| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 493.02296211251434, | |
| "global_step": 21200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 4.3982, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "eval_cer": 0.14366206687135194, | |
| "eval_loss": 0.5217852592468262, | |
| "eval_runtime": 117.8175, | |
| "eval_samples_per_second": 26.414, | |
| "eval_steps_per_second": 3.302, | |
| "eval_wer": 0.6506676906011348, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "learning_rate": 0.00029461077844311373, | |
| "loss": 0.229, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "eval_cer": 0.08482742332737273, | |
| "eval_loss": 0.36793914437294006, | |
| "eval_runtime": 118.0263, | |
| "eval_samples_per_second": 26.367, | |
| "eval_steps_per_second": 3.296, | |
| "eval_wer": 0.40479542642604205, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 27.9, | |
| "learning_rate": 0.00028742514970059877, | |
| "loss": 0.1054, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 27.9, | |
| "eval_cer": 0.07780964270049597, | |
| "eval_loss": 0.38127270340919495, | |
| "eval_runtime": 116.0512, | |
| "eval_samples_per_second": 26.816, | |
| "eval_steps_per_second": 3.352, | |
| "eval_wer": 0.367037842911387, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 37.21, | |
| "learning_rate": 0.0002802395209580838, | |
| "loss": 0.0784, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 37.21, | |
| "eval_cer": 0.07465164141840143, | |
| "eval_loss": 0.3839242458343506, | |
| "eval_runtime": 116.2982, | |
| "eval_samples_per_second": 26.759, | |
| "eval_steps_per_second": 3.345, | |
| "eval_wer": 0.35496394897393235, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 46.51, | |
| "learning_rate": 0.00027305389221556883, | |
| "loss": 0.066, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 46.51, | |
| "eval_cer": 0.07363946152029421, | |
| "eval_loss": 0.3969549238681793, | |
| "eval_runtime": 118.9877, | |
| "eval_samples_per_second": 26.154, | |
| "eval_steps_per_second": 3.269, | |
| "eval_wer": 0.3443406288664192, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 55.8, | |
| "learning_rate": 0.00026586826347305386, | |
| "loss": 0.0603, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 55.8, | |
| "eval_cer": 0.07218192246701981, | |
| "eval_loss": 0.3702129125595093, | |
| "eval_runtime": 118.9492, | |
| "eval_samples_per_second": 26.162, | |
| "eval_steps_per_second": 3.27, | |
| "eval_wer": 0.3393489483339733, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 65.11, | |
| "learning_rate": 0.0002586826347305389, | |
| "loss": 0.0539, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 65.11, | |
| "eval_cer": 0.07241809777657816, | |
| "eval_loss": 0.3762107491493225, | |
| "eval_runtime": 117.3723, | |
| "eval_samples_per_second": 26.514, | |
| "eval_steps_per_second": 3.314, | |
| "eval_wer": 0.33875165322752676, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 74.41, | |
| "learning_rate": 0.00025149700598802393, | |
| "loss": 0.0497, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 74.41, | |
| "eval_cer": 0.07128445629069807, | |
| "eval_loss": 0.36228740215301514, | |
| "eval_runtime": 116.3914, | |
| "eval_samples_per_second": 26.737, | |
| "eval_steps_per_second": 3.342, | |
| "eval_wer": 0.3413541533341866, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 83.71, | |
| "learning_rate": 0.00024431137724550896, | |
| "loss": 0.0432, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 83.71, | |
| "eval_cer": 0.07248557643645197, | |
| "eval_loss": 0.3847475051879883, | |
| "eval_runtime": 116.3314, | |
| "eval_samples_per_second": 26.751, | |
| "eval_steps_per_second": 3.344, | |
| "eval_wer": 0.3346132514185759, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 93.02, | |
| "learning_rate": 0.000237125748502994, | |
| "loss": 0.0438, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 93.02, | |
| "eval_cer": 0.07503626977968217, | |
| "eval_loss": 0.4057835340499878, | |
| "eval_runtime": 116.616, | |
| "eval_samples_per_second": 26.686, | |
| "eval_steps_per_second": 3.336, | |
| "eval_wer": 0.3393489483339733, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 102.32, | |
| "learning_rate": 0.00022994011976047902, | |
| "loss": 0.0413, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 102.32, | |
| "eval_cer": 0.07270825601403556, | |
| "eval_loss": 0.3957100510597229, | |
| "eval_runtime": 116.9156, | |
| "eval_samples_per_second": 26.617, | |
| "eval_steps_per_second": 3.327, | |
| "eval_wer": 0.3362771449293912, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 111.62, | |
| "learning_rate": 0.00022275449101796406, | |
| "loss": 0.039, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 111.62, | |
| "eval_cer": 0.07179729410573905, | |
| "eval_loss": 0.386459618806839, | |
| "eval_runtime": 119.0004, | |
| "eval_samples_per_second": 26.151, | |
| "eval_steps_per_second": 3.269, | |
| "eval_wer": 0.3330346857801101, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 120.92, | |
| "learning_rate": 0.0002155688622754491, | |
| "loss": 0.0356, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 120.92, | |
| "eval_cer": 0.07110226390903876, | |
| "eval_loss": 0.38599926233291626, | |
| "eval_runtime": 118.9882, | |
| "eval_samples_per_second": 26.154, | |
| "eval_steps_per_second": 3.269, | |
| "eval_wer": 0.33192542343956655, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 130.23, | |
| "learning_rate": 0.00020838323353293412, | |
| "loss": 0.0336, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 130.23, | |
| "eval_cer": 0.07001585748507035, | |
| "eval_loss": 0.3902195692062378, | |
| "eval_runtime": 116.3404, | |
| "eval_samples_per_second": 26.749, | |
| "eval_steps_per_second": 3.344, | |
| "eval_wer": 0.3241605870557618, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 139.53, | |
| "learning_rate": 0.00020119760479041913, | |
| "loss": 0.034, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 139.53, | |
| "eval_cer": 0.07322109382907656, | |
| "eval_loss": 0.39304569363594055, | |
| "eval_runtime": 116.7151, | |
| "eval_samples_per_second": 26.663, | |
| "eval_steps_per_second": 3.333, | |
| "eval_wer": 0.3337173087589061, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 148.83, | |
| "learning_rate": 0.00019402994011976046, | |
| "loss": 0.0273, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 148.83, | |
| "eval_cer": 0.07479334660413645, | |
| "eval_loss": 0.39119070768356323, | |
| "eval_runtime": 116.4756, | |
| "eval_samples_per_second": 26.718, | |
| "eval_steps_per_second": 3.34, | |
| "eval_wer": 0.33747173514228423, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 158.14, | |
| "learning_rate": 0.0001868443113772455, | |
| "loss": 0.027, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 158.14, | |
| "eval_cer": 0.07523870575930362, | |
| "eval_loss": 0.42656052112579346, | |
| "eval_runtime": 116.5029, | |
| "eval_samples_per_second": 26.712, | |
| "eval_steps_per_second": 3.339, | |
| "eval_wer": 0.34344468620674945, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 167.44, | |
| "learning_rate": 0.0001796586826347305, | |
| "loss": 0.028, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 167.44, | |
| "eval_cer": 0.0707648706096697, | |
| "eval_loss": 0.38949263095855713, | |
| "eval_runtime": 116.3646, | |
| "eval_samples_per_second": 26.744, | |
| "eval_steps_per_second": 3.343, | |
| "eval_wer": 0.32266734928964547, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 176.73, | |
| "learning_rate": 0.00017247305389221556, | |
| "loss": 0.0241, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 176.73, | |
| "eval_cer": 0.07274874320995985, | |
| "eval_loss": 0.3967472016811371, | |
| "eval_runtime": 116.4729, | |
| "eval_samples_per_second": 26.719, | |
| "eval_steps_per_second": 3.34, | |
| "eval_wer": 0.32936558726908144, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 186.05, | |
| "learning_rate": 0.00016530538922155687, | |
| "loss": 0.0241, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 186.05, | |
| "eval_cer": 0.07122372549681164, | |
| "eval_loss": 0.4058191776275635, | |
| "eval_runtime": 116.3498, | |
| "eval_samples_per_second": 26.747, | |
| "eval_steps_per_second": 3.343, | |
| "eval_wer": 0.32548316907717906, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 195.34, | |
| "learning_rate": 0.0001581197604790419, | |
| "loss": 0.0209, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 195.34, | |
| "eval_cer": 0.07019804986672964, | |
| "eval_loss": 0.4101807177066803, | |
| "eval_runtime": 116.6051, | |
| "eval_samples_per_second": 26.688, | |
| "eval_steps_per_second": 3.336, | |
| "eval_wer": 0.3233499722684415, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 204.64, | |
| "learning_rate": 0.00015093413173652694, | |
| "loss": 0.0206, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 204.64, | |
| "eval_cer": 0.06987415229933533, | |
| "eval_loss": 0.40751102566719055, | |
| "eval_runtime": 117.508, | |
| "eval_samples_per_second": 26.483, | |
| "eval_steps_per_second": 3.31, | |
| "eval_wer": 0.3193822262041896, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 213.94, | |
| "learning_rate": 0.00014376646706586825, | |
| "loss": 0.0172, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 213.94, | |
| "eval_cer": 0.06948952393805459, | |
| "eval_loss": 0.42218008637428284, | |
| "eval_runtime": 116.7394, | |
| "eval_samples_per_second": 26.658, | |
| "eval_steps_per_second": 3.332, | |
| "eval_wer": 0.31912624258714106, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 223.25, | |
| "learning_rate": 0.00013658083832335328, | |
| "loss": 0.0166, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 223.25, | |
| "eval_cer": 0.06777556597725969, | |
| "eval_loss": 0.38604938983917236, | |
| "eval_runtime": 116.6232, | |
| "eval_samples_per_second": 26.684, | |
| "eval_steps_per_second": 3.336, | |
| "eval_wer": 0.31345193907589913, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 232.55, | |
| "learning_rate": 0.0001293952095808383, | |
| "loss": 0.0156, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 232.55, | |
| "eval_cer": 0.0677013394513985, | |
| "eval_loss": 0.40345117449760437, | |
| "eval_runtime": 117.639, | |
| "eval_samples_per_second": 26.454, | |
| "eval_steps_per_second": 3.307, | |
| "eval_wer": 0.31170271769273433, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 241.85, | |
| "learning_rate": 0.00012220958083832334, | |
| "loss": 0.0149, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 241.85, | |
| "eval_cer": 0.0677013394513985, | |
| "eval_loss": 0.39512303471565247, | |
| "eval_runtime": 120.4059, | |
| "eval_samples_per_second": 25.846, | |
| "eval_steps_per_second": 3.231, | |
| "eval_wer": 0.30867357822432695, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 251.16, | |
| "learning_rate": 0.00011502395209580837, | |
| "loss": 0.0142, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 251.16, | |
| "eval_cer": 0.06735045042005466, | |
| "eval_loss": 0.3971852958202362, | |
| "eval_runtime": 118.3102, | |
| "eval_samples_per_second": 26.304, | |
| "eval_steps_per_second": 3.288, | |
| "eval_wer": 0.309697512692521, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 260.46, | |
| "learning_rate": 0.0001078383233532934, | |
| "loss": 0.0134, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 260.46, | |
| "eval_cer": 0.06749215560578967, | |
| "eval_loss": 0.40693503618240356, | |
| "eval_runtime": 120.6637, | |
| "eval_samples_per_second": 25.791, | |
| "eval_steps_per_second": 3.224, | |
| "eval_wer": 0.31114808652246256, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 269.76, | |
| "learning_rate": 0.00010065269461077844, | |
| "loss": 0.0116, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 269.76, | |
| "eval_cer": 0.06968521205168865, | |
| "eval_loss": 0.41885173320770264, | |
| "eval_runtime": 118.3717, | |
| "eval_samples_per_second": 26.29, | |
| "eval_steps_per_second": 3.286, | |
| "eval_wer": 0.31609710311873374, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 279.07, | |
| "learning_rate": 9.346706586826346e-05, | |
| "loss": 0.0119, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 279.07, | |
| "eval_cer": 0.0648200006747866, | |
| "eval_loss": 0.3901657462120056, | |
| "eval_runtime": 119.8759, | |
| "eval_samples_per_second": 25.96, | |
| "eval_steps_per_second": 3.245, | |
| "eval_wer": 0.3008234139681727, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 288.37, | |
| "learning_rate": 8.62814371257485e-05, | |
| "loss": 0.0098, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 288.37, | |
| "eval_cer": 0.06515064610816829, | |
| "eval_loss": 0.40946489572525024, | |
| "eval_runtime": 120.8583, | |
| "eval_samples_per_second": 25.749, | |
| "eval_steps_per_second": 3.219, | |
| "eval_wer": 0.30018345492555143, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 297.67, | |
| "learning_rate": 7.909580838323352e-05, | |
| "loss": 0.0091, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 297.67, | |
| "eval_cer": 0.06441512871554371, | |
| "eval_loss": 0.3892023265361786, | |
| "eval_runtime": 118.1231, | |
| "eval_samples_per_second": 26.345, | |
| "eval_steps_per_second": 3.293, | |
| "eval_wer": 0.2989888647126584, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 306.96, | |
| "learning_rate": 7.191017964071855e-05, | |
| "loss": 0.0094, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 306.96, | |
| "eval_cer": 0.06469853908701373, | |
| "eval_loss": 0.40261197090148926, | |
| "eval_runtime": 118.2919, | |
| "eval_samples_per_second": 26.308, | |
| "eval_steps_per_second": 3.288, | |
| "eval_wer": 0.29834890567003713, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 316.28, | |
| "learning_rate": 6.474251497005988e-05, | |
| "loss": 0.0081, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 316.28, | |
| "eval_cer": 0.06462431256115253, | |
| "eval_loss": 0.4302999675273895, | |
| "eval_runtime": 118.0293, | |
| "eval_samples_per_second": 26.366, | |
| "eval_steps_per_second": 3.296, | |
| "eval_wer": 0.29779427449976537, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 325.57, | |
| "learning_rate": 5.7556886227544904e-05, | |
| "loss": 0.0079, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 325.57, | |
| "eval_cer": 0.06431391072573299, | |
| "eval_loss": 0.40440893173217773, | |
| "eval_runtime": 118.2184, | |
| "eval_samples_per_second": 26.324, | |
| "eval_steps_per_second": 3.291, | |
| "eval_wer": 0.29796493024446435, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 334.87, | |
| "learning_rate": 5.038922155688622e-05, | |
| "loss": 0.0072, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 334.87, | |
| "eval_cer": 0.06546779580957522, | |
| "eval_loss": 0.382755845785141, | |
| "eval_runtime": 118.6873, | |
| "eval_samples_per_second": 26.22, | |
| "eval_steps_per_second": 3.278, | |
| "eval_wer": 0.29992747130850295, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 344.18, | |
| "learning_rate": 9.578571428571428e-05, | |
| "loss": 0.0081, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 344.18, | |
| "eval_cer": 0.06676338607915247, | |
| "eval_loss": 0.4108315706253052, | |
| "eval_runtime": 114.661, | |
| "eval_samples_per_second": 27.141, | |
| "eval_steps_per_second": 3.393, | |
| "eval_wer": 0.30457784035155083, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 353.48, | |
| "learning_rate": 9.007142857142856e-05, | |
| "loss": 0.0088, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 353.48, | |
| "eval_cer": 0.06539356928371402, | |
| "eval_loss": 0.40191251039505005, | |
| "eval_runtime": 117.0774, | |
| "eval_samples_per_second": 26.581, | |
| "eval_steps_per_second": 3.323, | |
| "eval_wer": 0.2993301762020564, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 362.78, | |
| "learning_rate": 8.435714285714286e-05, | |
| "loss": 0.0088, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 362.78, | |
| "eval_cer": 0.06814669860656568, | |
| "eval_loss": 0.4072999954223633, | |
| "eval_runtime": 114.628, | |
| "eval_samples_per_second": 27.149, | |
| "eval_steps_per_second": 3.394, | |
| "eval_wer": 0.3091002175860745, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 372.09, | |
| "learning_rate": 7.864285714285714e-05, | |
| "loss": 0.0079, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 372.09, | |
| "eval_cer": 0.0667161510172408, | |
| "eval_loss": 0.42044562101364136, | |
| "eval_runtime": 115.1974, | |
| "eval_samples_per_second": 27.014, | |
| "eval_steps_per_second": 3.377, | |
| "eval_wer": 0.30547378301122063, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 381.39, | |
| "learning_rate": 7.292857142857142e-05, | |
| "loss": 0.0072, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 381.39, | |
| "eval_cer": 0.06564998819123452, | |
| "eval_loss": 0.40300747752189636, | |
| "eval_runtime": 114.5196, | |
| "eval_samples_per_second": 27.174, | |
| "eval_steps_per_second": 3.397, | |
| "eval_wer": 0.3027859550322113, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 390.69, | |
| "learning_rate": 6.721428571428571e-05, | |
| "loss": 0.0073, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 390.69, | |
| "eval_cer": 0.0677350787813354, | |
| "eval_loss": 0.4031626284122467, | |
| "eval_runtime": 114.7831, | |
| "eval_samples_per_second": 27.112, | |
| "eval_steps_per_second": 3.389, | |
| "eval_wer": 0.30807628311788043, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 399.99, | |
| "learning_rate": 6.151428571428571e-05, | |
| "loss": 0.0069, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 399.99, | |
| "eval_cer": 0.06693208272883701, | |
| "eval_loss": 0.41302183270454407, | |
| "eval_runtime": 114.8457, | |
| "eval_samples_per_second": 27.097, | |
| "eval_steps_per_second": 3.387, | |
| "eval_wer": 0.30214599598959, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 409.3, | |
| "learning_rate": 5.5799999999999994e-05, | |
| "loss": 0.0063, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 409.3, | |
| "eval_cer": 0.06513715037619353, | |
| "eval_loss": 0.4071926772594452, | |
| "eval_runtime": 114.6356, | |
| "eval_samples_per_second": 27.147, | |
| "eval_steps_per_second": 3.393, | |
| "eval_wer": 0.2979222663082896, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 418.6, | |
| "learning_rate": 5.008571428571428e-05, | |
| "loss": 0.0059, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 418.6, | |
| "eval_cer": 0.06403724822025035, | |
| "eval_loss": 0.41102761030197144, | |
| "eval_runtime": 116.5576, | |
| "eval_samples_per_second": 26.699, | |
| "eval_steps_per_second": 3.337, | |
| "eval_wer": 0.2969409957762703, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 427.9, | |
| "learning_rate": 4.437142857142857e-05, | |
| "loss": 0.0056, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 427.9, | |
| "eval_cer": 0.06465805189108945, | |
| "eval_loss": 0.4228787422180176, | |
| "eval_runtime": 114.7096, | |
| "eval_samples_per_second": 27.129, | |
| "eval_steps_per_second": 3.391, | |
| "eval_wer": 0.29945816801058067, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 437.21, | |
| "learning_rate": 3.8657142857142856e-05, | |
| "loss": 0.005, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 437.21, | |
| "eval_cer": 0.062370525321367117, | |
| "eval_loss": 0.41175001859664917, | |
| "eval_runtime": 115.5821, | |
| "eval_samples_per_second": 26.925, | |
| "eval_steps_per_second": 3.366, | |
| "eval_wer": 0.2884508724774948, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 446.51, | |
| "learning_rate": 3.294285714285714e-05, | |
| "loss": 0.0046, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 446.51, | |
| "eval_cer": 0.06147305914504538, | |
| "eval_loss": 0.41112595796585083, | |
| "eval_runtime": 115.5282, | |
| "eval_samples_per_second": 26.937, | |
| "eval_steps_per_second": 3.367, | |
| "eval_wer": 0.28409915098767013, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 455.8, | |
| "learning_rate": 2.7228571428571427e-05, | |
| "loss": 0.0043, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 455.8, | |
| "eval_cer": 0.06160126859880563, | |
| "eval_loss": 0.40707847476005554, | |
| "eval_runtime": 117.681, | |
| "eval_samples_per_second": 26.444, | |
| "eval_steps_per_second": 3.306, | |
| "eval_wer": 0.28495242971116513, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 465.11, | |
| "learning_rate": 2.1514285714285714e-05, | |
| "loss": 0.0038, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 465.11, | |
| "eval_cer": 0.062363777455379736, | |
| "eval_loss": 0.4267757534980774, | |
| "eval_runtime": 115.119, | |
| "eval_samples_per_second": 27.033, | |
| "eval_steps_per_second": 3.379, | |
| "eval_wer": 0.28670165109433, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 474.41, | |
| "learning_rate": 1.5799999999999998e-05, | |
| "loss": 0.0035, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 474.41, | |
| "eval_cer": 0.06053510577279935, | |
| "eval_loss": 0.4116959869861603, | |
| "eval_runtime": 115.3416, | |
| "eval_samples_per_second": 26.981, | |
| "eval_steps_per_second": 3.373, | |
| "eval_wer": 0.2820086181151073, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 483.71, | |
| "learning_rate": 1.0085714285714285e-05, | |
| "loss": 0.0035, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 483.71, | |
| "eval_cer": 0.060238199669354564, | |
| "eval_loss": 0.4154604375362396, | |
| "eval_runtime": 115.2471, | |
| "eval_samples_per_second": 27.003, | |
| "eval_steps_per_second": 3.375, | |
| "eval_wer": 0.2819232902427578, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 493.02, | |
| "learning_rate": 4.371428571428571e-06, | |
| "loss": 0.0034, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 493.02, | |
| "eval_cer": 0.06007625088565741, | |
| "eval_loss": 0.41654759645462036, | |
| "eval_runtime": 115.3904, | |
| "eval_samples_per_second": 26.969, | |
| "eval_steps_per_second": 3.371, | |
| "eval_wer": 0.27991808524254447, | |
| "step": 21200 | |
| } | |
| ], | |
| "max_steps": 21500, | |
| "num_train_epochs": 500, | |
| "total_flos": 5.302246573116527e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |