| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.99892202659001, | |
| "global_step": 34750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.675e-06, | |
| "loss": 47.2908, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.425e-06, | |
| "loss": 33.9125, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.1174999999999999e-05, | |
| "loss": 26.6068, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.4925e-05, | |
| "loss": 23.2775, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.8675e-05, | |
| "loss": 19.7138, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_cer": 1.0, | |
| "eval_loss": 19.642736434936523, | |
| "eval_runtime": 41.3907, | |
| "eval_samples_per_second": 11.017, | |
| "eval_steps_per_second": 1.377, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.2424999999999996e-05, | |
| "loss": 15.7715, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.6174999999999996e-05, | |
| "loss": 11.4061, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 2.9925e-05, | |
| "loss": 7.4329, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.3675e-05, | |
| "loss": 5.3081, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 3.7424999999999995e-05, | |
| "loss": 4.8039, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_cer": 1.0, | |
| "eval_loss": 4.784187316894531, | |
| "eval_runtime": 42.2256, | |
| "eval_samples_per_second": 10.799, | |
| "eval_steps_per_second": 1.35, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.1175e-05, | |
| "loss": 4.762, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 4.4924999999999994e-05, | |
| "loss": 4.6928, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 4.8675e-05, | |
| "loss": 4.6292, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 5.2424999999999994e-05, | |
| "loss": 4.6321, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.6175e-05, | |
| "loss": 4.5619, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "eval_cer": 0.9598094788222327, | |
| "eval_loss": 4.560794830322266, | |
| "eval_runtime": 41.0352, | |
| "eval_samples_per_second": 11.112, | |
| "eval_steps_per_second": 1.389, | |
| "eval_wer": 0.9992449411054062, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 5.9925e-05, | |
| "loss": 4.4704, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 6.367499999999999e-05, | |
| "loss": 4.3806, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 6.7425e-05, | |
| "loss": 4.3092, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 7.1175e-05, | |
| "loss": 4.2794, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 7.492499999999999e-05, | |
| "loss": 4.254, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_cer": 0.906274602424815, | |
| "eval_loss": 4.272861003875732, | |
| "eval_runtime": 40.8387, | |
| "eval_samples_per_second": 11.166, | |
| "eval_steps_per_second": 1.396, | |
| "eval_wer": 0.9954696466324373, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 7.477557251908395e-05, | |
| "loss": 4.2616, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 7.454656488549618e-05, | |
| "loss": 4.2184, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 7.43175572519084e-05, | |
| "loss": 4.227, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 7.408854961832061e-05, | |
| "loss": 4.1985, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 7.385954198473281e-05, | |
| "loss": 4.1905, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "eval_cer": 0.8758463234136357, | |
| "eval_loss": 4.225706100463867, | |
| "eval_runtime": 40.6017, | |
| "eval_samples_per_second": 11.231, | |
| "eval_steps_per_second": 1.404, | |
| "eval_wer": 0.9903352461491997, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 7.36328244274809e-05, | |
| "loss": 4.1873, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 7.340381679389312e-05, | |
| "loss": 4.1615, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.317480916030534e-05, | |
| "loss": 4.157, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 7.294580152671756e-05, | |
| "loss": 4.1124, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 7.271679389312976e-05, | |
| "loss": 4.0683, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "eval_cer": 0.7911352542906629, | |
| "eval_loss": 3.929443120956421, | |
| "eval_runtime": 41.2715, | |
| "eval_samples_per_second": 11.049, | |
| "eval_steps_per_second": 1.381, | |
| "eval_wer": 0.9936575052854123, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 7.248778625954197e-05, | |
| "loss": 4.0704, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 7.22587786259542e-05, | |
| "loss": 3.9616, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 7.202977099236641e-05, | |
| "loss": 3.7798, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 7.180076335877862e-05, | |
| "loss": 3.6123, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 7.15740458015267e-05, | |
| "loss": 3.486, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_cer": 0.5933711226578492, | |
| "eval_loss": 2.704545497894287, | |
| "eval_runtime": 40.8408, | |
| "eval_samples_per_second": 11.165, | |
| "eval_steps_per_second": 1.396, | |
| "eval_wer": 1.0012080942313502, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 7.134503816793892e-05, | |
| "loss": 3.3283, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 7.111603053435114e-05, | |
| "loss": 3.2091, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 7.088931297709923e-05, | |
| "loss": 3.1158, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 7.066030534351145e-05, | |
| "loss": 2.9983, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 7.043129770992365e-05, | |
| "loss": 2.946, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "eval_cer": 0.4634309557549992, | |
| "eval_loss": 1.9690674543380737, | |
| "eval_runtime": 40.879, | |
| "eval_samples_per_second": 11.155, | |
| "eval_steps_per_second": 1.394, | |
| "eval_wer": 0.942464512231954, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 7.020229007633587e-05, | |
| "loss": 2.8545, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 6.997328244274808e-05, | |
| "loss": 2.8092, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 6.97442748091603e-05, | |
| "loss": 2.7229, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 6.951526717557252e-05, | |
| "loss": 2.7053, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 6.928625954198472e-05, | |
| "loss": 2.634, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "eval_cer": 0.38501810738466385, | |
| "eval_loss": 1.521231770515442, | |
| "eval_runtime": 41.5435, | |
| "eval_samples_per_second": 10.976, | |
| "eval_steps_per_second": 1.372, | |
| "eval_wer": 0.880700694654183, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 6.905725190839693e-05, | |
| "loss": 2.5996, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 6.882824427480916e-05, | |
| "loss": 2.5472, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 6.859923664122137e-05, | |
| "loss": 2.4959, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 6.837022900763359e-05, | |
| "loss": 2.4554, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 6.814122137404579e-05, | |
| "loss": 2.4066, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "eval_cer": 0.36014013541174617, | |
| "eval_loss": 1.2550952434539795, | |
| "eval_runtime": 41.0408, | |
| "eval_samples_per_second": 11.111, | |
| "eval_steps_per_second": 1.389, | |
| "eval_wer": 0.8177287828450619, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 6.791221374045801e-05, | |
| "loss": 2.3768, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 6.768320610687023e-05, | |
| "loss": 2.3557, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 6.745419847328244e-05, | |
| "loss": 2.3109, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 6.722519083969465e-05, | |
| "loss": 2.2953, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 6.699618320610687e-05, | |
| "loss": 2.2651, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "eval_cer": 0.30392851519445757, | |
| "eval_loss": 1.0423332452774048, | |
| "eval_runtime": 40.9098, | |
| "eval_samples_per_second": 11.146, | |
| "eval_steps_per_second": 1.393, | |
| "eval_wer": 0.7650256720024162, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 6.676717557251908e-05, | |
| "loss": 2.2589, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 6.654045801526718e-05, | |
| "loss": 2.2122, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 6.631145038167939e-05, | |
| "loss": 2.2017, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 6.60824427480916e-05, | |
| "loss": 2.1814, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 6.58534351145038e-05, | |
| "loss": 2.1828, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "eval_cer": 0.3106203747441348, | |
| "eval_loss": 0.9598844051361084, | |
| "eval_runtime": 41.4743, | |
| "eval_samples_per_second": 10.995, | |
| "eval_steps_per_second": 1.374, | |
| "eval_wer": 0.7272727272727273, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 6.562442748091603e-05, | |
| "loss": 2.1714, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "learning_rate": 6.539541984732824e-05, | |
| "loss": 2.1422, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 6.516641221374046e-05, | |
| "loss": 2.1546, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 6.493740458015267e-05, | |
| "loss": 2.12, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 6.470839694656488e-05, | |
| "loss": 2.1023, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "eval_cer": 0.30632971185640057, | |
| "eval_loss": 0.9481843113899231, | |
| "eval_runtime": 41.1867, | |
| "eval_samples_per_second": 11.072, | |
| "eval_steps_per_second": 1.384, | |
| "eval_wer": 0.7160978556327393, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 6.44793893129771e-05, | |
| "loss": 2.1104, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 6.425038167938931e-05, | |
| "loss": 2.0879, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 6.402137404580152e-05, | |
| "loss": 2.0724, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 6.379236641221374e-05, | |
| "loss": 2.0622, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 6.356335877862595e-05, | |
| "loss": 2.0536, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "eval_cer": 0.28597858604944104, | |
| "eval_loss": 0.8241907954216003, | |
| "eval_runtime": 41.2837, | |
| "eval_samples_per_second": 11.046, | |
| "eval_steps_per_second": 1.381, | |
| "eval_wer": 0.6766837813349441, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 6.333435114503816e-05, | |
| "loss": 2.0258, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "learning_rate": 6.310534351145038e-05, | |
| "loss": 2.038, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 6.287633587786259e-05, | |
| "loss": 2.0093, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 6.26473282442748e-05, | |
| "loss": 1.9839, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 10.79, | |
| "learning_rate": 6.241832061068702e-05, | |
| "loss": 1.9803, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.79, | |
| "eval_cer": 0.2636592662572823, | |
| "eval_loss": 0.7643126845359802, | |
| "eval_runtime": 41.3574, | |
| "eval_samples_per_second": 11.026, | |
| "eval_steps_per_second": 1.378, | |
| "eval_wer": 0.6562971911809121, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 6.218931297709923e-05, | |
| "loss": 1.9704, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 6.196030534351144e-05, | |
| "loss": 1.9923, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 6.173129770992366e-05, | |
| "loss": 1.9549, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 11.37, | |
| "learning_rate": 6.150229007633587e-05, | |
| "loss": 1.9339, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 6.127328244274808e-05, | |
| "loss": 1.9468, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "eval_cer": 0.25051173043615177, | |
| "eval_loss": 0.7318933606147766, | |
| "eval_runtime": 40.8808, | |
| "eval_samples_per_second": 11.154, | |
| "eval_steps_per_second": 1.394, | |
| "eval_wer": 0.644065237088493, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.65, | |
| "learning_rate": 6.10442748091603e-05, | |
| "loss": 1.9691, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 11.8, | |
| "learning_rate": 6.081526717557252e-05, | |
| "loss": 1.9845, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 6.0586259541984725e-05, | |
| "loss": 1.9561, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "learning_rate": 6.035725190839694e-05, | |
| "loss": 1.9486, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 6.012824427480916e-05, | |
| "loss": 1.9178, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "eval_cer": 0.24893717524799244, | |
| "eval_loss": 0.6936821937561035, | |
| "eval_runtime": 41.1829, | |
| "eval_samples_per_second": 11.073, | |
| "eval_steps_per_second": 1.384, | |
| "eval_wer": 0.6319842947749924, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 5.989923664122137e-05, | |
| "loss": 1.9133, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 12.52, | |
| "learning_rate": 5.9670229007633586e-05, | |
| "loss": 1.9327, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "learning_rate": 5.944122137404579e-05, | |
| "loss": 1.8749, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 5.9212213740458006e-05, | |
| "loss": 1.8775, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 12.95, | |
| "learning_rate": 5.8983206106870226e-05, | |
| "loss": 1.8515, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.95, | |
| "eval_cer": 0.21961108486852465, | |
| "eval_loss": 0.6443303823471069, | |
| "eval_runtime": 40.5279, | |
| "eval_samples_per_second": 11.252, | |
| "eval_steps_per_second": 1.406, | |
| "eval_wer": 0.6052552099063727, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "learning_rate": 5.875419847328244e-05, | |
| "loss": 1.8554, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 5.852519083969465e-05, | |
| "loss": 1.8568, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 5.829618320610686e-05, | |
| "loss": 1.8477, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 5.806717557251908e-05, | |
| "loss": 1.8328, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 5.783816793893129e-05, | |
| "loss": 1.8083, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "eval_cer": 0.21484805542434263, | |
| "eval_loss": 0.6285760402679443, | |
| "eval_runtime": 41.6653, | |
| "eval_samples_per_second": 10.944, | |
| "eval_steps_per_second": 1.368, | |
| "eval_wer": 0.6122017517366355, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 13.81, | |
| "learning_rate": 5.760916030534351e-05, | |
| "loss": 1.8236, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "learning_rate": 5.738015267175571e-05, | |
| "loss": 1.8199, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 5.7151145038167934e-05, | |
| "loss": 1.8285, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "learning_rate": 5.692213740458015e-05, | |
| "loss": 1.817, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "learning_rate": 5.669312977099236e-05, | |
| "loss": 1.819, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "eval_cer": 0.2074476460399937, | |
| "eval_loss": 0.6015097498893738, | |
| "eval_runtime": 41.6458, | |
| "eval_samples_per_second": 10.949, | |
| "eval_steps_per_second": 1.369, | |
| "eval_wer": 0.5986106916339474, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 14.53, | |
| "learning_rate": 5.6466412213740455e-05, | |
| "loss": 1.7952, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 5.623740458015266e-05, | |
| "loss": 1.7955, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 14.82, | |
| "learning_rate": 5.600839694656488e-05, | |
| "loss": 1.7878, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 5.5779389312977095e-05, | |
| "loss": 1.769, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 5.555267175572519e-05, | |
| "loss": 1.7684, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "eval_cer": 0.19815777042985355, | |
| "eval_loss": 0.5682193636894226, | |
| "eval_runtime": 41.2484, | |
| "eval_samples_per_second": 11.055, | |
| "eval_steps_per_second": 1.382, | |
| "eval_wer": 0.5741467834491091, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 5.53236641221374e-05, | |
| "loss": 1.7626, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 15.4, | |
| "learning_rate": 5.5094656488549616e-05, | |
| "loss": 1.7582, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "learning_rate": 5.486564885496182e-05, | |
| "loss": 1.75, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 15.68, | |
| "learning_rate": 5.463664122137404e-05, | |
| "loss": 1.735, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 15.83, | |
| "learning_rate": 5.4407633587786256e-05, | |
| "loss": 1.7195, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 15.83, | |
| "eval_cer": 0.20067705873090852, | |
| "eval_loss": 0.5385124683380127, | |
| "eval_runtime": 41.6481, | |
| "eval_samples_per_second": 10.949, | |
| "eval_steps_per_second": 1.369, | |
| "eval_wer": 0.5591966173361522, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 15.97, | |
| "learning_rate": 5.417862595419847e-05, | |
| "loss": 1.7274, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 16.11, | |
| "learning_rate": 5.3949618320610677e-05, | |
| "loss": 1.7183, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 16.26, | |
| "learning_rate": 5.37206106870229e-05, | |
| "loss": 1.7117, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 5.349160305343511e-05, | |
| "loss": 1.6918, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 5.3262595419847324e-05, | |
| "loss": 1.7044, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "eval_cer": 0.20965202330341678, | |
| "eval_loss": 0.5361923575401306, | |
| "eval_runtime": 41.5242, | |
| "eval_samples_per_second": 10.982, | |
| "eval_steps_per_second": 1.373, | |
| "eval_wer": 0.5524010872848082, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 16.69, | |
| "learning_rate": 5.303358778625954e-05, | |
| "loss": 1.7134, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 5.280458015267176e-05, | |
| "loss": 1.7016, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 5.2575572519083964e-05, | |
| "loss": 1.7069, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 17.12, | |
| "learning_rate": 5.234656488549618e-05, | |
| "loss": 1.7046, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "learning_rate": 5.211755725190839e-05, | |
| "loss": 1.6879, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "eval_cer": 0.20831365139348135, | |
| "eval_loss": 0.5119141936302185, | |
| "eval_runtime": 40.4618, | |
| "eval_samples_per_second": 11.27, | |
| "eval_steps_per_second": 1.409, | |
| "eval_wer": 0.5489278163696768, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 17.41, | |
| "learning_rate": 5.188854961832061e-05, | |
| "loss": 1.681, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 17.55, | |
| "learning_rate": 5.1659541984732825e-05, | |
| "loss": 1.6683, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 5.143053435114503e-05, | |
| "loss": 1.655, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 17.84, | |
| "learning_rate": 5.1201526717557245e-05, | |
| "loss": 1.6604, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 17.98, | |
| "learning_rate": 5.0972519083969465e-05, | |
| "loss": 1.656, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 17.98, | |
| "eval_cer": 0.19678003464021415, | |
| "eval_loss": 0.4990406930446625, | |
| "eval_runtime": 40.5826, | |
| "eval_samples_per_second": 11.236, | |
| "eval_steps_per_second": 1.405, | |
| "eval_wer": 0.5362428269405014, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "learning_rate": 5.074351145038168e-05, | |
| "loss": 1.6645, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 18.27, | |
| "learning_rate": 5.051450381679389e-05, | |
| "loss": 1.6269, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "learning_rate": 5.02854961832061e-05, | |
| "loss": 1.6306, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 5.005877862595419e-05, | |
| "loss": 1.6191, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 18.7, | |
| "learning_rate": 4.9829770992366406e-05, | |
| "loss": 1.6122, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 18.7, | |
| "eval_cer": 0.18997008345142496, | |
| "eval_loss": 0.45614466071128845, | |
| "eval_runtime": 41.2927, | |
| "eval_samples_per_second": 11.043, | |
| "eval_steps_per_second": 1.38, | |
| "eval_wer": 0.5092117185140441, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 18.85, | |
| "learning_rate": 4.9600763358778626e-05, | |
| "loss": 1.622, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 4.937175572519084e-05, | |
| "loss": 1.6305, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "learning_rate": 4.9142748091603046e-05, | |
| "loss": 1.6134, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 19.28, | |
| "learning_rate": 4.891374045801526e-05, | |
| "loss": 1.6044, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 4.868473282442748e-05, | |
| "loss": 1.5919, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "eval_cer": 0.19752794835458984, | |
| "eval_loss": 0.47778981924057007, | |
| "eval_runtime": 41.5758, | |
| "eval_samples_per_second": 10.968, | |
| "eval_steps_per_second": 1.371, | |
| "eval_wer": 0.5225007550588946, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 4.8455725190839694e-05, | |
| "loss": 1.595, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 19.71, | |
| "learning_rate": 4.822671755725191e-05, | |
| "loss": 1.5959, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 4.7997709923664114e-05, | |
| "loss": 1.6006, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 4.7768702290076334e-05, | |
| "loss": 1.5913, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 20.14, | |
| "learning_rate": 4.753969465648855e-05, | |
| "loss": 1.5896, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 20.14, | |
| "eval_cer": 0.18591560384191466, | |
| "eval_loss": 0.4563109278678894, | |
| "eval_runtime": 40.8794, | |
| "eval_samples_per_second": 11.155, | |
| "eval_steps_per_second": 1.394, | |
| "eval_wer": 0.5098157656297191, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 4.731068702290076e-05, | |
| "loss": 1.5823, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 20.43, | |
| "learning_rate": 4.708167938931297e-05, | |
| "loss": 1.5634, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 20.57, | |
| "learning_rate": 4.685267175572519e-05, | |
| "loss": 1.5573, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 20.72, | |
| "learning_rate": 4.66236641221374e-05, | |
| "loss": 1.5689, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 20.86, | |
| "learning_rate": 4.6394656488549615e-05, | |
| "loss": 1.5589, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 20.86, | |
| "eval_cer": 0.17249252086285624, | |
| "eval_loss": 0.43622052669525146, | |
| "eval_runtime": 41.7277, | |
| "eval_samples_per_second": 10.928, | |
| "eval_steps_per_second": 1.366, | |
| "eval_wer": 0.4939595288432498, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 4.616564885496183e-05, | |
| "loss": 1.5697, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "learning_rate": 4.593664122137405e-05, | |
| "loss": 1.5336, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 21.29, | |
| "learning_rate": 4.5707633587786255e-05, | |
| "loss": 1.5425, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 21.44, | |
| "learning_rate": 4.547862595419847e-05, | |
| "loss": 1.5461, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 21.58, | |
| "learning_rate": 4.524961832061068e-05, | |
| "loss": 1.5353, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 21.58, | |
| "eval_cer": 0.15804597701149425, | |
| "eval_loss": 0.41395294666290283, | |
| "eval_runtime": 40.5257, | |
| "eval_samples_per_second": 11.252, | |
| "eval_steps_per_second": 1.407, | |
| "eval_wer": 0.4826336454243431, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 21.73, | |
| "learning_rate": 4.5020610687022895e-05, | |
| "loss": 1.5348, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 21.87, | |
| "learning_rate": 4.4791603053435116e-05, | |
| "loss": 1.5279, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "learning_rate": 4.456259541984732e-05, | |
| "loss": 1.5492, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 22.16, | |
| "learning_rate": 4.4333587786259536e-05, | |
| "loss": 1.5291, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 22.3, | |
| "learning_rate": 4.410458015267175e-05, | |
| "loss": 1.5441, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 22.3, | |
| "eval_cer": 0.15501495827428752, | |
| "eval_loss": 0.40313535928726196, | |
| "eval_runtime": 41.0848, | |
| "eval_samples_per_second": 11.099, | |
| "eval_steps_per_second": 1.387, | |
| "eval_wer": 0.47417698580489276, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 22.45, | |
| "learning_rate": 4.387557251908397e-05, | |
| "loss": 1.518, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 22.59, | |
| "learning_rate": 4.364656488549618e-05, | |
| "loss": 1.5081, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "learning_rate": 4.341755725190839e-05, | |
| "loss": 1.4959, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 22.88, | |
| "learning_rate": 4.31885496183206e-05, | |
| "loss": 1.5097, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "learning_rate": 4.295954198473282e-05, | |
| "loss": 1.5116, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "eval_cer": 0.15450322783813572, | |
| "eval_loss": 0.39162585139274597, | |
| "eval_runtime": 40.7373, | |
| "eval_samples_per_second": 11.194, | |
| "eval_steps_per_second": 1.399, | |
| "eval_wer": 0.4747810329205678, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 23.17, | |
| "learning_rate": 4.273053435114504e-05, | |
| "loss": 1.4951, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 23.31, | |
| "learning_rate": 4.250152671755724e-05, | |
| "loss": 1.4974, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 23.45, | |
| "learning_rate": 4.227480916030534e-05, | |
| "loss": 1.5045, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 23.6, | |
| "learning_rate": 4.204580152671755e-05, | |
| "loss": 1.4944, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 23.74, | |
| "learning_rate": 4.181679389312977e-05, | |
| "loss": 1.4731, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 23.74, | |
| "eval_cer": 0.15422768068020784, | |
| "eval_loss": 0.3840835392475128, | |
| "eval_runtime": 40.8763, | |
| "eval_samples_per_second": 11.156, | |
| "eval_steps_per_second": 1.394, | |
| "eval_wer": 0.4809725158562368, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 23.88, | |
| "learning_rate": 4.1587786259541985e-05, | |
| "loss": 1.472, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 24.03, | |
| "learning_rate": 4.13587786259542e-05, | |
| "loss": 1.4847, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 24.17, | |
| "learning_rate": 4.1129770992366405e-05, | |
| "loss": 1.4603, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 24.32, | |
| "learning_rate": 4.090076335877862e-05, | |
| "loss": 1.4563, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 24.46, | |
| "learning_rate": 4.067175572519084e-05, | |
| "loss": 1.4647, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 24.46, | |
| "eval_cer": 0.14753582113053063, | |
| "eval_loss": 0.37518319487571716, | |
| "eval_runtime": 41.0205, | |
| "eval_samples_per_second": 11.116, | |
| "eval_steps_per_second": 1.39, | |
| "eval_wer": 0.452431289640592, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 24.6, | |
| "learning_rate": 4.044274809160305e-05, | |
| "loss": 1.4585, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 24.75, | |
| "learning_rate": 4.021374045801526e-05, | |
| "loss": 1.4692, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 24.89, | |
| "learning_rate": 3.998473282442747e-05, | |
| "loss": 1.444, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 25.04, | |
| "learning_rate": 3.975572519083969e-05, | |
| "loss": 1.4717, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 25.18, | |
| "learning_rate": 3.9526717557251906e-05, | |
| "loss": 1.4328, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 25.18, | |
| "eval_cer": 0.1461187214611872, | |
| "eval_loss": 0.35870596766471863, | |
| "eval_runtime": 40.6723, | |
| "eval_samples_per_second": 11.212, | |
| "eval_steps_per_second": 1.401, | |
| "eval_wer": 0.4475989127151918, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 25.32, | |
| "learning_rate": 3.929770992366412e-05, | |
| "loss": 1.4329, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "learning_rate": 3.9068702290076326e-05, | |
| "loss": 1.4209, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 25.61, | |
| "learning_rate": 3.884198473282442e-05, | |
| "loss": 1.4188, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 25.75, | |
| "learning_rate": 3.861297709923664e-05, | |
| "loss": 1.4301, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 25.9, | |
| "learning_rate": 3.8383969465648854e-05, | |
| "loss": 1.4129, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 25.9, | |
| "eval_cer": 0.13663202645252717, | |
| "eval_loss": 0.3428773581981659, | |
| "eval_runtime": 42.0192, | |
| "eval_samples_per_second": 10.852, | |
| "eval_steps_per_second": 1.357, | |
| "eval_wer": 0.42419208698278466, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 26.04, | |
| "learning_rate": 3.815496183206107e-05, | |
| "loss": 1.4266, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 26.19, | |
| "learning_rate": 3.7925954198473274e-05, | |
| "loss": 1.4166, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 26.33, | |
| "learning_rate": 3.7696946564885494e-05, | |
| "loss": 1.4157, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 26.47, | |
| "learning_rate": 3.746793893129771e-05, | |
| "loss": 1.4285, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 26.62, | |
| "learning_rate": 3.723893129770992e-05, | |
| "loss": 1.4062, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 26.62, | |
| "eval_cer": 0.13549047394111163, | |
| "eval_loss": 0.34499478340148926, | |
| "eval_runtime": 41.0336, | |
| "eval_samples_per_second": 11.113, | |
| "eval_steps_per_second": 1.389, | |
| "eval_wer": 0.4250981576562972, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 26.76, | |
| "learning_rate": 3.7009923664122134e-05, | |
| "loss": 1.4163, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 26.91, | |
| "learning_rate": 3.678091603053435e-05, | |
| "loss": 1.404, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 27.05, | |
| "learning_rate": 3.655190839694656e-05, | |
| "loss": 1.4134, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 27.19, | |
| "learning_rate": 3.6322900763358775e-05, | |
| "loss": 1.4001, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 27.34, | |
| "learning_rate": 3.609389312977099e-05, | |
| "loss": 1.3928, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 27.34, | |
| "eval_cer": 0.13218390804597702, | |
| "eval_loss": 0.32969579100608826, | |
| "eval_runtime": 41.0801, | |
| "eval_samples_per_second": 11.1, | |
| "eval_steps_per_second": 1.388, | |
| "eval_wer": 0.4145273331319843, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 27.48, | |
| "learning_rate": 3.58648854961832e-05, | |
| "loss": 1.3979, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 27.63, | |
| "learning_rate": 3.5635877862595415e-05, | |
| "loss": 1.3971, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 27.77, | |
| "learning_rate": 3.540687022900763e-05, | |
| "loss": 1.3934, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "learning_rate": 3.517786259541984e-05, | |
| "loss": 1.3866, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 28.06, | |
| "learning_rate": 3.4948854961832055e-05, | |
| "loss": 1.3906, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 28.06, | |
| "eval_cer": 0.1336403715950244, | |
| "eval_loss": 0.32101842761039734, | |
| "eval_runtime": 41.0367, | |
| "eval_samples_per_second": 11.112, | |
| "eval_steps_per_second": 1.389, | |
| "eval_wer": 0.4184536393838719, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 28.2, | |
| "learning_rate": 3.471984732824427e-05, | |
| "loss": 1.3689, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 28.34, | |
| "learning_rate": 3.449083969465649e-05, | |
| "loss": 1.3715, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 28.49, | |
| "learning_rate": 3.4261832061068696e-05, | |
| "loss": 1.3527, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 28.63, | |
| "learning_rate": 3.4032824427480916e-05, | |
| "loss": 1.3532, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 28.78, | |
| "learning_rate": 3.380381679389312e-05, | |
| "loss": 1.358, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 28.78, | |
| "eval_cer": 0.12753897024090693, | |
| "eval_loss": 0.31306591629981995, | |
| "eval_runtime": 41.2359, | |
| "eval_samples_per_second": 11.058, | |
| "eval_steps_per_second": 1.382, | |
| "eval_wer": 0.39700996677740863, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "learning_rate": 3.357480916030534e-05, | |
| "loss": 1.3582, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 29.06, | |
| "learning_rate": 3.334580152671755e-05, | |
| "loss": 1.3587, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 29.21, | |
| "learning_rate": 3.311679389312977e-05, | |
| "loss": 1.3392, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 29.35, | |
| "learning_rate": 3.288778625954198e-05, | |
| "loss": 1.3486, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 29.5, | |
| "learning_rate": 3.26587786259542e-05, | |
| "loss": 1.3445, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 29.5, | |
| "eval_cer": 0.12761769800031492, | |
| "eval_loss": 0.3069218099117279, | |
| "eval_runtime": 41.0687, | |
| "eval_samples_per_second": 11.103, | |
| "eval_steps_per_second": 1.388, | |
| "eval_wer": 0.3920265780730897, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 29.64, | |
| "learning_rate": 3.242977099236641e-05, | |
| "loss": 1.3354, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 29.78, | |
| "learning_rate": 3.2200763358778624e-05, | |
| "loss": 1.3334, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 29.93, | |
| "learning_rate": 3.197175572519084e-05, | |
| "loss": 1.3305, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 30.07, | |
| "learning_rate": 3.174274809160305e-05, | |
| "loss": 1.3354, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 30.22, | |
| "learning_rate": 3.1513740458015264e-05, | |
| "loss": 1.3159, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 30.22, | |
| "eval_cer": 0.1254920484962998, | |
| "eval_loss": 0.30346596240997314, | |
| "eval_runtime": 41.0784, | |
| "eval_samples_per_second": 11.101, | |
| "eval_steps_per_second": 1.388, | |
| "eval_wer": 0.3961038961038961, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 30.36, | |
| "learning_rate": 3.128473282442748e-05, | |
| "loss": 1.3376, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 30.5, | |
| "learning_rate": 3.105572519083969e-05, | |
| "loss": 1.324, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 30.65, | |
| "learning_rate": 3.0826717557251904e-05, | |
| "loss": 1.3091, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 30.79, | |
| "learning_rate": 3.059770992366412e-05, | |
| "loss": 1.3213, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 30.93, | |
| "learning_rate": 3.0368702290076335e-05, | |
| "loss": 1.3044, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 30.93, | |
| "eval_cer": 0.12423240434577232, | |
| "eval_loss": 0.29519879817962646, | |
| "eval_runtime": 41.1753, | |
| "eval_samples_per_second": 11.075, | |
| "eval_steps_per_second": 1.384, | |
| "eval_wer": 0.3853820598006645, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 31.08, | |
| "learning_rate": 3.0139694656488545e-05, | |
| "loss": 1.3033, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 31.22, | |
| "learning_rate": 2.991068702290076e-05, | |
| "loss": 1.2995, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 31.37, | |
| "learning_rate": 2.9681679389312975e-05, | |
| "loss": 1.3101, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 31.51, | |
| "learning_rate": 2.945267175572519e-05, | |
| "loss": 1.304, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 31.65, | |
| "learning_rate": 2.9223664122137402e-05, | |
| "loss": 1.3034, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 31.65, | |
| "eval_cer": 0.12273657691702095, | |
| "eval_loss": 0.29660850763320923, | |
| "eval_runtime": 41.8912, | |
| "eval_samples_per_second": 10.885, | |
| "eval_steps_per_second": 1.361, | |
| "eval_wer": 0.37722742373905166, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 31.8, | |
| "learning_rate": 2.8994656488549615e-05, | |
| "loss": 1.2912, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 31.94, | |
| "learning_rate": 2.876564885496183e-05, | |
| "loss": 1.299, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 32.09, | |
| "learning_rate": 2.8536641221374046e-05, | |
| "loss": 1.3042, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 32.23, | |
| "learning_rate": 2.8307633587786256e-05, | |
| "loss": 1.294, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 32.37, | |
| "learning_rate": 2.8078625954198472e-05, | |
| "loss": 1.2963, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 32.37, | |
| "eval_cer": 0.12080774681152574, | |
| "eval_loss": 0.2843906879425049, | |
| "eval_runtime": 41.7644, | |
| "eval_samples_per_second": 10.918, | |
| "eval_steps_per_second": 1.365, | |
| "eval_wer": 0.3705829054666264, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 32.52, | |
| "learning_rate": 2.7849618320610682e-05, | |
| "loss": 1.2769, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 32.66, | |
| "learning_rate": 2.76206106870229e-05, | |
| "loss": 1.2812, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "learning_rate": 2.7391603053435113e-05, | |
| "loss": 1.2827, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 32.95, | |
| "learning_rate": 2.7162595419847326e-05, | |
| "loss": 1.2747, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 33.09, | |
| "learning_rate": 2.6935877862595417e-05, | |
| "loss": 1.2765, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 33.09, | |
| "eval_cer": 0.11726499763816722, | |
| "eval_loss": 0.28407707810401917, | |
| "eval_runtime": 40.9894, | |
| "eval_samples_per_second": 11.125, | |
| "eval_steps_per_second": 1.391, | |
| "eval_wer": 0.35668982180610087, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 33.24, | |
| "learning_rate": 2.670687022900763e-05, | |
| "loss": 1.2785, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 33.38, | |
| "learning_rate": 2.6477862595419844e-05, | |
| "loss": 1.2644, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 33.52, | |
| "learning_rate": 2.624885496183206e-05, | |
| "loss": 1.2724, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 33.67, | |
| "learning_rate": 2.601984732824427e-05, | |
| "loss": 1.2551, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 33.81, | |
| "learning_rate": 2.5790839694656488e-05, | |
| "loss": 1.2438, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 33.81, | |
| "eval_cer": 0.11372224846480869, | |
| "eval_loss": 0.2734295129776001, | |
| "eval_runtime": 41.7199, | |
| "eval_samples_per_second": 10.93, | |
| "eval_steps_per_second": 1.366, | |
| "eval_wer": 0.35517970401691334, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "learning_rate": 2.5561832061068698e-05, | |
| "loss": 1.2491, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 34.1, | |
| "learning_rate": 2.5332824427480915e-05, | |
| "loss": 1.252, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 34.24, | |
| "learning_rate": 2.5103816793893128e-05, | |
| "loss": 1.2467, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 34.39, | |
| "learning_rate": 2.487480916030534e-05, | |
| "loss": 1.2406, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 34.53, | |
| "learning_rate": 2.4645801526717555e-05, | |
| "loss": 1.2487, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 34.53, | |
| "eval_cer": 0.11179341835931349, | |
| "eval_loss": 0.2702818512916565, | |
| "eval_runtime": 41.8515, | |
| "eval_samples_per_second": 10.896, | |
| "eval_steps_per_second": 1.362, | |
| "eval_wer": 0.3501963153125944, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 34.68, | |
| "learning_rate": 2.441679389312977e-05, | |
| "loss": 1.2504, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 34.82, | |
| "learning_rate": 2.4187786259541982e-05, | |
| "loss": 1.2341, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 34.96, | |
| "learning_rate": 2.39587786259542e-05, | |
| "loss": 1.2477, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 35.11, | |
| "learning_rate": 2.372977099236641e-05, | |
| "loss": 1.2427, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 35.25, | |
| "learning_rate": 2.3500763358778626e-05, | |
| "loss": 1.2249, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 35.25, | |
| "eval_cer": 0.11423397890096047, | |
| "eval_loss": 0.2650163471698761, | |
| "eval_runtime": 41.2103, | |
| "eval_samples_per_second": 11.065, | |
| "eval_steps_per_second": 1.383, | |
| "eval_wer": 0.3483841739655693, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 35.4, | |
| "learning_rate": 2.3271755725190836e-05, | |
| "loss": 1.2265, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 35.54, | |
| "learning_rate": 2.3042748091603052e-05, | |
| "loss": 1.2276, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 35.68, | |
| "learning_rate": 2.2816030534351143e-05, | |
| "loss": 1.2332, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 35.83, | |
| "learning_rate": 2.2587022900763357e-05, | |
| "loss": 1.2249, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 35.97, | |
| "learning_rate": 2.235801526717557e-05, | |
| "loss": 1.2229, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 35.97, | |
| "eval_cer": 0.10970713273500236, | |
| "eval_loss": 0.25843024253845215, | |
| "eval_runtime": 42.815, | |
| "eval_samples_per_second": 10.65, | |
| "eval_steps_per_second": 1.331, | |
| "eval_wer": 0.3373603141045001, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 36.11, | |
| "learning_rate": 2.2129007633587784e-05, | |
| "loss": 1.2412, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 36.26, | |
| "learning_rate": 2.1899999999999997e-05, | |
| "loss": 1.212, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 36.4, | |
| "learning_rate": 2.1670992366412214e-05, | |
| "loss": 1.2151, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 36.55, | |
| "learning_rate": 2.1441984732824424e-05, | |
| "loss": 1.2303, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 36.69, | |
| "learning_rate": 2.121297709923664e-05, | |
| "loss": 1.2374, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 36.69, | |
| "eval_cer": 0.10951031333648244, | |
| "eval_loss": 0.2568279504776001, | |
| "eval_runtime": 41.6839, | |
| "eval_samples_per_second": 10.939, | |
| "eval_steps_per_second": 1.367, | |
| "eval_wer": 0.33373603141045, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 36.83, | |
| "learning_rate": 2.098396946564885e-05, | |
| "loss": 1.2152, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 36.98, | |
| "learning_rate": 2.0754961832061068e-05, | |
| "loss": 1.2089, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 37.12, | |
| "learning_rate": 2.052595419847328e-05, | |
| "loss": 1.2201, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 37.27, | |
| "learning_rate": 2.0296946564885495e-05, | |
| "loss": 1.2006, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 37.41, | |
| "learning_rate": 2.0067938931297708e-05, | |
| "loss": 1.2153, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 37.41, | |
| "eval_cer": 0.10710911667453944, | |
| "eval_loss": 0.24941784143447876, | |
| "eval_runtime": 41.3494, | |
| "eval_samples_per_second": 11.028, | |
| "eval_steps_per_second": 1.378, | |
| "eval_wer": 0.33267894895801875, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 37.55, | |
| "learning_rate": 1.983893129770992e-05, | |
| "loss": 1.2071, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 37.7, | |
| "learning_rate": 1.9609923664122135e-05, | |
| "loss": 1.2042, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 37.84, | |
| "learning_rate": 1.9380916030534352e-05, | |
| "loss": 1.2037, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 37.98, | |
| "learning_rate": 1.9151908396946562e-05, | |
| "loss": 1.1962, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 38.13, | |
| "learning_rate": 1.892290076335878e-05, | |
| "loss": 1.1925, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 38.13, | |
| "eval_cer": 0.1076995748700992, | |
| "eval_loss": 0.2518324553966522, | |
| "eval_runtime": 40.748, | |
| "eval_samples_per_second": 11.191, | |
| "eval_steps_per_second": 1.399, | |
| "eval_wer": 0.33660525520990636, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 38.27, | |
| "learning_rate": 1.869389312977099e-05, | |
| "loss": 1.1969, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 38.42, | |
| "learning_rate": 1.8464885496183202e-05, | |
| "loss": 1.1947, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 38.56, | |
| "learning_rate": 1.823587786259542e-05, | |
| "loss": 1.2005, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 38.7, | |
| "learning_rate": 1.8006870229007632e-05, | |
| "loss": 1.1961, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 38.85, | |
| "learning_rate": 1.7777862595419846e-05, | |
| "loss": 1.1908, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 38.85, | |
| "eval_cer": 0.10565265312549205, | |
| "eval_loss": 0.24367305636405945, | |
| "eval_runtime": 41.2308, | |
| "eval_samples_per_second": 11.06, | |
| "eval_steps_per_second": 1.382, | |
| "eval_wer": 0.3272425249169435, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 38.99, | |
| "learning_rate": 1.754885496183206e-05, | |
| "loss": 1.1762, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 39.14, | |
| "learning_rate": 1.7319847328244273e-05, | |
| "loss": 1.2018, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 39.28, | |
| "learning_rate": 1.7090839694656486e-05, | |
| "loss": 1.1822, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 39.42, | |
| "learning_rate": 1.68618320610687e-05, | |
| "loss": 1.1745, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 39.57, | |
| "learning_rate": 1.6632824427480913e-05, | |
| "loss": 1.1858, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 39.57, | |
| "eval_cer": 0.10443237285466855, | |
| "eval_loss": 0.23960824310779572, | |
| "eval_runtime": 42.517, | |
| "eval_samples_per_second": 10.725, | |
| "eval_steps_per_second": 1.341, | |
| "eval_wer": 0.32648746602234974, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 39.71, | |
| "learning_rate": 1.6403816793893127e-05, | |
| "loss": 1.1866, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 39.86, | |
| "learning_rate": 1.617480916030534e-05, | |
| "loss": 1.1878, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 1.5945801526717557e-05, | |
| "loss": 1.1817, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 40.14, | |
| "learning_rate": 1.571679389312977e-05, | |
| "loss": 1.1851, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 40.29, | |
| "learning_rate": 1.5487786259541984e-05, | |
| "loss": 1.1808, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 40.29, | |
| "eval_cer": 0.10277908990710125, | |
| "eval_loss": 0.2373325228691101, | |
| "eval_runtime": 41.3513, | |
| "eval_samples_per_second": 11.027, | |
| "eval_steps_per_second": 1.378, | |
| "eval_wer": 0.31561461794019935, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 40.43, | |
| "learning_rate": 1.5258778625954197e-05, | |
| "loss": 1.1558, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "learning_rate": 1.502977099236641e-05, | |
| "loss": 1.1804, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 40.72, | |
| "learning_rate": 1.4800763358778624e-05, | |
| "loss": 1.1736, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 40.86, | |
| "learning_rate": 1.4571755725190838e-05, | |
| "loss": 1.1782, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 41.01, | |
| "learning_rate": 1.4342748091603053e-05, | |
| "loss": 1.1842, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 41.01, | |
| "eval_cer": 0.10258227050858132, | |
| "eval_loss": 0.23562349379062653, | |
| "eval_runtime": 40.492, | |
| "eval_samples_per_second": 11.261, | |
| "eval_steps_per_second": 1.408, | |
| "eval_wer": 0.31516158260344307, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 41.15, | |
| "learning_rate": 1.4113740458015266e-05, | |
| "loss": 1.1595, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 41.29, | |
| "learning_rate": 1.388473282442748e-05, | |
| "loss": 1.1527, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 41.44, | |
| "learning_rate": 1.3655725190839693e-05, | |
| "loss": 1.1517, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 41.58, | |
| "learning_rate": 1.3426717557251907e-05, | |
| "loss": 1.1609, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 41.73, | |
| "learning_rate": 1.3197709923664122e-05, | |
| "loss": 1.1668, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 41.73, | |
| "eval_cer": 0.10246417886946937, | |
| "eval_loss": 0.23187227547168732, | |
| "eval_runtime": 40.5813, | |
| "eval_samples_per_second": 11.237, | |
| "eval_steps_per_second": 1.405, | |
| "eval_wer": 0.3187858652974932, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 41.87, | |
| "learning_rate": 1.2968702290076335e-05, | |
| "loss": 1.1536, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 42.01, | |
| "learning_rate": 1.2739694656488549e-05, | |
| "loss": 1.1649, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 42.16, | |
| "learning_rate": 1.2510687022900762e-05, | |
| "loss": 1.1459, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 42.3, | |
| "learning_rate": 1.2281679389312975e-05, | |
| "loss": 1.1495, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "learning_rate": 1.205267175572519e-05, | |
| "loss": 1.1448, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "eval_cer": 0.09947252401196661, | |
| "eval_loss": 0.2292834371328354, | |
| "eval_runtime": 41.732, | |
| "eval_samples_per_second": 10.927, | |
| "eval_steps_per_second": 1.366, | |
| "eval_wer": 0.3098761703412866, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 42.59, | |
| "learning_rate": 1.1823664122137404e-05, | |
| "loss": 1.1408, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 42.73, | |
| "learning_rate": 1.1594656488549618e-05, | |
| "loss": 1.1458, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 42.88, | |
| "learning_rate": 1.1365648854961831e-05, | |
| "loss": 1.1358, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "learning_rate": 1.1136641221374044e-05, | |
| "loss": 1.1519, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 43.17, | |
| "learning_rate": 1.0909923664122137e-05, | |
| "loss": 1.1327, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 43.17, | |
| "eval_cer": 0.09793733270351125, | |
| "eval_loss": 0.2265164852142334, | |
| "eval_runtime": 40.9338, | |
| "eval_samples_per_second": 11.14, | |
| "eval_steps_per_second": 1.392, | |
| "eval_wer": 0.3047417698580489, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 43.31, | |
| "learning_rate": 1.068091603053435e-05, | |
| "loss": 1.1322, | |
| "step": 30100 | |
| }, | |
| { | |
| "epoch": 43.45, | |
| "learning_rate": 1.0451908396946564e-05, | |
| "loss": 1.1392, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 43.6, | |
| "learning_rate": 1.0222900763358777e-05, | |
| "loss": 1.1318, | |
| "step": 30300 | |
| }, | |
| { | |
| "epoch": 43.74, | |
| "learning_rate": 9.99389312977099e-06, | |
| "loss": 1.1321, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 43.88, | |
| "learning_rate": 9.764885496183206e-06, | |
| "loss": 1.1307, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 43.88, | |
| "eval_cer": 0.09888206581640686, | |
| "eval_loss": 0.22221311926841736, | |
| "eval_runtime": 40.9097, | |
| "eval_samples_per_second": 11.147, | |
| "eval_steps_per_second": 1.393, | |
| "eval_wer": 0.30776200543642407, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 44.03, | |
| "learning_rate": 9.53587786259542e-06, | |
| "loss": 1.1358, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 44.17, | |
| "learning_rate": 9.306870229007633e-06, | |
| "loss": 1.1342, | |
| "step": 30700 | |
| }, | |
| { | |
| "epoch": 44.32, | |
| "learning_rate": 9.077862595419846e-06, | |
| "loss": 1.1348, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 44.46, | |
| "learning_rate": 8.84885496183206e-06, | |
| "loss": 1.1294, | |
| "step": 30900 | |
| }, | |
| { | |
| "epoch": 44.6, | |
| "learning_rate": 8.619847328244275e-06, | |
| "loss": 1.1419, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 44.6, | |
| "eval_cer": 0.09813415210203118, | |
| "eval_loss": 0.22149430215358734, | |
| "eval_runtime": 40.8027, | |
| "eval_samples_per_second": 11.176, | |
| "eval_steps_per_second": 1.397, | |
| "eval_wer": 0.3038356991845364, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 44.75, | |
| "learning_rate": 8.390839694656488e-06, | |
| "loss": 1.1191, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 44.89, | |
| "learning_rate": 8.161832061068702e-06, | |
| "loss": 1.1223, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 45.04, | |
| "learning_rate": 7.932824427480915e-06, | |
| "loss": 1.1393, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 45.18, | |
| "learning_rate": 7.703816793893129e-06, | |
| "loss": 1.1172, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 45.32, | |
| "learning_rate": 7.474809160305343e-06, | |
| "loss": 1.1231, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 45.32, | |
| "eval_cer": 0.09722878286883956, | |
| "eval_loss": 0.2193477302789688, | |
| "eval_runtime": 40.6396, | |
| "eval_samples_per_second": 11.221, | |
| "eval_steps_per_second": 1.403, | |
| "eval_wer": 0.3012684989429176, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 45.47, | |
| "learning_rate": 7.245801526717557e-06, | |
| "loss": 1.1289, | |
| "step": 31600 | |
| }, | |
| { | |
| "epoch": 45.61, | |
| "learning_rate": 7.016793893129771e-06, | |
| "loss": 1.1083, | |
| "step": 31700 | |
| }, | |
| { | |
| "epoch": 45.75, | |
| "learning_rate": 6.787786259541984e-06, | |
| "loss": 1.109, | |
| "step": 31800 | |
| }, | |
| { | |
| "epoch": 45.9, | |
| "learning_rate": 6.558778625954198e-06, | |
| "loss": 1.1218, | |
| "step": 31900 | |
| }, | |
| { | |
| "epoch": 46.04, | |
| "learning_rate": 6.329770992366412e-06, | |
| "loss": 1.139, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 46.04, | |
| "eval_cer": 0.09683514407179972, | |
| "eval_loss": 0.2162453532218933, | |
| "eval_runtime": 41.1951, | |
| "eval_samples_per_second": 11.069, | |
| "eval_steps_per_second": 1.384, | |
| "eval_wer": 0.30066445182724255, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 46.19, | |
| "learning_rate": 6.100763358778626e-06, | |
| "loss": 1.1155, | |
| "step": 32100 | |
| }, | |
| { | |
| "epoch": 46.33, | |
| "learning_rate": 5.8717557251908395e-06, | |
| "loss": 1.1212, | |
| "step": 32200 | |
| }, | |
| { | |
| "epoch": 46.47, | |
| "learning_rate": 5.642748091603053e-06, | |
| "loss": 1.1149, | |
| "step": 32300 | |
| }, | |
| { | |
| "epoch": 46.62, | |
| "learning_rate": 5.413740458015267e-06, | |
| "loss": 1.1183, | |
| "step": 32400 | |
| }, | |
| { | |
| "epoch": 46.76, | |
| "learning_rate": 5.184732824427481e-06, | |
| "loss": 1.1114, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 46.76, | |
| "eval_cer": 0.09596913871831207, | |
| "eval_loss": 0.2121613770723343, | |
| "eval_runtime": 40.6982, | |
| "eval_samples_per_second": 11.204, | |
| "eval_steps_per_second": 1.401, | |
| "eval_wer": 0.2982482633645424, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 46.91, | |
| "learning_rate": 4.955725190839695e-06, | |
| "loss": 1.1091, | |
| "step": 32600 | |
| }, | |
| { | |
| "epoch": 47.05, | |
| "learning_rate": 4.726717557251908e-06, | |
| "loss": 1.1148, | |
| "step": 32700 | |
| }, | |
| { | |
| "epoch": 47.19, | |
| "learning_rate": 4.497709923664122e-06, | |
| "loss": 1.0962, | |
| "step": 32800 | |
| }, | |
| { | |
| "epoch": 47.34, | |
| "learning_rate": 4.268702290076335e-06, | |
| "loss": 1.0984, | |
| "step": 32900 | |
| }, | |
| { | |
| "epoch": 47.48, | |
| "learning_rate": 4.03969465648855e-06, | |
| "loss": 1.111, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 47.48, | |
| "eval_cer": 0.09482758620689655, | |
| "eval_loss": 0.21248506009578705, | |
| "eval_runtime": 40.6368, | |
| "eval_samples_per_second": 11.221, | |
| "eval_steps_per_second": 1.403, | |
| "eval_wer": 0.2946239806704923, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 47.63, | |
| "learning_rate": 3.810687022900763e-06, | |
| "loss": 1.1031, | |
| "step": 33100 | |
| }, | |
| { | |
| "epoch": 47.77, | |
| "learning_rate": 3.581679389312977e-06, | |
| "loss": 1.1159, | |
| "step": 33200 | |
| }, | |
| { | |
| "epoch": 47.91, | |
| "learning_rate": 3.352671755725191e-06, | |
| "loss": 1.0905, | |
| "step": 33300 | |
| }, | |
| { | |
| "epoch": 48.06, | |
| "learning_rate": 3.1236641221374048e-06, | |
| "loss": 1.1087, | |
| "step": 33400 | |
| }, | |
| { | |
| "epoch": 48.2, | |
| "learning_rate": 2.894656488549618e-06, | |
| "loss": 1.0982, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 48.2, | |
| "eval_cer": 0.09533931664304834, | |
| "eval_loss": 0.2098563313484192, | |
| "eval_runtime": 40.5946, | |
| "eval_samples_per_second": 11.233, | |
| "eval_steps_per_second": 1.404, | |
| "eval_wer": 0.2956810631229236, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 48.34, | |
| "learning_rate": 2.66793893129771e-06, | |
| "loss": 1.0947, | |
| "step": 33600 | |
| }, | |
| { | |
| "epoch": 48.49, | |
| "learning_rate": 2.4389312977099237e-06, | |
| "loss": 1.1102, | |
| "step": 33700 | |
| }, | |
| { | |
| "epoch": 48.63, | |
| "learning_rate": 2.209923664122137e-06, | |
| "loss": 1.0891, | |
| "step": 33800 | |
| }, | |
| { | |
| "epoch": 48.78, | |
| "learning_rate": 1.980916030534351e-06, | |
| "loss": 1.0937, | |
| "step": 33900 | |
| }, | |
| { | |
| "epoch": 48.92, | |
| "learning_rate": 1.7519083969465647e-06, | |
| "loss": 1.109, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 48.92, | |
| "eval_cer": 0.09545740828216029, | |
| "eval_loss": 0.20918497443199158, | |
| "eval_runtime": 40.877, | |
| "eval_samples_per_second": 11.155, | |
| "eval_steps_per_second": 1.394, | |
| "eval_wer": 0.29553005134400484, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "learning_rate": 1.5229007633587786e-06, | |
| "loss": 1.097, | |
| "step": 34100 | |
| }, | |
| { | |
| "epoch": 49.21, | |
| "learning_rate": 1.2938931297709922e-06, | |
| "loss": 1.0909, | |
| "step": 34200 | |
| }, | |
| { | |
| "epoch": 49.35, | |
| "learning_rate": 1.0648854961832059e-06, | |
| "loss": 1.1008, | |
| "step": 34300 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "learning_rate": 8.358778625954198e-07, | |
| "loss": 1.0904, | |
| "step": 34400 | |
| }, | |
| { | |
| "epoch": 49.64, | |
| "learning_rate": 6.068702290076335e-07, | |
| "loss": 1.0905, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 49.64, | |
| "eval_cer": 0.09526058888364038, | |
| "eval_loss": 0.20883652567863464, | |
| "eval_runtime": 40.602, | |
| "eval_samples_per_second": 11.231, | |
| "eval_steps_per_second": 1.404, | |
| "eval_wer": 0.2953790395650861, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 49.78, | |
| "learning_rate": 3.778625954198473e-07, | |
| "loss": 1.0961, | |
| "step": 34600 | |
| }, | |
| { | |
| "epoch": 49.93, | |
| "learning_rate": 1.4885496183206107e-07, | |
| "loss": 1.095, | |
| "step": 34700 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 34750, | |
| "total_flos": 2.8392187465644065e+20, | |
| "train_loss": 2.2316733406121783, | |
| "train_runtime": 114311.9751, | |
| "train_samples_per_second": 9.737, | |
| "train_steps_per_second": 0.304 | |
| } | |
| ], | |
| "max_steps": 34750, | |
| "num_train_epochs": 50, | |
| "total_flos": 2.8392187465644065e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |