| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.930420896543858, | |
| "eval_steps": 300, | |
| "global_step": 24600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.900000000000002e-06, | |
| "loss": 19.7382, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_cer": 1.0, | |
| "eval_loss": 6.821648120880127, | |
| "eval_runtime": 52.3193, | |
| "eval_samples_per_second": 42.451, | |
| "eval_steps_per_second": 5.314, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.9900000000000003e-05, | |
| "loss": 9.8181, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_cer": 1.0, | |
| "eval_loss": 6.651111602783203, | |
| "eval_runtime": 43.577, | |
| "eval_samples_per_second": 50.967, | |
| "eval_steps_per_second": 6.38, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.9900000000000002e-05, | |
| "loss": 9.5191, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_cer": 0.9806448402826152, | |
| "eval_loss": 6.58424186706543, | |
| "eval_runtime": 43.1655, | |
| "eval_samples_per_second": 51.453, | |
| "eval_steps_per_second": 6.44, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3.99e-05, | |
| "loss": 8.6238, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_cer": 0.9216837496268285, | |
| "eval_loss": 6.142301082611084, | |
| "eval_runtime": 42.9764, | |
| "eval_samples_per_second": 51.68, | |
| "eval_steps_per_second": 6.469, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 4.99e-05, | |
| "loss": 6.883, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_cer": 0.850830928450592, | |
| "eval_loss": 3.596842050552368, | |
| "eval_runtime": 42.8348, | |
| "eval_samples_per_second": 51.85, | |
| "eval_steps_per_second": 6.49, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 4.93579766536965e-05, | |
| "loss": 4.0838, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_cer": 0.48343118718280426, | |
| "eval_loss": 2.5516390800476074, | |
| "eval_runtime": 42.9354, | |
| "eval_samples_per_second": 51.729, | |
| "eval_steps_per_second": 6.475, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.87094682230869e-05, | |
| "loss": 3.167, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_cer": 0.4450691611105583, | |
| "eval_loss": 2.2739391326904297, | |
| "eval_runtime": 42.8894, | |
| "eval_samples_per_second": 51.784, | |
| "eval_steps_per_second": 6.482, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 4.806312148724601e-05, | |
| "loss": 2.826, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_cer": 0.4178525226390686, | |
| "eval_loss": 2.0223917961120605, | |
| "eval_runtime": 42.9677, | |
| "eval_samples_per_second": 51.69, | |
| "eval_steps_per_second": 6.47, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 4.7414613056636405e-05, | |
| "loss": 2.6955, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "eval_cer": 0.4174544730818987, | |
| "eval_loss": 1.9600275754928589, | |
| "eval_runtime": 42.8412, | |
| "eval_samples_per_second": 51.843, | |
| "eval_steps_per_second": 6.489, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 4.676610462602681e-05, | |
| "loss": 2.5812, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "eval_cer": 0.40994128769031746, | |
| "eval_loss": 1.769142985343933, | |
| "eval_runtime": 42.9361, | |
| "eval_samples_per_second": 51.728, | |
| "eval_steps_per_second": 6.475, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 4.611975789018591e-05, | |
| "loss": 2.4952, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_cer": 0.4013832222111653, | |
| "eval_loss": 1.8323670625686646, | |
| "eval_runtime": 42.7115, | |
| "eval_samples_per_second": 52.0, | |
| "eval_steps_per_second": 6.509, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 4.547124945957631e-05, | |
| "loss": 2.3938, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "eval_cer": 0.37799781072743555, | |
| "eval_loss": 1.7351980209350586, | |
| "eval_runtime": 42.7436, | |
| "eval_samples_per_second": 51.961, | |
| "eval_steps_per_second": 6.504, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 4.482490272373541e-05, | |
| "loss": 2.3584, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_cer": 0.3678475470196039, | |
| "eval_loss": 1.64540696144104, | |
| "eval_runtime": 42.7445, | |
| "eval_samples_per_second": 51.96, | |
| "eval_steps_per_second": 6.504, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 4.417639429312581e-05, | |
| "loss": 2.325, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "eval_cer": 0.38352074833316746, | |
| "eval_loss": 1.6946874856948853, | |
| "eval_runtime": 42.4882, | |
| "eval_samples_per_second": 52.273, | |
| "eval_steps_per_second": 6.543, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 4.3527885862516214e-05, | |
| "loss": 2.2454, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "eval_cer": 0.34078017713205294, | |
| "eval_loss": 1.5765234231948853, | |
| "eval_runtime": 42.1243, | |
| "eval_samples_per_second": 52.725, | |
| "eval_steps_per_second": 6.6, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 4.287937743190661e-05, | |
| "loss": 2.1954, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "eval_cer": 0.37093243108767043, | |
| "eval_loss": 1.603211760520935, | |
| "eval_runtime": 42.6116, | |
| "eval_samples_per_second": 52.122, | |
| "eval_steps_per_second": 6.524, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 4.223086900129702e-05, | |
| "loss": 2.1492, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_cer": 0.3447606727037516, | |
| "eval_loss": 1.6078611612319946, | |
| "eval_runtime": 42.9188, | |
| "eval_samples_per_second": 51.749, | |
| "eval_steps_per_second": 6.477, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 4.1582360570687426e-05, | |
| "loss": 2.1655, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "eval_cer": 0.33555577669419845, | |
| "eval_loss": 1.4955742359161377, | |
| "eval_runtime": 42.6136, | |
| "eval_samples_per_second": 52.12, | |
| "eval_steps_per_second": 6.524, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 4.093385214007782e-05, | |
| "loss": 2.1393, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "eval_cer": 0.331625037317146, | |
| "eval_loss": 1.4772567749023438, | |
| "eval_runtime": 42.6929, | |
| "eval_samples_per_second": 52.023, | |
| "eval_steps_per_second": 6.512, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "learning_rate": 4.028534370946823e-05, | |
| "loss": 2.1027, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "eval_cer": 0.341427007662454, | |
| "eval_loss": 1.5089548826217651, | |
| "eval_runtime": 42.7699, | |
| "eval_samples_per_second": 51.929, | |
| "eval_steps_per_second": 6.5, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 3.9636835278858624e-05, | |
| "loss": 2.0824, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "eval_cer": 0.34575579659667627, | |
| "eval_loss": 1.5948169231414795, | |
| "eval_runtime": 42.6031, | |
| "eval_samples_per_second": 52.132, | |
| "eval_steps_per_second": 6.525, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 3.899048854301773e-05, | |
| "loss": 2.061, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "eval_cer": 0.35058214747736094, | |
| "eval_loss": 1.4923882484436035, | |
| "eval_runtime": 42.6516, | |
| "eval_samples_per_second": 52.073, | |
| "eval_steps_per_second": 6.518, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 3.8341980112408135e-05, | |
| "loss": 2.0212, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "eval_cer": 0.33247089262613194, | |
| "eval_loss": 1.4590569734573364, | |
| "eval_runtime": 42.5489, | |
| "eval_samples_per_second": 52.199, | |
| "eval_steps_per_second": 6.534, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 3.769347168179853e-05, | |
| "loss": 2.0504, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "eval_cer": 0.3344611404119813, | |
| "eval_loss": 1.4551000595092773, | |
| "eval_runtime": 42.7689, | |
| "eval_samples_per_second": 51.93, | |
| "eval_steps_per_second": 6.5, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "learning_rate": 3.7044963251188936e-05, | |
| "loss": 2.0113, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "eval_cer": 0.3344113842173351, | |
| "eval_loss": 1.4068984985351562, | |
| "eval_runtime": 42.6741, | |
| "eval_samples_per_second": 52.046, | |
| "eval_steps_per_second": 6.514, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "learning_rate": 3.639645482057933e-05, | |
| "loss": 2.0057, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "eval_cer": 0.3230669718379938, | |
| "eval_loss": 1.443265676498413, | |
| "eval_runtime": 42.7855, | |
| "eval_samples_per_second": 51.91, | |
| "eval_steps_per_second": 6.498, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 3.574794638996974e-05, | |
| "loss": 1.9741, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "eval_cer": 0.3216240421932531, | |
| "eval_loss": 1.426885724067688, | |
| "eval_runtime": 42.8301, | |
| "eval_samples_per_second": 51.856, | |
| "eval_steps_per_second": 6.491, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 3.509943795936014e-05, | |
| "loss": 1.936, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "eval_cer": 0.3227186784754702, | |
| "eval_loss": 1.3612221479415894, | |
| "eval_runtime": 43.0542, | |
| "eval_samples_per_second": 51.586, | |
| "eval_steps_per_second": 6.457, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 3.4450929528750544e-05, | |
| "loss": 1.9416, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "eval_cer": 0.3027664444223306, | |
| "eval_loss": 1.363100290298462, | |
| "eval_runtime": 43.1271, | |
| "eval_samples_per_second": 51.499, | |
| "eval_steps_per_second": 6.446, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 10.95, | |
| "learning_rate": 3.380242109814095e-05, | |
| "loss": 1.9425, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.95, | |
| "eval_cer": 0.3038610807045477, | |
| "eval_loss": 1.3716000318527222, | |
| "eval_runtime": 43.286, | |
| "eval_samples_per_second": 51.31, | |
| "eval_steps_per_second": 6.422, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 3.3153912667531345e-05, | |
| "loss": 1.9351, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "eval_cer": 0.31490695591601153, | |
| "eval_loss": 1.3932286500930786, | |
| "eval_runtime": 43.3029, | |
| "eval_samples_per_second": 51.29, | |
| "eval_steps_per_second": 6.42, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 11.68, | |
| "learning_rate": 3.250540423692175e-05, | |
| "loss": 1.9046, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 11.68, | |
| "eval_cer": 0.3329684545725943, | |
| "eval_loss": 1.4470584392547607, | |
| "eval_runtime": 42.9923, | |
| "eval_samples_per_second": 51.66, | |
| "eval_steps_per_second": 6.466, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 3.185905750108085e-05, | |
| "loss": 1.8587, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "eval_cer": 0.3056523037118121, | |
| "eval_loss": 1.3519924879074097, | |
| "eval_runtime": 42.7004, | |
| "eval_samples_per_second": 52.014, | |
| "eval_steps_per_second": 6.51, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 3.1210549070471253e-05, | |
| "loss": 1.8699, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_cer": 0.3289879590008956, | |
| "eval_loss": 1.4434651136398315, | |
| "eval_runtime": 42.904, | |
| "eval_samples_per_second": 51.767, | |
| "eval_steps_per_second": 6.48, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "learning_rate": 3.056204063986166e-05, | |
| "loss": 1.8328, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "eval_cer": 0.31356353866056325, | |
| "eval_loss": 1.361649751663208, | |
| "eval_runtime": 42.7673, | |
| "eval_samples_per_second": 51.932, | |
| "eval_steps_per_second": 6.5, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 13.14, | |
| "learning_rate": 2.9913532209252054e-05, | |
| "loss": 1.8136, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 13.14, | |
| "eval_cer": 0.2943078913324709, | |
| "eval_loss": 1.3512203693389893, | |
| "eval_runtime": 42.5723, | |
| "eval_samples_per_second": 52.17, | |
| "eval_steps_per_second": 6.53, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 2.9265023778642458e-05, | |
| "loss": 1.8099, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "eval_cer": 0.2956513085879192, | |
| "eval_loss": 1.3534834384918213, | |
| "eval_runtime": 42.854, | |
| "eval_samples_per_second": 51.827, | |
| "eval_steps_per_second": 6.487, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "learning_rate": 2.861651534803286e-05, | |
| "loss": 1.8021, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "eval_cer": 0.2981888745148771, | |
| "eval_loss": 1.3732918500900269, | |
| "eval_runtime": 42.5792, | |
| "eval_samples_per_second": 52.162, | |
| "eval_steps_per_second": 6.529, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "learning_rate": 2.7968006917423263e-05, | |
| "loss": 1.7809, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "eval_cer": 0.30804060105483133, | |
| "eval_loss": 1.3088232278823853, | |
| "eval_runtime": 42.9072, | |
| "eval_samples_per_second": 51.763, | |
| "eval_steps_per_second": 6.479, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 14.6, | |
| "learning_rate": 2.731949848681366e-05, | |
| "loss": 1.7734, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.6, | |
| "eval_cer": 0.28858592894815405, | |
| "eval_loss": 1.320089340209961, | |
| "eval_runtime": 42.6671, | |
| "eval_samples_per_second": 52.054, | |
| "eval_steps_per_second": 6.516, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 2.6670990056204063e-05, | |
| "loss": 1.7646, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_cer": 0.3268981988257538, | |
| "eval_loss": 1.3471167087554932, | |
| "eval_runtime": 42.7924, | |
| "eval_samples_per_second": 51.902, | |
| "eval_steps_per_second": 6.496, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "learning_rate": 2.602248162559447e-05, | |
| "loss": 1.733, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "eval_cer": 0.30321425017414666, | |
| "eval_loss": 1.3437916040420532, | |
| "eval_runtime": 42.7066, | |
| "eval_samples_per_second": 52.006, | |
| "eval_steps_per_second": 6.51, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "learning_rate": 2.5373973194984868e-05, | |
| "loss": 1.7182, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "eval_cer": 0.2999800975221415, | |
| "eval_loss": 1.3310909271240234, | |
| "eval_runtime": 42.787, | |
| "eval_samples_per_second": 51.908, | |
| "eval_steps_per_second": 6.497, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "learning_rate": 2.472546476437527e-05, | |
| "loss": 1.7071, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "eval_cer": 0.3073937705244303, | |
| "eval_loss": 1.2641910314559937, | |
| "eval_runtime": 42.6973, | |
| "eval_samples_per_second": 52.017, | |
| "eval_steps_per_second": 6.511, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 16.43, | |
| "learning_rate": 2.4076956333765675e-05, | |
| "loss": 1.7196, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 16.43, | |
| "eval_cer": 0.2859488506319037, | |
| "eval_loss": 1.2662409543991089, | |
| "eval_runtime": 42.6819, | |
| "eval_samples_per_second": 52.036, | |
| "eval_steps_per_second": 6.513, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 2.3428447903156076e-05, | |
| "loss": 1.7264, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "eval_cer": 0.2878893422231068, | |
| "eval_loss": 1.2460156679153442, | |
| "eval_runtime": 42.7771, | |
| "eval_samples_per_second": 51.92, | |
| "eval_steps_per_second": 6.499, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 2.2782101167315176e-05, | |
| "loss": 1.6875, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "eval_cer": 0.2931137426609613, | |
| "eval_loss": 1.3022774457931519, | |
| "eval_runtime": 42.5345, | |
| "eval_samples_per_second": 52.216, | |
| "eval_steps_per_second": 6.536, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "learning_rate": 2.2133592736705577e-05, | |
| "loss": 1.6659, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "eval_cer": 0.2927654492984377, | |
| "eval_loss": 1.32107675075531, | |
| "eval_runtime": 42.821, | |
| "eval_samples_per_second": 51.867, | |
| "eval_steps_per_second": 6.492, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 2.148508430609598e-05, | |
| "loss": 1.6694, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "eval_cer": 0.2882873917802766, | |
| "eval_loss": 1.3291140794754028, | |
| "eval_runtime": 42.7715, | |
| "eval_samples_per_second": 51.927, | |
| "eval_steps_per_second": 6.5, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 18.25, | |
| "learning_rate": 2.0836575875486384e-05, | |
| "loss": 1.643, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 18.25, | |
| "eval_cer": 0.294755697084287, | |
| "eval_loss": 1.2615532875061035, | |
| "eval_runtime": 42.8646, | |
| "eval_samples_per_second": 51.814, | |
| "eval_steps_per_second": 6.486, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 2.0188067444876785e-05, | |
| "loss": 1.676, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "eval_cer": 0.2835107970942382, | |
| "eval_loss": 1.2185758352279663, | |
| "eval_runtime": 42.7823, | |
| "eval_samples_per_second": 51.914, | |
| "eval_steps_per_second": 6.498, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "learning_rate": 1.9539559014267185e-05, | |
| "loss": 1.6397, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_cer": 0.2810727435565728, | |
| "eval_loss": 1.3059513568878174, | |
| "eval_runtime": 42.9668, | |
| "eval_samples_per_second": 51.691, | |
| "eval_steps_per_second": 6.47, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 1.8893212278426286e-05, | |
| "loss": 1.6347, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_cer": 0.28838690416956914, | |
| "eval_loss": 1.2377227544784546, | |
| "eval_runtime": 42.8094, | |
| "eval_samples_per_second": 51.881, | |
| "eval_steps_per_second": 6.494, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 19.71, | |
| "learning_rate": 1.824470384781669e-05, | |
| "loss": 1.6328, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 19.71, | |
| "eval_cer": 0.27599761170265696, | |
| "eval_loss": 1.2721112966537476, | |
| "eval_runtime": 42.6795, | |
| "eval_samples_per_second": 52.039, | |
| "eval_steps_per_second": 6.514, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 20.08, | |
| "learning_rate": 1.7596195417207094e-05, | |
| "loss": 1.6092, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 20.08, | |
| "eval_cer": 0.28231664842272863, | |
| "eval_loss": 1.2696741819381714, | |
| "eval_runtime": 42.6768, | |
| "eval_samples_per_second": 52.042, | |
| "eval_steps_per_second": 6.514, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 20.44, | |
| "learning_rate": 1.6947686986597494e-05, | |
| "loss": 1.5737, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 20.44, | |
| "eval_cer": 0.28306299134242213, | |
| "eval_loss": 1.2230887413024902, | |
| "eval_runtime": 42.9425, | |
| "eval_samples_per_second": 51.72, | |
| "eval_steps_per_second": 6.474, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 20.81, | |
| "learning_rate": 1.6299178555987894e-05, | |
| "loss": 1.6166, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 20.81, | |
| "eval_cer": 0.2663449099412877, | |
| "eval_loss": 1.2277541160583496, | |
| "eval_runtime": 42.813, | |
| "eval_samples_per_second": 51.877, | |
| "eval_steps_per_second": 6.493, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 21.17, | |
| "learning_rate": 1.56506701253783e-05, | |
| "loss": 1.5964, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 21.17, | |
| "eval_cer": 0.27355955816499156, | |
| "eval_loss": 1.2313120365142822, | |
| "eval_runtime": 42.7309, | |
| "eval_samples_per_second": 51.976, | |
| "eval_steps_per_second": 6.506, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 1.5002161694768699e-05, | |
| "loss": 1.5237, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "eval_cer": 0.27863469001890734, | |
| "eval_loss": 1.2411593198776245, | |
| "eval_runtime": 42.9368, | |
| "eval_samples_per_second": 51.727, | |
| "eval_steps_per_second": 6.475, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 21.9, | |
| "learning_rate": 1.4353653264159101e-05, | |
| "loss": 1.5419, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 21.9, | |
| "eval_cer": 0.28634690018907355, | |
| "eval_loss": 1.2718561887741089, | |
| "eval_runtime": 42.5781, | |
| "eval_samples_per_second": 52.163, | |
| "eval_steps_per_second": 6.529, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 22.27, | |
| "learning_rate": 1.3705144833549505e-05, | |
| "loss": 1.5654, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 22.27, | |
| "eval_cer": 0.26734003383421234, | |
| "eval_loss": 1.2373576164245605, | |
| "eval_runtime": 42.5574, | |
| "eval_samples_per_second": 52.188, | |
| "eval_steps_per_second": 6.532, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "learning_rate": 1.3060959792477304e-05, | |
| "loss": 1.5331, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "eval_cer": 0.2708727236540949, | |
| "eval_loss": 1.197614073753357, | |
| "eval_runtime": 42.6921, | |
| "eval_samples_per_second": 52.024, | |
| "eval_steps_per_second": 6.512, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 1.2412451361867706e-05, | |
| "loss": 1.5378, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_cer": 0.26838491392178326, | |
| "eval_loss": 1.1672557592391968, | |
| "eval_runtime": 42.9497, | |
| "eval_samples_per_second": 51.712, | |
| "eval_steps_per_second": 6.473, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 23.36, | |
| "learning_rate": 1.1763942931258106e-05, | |
| "loss": 1.4972, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 23.36, | |
| "eval_cer": 0.26938003781470793, | |
| "eval_loss": 1.1548832654953003, | |
| "eval_runtime": 42.5425, | |
| "eval_samples_per_second": 52.207, | |
| "eval_steps_per_second": 6.535, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "learning_rate": 1.1115434500648508e-05, | |
| "loss": 1.5112, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "eval_cer": 0.2684844263110757, | |
| "eval_loss": 1.2580962181091309, | |
| "eval_runtime": 43.0178, | |
| "eval_samples_per_second": 51.63, | |
| "eval_steps_per_second": 6.462, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 24.09, | |
| "learning_rate": 1.046692607003891e-05, | |
| "loss": 1.5026, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 24.09, | |
| "eval_cer": 0.26957906259329284, | |
| "eval_loss": 1.2475780248641968, | |
| "eval_runtime": 42.8521, | |
| "eval_samples_per_second": 51.829, | |
| "eval_steps_per_second": 6.487, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 24.46, | |
| "learning_rate": 9.818417639429313e-06, | |
| "loss": 1.5062, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 24.46, | |
| "eval_cer": 0.2755995621454871, | |
| "eval_loss": 1.2111254930496216, | |
| "eval_runtime": 42.9059, | |
| "eval_samples_per_second": 51.764, | |
| "eval_steps_per_second": 6.479, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 24.82, | |
| "learning_rate": 9.169909208819715e-06, | |
| "loss": 1.4816, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 24.82, | |
| "eval_cer": 0.26345905065180614, | |
| "eval_loss": 1.2007070779800415, | |
| "eval_runtime": 42.6871, | |
| "eval_samples_per_second": 52.03, | |
| "eval_steps_per_second": 6.513, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 25.19, | |
| "learning_rate": 8.521400778210117e-06, | |
| "loss": 1.4836, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 25.19, | |
| "eval_cer": 0.27435565727933126, | |
| "eval_loss": 1.2548900842666626, | |
| "eval_runtime": 42.6925, | |
| "eval_samples_per_second": 52.023, | |
| "eval_steps_per_second": 6.512, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 25.55, | |
| "learning_rate": 7.87289234760052e-06, | |
| "loss": 1.479, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 25.55, | |
| "eval_cer": 0.26987759976117026, | |
| "eval_loss": 1.1535056829452515, | |
| "eval_runtime": 42.5922, | |
| "eval_samples_per_second": 52.146, | |
| "eval_steps_per_second": 6.527, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 25.92, | |
| "learning_rate": 7.2243839169909205e-06, | |
| "loss": 1.493, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 25.92, | |
| "eval_cer": 0.26972833117723155, | |
| "eval_loss": 1.198728084564209, | |
| "eval_runtime": 42.6034, | |
| "eval_samples_per_second": 52.132, | |
| "eval_steps_per_second": 6.525, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 26.28, | |
| "learning_rate": 6.5758754863813235e-06, | |
| "loss": 1.4524, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 26.28, | |
| "eval_cer": 0.27470395064185493, | |
| "eval_loss": 1.2245545387268066, | |
| "eval_runtime": 42.9242, | |
| "eval_samples_per_second": 51.742, | |
| "eval_steps_per_second": 6.477, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 26.65, | |
| "learning_rate": 5.927367055771725e-06, | |
| "loss": 1.4569, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 26.65, | |
| "eval_cer": 0.2605234351676784, | |
| "eval_loss": 1.1879122257232666, | |
| "eval_runtime": 42.634, | |
| "eval_samples_per_second": 52.095, | |
| "eval_steps_per_second": 6.521, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "learning_rate": 5.278858625162128e-06, | |
| "loss": 1.4535, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "eval_cer": 0.27281321524529806, | |
| "eval_loss": 1.2265853881835938, | |
| "eval_runtime": 42.7816, | |
| "eval_samples_per_second": 51.915, | |
| "eval_steps_per_second": 6.498, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 27.38, | |
| "learning_rate": 4.63035019455253e-06, | |
| "loss": 1.4452, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 27.38, | |
| "eval_cer": 0.2566424519852722, | |
| "eval_loss": 1.1812487840652466, | |
| "eval_runtime": 42.5285, | |
| "eval_samples_per_second": 52.224, | |
| "eval_steps_per_second": 6.537, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 27.74, | |
| "learning_rate": 3.981841763942931e-06, | |
| "loss": 1.4513, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 27.74, | |
| "eval_cer": 0.26286197631605135, | |
| "eval_loss": 1.1672886610031128, | |
| "eval_runtime": 43.0029, | |
| "eval_samples_per_second": 51.648, | |
| "eval_steps_per_second": 6.465, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 28.11, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.4561, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 28.11, | |
| "eval_cer": 0.26181709622848043, | |
| "eval_loss": 1.1963270902633667, | |
| "eval_runtime": 42.7473, | |
| "eval_samples_per_second": 51.957, | |
| "eval_steps_per_second": 6.503, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 28.47, | |
| "learning_rate": 2.6848249027237355e-06, | |
| "loss": 1.4357, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 28.47, | |
| "eval_cer": 0.26913125684147676, | |
| "eval_loss": 1.201293706893921, | |
| "eval_runtime": 42.6759, | |
| "eval_samples_per_second": 52.043, | |
| "eval_steps_per_second": 6.514, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 28.84, | |
| "learning_rate": 2.0363164721141376e-06, | |
| "loss": 1.4427, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 28.84, | |
| "eval_cer": 0.2726141904667131, | |
| "eval_loss": 1.2448346614837646, | |
| "eval_runtime": 42.4761, | |
| "eval_samples_per_second": 52.288, | |
| "eval_steps_per_second": 6.545, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 29.2, | |
| "learning_rate": 1.3899697362732382e-06, | |
| "loss": 1.4171, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 29.2, | |
| "eval_cer": 0.26684247188775, | |
| "eval_loss": 1.2063277959823608, | |
| "eval_runtime": 42.7033, | |
| "eval_samples_per_second": 52.01, | |
| "eval_steps_per_second": 6.51, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 29.57, | |
| "learning_rate": 7.414613056636403e-07, | |
| "loss": 1.4639, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 29.57, | |
| "eval_cer": 0.26694198427704247, | |
| "eval_loss": 1.2228556871414185, | |
| "eval_runtime": 42.5785, | |
| "eval_samples_per_second": 52.162, | |
| "eval_steps_per_second": 6.529, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 29.93, | |
| "learning_rate": 9.295287505404236e-08, | |
| "loss": 1.4234, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 29.93, | |
| "eval_cer": 0.2594785550801075, | |
| "eval_loss": 1.1955249309539795, | |
| "eval_runtime": 43.02, | |
| "eval_samples_per_second": 51.627, | |
| "eval_steps_per_second": 6.462, | |
| "step": 24600 | |
| } | |
| ], | |
| "logging_steps": 300, | |
| "max_steps": 24630, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 600, | |
| "total_flos": 3.9184197928838064e+20, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |