "grad_norm": 0.08325325697660446, |
| "learning_rate": 0.00023600294985250735, |
| "loss": 0.6199, |
| "step": 6600 |
| }, |
| { |
| "epoch": 14.713178294573643, |
| "grad_norm": 0.10860442370176315, |
| "learning_rate": 0.0002352654867256637, |
| "loss": 0.6196, |
| "step": 6650 |
| }, |
| { |
| "epoch": 14.82392026578073, |
| "grad_norm": 0.08366755396127701, |
| "learning_rate": 0.00023452802359882004, |
| "loss": 0.6137, |
| "step": 6700 |
| }, |
| { |
| "epoch": 14.934662236987819, |
| "grad_norm": 0.09220071136951447, |
| "learning_rate": 0.00023379056047197636, |
| "loss": 0.637, |
| "step": 6750 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_cer": 0.10071360712446788, |
| "eval_loss": 0.3556392192840576, |
| "eval_runtime": 555.5973, |
| "eval_samples_per_second": 18.324, |
| "eval_steps_per_second": 2.291, |
| "eval_wer": 0.39076672399746537, |
| "step": 6780 |
| }, |
| { |
| "epoch": 15.044296788482836, |
| "grad_norm": 0.08812420070171356, |
| "learning_rate": 0.00023305309734513273, |
| "loss": 0.652, |
| "step": 6800 |
| }, |
| { |
| "epoch": 15.155038759689923, |
| "grad_norm": 0.15457949042320251, |
| "learning_rate": 0.00023231563421828905, |
| "loss": 0.6424, |
| "step": 6850 |
| }, |
| { |
| "epoch": 15.26578073089701, |
| "grad_norm": 0.14249703288078308, |
| "learning_rate": 0.00023157817109144542, |
| "loss": 0.6853, |
| "step": 6900 |
| }, |
| { |
| "epoch": 15.376522702104097, |
| "grad_norm": 0.1101984977722168, |
| "learning_rate": 0.00023084070796460174, |
| "loss": 0.6225, |
| "step": 6950 |
| }, |
| { |
| "epoch": 15.487264673311184, |
| "grad_norm": 0.09469130635261536, |
| "learning_rate": 0.00023010324483775812, |
| "loss": 0.612, |
| "step": 7000 |
| }, |
| { |
| "epoch": 15.598006644518273, |
| "grad_norm": 0.137269526720047, |
| "learning_rate": 0.00022936578171091443, |
| "loss": 0.6187, |
| "step": 7050 |
| }, |
| { |
| "epoch": 15.70874861572536, |
| "grad_norm": 0.06478449702262878, |
| "learning_rate": 0.00022862831858407078, |
| "loss": 0.6361, |
| "step": 7100 |
| }, |
| { |
| "epoch": 15.819490586932448, |
| "grad_norm": 0.1404709815979004, |
| "learning_rate": 0.00022789085545722713, |
| "loss": 0.6386, |
| "step": 7150 |
| }, |
| { |
| "epoch": 15.930232558139535, |
| "grad_norm": 0.1699485033750534, |
| "learning_rate": 0.00022715339233038347, |
| "loss": 0.6606, |
| "step": 7200 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_cer": 0.10081513659339623, |
| "eval_loss": 0.35487398505210876, |
| "eval_runtime": 558.9135, |
| "eval_samples_per_second": 18.216, |
| "eval_steps_per_second": 2.278, |
| "eval_wer": 0.39145469358196794, |
| "step": 7232 |
| }, |
| { |
| "epoch": 16.039867109634553, |
| "grad_norm": 0.09163700044155121, |
| "learning_rate": 0.00022641592920353982, |
| "loss": 0.6593, |
| "step": 7250 |
| }, |
| { |
| "epoch": 16.15060908084164, |
| "grad_norm": 0.09030932933092117, |
| "learning_rate": 0.00022567846607669613, |
| "loss": 0.6503, |
| "step": 7300 |
| }, |
| { |
| "epoch": 16.261351052048727, |
| "grad_norm": 0.08114341646432877, |
| "learning_rate": 0.00022494100294985248, |
| "loss": 0.6315, |
| "step": 7350 |
| }, |
| { |
| "epoch": 16.372093023255815, |
| "grad_norm": 0.0924312174320221, |
| "learning_rate": 0.00022420353982300883, |
| "loss": 0.6149, |
| "step": 7400 |
| }, |
| { |
| "epoch": 16.482834994462902, |
| "grad_norm": 0.16909778118133545, |
| "learning_rate": 0.00022346607669616517, |
| "loss": 0.6639, |
| "step": 7450 |
| }, |
| { |
| "epoch": 16.59357696566999, |
| "grad_norm": 0.0675007775425911, |
| "learning_rate": 0.0002227286135693215, |
| "loss": 0.6149, |
| "step": 7500 |
| }, |
| { |
| "epoch": 16.704318936877076, |
| "grad_norm": 0.09390339255332947, |
| "learning_rate": 0.00022199115044247786, |
| "loss": 0.609, |
| "step": 7550 |
| }, |
| { |
| "epoch": 16.815060908084163, |
| "grad_norm": 0.16146640479564667, |
| "learning_rate": 0.00022125368731563418, |
| "loss": 0.6256, |
| "step": 7600 |
| }, |
| { |
| "epoch": 16.92580287929125, |
| "grad_norm": 0.1857237070798874, |
| "learning_rate": 0.00022051622418879055, |
| "loss": 0.6198, |
| "step": 7650 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_cer": 0.10072811133431478, |
| "eval_loss": 0.3537963926792145, |
| "eval_runtime": 558.7168, |
| "eval_samples_per_second": 18.222, |
| "eval_steps_per_second": 2.278, |
| "eval_wer": 0.3909839775504662, |
| "step": 7684 |
| }, |
| { |
| "epoch": 17.035437430786267, |
| "grad_norm": 0.08414191007614136, |
| "learning_rate": 0.00021977876106194687, |
| "loss": 0.6297, |
| "step": 7700 |
| }, |
| { |
| "epoch": 17.146179401993354, |
| "grad_norm": 0.07838060706853867, |
| "learning_rate": 0.00021904129793510324, |
| "loss": 0.6313, |
| "step": 7750 |
| }, |
| { |
| "epoch": 17.25692137320044, |
| "grad_norm": 0.09409710764884949, |
| "learning_rate": 0.00021830383480825956, |
| "loss": 0.5905, |
| "step": 7800 |
| }, |
| { |
| "epoch": 17.367663344407532, |
| "grad_norm": 0.08049694448709488, |
| "learning_rate": 0.0002175663716814159, |
| "loss": 0.6369, |
| "step": 7850 |
| }, |
| { |
| "epoch": 17.47840531561462, |
| "grad_norm": 0.07696959376335144, |
| "learning_rate": 0.00021682890855457225, |
| "loss": 0.6695, |
| "step": 7900 |
| }, |
| { |
| "epoch": 17.589147286821706, |
| "grad_norm": 0.126460462808609, |
| "learning_rate": 0.0002160914454277286, |
| "loss": 0.6243, |
| "step": 7950 |
| }, |
| { |
| "epoch": 17.699889258028794, |
| "grad_norm": 0.1687740832567215, |
| "learning_rate": 0.00021535398230088494, |
| "loss": 0.6513, |
| "step": 8000 |
| }, |
| { |
| "epoch": 17.81063122923588, |
| "grad_norm": 0.11857572942972183, |
| "learning_rate": 0.0002146165191740413, |
| "loss": 0.6224, |
| "step": 8050 |
| }, |
| { |
| "epoch": 17.921373200442968, |
| "grad_norm": 0.0586530826985836, |
| "learning_rate": 0.0002138790560471976, |
| "loss": 0.6195, |
| "step": 8100 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_cer": 0.10032561951106309, |
| "eval_loss": 0.3545036017894745, |
| "eval_runtime": 557.6332, |
| "eval_samples_per_second": 18.258, |
| "eval_steps_per_second": 2.283, |
| "eval_wer": 0.3897890830089617, |
| "step": 8136 |
| }, |
| { |
| "epoch": 18.031007751937985, |
| "grad_norm": 0.09959647804498672, |
| "learning_rate": 0.00021314159292035398, |
| "loss": 0.67, |
| "step": 8150 |
| }, |
| { |
| "epoch": 18.14174972314507, |
| "grad_norm": 0.06997577100992203, |
| "learning_rate": 0.0002124041297935103, |
| "loss": 0.6199, |
| "step": 8200 |
| }, |
| { |
| "epoch": 18.25249169435216, |
| "grad_norm": 0.08038543164730072, |
| "learning_rate": 0.00021166666666666667, |
| "loss": 0.6045, |
| "step": 8250 |
| }, |
| { |
| "epoch": 18.363233665559246, |
| "grad_norm": 0.2446935921907425, |
| "learning_rate": 0.000210929203539823, |
| "loss": 0.6714, |
| "step": 8300 |
| }, |
| { |
| "epoch": 18.473975636766333, |
| "grad_norm": 0.10643508285284042, |
| "learning_rate": 0.00021019174041297936, |
| "loss": 0.634, |
| "step": 8350 |
| }, |
| { |
| "epoch": 18.58471760797342, |
| "grad_norm": 0.1042015478014946, |
| "learning_rate": 0.00020945427728613568, |
| "loss": 0.6283, |
| "step": 8400 |
| }, |
| { |
| "epoch": 18.695459579180508, |
| "grad_norm": 0.07514345645904541, |
| "learning_rate": 0.000208716814159292, |
| "loss": 0.6172, |
| "step": 8450 |
| }, |
| { |
| "epoch": 18.8062015503876, |
| "grad_norm": 0.0896649956703186, |
| "learning_rate": 0.00020797935103244837, |
| "loss": 0.6445, |
| "step": 8500 |
| }, |
| { |
| "epoch": 18.916943521594686, |
| "grad_norm": 0.07546442002058029, |
| "learning_rate": 0.0002072418879056047, |
| "loss": 0.6122, |
| "step": 8550 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_cer": 0.1005032960816877, |
| "eval_loss": 0.3527389466762543, |
| "eval_runtime": 561.1035, |
| "eval_samples_per_second": 18.145, |
| "eval_steps_per_second": 2.269, |
| "eval_wer": 0.39015117226396306, |
| "step": 8588 |
| }, |
| { |
| "epoch": 19.026578073089702, |
| "grad_norm": 0.10488445311784744, |
| "learning_rate": 0.00020650442477876103, |
| "loss": 0.6317, |
| "step": 8600 |
| }, |
| { |
| "epoch": 19.13732004429679, |
| "grad_norm": 0.10324478149414062, |
| "learning_rate": 0.00020576696165191738, |
| "loss": 0.619, |
| "step": 8650 |
| }, |
| { |
| "epoch": 19.248062015503876, |
| "grad_norm": 0.06274005770683289, |
| "learning_rate": 0.00020502949852507373, |
| "loss": 0.618, |
| "step": 8700 |
| }, |
| { |
| "epoch": 19.358803986710964, |
| "grad_norm": 0.0947486013174057, |
| "learning_rate": 0.00020429203539823007, |
| "loss": 0.614, |
| "step": 8750 |
| }, |
| { |
| "epoch": 19.46954595791805, |
| "grad_norm": 0.07655549049377441, |
| "learning_rate": 0.00020355457227138642, |
| "loss": 0.6212, |
| "step": 8800 |
| }, |
| { |
| "epoch": 19.580287929125138, |
| "grad_norm": 0.10087399184703827, |
| "learning_rate": 0.00020281710914454273, |
| "loss": 0.6047, |
| "step": 8850 |
| }, |
| { |
| "epoch": 19.691029900332225, |
| "grad_norm": 0.08093441277742386, |
| "learning_rate": 0.0002020796460176991, |
| "loss": 0.6644, |
| "step": 8900 |
| }, |
| { |
| "epoch": 19.801771871539312, |
| "grad_norm": 0.10238572210073471, |
| "learning_rate": 0.00020134218289085543, |
| "loss": 0.6266, |
| "step": 8950 |
| }, |
| { |
| "epoch": 19.9125138427464, |
| "grad_norm": 0.08402615040540695, |
| "learning_rate": 0.0002006047197640118, |
| "loss": 0.6429, |
| "step": 9000 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_cer": 0.10031836740613963, |
| "eval_loss": 0.35210391879081726, |
| "eval_runtime": 558.1145, |
| "eval_samples_per_second": 18.242, |
| "eval_steps_per_second": 2.281, |
| "eval_wer": 0.3902597990404635, |
| "step": 9040 |
| }, |
| { |
| "epoch": 20.022148394241416, |
| "grad_norm": 0.1166577860713005, |
| "learning_rate": 0.00019986725663716812, |
| "loss": 0.6331, |
| "step": 9050 |
| }, |
| { |
| "epoch": 20.132890365448507, |
| "grad_norm": 0.10892694443464279, |
| "learning_rate": 0.0001991297935103245, |
| "loss": 0.632, |
| "step": 9100 |
| }, |
| { |
| "epoch": 20.243632336655594, |
| "grad_norm": 0.07179673761129379, |
| "learning_rate": 0.0001983923303834808, |
| "loss": 0.6743, |
| "step": 9150 |
| }, |
| { |
| "epoch": 20.35437430786268, |
| "grad_norm": 0.08332252502441406, |
| "learning_rate": 0.00019765486725663715, |
| "loss": 0.622, |
| "step": 9200 |
| }, |
| { |
| "epoch": 20.46511627906977, |
| "grad_norm": 0.08446131646633148, |
| "learning_rate": 0.0001969174041297935, |
| "loss": 0.6225, |
| "step": 9250 |
| }, |
| { |
| "epoch": 20.575858250276855, |
| "grad_norm": 0.07003038376569748, |
| "learning_rate": 0.00019617994100294984, |
| "loss": 0.6013, |
| "step": 9300 |
| }, |
| { |
| "epoch": 20.686600221483943, |
| "grad_norm": 0.0795166864991188, |
| "learning_rate": 0.00019544247787610616, |
| "loss": 0.6369, |
| "step": 9350 |
| }, |
| { |
| "epoch": 20.79734219269103, |
| "grad_norm": 0.07595863193273544, |
| "learning_rate": 0.00019470501474926253, |
| "loss": 0.6167, |
| "step": 9400 |
| }, |
| { |
| "epoch": 20.908084163898117, |
| "grad_norm": 0.06961461901664734, |
| "learning_rate": 0.00019396755162241885, |
| "loss": 0.6042, |
| "step": 9450 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_cer": 0.10031836740613963, |
| "eval_loss": 0.352720707654953, |
| "eval_runtime": 560.7163, |
| "eval_samples_per_second": 18.157, |
| "eval_steps_per_second": 2.27, |
| "eval_wer": 0.3894813071422106, |
| "step": 9492 |
| }, |
| { |
| "epoch": 21.017718715393134, |
| "grad_norm": 0.08535457402467728, |
| "learning_rate": 0.00019323008849557523, |
| "loss": 0.6546, |
| "step": 9500 |
| }, |
| { |
| "epoch": 21.12846068660022, |
| "grad_norm": 0.06741978973150253, |
| "learning_rate": 0.00019249262536873154, |
| "loss": 0.6321, |
| "step": 9550 |
| }, |
| { |
| "epoch": 21.239202657807308, |
| "grad_norm": 0.04948314651846886, |
| "learning_rate": 0.00019175516224188792, |
| "loss": 0.6126, |
| "step": 9600 |
| }, |
| { |
| "epoch": 21.349944629014395, |
| "grad_norm": 0.07996458560228348, |
| "learning_rate": 0.00019101769911504423, |
| "loss": 0.6155, |
| "step": 9650 |
| }, |
| { |
| "epoch": 21.460686600221482, |
| "grad_norm": 0.11527855694293976, |
| "learning_rate": 0.00019028023598820055, |
| "loss": 0.6386, |
| "step": 9700 |
| }, |
| { |
| "epoch": 21.571428571428573, |
| "grad_norm": 0.10837151110172272, |
| "learning_rate": 0.00018954277286135693, |
| "loss": 0.6192, |
| "step": 9750 |
| }, |
| { |
| "epoch": 21.68217054263566, |
| "grad_norm": 0.0754859447479248, |
| "learning_rate": 0.00018880530973451324, |
| "loss": 0.632, |
| "step": 9800 |
| }, |
| { |
| "epoch": 21.792912513842747, |
| "grad_norm": 0.10383785516023636, |
| "learning_rate": 0.00018806784660766962, |
| "loss": 0.6266, |
| "step": 9850 |
| }, |
| { |
| "epoch": 21.903654485049834, |
| "grad_norm": 0.106484055519104, |
| "learning_rate": 0.00018733038348082593, |
| "loss": 0.6689, |
| "step": 9900 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_cer": 0.10026397661921373, |
| "eval_loss": 0.3514802157878876, |
| "eval_runtime": 561.5823, |
| "eval_samples_per_second": 18.129, |
| "eval_steps_per_second": 2.267, |
| "eval_wer": 0.3896442473069612, |
| "step": 9944 |
| }, |
| { |
| "epoch": 22.01328903654485, |
| "grad_norm": 0.05361666530370712, |
| "learning_rate": 0.00018659292035398228, |
| "loss": 0.6447, |
| "step": 9950 |
| }, |
| { |
| "epoch": 22.124031007751938, |
| "grad_norm": 0.2772011160850525, |
| "learning_rate": 0.00018585545722713863, |
| "loss": 0.6217, |
| "step": 10000 |
| }, |
| { |
| "epoch": 22.234772978959025, |
| "grad_norm": 0.07637261599302292, |
| "learning_rate": 0.00018511799410029497, |
| "loss": 0.628, |
| "step": 10050 |
| }, |
| { |
| "epoch": 22.345514950166113, |
| "grad_norm": 0.10416372865438461, |
| "learning_rate": 0.0001843805309734513, |
| "loss": 0.6205, |
| "step": 10100 |
| }, |
| { |
| "epoch": 22.4562569213732, |
| "grad_norm": 0.07692105323076248, |
| "learning_rate": 0.00018364306784660766, |
| "loss": 0.6378, |
| "step": 10150 |
| }, |
| { |
| "epoch": 22.566998892580287, |
| "grad_norm": 0.06518734246492386, |
| "learning_rate": 0.00018290560471976398, |
| "loss": 0.6498, |
| "step": 10200 |
| }, |
| { |
| "epoch": 22.677740863787374, |
| "grad_norm": 0.08064991235733032, |
| "learning_rate": 0.00018216814159292035, |
| "loss": 0.6819, |
| "step": 10250 |
| }, |
| { |
| "epoch": 22.78848283499446, |
| "grad_norm": 0.0814441442489624, |
| "learning_rate": 0.00018143067846607667, |
| "loss": 0.6429, |
| "step": 10300 |
| }, |
| { |
| "epoch": 22.899224806201552, |
| "grad_norm": 0.07057317346334457, |
| "learning_rate": 0.00018069321533923304, |
| "loss": 0.6174, |
| "step": 10350 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_cer": 0.10031836740613963, |
| "eval_loss": 0.3517817556858063, |
| "eval_runtime": 562.2584, |
| "eval_samples_per_second": 18.107, |
| "eval_steps_per_second": 2.264, |
| "eval_wer": 0.3896804562324613, |
| "step": 10396 |
| }, |
| { |
| "epoch": 23.00885935769657, |
| "grad_norm": 0.11094529181718826, |
| "learning_rate": 0.00017995575221238936, |
| "loss": 0.6189, |
| "step": 10400 |
| }, |
| { |
| "epoch": 23.119601328903656, |
| "grad_norm": 0.05448583886027336, |
| "learning_rate": 0.0001792182890855457, |
| "loss": 0.6159, |
| "step": 10450 |
| }, |
| { |
| "epoch": 23.230343300110743, |
| "grad_norm": 0.07730862498283386, |
| "learning_rate": 0.00017848082595870205, |
| "loss": 0.6178, |
| "step": 10500 |
| }, |
| { |
| "epoch": 23.34108527131783, |
| "grad_norm": 0.11480893194675446, |
| "learning_rate": 0.0001777433628318584, |
| "loss": 0.6157, |
| "step": 10550 |
| }, |
| { |
| "epoch": 23.451827242524917, |
| "grad_norm": 0.11411404609680176, |
| "learning_rate": 0.00017700589970501474, |
| "loss": 0.6538, |
| "step": 10600 |
| }, |
| { |
| "epoch": 23.562569213732004, |
| "grad_norm": 0.05760791897773743, |
| "learning_rate": 0.0001762684365781711, |
| "loss": 0.614, |
| "step": 10650 |
| }, |
| { |
| "epoch": 23.67331118493909, |
| "grad_norm": 0.10577227920293808, |
| "learning_rate": 0.0001755309734513274, |
| "loss": 0.6234, |
| "step": 10700 |
| }, |
| { |
| "epoch": 23.78405315614618, |
| "grad_norm": 0.0690847784280777, |
| "learning_rate": 0.00017479351032448378, |
| "loss": 0.6941, |
| "step": 10750 |
| }, |
| { |
| "epoch": 23.894795127353266, |
| "grad_norm": 0.10800562053918839, |
| "learning_rate": 0.0001740560471976401, |
| "loss": 0.6183, |
| "step": 10800 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_cer": 0.10024584635690509, |
| "eval_loss": 0.3515095114707947, |
| "eval_runtime": 561.1662, |
| "eval_samples_per_second": 18.143, |
| "eval_steps_per_second": 2.268, |
| "eval_wer": 0.38939078482846023, |
| "step": 10848 |
| }, |
| { |
| "epoch": 24.004429678848282, |
| "grad_norm": 0.07897496223449707, |
| "learning_rate": 0.00017331858407079642, |
| "loss": 0.6165, |
| "step": 10850 |
| }, |
| { |
| "epoch": 24.11517165005537, |
| "grad_norm": 0.08898892253637314, |
| "learning_rate": 0.0001725811209439528, |
| "loss": 0.6076, |
| "step": 10900 |
| }, |
| { |
| "epoch": 24.225913621262457, |
| "grad_norm": 0.06152245029807091, |
| "learning_rate": 0.0001718436578171091, |
| "loss": 0.6404, |
| "step": 10950 |
| }, |
| { |
| "epoch": 24.336655592469548, |
| "grad_norm": 0.12901481986045837, |
| "learning_rate": 0.00017110619469026548, |
| "loss": 0.617, |
| "step": 11000 |
| }, |
| { |
| "epoch": 24.447397563676635, |
| "grad_norm": 0.06419184058904648, |
| "learning_rate": 0.0001703687315634218, |
| "loss": 0.6144, |
| "step": 11050 |
| }, |
| { |
| "epoch": 24.558139534883722, |
| "grad_norm": 0.11005530506372452, |
| "learning_rate": 0.00016963126843657817, |
| "loss": 0.6257, |
| "step": 11100 |
| }, |
| { |
| "epoch": 24.66888150609081, |
| "grad_norm": 0.10072950273752213, |
| "learning_rate": 0.0001688938053097345, |
| "loss": 0.6658, |
| "step": 11150 |
| }, |
| { |
| "epoch": 24.779623477297896, |
| "grad_norm": 0.06750478595495224, |
| "learning_rate": 0.00016815634218289084, |
| "loss": 0.6114, |
| "step": 11200 |
| }, |
| { |
| "epoch": 24.890365448504983, |
| "grad_norm": 0.08653440326452255, |
| "learning_rate": 0.00016741887905604718, |
| "loss": 0.6513, |
| "step": 11250 |
| }, |
| { |
| "epoch": 25.0, |
| "grad_norm": 0.2293446809053421, |
| "learning_rate": 0.00016668141592920353, |
| "loss": 0.612, |
| "step": 11300 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_cer": 0.10022409004213473, |
| "eval_loss": 0.35145843029022217, |
| "eval_runtime": 561.6713, |
| "eval_samples_per_second": 18.126, |
| "eval_steps_per_second": 2.266, |
| "eval_wer": 0.38926405358920974, |
| "step": 11300 |
| }, |
| { |
| "epoch": 25.110741971207087, |
| "grad_norm": 0.15661917626857758, |
| "learning_rate": 0.00016594395280235987, |
| "loss": 0.6236, |
| "step": 11350 |
| }, |
| { |
| "epoch": 25.221483942414174, |
| "grad_norm": 0.12784335017204285, |
| "learning_rate": 0.00016520648967551622, |
| "loss": 0.6493, |
| "step": 11400 |
| }, |
| { |
| "epoch": 25.33222591362126, |
| "grad_norm": 0.1909065544605255, |
| "learning_rate": 0.00016446902654867254, |
| "loss": 0.6244, |
| "step": 11450 |
| }, |
| { |
| "epoch": 25.44296788482835, |
| "grad_norm": 0.12139736860990524, |
| "learning_rate": 0.0001637315634218289, |
| "loss": 0.6028, |
| "step": 11500 |
| }, |
| { |
| "epoch": 25.553709856035436, |
| "grad_norm": 0.161128968000412, |
| "learning_rate": 0.00016299410029498523, |
| "loss": 0.6146, |
| "step": 11550 |
| }, |
| { |
| "epoch": 25.664451827242527, |
| "grad_norm": 0.1443365067243576, |
| "learning_rate": 0.0001622566371681416, |
| "loss": 0.6052, |
| "step": 11600 |
| }, |
| { |
| "epoch": 25.775193798449614, |
| "grad_norm": 0.12779797613620758, |
| "learning_rate": 0.00016151917404129792, |
| "loss": 0.6123, |
| "step": 11650 |
| }, |
| { |
| "epoch": 25.8859357696567, |
| "grad_norm": 0.13696281611919403, |
| "learning_rate": 0.00016078171091445426, |
| "loss": 0.6373, |
| "step": 11700 |
| }, |
| { |
| "epoch": 25.996677740863788, |
| "grad_norm": 0.13164596259593964, |
| "learning_rate": 0.0001600442477876106, |
| "loss": 0.6623, |
| "step": 11750 |
| }, |
| { |
| "epoch": 26.0, |
| "eval_cer": 0.10007904794366565, |
| "eval_loss": 0.35043439269065857, |
| "eval_runtime": 565.9548, |
| "eval_samples_per_second": 17.989, |
| "eval_steps_per_second": 2.249, |
| "eval_wer": 0.3891554268127093, |
| "step": 11752 |
| }, |
| { |
| "epoch": 26.106312292358805, |
| "grad_norm": 0.2625507712364197, |
| "learning_rate": 0.00015930678466076695, |
| "loss": 0.6105, |
| "step": 11800 |
| }, |
| { |
| "epoch": 26.217054263565892, |
| "grad_norm": 0.16177362203598022, |
| "learning_rate": 0.0001585693215339233, |
| "loss": 0.6126, |
| "step": 11850 |
| }, |
| { |
| "epoch": 26.32779623477298, |
| "grad_norm": 0.10866735130548477, |
| "learning_rate": 0.00015783185840707964, |
| "loss": 0.6576, |
| "step": 11900 |
| }, |
| { |
| "epoch": 26.438538205980066, |
| "grad_norm": 0.14993871748447418, |
| "learning_rate": 0.00015709439528023596, |
| "loss": 0.6162, |
| "step": 11950 |
| }, |
| { |
| "epoch": 26.549280177187153, |
| "grad_norm": 0.15170776844024658, |
| "learning_rate": 0.0001563569321533923, |
| "loss": 0.6001, |
| "step": 12000 |
| }, |
| { |
| "epoch": 26.66002214839424, |
| "grad_norm": 0.12818017601966858, |
| "learning_rate": 0.00015561946902654865, |
| "loss": 0.6013, |
| "step": 12050 |
| }, |
| { |
| "epoch": 26.770764119601328, |
| "grad_norm": 0.11504745483398438, |
| "learning_rate": 0.000154882005899705, |
| "loss": 0.6172, |
| "step": 12100 |
| }, |
| { |
| "epoch": 26.881506090808415, |
| "grad_norm": 0.1459517925977707, |
| "learning_rate": 0.00015414454277286134, |
| "loss": 0.6863, |
| "step": 12150 |
| }, |
| { |
| "epoch": 26.992248062015506, |
| "grad_norm": 0.17688819766044617, |
| "learning_rate": 0.00015340707964601766, |
| "loss": 0.6361, |
| "step": 12200 |
| }, |
| { |
| "epoch": 27.0, |
| "eval_cer": 0.10016244715028537, |
| "eval_loss": 0.3501330316066742, |
| "eval_runtime": 564.3824, |
| "eval_samples_per_second": 18.039, |
| "eval_steps_per_second": 2.256, |
| "eval_wer": 0.3891554268127093, |
| "step": 12204 |
| }, |
| { |
| "epoch": 27.101882613510522, |
| "grad_norm": 0.12469103187322617, |
| "learning_rate": 0.00015266961651917404, |
| "loss": 0.6121, |
| "step": 12250 |
| }, |
| { |
| "epoch": 27.21262458471761, |
| "grad_norm": 0.10200977325439453, |
| "learning_rate": 0.00015193215339233035, |
| "loss": 0.6549, |
| "step": 12300 |
| }, |
| { |
| "epoch": 27.323366555924697, |
| "grad_norm": 0.12290172278881073, |
| "learning_rate": 0.00015119469026548673, |
| "loss": 0.6097, |
| "step": 12350 |
| }, |
| { |
| "epoch": 27.434108527131784, |
| "grad_norm": 0.1658310741186142, |
| "learning_rate": 0.00015045722713864304, |
| "loss": 0.6106, |
| "step": 12400 |
| }, |
| { |
| "epoch": 27.54485049833887, |
| "grad_norm": 0.1108793318271637, |
| "learning_rate": 0.0001497197640117994, |
| "loss": 0.6021, |
| "step": 12450 |
| }, |
| { |
| "epoch": 27.655592469545958, |
| "grad_norm": 0.17435210943222046, |
| "learning_rate": 0.00014898230088495574, |
| "loss": 0.6241, |
| "step": 12500 |
| }, |
| { |
| "epoch": 27.766334440753045, |
| "grad_norm": 0.13998527824878693, |
| "learning_rate": 0.00014824483775811208, |
| "loss": 0.6192, |
| "step": 12550 |
| }, |
| { |
| "epoch": 27.877076411960132, |
| "grad_norm": 0.1330147385597229, |
| "learning_rate": 0.00014750737463126843, |
| "loss": 0.609, |
| "step": 12600 |
| }, |
| { |
| "epoch": 27.98781838316722, |
| "grad_norm": 0.15310999751091003, |
| "learning_rate": 0.00014676991150442477, |
| "loss": 0.6477, |
| "step": 12650 |
| }, |
| { |
| "epoch": 28.0, |
| "eval_cer": 0.09993037979273484, |
| "eval_loss": 0.35032567381858826, |
| "eval_runtime": 564.6857, |
| "eval_samples_per_second": 18.029, |
| "eval_steps_per_second": 2.254, |
| "eval_wer": 0.388268308137956, |
| "step": 12656 |
| }, |
| { |
| "epoch": 28.097452934662236, |
| "grad_norm": 0.16820304095745087, |
| "learning_rate": 0.0001460324483775811, |
| "loss": 0.6146, |
| "step": 12700 |
| }, |
| { |
| "epoch": 28.208194905869323, |
| "grad_norm": 0.1132291629910469, |
| "learning_rate": 0.00014529498525073744, |
| "loss": 0.6127, |
| "step": 12750 |
| }, |
| { |
| "epoch": 28.31893687707641, |
| "grad_norm": 0.6010872721672058, |
| "learning_rate": 0.00014455752212389378, |
| "loss": 0.6169, |
| "step": 12800 |
| }, |
| { |
| "epoch": 28.4296788482835, |
| "grad_norm": 0.11521100252866745, |
| "learning_rate": 0.00014382005899705013, |
| "loss": 0.6457, |
| "step": 12850 |
| }, |
| { |
| "epoch": 28.54042081949059, |
| "grad_norm": 0.11659065634012222, |
| "learning_rate": 0.00014308259587020647, |
| "loss": 0.606, |
| "step": 12900 |
| }, |
| { |
| "epoch": 28.651162790697676, |
| "grad_norm": 0.12543798983097076, |
| "learning_rate": 0.00014234513274336282, |
| "loss": 0.6266, |
| "step": 12950 |
| }, |
| { |
| "epoch": 28.761904761904763, |
| "grad_norm": 0.13689455389976501, |
| "learning_rate": 0.00014160766961651916, |
| "loss": 0.6227, |
| "step": 13000 |
| }, |
| { |
| "epoch": 28.87264673311185, |
| "grad_norm": 0.13249638676643372, |
| "learning_rate": 0.0001408702064896755, |
| "loss": 0.6692, |
| "step": 13050 |
| }, |
| { |
| "epoch": 28.983388704318937, |
| "grad_norm": 0.1185535416007042, |
| "learning_rate": 0.00014013274336283185, |
| "loss": 0.6328, |
| "step": 13100 |
| }, |
| { |
| "epoch": 29.0, |
| "eval_cer": 0.09988686716319412, |
| "eval_loss": 0.3504014313220978, |
| "eval_runtime": 566.291, |
| "eval_samples_per_second": 17.978, |
| "eval_steps_per_second": 2.248, |
| "eval_wer": 0.387743278718204, |
| "step": 13108 |
| }, |
| { |
| "epoch": 29.093023255813954, |
| "grad_norm": 0.09983638674020767, |
| "learning_rate": 0.0001393952802359882, |
| "loss": 0.618, |
| "step": 13150 |
| }, |
| { |
| "epoch": 29.20376522702104, |
| "grad_norm": 0.12238065153360367, |
| "learning_rate": 0.00013865781710914454, |
| "loss": 0.6077, |
| "step": 13200 |
| }, |
| { |
| "epoch": 29.314507198228128, |
| "grad_norm": 0.10367804765701294, |
| "learning_rate": 0.00013792035398230086, |
| "loss": 0.6725, |
| "step": 13250 |
| }, |
| { |
| "epoch": 29.425249169435215, |
| "grad_norm": 0.159180149435997, |
| "learning_rate": 0.0001371828908554572, |
| "loss": 0.6204, |
| "step": 13300 |
| }, |
| { |
| "epoch": 29.535991140642302, |
| "grad_norm": 0.1302846521139145, |
| "learning_rate": 0.00013644542772861355, |
| "loss": 0.6317, |
| "step": 13350 |
| }, |
| { |
| "epoch": 29.64673311184939, |
| "grad_norm": 0.10376603156328201, |
| "learning_rate": 0.0001357079646017699, |
| "loss": 0.6208, |
| "step": 13400 |
| }, |
| { |
| "epoch": 29.757475083056477, |
| "grad_norm": 0.10481042414903641, |
| "learning_rate": 0.00013497050147492624, |
| "loss": 0.5972, |
| "step": 13450 |
| }, |
| { |
| "epoch": 29.868217054263567, |
| "grad_norm": 0.12036976218223572, |
| "learning_rate": 0.0001342330383480826, |
| "loss": 0.6178, |
| "step": 13500 |
| }, |
| { |
| "epoch": 29.978959025470655, |
| "grad_norm": 0.19051200151443481, |
| "learning_rate": 0.00013349557522123894, |
| "loss": 0.6217, |
| "step": 13550 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_cer": 0.09986873690088548, |
| "eval_loss": 0.3510604202747345, |
| "eval_runtime": 564.4628, |
| "eval_samples_per_second": 18.037, |
| "eval_steps_per_second": 2.255, |
| "eval_wer": 0.38738118946320266, |
| "step": 13560 |
| }, |
| { |
| "epoch": 30.08859357696567, |
| "grad_norm": 0.11200982332229614, |
| "learning_rate": 0.00013275811209439528, |
| "loss": 0.6147, |
| "step": 13600 |
| }, |
| { |
| "epoch": 30.19933554817276, |
| "grad_norm": 0.16714787483215332, |
| "learning_rate": 0.00013202064896755163, |
| "loss": 0.6123, |
| "step": 13650 |
| }, |
| { |
| "epoch": 30.310077519379846, |
| "grad_norm": 0.10761655867099762, |
| "learning_rate": 0.00013128318584070797, |
| "loss": 0.5952, |
| "step": 13700 |
| }, |
| { |
| "epoch": 30.420819490586933, |
| "grad_norm": 0.08980994671583176, |
| "learning_rate": 0.0001305457227138643, |
| "loss": 0.5969, |
| "step": 13750 |
| }, |
| { |
| "epoch": 30.53156146179402, |
| "grad_norm": 0.14581350982189178, |
| "learning_rate": 0.00012980825958702064, |
| "loss": 0.6011, |
| "step": 13800 |
| }, |
| { |
| "epoch": 30.642303433001107, |
| "grad_norm": 0.1294136941432953, |
| "learning_rate": 0.00012907079646017698, |
| "loss": 0.6833, |
| "step": 13850 |
| }, |
| { |
| "epoch": 30.753045404208194, |
| "grad_norm": 0.1090904027223587, |
| "learning_rate": 0.00012833333333333333, |
| "loss": 0.633, |
| "step": 13900 |
| }, |
| { |
| "epoch": 30.86378737541528, |
| "grad_norm": 0.10443566739559174, |
| "learning_rate": 0.00012759587020648967, |
| "loss": 0.6415, |
| "step": 13950 |
| }, |
| { |
| "epoch": 30.97452934662237, |
| "grad_norm": 0.10036402940750122, |
| "learning_rate": 0.000126858407079646, |
| "loss": 0.624, |
| "step": 14000 |
| }, |
| { |
| "epoch": 31.0, |
| "eval_cer": 0.10020958583228783, |
| "eval_loss": 0.34915345907211304, |
| "eval_runtime": 563.9167, |
| "eval_samples_per_second": 18.054, |
| "eval_steps_per_second": 2.257, |
| "eval_wer": 0.3891373223499593, |
| "step": 14012 |
| }, |
| { |
| "epoch": 31.084163898117385, |
| "grad_norm": 0.1317962110042572, |
| "learning_rate": 0.00012612094395280234, |
| "loss": 0.6146, |
| "step": 14050 |
| }, |
| { |
| "epoch": 31.194905869324472, |
| "grad_norm": 0.12530717253684998, |
| "learning_rate": 0.00012538348082595868, |
| "loss": 0.6844, |
| "step": 14100 |
| }, |
| { |
| "epoch": 31.305647840531563, |
| "grad_norm": 0.2920992970466614, |
| "learning_rate": 0.00012464601769911503, |
| "loss": 0.65, |
| "step": 14150 |
| }, |
| { |
| "epoch": 31.41638981173865, |
| "grad_norm": 0.1749795526266098, |
| "learning_rate": 0.00012390855457227137, |
| "loss": 0.6423, |
| "step": 14200 |
| }, |
| { |
| "epoch": 31.527131782945737, |
| "grad_norm": 0.13955280184745789, |
| "learning_rate": 0.00012317109144542772, |
| "loss": 0.6209, |
| "step": 14250 |
| }, |
| { |
| "epoch": 31.637873754152825, |
| "grad_norm": 0.09780047833919525, |
| "learning_rate": 0.00012243362831858406, |
| "loss": 0.6038, |
| "step": 14300 |
| }, |
| { |
| "epoch": 31.74861572535991, |
| "grad_norm": 0.1295449286699295, |
| "learning_rate": 0.00012169616519174041, |
| "loss": 0.6072, |
| "step": 14350 |
| }, |
| { |
| "epoch": 31.859357696567, |
| "grad_norm": 0.09745237976312637, |
| "learning_rate": 0.00012095870206489674, |
| "loss": 0.6012, |
| "step": 14400 |
| }, |
| { |
| "epoch": 31.970099667774086, |
| "grad_norm": 0.19207540154457092, |
| "learning_rate": 0.00012022123893805309, |
| "loss": 0.6161, |
| "step": 14450 |
| }, |
| { |
| "epoch": 32.0, |
| "eval_cer": 0.10011168241582119, |
| "eval_loss": 0.3491465151309967, |
| "eval_runtime": 567.5493, |
| "eval_samples_per_second": 17.939, |
| "eval_steps_per_second": 2.243, |
| "eval_wer": 0.38892006879695845, |
| "step": 14464 |
| }, |
| { |
| "epoch": 32.079734219269106, |
| "grad_norm": 0.09013010561466217, |
| "learning_rate": 0.00011948377581120943, |
| "loss": 0.6168, |
| "step": 14500 |
| }, |
| { |
| "epoch": 32.19047619047619, |
| "grad_norm": 0.1108330637216568, |
| "learning_rate": 0.00011874631268436578, |
| "loss": 0.6141, |
| "step": 14550 |
| }, |
| { |
| "epoch": 32.30121816168328, |
| "grad_norm": 0.08618021756410599, |
| "learning_rate": 0.00011800884955752212, |
| "loss": 0.6029, |
| "step": 14600 |
| }, |
| { |
| "epoch": 32.411960132890364, |
| "grad_norm": 0.10959706455469131, |
| "learning_rate": 0.00011727138643067845, |
| "loss": 0.6046, |
| "step": 14650 |
| }, |
| { |
| "epoch": 32.522702104097455, |
| "grad_norm": 0.12988609075546265, |
| "learning_rate": 0.0001165339233038348, |
| "loss": 0.6135, |
| "step": 14700 |
| }, |
| { |
| "epoch": 32.63344407530454, |
| "grad_norm": 0.09731896221637726, |
| "learning_rate": 0.00011579646017699114, |
| "loss": 0.6522, |
| "step": 14750 |
| }, |
| { |
| "epoch": 32.74418604651163, |
| "grad_norm": 0.09596145898103714, |
| "learning_rate": 0.00011505899705014749, |
| "loss": 0.6053, |
| "step": 14800 |
| }, |
| { |
| "epoch": 32.85492801771871, |
| "grad_norm": 0.14724671840667725, |
| "learning_rate": 0.00011432153392330384, |
| "loss": 0.6345, |
| "step": 14850 |
| }, |
| { |
| "epoch": 32.965669988925804, |
| "grad_norm": 0.18355178833007812, |
| "learning_rate": 0.00011358407079646018, |
| "loss": 0.6264, |
| "step": 14900 |
| }, |
| { |
| "epoch": 33.0, |
| "eval_cer": 0.10014794294043847, |
| "eval_loss": 0.3491411507129669, |
| "eval_runtime": 570.1704, |
| "eval_samples_per_second": 17.856, |
| "eval_steps_per_second": 2.233, |
| "eval_wer": 0.3890105911107088, |
| "step": 14916 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 22600, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 50, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 3, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 3 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.2295074179858788e+20, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
