| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 23250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.8500000000000002e-06, | |
| "loss": 7.8132, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 5.82e-06, | |
| "loss": 3.053, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 8.82e-06, | |
| "loss": 2.9223, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.182e-05, | |
| "loss": 2.6894, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.482e-05, | |
| "loss": 1.6894, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 1.2422521114349365, | |
| "eval_runtime": 81.2578, | |
| "eval_samples_per_second": 10.177, | |
| "eval_steps_per_second": 1.28, | |
| "eval_wer": 0.8619092627599244, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.782e-05, | |
| "loss": 1.3281, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.082e-05, | |
| "loss": 1.1146, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.3820000000000002e-05, | |
| "loss": 0.9295, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 2.682e-05, | |
| "loss": 0.8392, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.982e-05, | |
| "loss": 0.7543, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "eval_loss": 0.5956054329872131, | |
| "eval_runtime": 81.5082, | |
| "eval_samples_per_second": 10.146, | |
| "eval_steps_per_second": 1.276, | |
| "eval_wer": 0.38166351606805293, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 3.282e-05, | |
| "loss": 0.7128, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.582e-05, | |
| "loss": 0.6859, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.8820000000000004e-05, | |
| "loss": 0.6024, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.1819999999999996e-05, | |
| "loss": 0.6365, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 4.482e-05, | |
| "loss": 0.5481, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "eval_loss": 0.5042839646339417, | |
| "eval_runtime": 83.6557, | |
| "eval_samples_per_second": 9.886, | |
| "eval_steps_per_second": 1.243, | |
| "eval_wer": 0.32457466918714556, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 4.7820000000000006e-05, | |
| "loss": 0.5648, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 5.082e-05, | |
| "loss": 0.555, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 5.382e-05, | |
| "loss": 0.5723, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 5.682e-05, | |
| "loss": 0.4716, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 5.982e-05, | |
| "loss": 0.4661, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "eval_loss": 0.48127105832099915, | |
| "eval_runtime": 81.5918, | |
| "eval_samples_per_second": 10.136, | |
| "eval_steps_per_second": 1.275, | |
| "eval_wer": 0.2793005671077505, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 5.973458823529412e-05, | |
| "loss": 0.4756, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 5.945223529411765e-05, | |
| "loss": 0.4022, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 5.9169882352941175e-05, | |
| "loss": 0.4314, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 5.8887529411764706e-05, | |
| "loss": 0.4154, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 5.8605176470588236e-05, | |
| "loss": 0.3901, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "eval_loss": 0.4370500147342682, | |
| "eval_runtime": 81.2387, | |
| "eval_samples_per_second": 10.18, | |
| "eval_steps_per_second": 1.28, | |
| "eval_wer": 0.25916824196597354, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 5.8322823529411767e-05, | |
| "loss": 0.4014, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 5.80404705882353e-05, | |
| "loss": 0.3762, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 5.775811764705882e-05, | |
| "loss": 0.3777, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 5.747576470588236e-05, | |
| "loss": 0.3342, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 5.719341176470588e-05, | |
| "loss": 0.3512, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_loss": 0.4216073155403137, | |
| "eval_runtime": 81.8015, | |
| "eval_samples_per_second": 10.11, | |
| "eval_steps_per_second": 1.271, | |
| "eval_wer": 0.24584120982986768, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 5.691105882352942e-05, | |
| "loss": 0.3204, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 5.662870588235294e-05, | |
| "loss": 0.3232, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 5.634635294117647e-05, | |
| "loss": 0.2976, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 5.6064000000000004e-05, | |
| "loss": 0.296, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 5.578164705882353e-05, | |
| "loss": 0.3016, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "eval_loss": 0.3813554048538208, | |
| "eval_runtime": 83.1204, | |
| "eval_samples_per_second": 9.949, | |
| "eval_steps_per_second": 1.251, | |
| "eval_wer": 0.22570888468809075, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 5.549929411764706e-05, | |
| "loss": 0.2726, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 5.521694117647059e-05, | |
| "loss": 0.2953, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 5.493458823529412e-05, | |
| "loss": 0.2647, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 5.465223529411765e-05, | |
| "loss": 0.2524, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 5.436988235294118e-05, | |
| "loss": 0.278, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "eval_loss": 0.41506877541542053, | |
| "eval_runtime": 81.9008, | |
| "eval_samples_per_second": 10.098, | |
| "eval_steps_per_second": 1.27, | |
| "eval_wer": 0.21446124763705104, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 5.4087529411764704e-05, | |
| "loss": 0.2559, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 5.380517647058824e-05, | |
| "loss": 0.2669, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 5.3522823529411765e-05, | |
| "loss": 0.2189, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 5.3240470588235296e-05, | |
| "loss": 0.2321, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 5.2958117647058826e-05, | |
| "loss": 0.2435, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "eval_loss": 0.48163077235221863, | |
| "eval_runtime": 80.9004, | |
| "eval_samples_per_second": 10.222, | |
| "eval_steps_per_second": 1.286, | |
| "eval_wer": 0.21304347826086956, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 5.267576470588235e-05, | |
| "loss": 0.2348, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 5.239341176470589e-05, | |
| "loss": 0.2266, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 5.211105882352941e-05, | |
| "loss": 0.2345, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 5.182870588235294e-05, | |
| "loss": 0.2266, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 10.75, | |
| "learning_rate": 5.154635294117647e-05, | |
| "loss": 0.2122, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.75, | |
| "eval_loss": 0.4489321708679199, | |
| "eval_runtime": 81.0474, | |
| "eval_samples_per_second": 10.204, | |
| "eval_steps_per_second": 1.283, | |
| "eval_wer": 0.21370510396975426, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 5.1264e-05, | |
| "loss": 0.2038, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 5.098164705882353e-05, | |
| "loss": 0.1916, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 5.069929411764706e-05, | |
| "loss": 0.1896, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 5.041694117647059e-05, | |
| "loss": 0.207, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 5.013458823529412e-05, | |
| "loss": 0.1949, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "eval_loss": 0.3977554738521576, | |
| "eval_runtime": 80.708, | |
| "eval_samples_per_second": 10.247, | |
| "eval_steps_per_second": 1.289, | |
| "eval_wer": 0.20633270321361058, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 4.985223529411765e-05, | |
| "loss": 0.2177, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 4.956988235294118e-05, | |
| "loss": 0.182, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 12.47, | |
| "learning_rate": 4.928752941176471e-05, | |
| "loss": 0.1809, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 4.900517647058823e-05, | |
| "loss": 0.1876, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 4.872282352941177e-05, | |
| "loss": 0.1929, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_loss": 0.38229823112487793, | |
| "eval_runtime": 82.0352, | |
| "eval_samples_per_second": 10.081, | |
| "eval_steps_per_second": 1.268, | |
| "eval_wer": 0.20264650283553876, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 4.8440470588235294e-05, | |
| "loss": 0.1952, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 4.8158117647058825e-05, | |
| "loss": 0.1739, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 4.7875764705882355e-05, | |
| "loss": 0.1748, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 4.759341176470588e-05, | |
| "loss": 0.1625, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "learning_rate": 4.7311058823529416e-05, | |
| "loss": 0.1757, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "eval_loss": 0.3409084379673004, | |
| "eval_runtime": 80.5311, | |
| "eval_samples_per_second": 10.269, | |
| "eval_steps_per_second": 1.291, | |
| "eval_wer": 0.19650283553875236, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 4.702870588235294e-05, | |
| "loss": 0.1422, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 4.674635294117647e-05, | |
| "loss": 0.1571, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 4.6464e-05, | |
| "loss": 0.1565, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "learning_rate": 4.618164705882353e-05, | |
| "loss": 0.1717, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 4.589929411764706e-05, | |
| "loss": 0.1771, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "eval_loss": 0.3844490051269531, | |
| "eval_runtime": 81.332, | |
| "eval_samples_per_second": 10.168, | |
| "eval_steps_per_second": 1.279, | |
| "eval_wer": 0.19357277882797733, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 4.561694117647059e-05, | |
| "loss": 0.1408, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 4.5334588235294116e-05, | |
| "loss": 0.1512, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "learning_rate": 4.505223529411765e-05, | |
| "loss": 0.1505, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 4.476988235294118e-05, | |
| "loss": 0.1436, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 4.448752941176471e-05, | |
| "loss": 0.1452, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_loss": 0.37490135431289673, | |
| "eval_runtime": 81.4843, | |
| "eval_samples_per_second": 10.149, | |
| "eval_steps_per_second": 1.276, | |
| "eval_wer": 0.1899810964083176, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 4.420517647058824e-05, | |
| "loss": 0.1462, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "learning_rate": 4.392282352941176e-05, | |
| "loss": 0.1425, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 4.36404705882353e-05, | |
| "loss": 0.144, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 4.335811764705882e-05, | |
| "loss": 0.1622, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 17.2, | |
| "learning_rate": 4.3075764705882354e-05, | |
| "loss": 0.1341, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 17.2, | |
| "eval_loss": 0.44071856141090393, | |
| "eval_runtime": 82.1207, | |
| "eval_samples_per_second": 10.071, | |
| "eval_steps_per_second": 1.266, | |
| "eval_wer": 0.20264650283553876, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 17.42, | |
| "learning_rate": 4.2793411764705884e-05, | |
| "loss": 0.1151, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 17.63, | |
| "learning_rate": 4.251105882352941e-05, | |
| "loss": 0.1477, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "learning_rate": 4.2228705882352945e-05, | |
| "loss": 0.1377, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 4.194635294117647e-05, | |
| "loss": 0.129, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 4.1664e-05, | |
| "loss": 0.13, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "eval_loss": 0.42531654238700867, | |
| "eval_runtime": 82.3744, | |
| "eval_samples_per_second": 10.04, | |
| "eval_steps_per_second": 1.263, | |
| "eval_wer": 0.1882797731568998, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 18.49, | |
| "learning_rate": 4.138164705882353e-05, | |
| "loss": 0.1353, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 18.71, | |
| "learning_rate": 4.109929411764706e-05, | |
| "loss": 0.1216, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 18.92, | |
| "learning_rate": 4.081694117647059e-05, | |
| "loss": 0.1301, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "learning_rate": 4.053458823529412e-05, | |
| "loss": 0.1195, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 4.0252235294117645e-05, | |
| "loss": 0.1183, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_loss": 0.43106988072395325, | |
| "eval_runtime": 80.8241, | |
| "eval_samples_per_second": 10.232, | |
| "eval_steps_per_second": 1.287, | |
| "eval_wer": 0.18799621928166352, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 3.996988235294118e-05, | |
| "loss": 0.1147, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "learning_rate": 3.9687529411764706e-05, | |
| "loss": 0.1178, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 3.940517647058823e-05, | |
| "loss": 0.1177, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 20.22, | |
| "learning_rate": 3.912282352941177e-05, | |
| "loss": 0.1142, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 20.43, | |
| "learning_rate": 3.884047058823529e-05, | |
| "loss": 0.118, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 20.43, | |
| "eval_loss": 0.4430650472640991, | |
| "eval_runtime": 81.2185, | |
| "eval_samples_per_second": 10.182, | |
| "eval_steps_per_second": 1.28, | |
| "eval_wer": 0.18818525519848772, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 3.855811764705883e-05, | |
| "loss": 0.1125, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 20.86, | |
| "learning_rate": 3.827576470588235e-05, | |
| "loss": 0.1231, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 21.08, | |
| "learning_rate": 3.799341176470588e-05, | |
| "loss": 0.1133, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 21.29, | |
| "learning_rate": 3.771105882352941e-05, | |
| "loss": 0.1152, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 21.51, | |
| "learning_rate": 3.7428705882352944e-05, | |
| "loss": 0.1123, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 21.51, | |
| "eval_loss": 0.4753414988517761, | |
| "eval_runtime": 81.7131, | |
| "eval_samples_per_second": 10.121, | |
| "eval_steps_per_second": 1.273, | |
| "eval_wer": 0.18204158790170133, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 21.72, | |
| "learning_rate": 3.7146352941176474e-05, | |
| "loss": 0.1186, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 3.6864e-05, | |
| "loss": 0.1113, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 22.15, | |
| "learning_rate": 3.658164705882353e-05, | |
| "loss": 0.1243, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 22.37, | |
| "learning_rate": 3.629929411764706e-05, | |
| "loss": 0.1056, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 22.58, | |
| "learning_rate": 3.601694117647059e-05, | |
| "loss": 0.1037, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 22.58, | |
| "eval_loss": 0.40868785977363586, | |
| "eval_runtime": 80.8904, | |
| "eval_samples_per_second": 10.224, | |
| "eval_steps_per_second": 1.286, | |
| "eval_wer": 0.1833648393194707, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "learning_rate": 3.573458823529411e-05, | |
| "loss": 0.1055, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 3.545223529411765e-05, | |
| "loss": 0.0977, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 3.5169882352941174e-05, | |
| "loss": 0.1076, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 23.44, | |
| "learning_rate": 3.488752941176471e-05, | |
| "loss": 0.1028, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 23.66, | |
| "learning_rate": 3.4605176470588235e-05, | |
| "loss": 0.1066, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 23.66, | |
| "eval_loss": 0.415127694606781, | |
| "eval_runtime": 81.0404, | |
| "eval_samples_per_second": 10.205, | |
| "eval_steps_per_second": 1.283, | |
| "eval_wer": 0.18449905482041587, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 23.87, | |
| "learning_rate": 3.4322823529411766e-05, | |
| "loss": 0.1091, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 24.09, | |
| "learning_rate": 3.4040470588235296e-05, | |
| "loss": 0.089, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 24.3, | |
| "learning_rate": 3.375811764705882e-05, | |
| "loss": 0.1014, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 3.347576470588236e-05, | |
| "loss": 0.0864, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 24.73, | |
| "learning_rate": 3.319341176470588e-05, | |
| "loss": 0.0977, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 24.73, | |
| "eval_loss": 0.43674391508102417, | |
| "eval_runtime": 81.0756, | |
| "eval_samples_per_second": 10.2, | |
| "eval_steps_per_second": 1.283, | |
| "eval_wer": 0.1782608695652174, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 24.95, | |
| "learning_rate": 3.291105882352941e-05, | |
| "loss": 0.0989, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 25.16, | |
| "learning_rate": 3.262870588235294e-05, | |
| "loss": 0.0977, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 25.38, | |
| "learning_rate": 3.234635294117647e-05, | |
| "loss": 0.0869, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 25.59, | |
| "learning_rate": 3.2064e-05, | |
| "loss": 0.0851, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 3.1781647058823534e-05, | |
| "loss": 0.0968, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "eval_loss": 0.4236660897731781, | |
| "eval_runtime": 80.9504, | |
| "eval_samples_per_second": 10.216, | |
| "eval_steps_per_second": 1.285, | |
| "eval_wer": 0.17561436672967864, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "learning_rate": 3.149929411764706e-05, | |
| "loss": 0.0866, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "learning_rate": 3.121694117647059e-05, | |
| "loss": 0.0774, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 26.45, | |
| "learning_rate": 3.093458823529412e-05, | |
| "loss": 0.0974, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 3.065223529411764e-05, | |
| "loss": 0.0807, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "learning_rate": 3.0372705882352942e-05, | |
| "loss": 0.0835, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "eval_loss": 0.4728855490684509, | |
| "eval_runtime": 81.9776, | |
| "eval_samples_per_second": 10.088, | |
| "eval_steps_per_second": 1.269, | |
| "eval_wer": 0.17807183364839319, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 3.009035294117647e-05, | |
| "loss": 0.0888, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 27.31, | |
| "learning_rate": 2.9808000000000003e-05, | |
| "loss": 0.0845, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 27.53, | |
| "learning_rate": 2.952564705882353e-05, | |
| "loss": 0.0888, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 27.74, | |
| "learning_rate": 2.9243294117647058e-05, | |
| "loss": 0.0816, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 27.96, | |
| "learning_rate": 2.8960941176470588e-05, | |
| "loss": 0.0919, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 27.96, | |
| "eval_loss": 0.4152912497520447, | |
| "eval_runtime": 81.9103, | |
| "eval_samples_per_second": 10.096, | |
| "eval_steps_per_second": 1.27, | |
| "eval_wer": 0.17013232514177692, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 28.17, | |
| "learning_rate": 2.867858823529412e-05, | |
| "loss": 0.0886, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 2.8396235294117646e-05, | |
| "loss": 0.0809, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 28.6, | |
| "learning_rate": 2.8113882352941176e-05, | |
| "loss": 0.0786, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 28.82, | |
| "learning_rate": 2.7831529411764707e-05, | |
| "loss": 0.0766, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "learning_rate": 2.7552e-05, | |
| "loss": 0.0677, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "eval_loss": 0.431657612323761, | |
| "eval_runtime": 81.4905, | |
| "eval_samples_per_second": 10.148, | |
| "eval_steps_per_second": 1.276, | |
| "eval_wer": 0.16928166351606805, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 2.726964705882353e-05, | |
| "loss": 0.0757, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 29.46, | |
| "learning_rate": 2.698729411764706e-05, | |
| "loss": 0.0719, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 2.670494117647059e-05, | |
| "loss": 0.0772, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 29.89, | |
| "learning_rate": 2.642258823529412e-05, | |
| "loss": 0.0762, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 30.11, | |
| "learning_rate": 2.614023529411765e-05, | |
| "loss": 0.0726, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 30.11, | |
| "eval_loss": 0.43801796436309814, | |
| "eval_runtime": 83.6487, | |
| "eval_samples_per_second": 9.887, | |
| "eval_steps_per_second": 1.243, | |
| "eval_wer": 0.17362948960302457, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 30.32, | |
| "learning_rate": 2.5857882352941176e-05, | |
| "loss": 0.0654, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 30.54, | |
| "learning_rate": 2.5575529411764707e-05, | |
| "loss": 0.0691, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 30.75, | |
| "learning_rate": 2.5293176470588234e-05, | |
| "loss": 0.0677, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 2.5010823529411764e-05, | |
| "loss": 0.0668, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 31.18, | |
| "learning_rate": 2.4728470588235295e-05, | |
| "loss": 0.066, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 31.18, | |
| "eval_loss": 0.4384245276451111, | |
| "eval_runtime": 81.1759, | |
| "eval_samples_per_second": 10.188, | |
| "eval_steps_per_second": 1.281, | |
| "eval_wer": 0.16814744801512288, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 31.4, | |
| "learning_rate": 2.4446117647058825e-05, | |
| "loss": 0.0663, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 31.61, | |
| "learning_rate": 2.4163764705882352e-05, | |
| "loss": 0.0688, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 31.83, | |
| "learning_rate": 2.3881411764705883e-05, | |
| "loss": 0.0677, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 32.04, | |
| "learning_rate": 2.3599058823529414e-05, | |
| "loss": 0.0718, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 2.3316705882352944e-05, | |
| "loss": 0.0713, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "eval_loss": 0.4215048551559448, | |
| "eval_runtime": 80.5465, | |
| "eval_samples_per_second": 10.267, | |
| "eval_steps_per_second": 1.291, | |
| "eval_wer": 0.16294896030245748, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 32.47, | |
| "learning_rate": 2.303435294117647e-05, | |
| "loss": 0.065, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 32.69, | |
| "learning_rate": 2.2752e-05, | |
| "loss": 0.0631, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "learning_rate": 2.246964705882353e-05, | |
| "loss": 0.0616, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 33.12, | |
| "learning_rate": 2.218729411764706e-05, | |
| "loss": 0.0812, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 2.190494117647059e-05, | |
| "loss": 0.0605, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 0.457427054643631, | |
| "eval_runtime": 80.6078, | |
| "eval_samples_per_second": 10.26, | |
| "eval_steps_per_second": 1.29, | |
| "eval_wer": 0.1713610586011342, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 2.1622588235294117e-05, | |
| "loss": 0.0566, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 33.76, | |
| "learning_rate": 2.1340235294117648e-05, | |
| "loss": 0.064, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 33.98, | |
| "learning_rate": 2.1057882352941178e-05, | |
| "loss": 0.0591, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 34.19, | |
| "learning_rate": 2.077552941176471e-05, | |
| "loss": 0.0581, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 34.41, | |
| "learning_rate": 2.0493176470588236e-05, | |
| "loss": 0.0632, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 34.41, | |
| "eval_loss": 0.43431583046913147, | |
| "eval_runtime": 81.6162, | |
| "eval_samples_per_second": 10.133, | |
| "eval_steps_per_second": 1.274, | |
| "eval_wer": 0.16417769376181474, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 34.62, | |
| "learning_rate": 2.0210823529411763e-05, | |
| "loss": 0.0578, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 1.9928470588235293e-05, | |
| "loss": 0.0616, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 35.05, | |
| "learning_rate": 1.9648941176470586e-05, | |
| "loss": 0.0516, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 35.27, | |
| "learning_rate": 1.9366588235294117e-05, | |
| "loss": 0.0533, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "learning_rate": 1.9084235294117647e-05, | |
| "loss": 0.0567, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "eval_loss": 0.42308300733566284, | |
| "eval_runtime": 82.0264, | |
| "eval_samples_per_second": 10.082, | |
| "eval_steps_per_second": 1.268, | |
| "eval_wer": 0.16011342155009453, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 35.7, | |
| "learning_rate": 1.8801882352941178e-05, | |
| "loss": 0.0562, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 35.91, | |
| "learning_rate": 1.8519529411764705e-05, | |
| "loss": 0.0554, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 1.8237176470588236e-05, | |
| "loss": 0.0564, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 36.34, | |
| "learning_rate": 1.7954823529411766e-05, | |
| "loss": 0.0617, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 36.56, | |
| "learning_rate": 1.7672470588235297e-05, | |
| "loss": 0.0556, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 36.56, | |
| "eval_loss": 0.4404306411743164, | |
| "eval_runtime": 81.4874, | |
| "eval_samples_per_second": 10.149, | |
| "eval_steps_per_second": 1.276, | |
| "eval_wer": 0.1667296786389414, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 36.77, | |
| "learning_rate": 1.7390117647058824e-05, | |
| "loss": 0.0524, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 36.99, | |
| "learning_rate": 1.7107764705882354e-05, | |
| "loss": 0.0602, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 37.2, | |
| "learning_rate": 1.682541176470588e-05, | |
| "loss": 0.0472, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 37.42, | |
| "learning_rate": 1.6543058823529412e-05, | |
| "loss": 0.047, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 37.63, | |
| "learning_rate": 1.6260705882352943e-05, | |
| "loss": 0.0426, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 37.63, | |
| "eval_loss": 0.4458593428134918, | |
| "eval_runtime": 80.8169, | |
| "eval_samples_per_second": 10.233, | |
| "eval_steps_per_second": 1.287, | |
| "eval_wer": 0.16247637051039698, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 37.85, | |
| "learning_rate": 1.597835294117647e-05, | |
| "loss": 0.0513, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 38.06, | |
| "learning_rate": 1.5696e-05, | |
| "loss": 0.056, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 38.28, | |
| "learning_rate": 1.541364705882353e-05, | |
| "loss": 0.0498, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 38.49, | |
| "learning_rate": 1.513129411764706e-05, | |
| "loss": 0.0473, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "learning_rate": 1.4848941176470588e-05, | |
| "loss": 0.0445, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "eval_loss": 0.4483908712863922, | |
| "eval_runtime": 82.388, | |
| "eval_samples_per_second": 10.038, | |
| "eval_steps_per_second": 1.262, | |
| "eval_wer": 0.16285444234404536, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 38.92, | |
| "learning_rate": 1.4566588235294117e-05, | |
| "loss": 0.0487, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 39.14, | |
| "learning_rate": 1.4284235294117648e-05, | |
| "loss": 0.0491, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 39.35, | |
| "learning_rate": 1.4001882352941177e-05, | |
| "loss": 0.044, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 39.57, | |
| "learning_rate": 1.3722352941176471e-05, | |
| "loss": 0.0467, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 39.78, | |
| "learning_rate": 1.344e-05, | |
| "loss": 0.0463, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 39.78, | |
| "eval_loss": 0.45076683163642883, | |
| "eval_runtime": 81.2197, | |
| "eval_samples_per_second": 10.182, | |
| "eval_steps_per_second": 1.28, | |
| "eval_wer": 0.15964083175803404, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 1.315764705882353e-05, | |
| "loss": 0.0464, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 40.22, | |
| "learning_rate": 1.2875294117647058e-05, | |
| "loss": 0.0395, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 40.43, | |
| "learning_rate": 1.2592941176470588e-05, | |
| "loss": 0.0421, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 40.65, | |
| "learning_rate": 1.2310588235294119e-05, | |
| "loss": 0.0463, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 40.86, | |
| "learning_rate": 1.2028235294117648e-05, | |
| "loss": 0.0448, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 40.86, | |
| "eval_loss": 0.4395386278629303, | |
| "eval_runtime": 82.035, | |
| "eval_samples_per_second": 10.081, | |
| "eval_steps_per_second": 1.268, | |
| "eval_wer": 0.1604914933837429, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 41.08, | |
| "learning_rate": 1.1745882352941178e-05, | |
| "loss": 0.0483, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 41.29, | |
| "learning_rate": 1.1463529411764705e-05, | |
| "loss": 0.0457, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "learning_rate": 1.1181176470588236e-05, | |
| "loss": 0.0363, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 41.72, | |
| "learning_rate": 1.0898823529411765e-05, | |
| "loss": 0.0428, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 41.94, | |
| "learning_rate": 1.0616470588235295e-05, | |
| "loss": 0.0434, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 41.94, | |
| "eval_loss": 0.4489993453025818, | |
| "eval_runtime": 82.599, | |
| "eval_samples_per_second": 10.012, | |
| "eval_steps_per_second": 1.259, | |
| "eval_wer": 0.16068052930056712, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 42.15, | |
| "learning_rate": 1.0334117647058824e-05, | |
| "loss": 0.0389, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 42.37, | |
| "learning_rate": 1.0051764705882353e-05, | |
| "loss": 0.0482, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 42.58, | |
| "learning_rate": 9.769411764705882e-06, | |
| "loss": 0.0467, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 42.8, | |
| "learning_rate": 9.487058823529412e-06, | |
| "loss": 0.0379, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 9.204705882352941e-06, | |
| "loss": 0.0347, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "eval_loss": 0.47717225551605225, | |
| "eval_runtime": 81.2937, | |
| "eval_samples_per_second": 10.173, | |
| "eval_steps_per_second": 1.279, | |
| "eval_wer": 0.15822306238185255, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 43.23, | |
| "learning_rate": 8.922352941176471e-06, | |
| "loss": 0.0327, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 43.44, | |
| "learning_rate": 8.64e-06, | |
| "loss": 0.0407, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 43.66, | |
| "learning_rate": 8.357647058823529e-06, | |
| "loss": 0.0368, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 43.87, | |
| "learning_rate": 8.07529411764706e-06, | |
| "loss": 0.0365, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 44.09, | |
| "learning_rate": 7.792941176470588e-06, | |
| "loss": 0.0332, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 44.09, | |
| "eval_loss": 0.4728728234767914, | |
| "eval_runtime": 81.4631, | |
| "eval_samples_per_second": 10.152, | |
| "eval_steps_per_second": 1.277, | |
| "eval_wer": 0.15822306238185255, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 44.3, | |
| "learning_rate": 7.510588235294118e-06, | |
| "loss": 0.0353, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 44.52, | |
| "learning_rate": 7.228235294117648e-06, | |
| "loss": 0.039, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 44.73, | |
| "learning_rate": 6.945882352941177e-06, | |
| "loss": 0.0395, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 44.95, | |
| "learning_rate": 6.663529411764706e-06, | |
| "loss": 0.0368, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "learning_rate": 6.381176470588236e-06, | |
| "loss": 0.037, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "eval_loss": 0.45591220259666443, | |
| "eval_runtime": 83.1729, | |
| "eval_samples_per_second": 9.943, | |
| "eval_steps_per_second": 1.25, | |
| "eval_wer": 0.15727788279773156, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 45.38, | |
| "learning_rate": 6.098823529411765e-06, | |
| "loss": 0.0328, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 45.59, | |
| "learning_rate": 5.8164705882352945e-06, | |
| "loss": 0.0541, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 45.81, | |
| "learning_rate": 5.534117647058823e-06, | |
| "loss": 0.0413, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 46.02, | |
| "learning_rate": 5.251764705882353e-06, | |
| "loss": 0.0403, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 46.24, | |
| "learning_rate": 4.969411764705883e-06, | |
| "loss": 0.0328, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 46.24, | |
| "eval_loss": 0.46641021966934204, | |
| "eval_runtime": 82.6563, | |
| "eval_samples_per_second": 10.005, | |
| "eval_steps_per_second": 1.258, | |
| "eval_wer": 0.1560491493383743, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "learning_rate": 4.6870588235294115e-06, | |
| "loss": 0.0334, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 46.67, | |
| "learning_rate": 4.404705882352941e-06, | |
| "loss": 0.0334, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 46.88, | |
| "learning_rate": 4.122352941176471e-06, | |
| "loss": 0.0324, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 47.1, | |
| "learning_rate": 3.8400000000000005e-06, | |
| "loss": 0.0337, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 47.31, | |
| "learning_rate": 3.5576470588235293e-06, | |
| "loss": 0.0366, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 47.31, | |
| "eval_loss": 0.454330176115036, | |
| "eval_runtime": 81.2727, | |
| "eval_samples_per_second": 10.176, | |
| "eval_steps_per_second": 1.28, | |
| "eval_wer": 0.15434782608695652, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 47.53, | |
| "learning_rate": 3.275294117647059e-06, | |
| "loss": 0.0332, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 47.74, | |
| "learning_rate": 2.9929411764705883e-06, | |
| "loss": 0.0295, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 47.96, | |
| "learning_rate": 2.710588235294118e-06, | |
| "loss": 0.037, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 48.17, | |
| "learning_rate": 2.428235294117647e-06, | |
| "loss": 0.0314, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "learning_rate": 2.1458823529411764e-06, | |
| "loss": 0.0377, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "eval_loss": 0.45068359375, | |
| "eval_runtime": 80.8954, | |
| "eval_samples_per_second": 10.223, | |
| "eval_steps_per_second": 1.286, | |
| "eval_wer": 0.15595463137996218, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 48.6, | |
| "learning_rate": 1.8635294117647059e-06, | |
| "loss": 0.0284, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 48.82, | |
| "learning_rate": 1.5811764705882354e-06, | |
| "loss": 0.0326, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 49.03, | |
| "learning_rate": 1.2988235294117648e-06, | |
| "loss": 0.034, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 49.25, | |
| "learning_rate": 1.016470588235294e-06, | |
| "loss": 0.0319, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 49.46, | |
| "learning_rate": 7.341176470588236e-07, | |
| "loss": 0.0331, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 49.46, | |
| "eval_loss": 0.4567229151725769, | |
| "eval_runtime": 80.8744, | |
| "eval_samples_per_second": 10.226, | |
| "eval_steps_per_second": 1.286, | |
| "eval_wer": 0.15330812854442344, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 49.68, | |
| "learning_rate": 4.51764705882353e-07, | |
| "loss": 0.0334, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 49.89, | |
| "learning_rate": 1.6941176470588237e-07, | |
| "loss": 0.027, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 23250, | |
| "total_flos": 2.4501256105279144e+20, | |
| "train_loss": 0.23119104407936014, | |
| "train_runtime": 64754.7357, | |
| "train_samples_per_second": 5.744, | |
| "train_steps_per_second": 0.359 | |
| } | |
| ], | |
| "max_steps": 23250, | |
| "num_train_epochs": 50, | |
| "total_flos": 2.4501256105279144e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |