| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 14800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3.6375e-06, | |
| "loss": 19.2806, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 7.3875e-06, | |
| "loss": 10.959, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.1137499999999998e-05, | |
| "loss": 6.9106, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1.48875e-05, | |
| "loss": 5.624, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 1.86375e-05, | |
| "loss": 4.6095, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "eval_loss": 4.5881452560424805, | |
| "eval_runtime": 92.8763, | |
| "eval_samples_per_second": 22.557, | |
| "eval_steps_per_second": 0.711, | |
| "eval_wer": 0.9999453103636861, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 2.23875e-05, | |
| "loss": 3.9225, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 2.6137499999999995e-05, | |
| "loss": 3.5503, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 2.9887499999999998e-05, | |
| "loss": 3.4262, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 3.36375e-05, | |
| "loss": 3.391, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 3.7387499999999994e-05, | |
| "loss": 3.3396, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "eval_loss": 3.3301022052764893, | |
| "eval_runtime": 93.143, | |
| "eval_samples_per_second": 22.492, | |
| "eval_steps_per_second": 0.709, | |
| "eval_wer": 1.0000546896363138, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 4.11375e-05, | |
| "loss": 3.2895, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 4.48875e-05, | |
| "loss": 3.228, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 4.8637499999999996e-05, | |
| "loss": 2.987, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 5.23875e-05, | |
| "loss": 2.4352, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 5.61375e-05, | |
| "loss": 2.0061, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "eval_loss": 1.20964777469635, | |
| "eval_runtime": 93.1759, | |
| "eval_samples_per_second": 22.484, | |
| "eval_steps_per_second": 0.708, | |
| "eval_wer": 1.0063439978124145, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 5.988749999999999e-05, | |
| "loss": 1.7983, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 6.36375e-05, | |
| "loss": 1.6684, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 6.738749999999999e-05, | |
| "loss": 1.6307, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.84, | |
| "learning_rate": 7.11375e-05, | |
| "loss": 1.5654, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 1.523, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "eval_loss": 0.7836298942565918, | |
| "eval_runtime": 93.1376, | |
| "eval_samples_per_second": 22.494, | |
| "eval_steps_per_second": 0.709, | |
| "eval_wer": 1.0051408258135084, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 7.4431640625e-05, | |
| "loss": 1.4937, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.86, | |
| "learning_rate": 7.384570312499999e-05, | |
| "loss": 1.4538, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "learning_rate": 7.325976562499999e-05, | |
| "loss": 1.4016, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 16.22, | |
| "learning_rate": 7.2673828125e-05, | |
| "loss": 1.403, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 7.2087890625e-05, | |
| "loss": 1.3868, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "eval_loss": 0.6836622953414917, | |
| "eval_runtime": 92.9915, | |
| "eval_samples_per_second": 22.529, | |
| "eval_steps_per_second": 0.71, | |
| "eval_wer": 1.007984686901832, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.57, | |
| "learning_rate": 7.150195312499999e-05, | |
| "loss": 1.3313, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 7.091601562499999e-05, | |
| "loss": 1.2931, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.92, | |
| "learning_rate": 7.0330078125e-05, | |
| "loss": 1.3319, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 6.9744140625e-05, | |
| "loss": 1.3133, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 6.915820312499999e-05, | |
| "loss": 1.2807, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "eval_loss": 0.656791090965271, | |
| "eval_runtime": 93.0318, | |
| "eval_samples_per_second": 22.519, | |
| "eval_steps_per_second": 0.709, | |
| "eval_wer": 1.0112113754443532, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 20.95, | |
| "learning_rate": 6.857226562499999e-05, | |
| "loss": 1.2626, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 21.62, | |
| "learning_rate": 6.7986328125e-05, | |
| "loss": 1.2535, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 22.3, | |
| "learning_rate": 6.7400390625e-05, | |
| "loss": 1.2297, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "learning_rate": 6.681445312499999e-05, | |
| "loss": 1.2074, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 23.65, | |
| "learning_rate": 6.6234375e-05, | |
| "loss": 1.231, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 23.65, | |
| "eval_loss": 0.612037718296051, | |
| "eval_runtime": 93.2672, | |
| "eval_samples_per_second": 22.462, | |
| "eval_steps_per_second": 0.708, | |
| "eval_wer": 1.0105004101722723, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 24.32, | |
| "learning_rate": 6.564843749999999e-05, | |
| "loss": 1.2092, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 6.50625e-05, | |
| "loss": 1.1775, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 25.68, | |
| "learning_rate": 6.44765625e-05, | |
| "loss": 1.1916, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 26.35, | |
| "learning_rate": 6.3890625e-05, | |
| "loss": 1.1798, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 27.03, | |
| "learning_rate": 6.330468749999999e-05, | |
| "loss": 1.1673, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 27.03, | |
| "eval_loss": 0.5971782207489014, | |
| "eval_runtime": 93.4057, | |
| "eval_samples_per_second": 22.429, | |
| "eval_steps_per_second": 0.707, | |
| "eval_wer": 1.0089144107191688, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 27.7, | |
| "learning_rate": 6.271875e-05, | |
| "loss": 1.1626, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 28.38, | |
| "learning_rate": 6.21328125e-05, | |
| "loss": 1.162, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 29.05, | |
| "learning_rate": 6.1546875e-05, | |
| "loss": 1.1471, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 29.73, | |
| "learning_rate": 6.09609375e-05, | |
| "loss": 1.1203, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 30.41, | |
| "learning_rate": 6.0375e-05, | |
| "loss": 1.1416, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 30.41, | |
| "eval_loss": 0.5780259966850281, | |
| "eval_runtime": 92.6708, | |
| "eval_samples_per_second": 22.607, | |
| "eval_steps_per_second": 0.712, | |
| "eval_wer": 1.0131802023516543, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 31.08, | |
| "learning_rate": 5.97890625e-05, | |
| "loss": 1.1175, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 31.76, | |
| "learning_rate": 5.9203125e-05, | |
| "loss": 1.1053, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 32.43, | |
| "learning_rate": 5.86171875e-05, | |
| "loss": 1.107, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 33.11, | |
| "learning_rate": 5.803125e-05, | |
| "loss": 1.115, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 33.78, | |
| "learning_rate": 5.74453125e-05, | |
| "loss": 1.0738, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 33.78, | |
| "eval_loss": 0.580643892288208, | |
| "eval_runtime": 95.0733, | |
| "eval_samples_per_second": 22.036, | |
| "eval_steps_per_second": 0.694, | |
| "eval_wer": 1.0123051681706317, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 34.46, | |
| "learning_rate": 5.686523437499999e-05, | |
| "loss": 1.0778, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 35.14, | |
| "learning_rate": 5.6279296874999994e-05, | |
| "loss": 1.0948, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 35.81, | |
| "learning_rate": 5.569335937499999e-05, | |
| "loss": 1.062, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 36.49, | |
| "learning_rate": 5.5107421874999994e-05, | |
| "loss": 1.0503, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "learning_rate": 5.452148437499999e-05, | |
| "loss": 1.0771, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "eval_loss": 0.5585715174674988, | |
| "eval_runtime": 92.7751, | |
| "eval_samples_per_second": 22.581, | |
| "eval_steps_per_second": 0.711, | |
| "eval_wer": 1.0066721356302981, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 37.84, | |
| "learning_rate": 5.3935546874999995e-05, | |
| "loss": 1.0368, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 38.51, | |
| "learning_rate": 5.334960937499999e-05, | |
| "loss": 1.0329, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 39.19, | |
| "learning_rate": 5.2763671874999995e-05, | |
| "loss": 1.0402, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 39.86, | |
| "learning_rate": 5.217773437499999e-05, | |
| "loss": 1.0346, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 40.54, | |
| "learning_rate": 5.1591796874999995e-05, | |
| "loss": 1.0287, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 40.54, | |
| "eval_loss": 0.5463963747024536, | |
| "eval_runtime": 92.708, | |
| "eval_samples_per_second": 22.598, | |
| "eval_steps_per_second": 0.712, | |
| "eval_wer": 1.0057971014492753, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 41.22, | |
| "learning_rate": 5.100585937499999e-05, | |
| "loss": 1.0196, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 41.89, | |
| "learning_rate": 5.0419921874999995e-05, | |
| "loss": 1.0198, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 42.57, | |
| "learning_rate": 4.983398437499999e-05, | |
| "loss": 0.9864, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 43.24, | |
| "learning_rate": 4.9248046874999996e-05, | |
| "loss": 0.999, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 43.92, | |
| "learning_rate": 4.866210937499999e-05, | |
| "loss": 1.0106, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 43.92, | |
| "eval_loss": 0.5407418608665466, | |
| "eval_runtime": 92.9267, | |
| "eval_samples_per_second": 22.545, | |
| "eval_steps_per_second": 0.71, | |
| "eval_wer": 1.0061799289034727, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.59, | |
| "learning_rate": 4.8076171874999996e-05, | |
| "loss": 0.9816, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 45.27, | |
| "learning_rate": 4.749023437499999e-05, | |
| "loss": 0.976, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 45.95, | |
| "learning_rate": 4.6904296874999996e-05, | |
| "loss": 0.9833, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 46.62, | |
| "learning_rate": 4.631835937499999e-05, | |
| "loss": 0.9905, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 47.3, | |
| "learning_rate": 4.5732421875e-05, | |
| "loss": 0.9538, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 47.3, | |
| "eval_loss": 0.5333988070487976, | |
| "eval_runtime": 92.3771, | |
| "eval_samples_per_second": 22.679, | |
| "eval_steps_per_second": 0.714, | |
| "eval_wer": 1.0089144107191688, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 47.97, | |
| "learning_rate": 4.5146484374999993e-05, | |
| "loss": 0.9578, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 48.65, | |
| "learning_rate": 4.4560546875e-05, | |
| "loss": 0.9607, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 49.32, | |
| "learning_rate": 4.3974609374999994e-05, | |
| "loss": 0.9451, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 4.3388671875e-05, | |
| "loss": 0.9453, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 50.68, | |
| "learning_rate": 4.2802734374999994e-05, | |
| "loss": 0.9607, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 50.68, | |
| "eval_loss": 0.5395269989967346, | |
| "eval_runtime": 92.6715, | |
| "eval_samples_per_second": 22.607, | |
| "eval_steps_per_second": 0.712, | |
| "eval_wer": 1.0110473065354115, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 51.35, | |
| "learning_rate": 4.2216796875e-05, | |
| "loss": 0.9445, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 52.03, | |
| "learning_rate": 4.1630859374999994e-05, | |
| "loss": 0.9314, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 52.7, | |
| "learning_rate": 4.1044921875e-05, | |
| "loss": 0.9166, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 53.38, | |
| "learning_rate": 4.0458984374999995e-05, | |
| "loss": 0.9264, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 54.05, | |
| "learning_rate": 3.987304687499999e-05, | |
| "loss": 0.9108, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 54.05, | |
| "eval_loss": 0.5501919388771057, | |
| "eval_runtime": 92.8418, | |
| "eval_samples_per_second": 22.565, | |
| "eval_steps_per_second": 0.711, | |
| "eval_wer": 1.0137270987147935, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 54.73, | |
| "learning_rate": 3.9287109374999995e-05, | |
| "loss": 0.9215, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 55.41, | |
| "learning_rate": 3.870117187499999e-05, | |
| "loss": 0.9326, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 56.08, | |
| "learning_rate": 3.8115234374999995e-05, | |
| "loss": 0.9199, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 56.76, | |
| "learning_rate": 3.752929687499999e-05, | |
| "loss": 0.9043, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 57.43, | |
| "learning_rate": 3.6943359374999996e-05, | |
| "loss": 0.9252, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 57.43, | |
| "eval_loss": 0.5498473644256592, | |
| "eval_runtime": 92.009, | |
| "eval_samples_per_second": 22.77, | |
| "eval_steps_per_second": 0.717, | |
| "eval_wer": 1.0062346185397868, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 58.11, | |
| "learning_rate": 3.635742187499999e-05, | |
| "loss": 0.9041, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 58.78, | |
| "learning_rate": 3.5771484374999996e-05, | |
| "loss": 0.8963, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 59.46, | |
| "learning_rate": 3.518554687499999e-05, | |
| "loss": 0.901, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 60.14, | |
| "learning_rate": 3.4599609374999996e-05, | |
| "loss": 0.8947, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 60.81, | |
| "learning_rate": 3.401367187499999e-05, | |
| "loss": 0.8943, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 60.81, | |
| "eval_loss": 0.5447660684585571, | |
| "eval_runtime": 92.1905, | |
| "eval_samples_per_second": 22.725, | |
| "eval_steps_per_second": 0.716, | |
| "eval_wer": 1.0158053048947224, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 61.49, | |
| "learning_rate": 3.3427734375e-05, | |
| "loss": 0.875, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 62.16, | |
| "learning_rate": 3.284765625e-05, | |
| "loss": 0.8961, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 62.84, | |
| "learning_rate": 3.2261718749999996e-05, | |
| "loss": 0.8848, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 63.51, | |
| "learning_rate": 3.167578125e-05, | |
| "loss": 0.8573, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 64.19, | |
| "learning_rate": 3.1089843749999996e-05, | |
| "loss": 0.8728, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 64.19, | |
| "eval_loss": 0.5256930589675903, | |
| "eval_runtime": 92.3173, | |
| "eval_samples_per_second": 22.693, | |
| "eval_steps_per_second": 0.715, | |
| "eval_wer": 1.0113207547169811, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 64.86, | |
| "learning_rate": 3.050390625e-05, | |
| "loss": 0.8819, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 65.54, | |
| "learning_rate": 2.991796875e-05, | |
| "loss": 0.8324, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 66.22, | |
| "learning_rate": 2.933203125e-05, | |
| "loss": 0.8461, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 66.89, | |
| "learning_rate": 2.874609375e-05, | |
| "loss": 0.8512, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 67.57, | |
| "learning_rate": 2.816015625e-05, | |
| "loss": 0.8577, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 67.57, | |
| "eval_loss": 0.554978609085083, | |
| "eval_runtime": 92.3777, | |
| "eval_samples_per_second": 22.679, | |
| "eval_steps_per_second": 0.714, | |
| "eval_wer": 1.0177741318020235, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 68.24, | |
| "learning_rate": 2.7574218749999997e-05, | |
| "loss": 0.845, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 68.92, | |
| "learning_rate": 2.6988281249999997e-05, | |
| "loss": 0.8622, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 69.59, | |
| "learning_rate": 2.6402343749999998e-05, | |
| "loss": 0.8534, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 70.27, | |
| "learning_rate": 2.5816406249999998e-05, | |
| "loss": 0.8328, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 70.95, | |
| "learning_rate": 2.5230468749999998e-05, | |
| "loss": 0.8332, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 70.95, | |
| "eval_loss": 0.5607455968856812, | |
| "eval_runtime": 91.9364, | |
| "eval_samples_per_second": 22.787, | |
| "eval_steps_per_second": 0.718, | |
| "eval_wer": 1.0165709598031174, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 71.62, | |
| "learning_rate": 2.4644531249999998e-05, | |
| "loss": 0.8389, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 72.3, | |
| "learning_rate": 2.4058593749999998e-05, | |
| "loss": 0.8225, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 72.97, | |
| "learning_rate": 2.347265625e-05, | |
| "loss": 0.8141, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 73.65, | |
| "learning_rate": 2.288671875e-05, | |
| "loss": 0.8215, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 74.32, | |
| "learning_rate": 2.230078125e-05, | |
| "loss": 0.8174, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 74.32, | |
| "eval_loss": 0.5428524613380432, | |
| "eval_runtime": 94.2641, | |
| "eval_samples_per_second": 22.225, | |
| "eval_steps_per_second": 0.7, | |
| "eval_wer": 1.0145474432595023, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 2.171484375e-05, | |
| "loss": 0.7876, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 75.68, | |
| "learning_rate": 2.112890625e-05, | |
| "loss": 0.8226, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 76.35, | |
| "learning_rate": 2.054296875e-05, | |
| "loss": 0.8177, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 77.03, | |
| "learning_rate": 1.995703125e-05, | |
| "loss": 0.7941, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 77.7, | |
| "learning_rate": 1.937109375e-05, | |
| "loss": 0.8168, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 77.7, | |
| "eval_loss": 0.5561283230781555, | |
| "eval_runtime": 94.9459, | |
| "eval_samples_per_second": 22.065, | |
| "eval_steps_per_second": 0.695, | |
| "eval_wer": 1.0116488925348646, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 78.38, | |
| "learning_rate": 1.878515625e-05, | |
| "loss": 0.8095, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 79.05, | |
| "learning_rate": 1.8199218749999996e-05, | |
| "loss": 0.791, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 79.73, | |
| "learning_rate": 1.7613281249999997e-05, | |
| "loss": 0.7812, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 80.41, | |
| "learning_rate": 1.7033203125e-05, | |
| "loss": 0.8103, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 81.08, | |
| "learning_rate": 1.6447265625e-05, | |
| "loss": 0.7872, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 81.08, | |
| "eval_loss": 0.5477647185325623, | |
| "eval_runtime": 95.0024, | |
| "eval_samples_per_second": 22.052, | |
| "eval_steps_per_second": 0.695, | |
| "eval_wer": 1.0163522012578616, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 81.76, | |
| "learning_rate": 1.5861328125e-05, | |
| "loss": 0.7658, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 82.43, | |
| "learning_rate": 1.5275390625e-05, | |
| "loss": 0.7891, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 83.11, | |
| "learning_rate": 1.4689453124999998e-05, | |
| "loss": 0.7723, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 83.78, | |
| "learning_rate": 1.4103515624999998e-05, | |
| "loss": 0.7773, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 84.46, | |
| "learning_rate": 1.3517578124999998e-05, | |
| "loss": 0.7707, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 84.46, | |
| "eval_loss": 0.5412248373031616, | |
| "eval_runtime": 95.413, | |
| "eval_samples_per_second": 21.957, | |
| "eval_steps_per_second": 0.692, | |
| "eval_wer": 1.021602406343998, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 85.14, | |
| "learning_rate": 1.2931640624999999e-05, | |
| "loss": 0.7876, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 85.81, | |
| "learning_rate": 1.2345703124999999e-05, | |
| "loss": 0.7707, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 86.49, | |
| "learning_rate": 1.1759765624999999e-05, | |
| "loss": 0.7654, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 87.16, | |
| "learning_rate": 1.1173828124999999e-05, | |
| "loss": 0.77, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 87.84, | |
| "learning_rate": 1.0587890625e-05, | |
| "loss": 0.7742, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 87.84, | |
| "eval_loss": 0.5391495227813721, | |
| "eval_runtime": 95.0784, | |
| "eval_samples_per_second": 22.034, | |
| "eval_steps_per_second": 0.694, | |
| "eval_wer": 1.0206726825266612, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 88.51, | |
| "learning_rate": 1.0007812499999998e-05, | |
| "loss": 0.7534, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 89.19, | |
| "learning_rate": 9.421874999999999e-06, | |
| "loss": 0.7765, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 89.86, | |
| "learning_rate": 8.835937499999999e-06, | |
| "loss": 0.7598, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 90.54, | |
| "learning_rate": 8.249999999999999e-06, | |
| "loss": 0.7564, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 91.22, | |
| "learning_rate": 7.664062499999999e-06, | |
| "loss": 0.7594, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 91.22, | |
| "eval_loss": 0.5379434823989868, | |
| "eval_runtime": 95.2181, | |
| "eval_samples_per_second": 22.002, | |
| "eval_steps_per_second": 0.693, | |
| "eval_wer": 1.020836751435603, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 91.89, | |
| "learning_rate": 7.078124999999999e-06, | |
| "loss": 0.7739, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 92.57, | |
| "learning_rate": 6.492187499999999e-06, | |
| "loss": 0.7646, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 93.24, | |
| "learning_rate": 5.9062499999999996e-06, | |
| "loss": 0.7321, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 93.92, | |
| "learning_rate": 5.3203125e-06, | |
| "loss": 0.7754, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 94.59, | |
| "learning_rate": 4.734375e-06, | |
| "loss": 0.7678, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 94.59, | |
| "eval_loss": 0.541484534740448, | |
| "eval_runtime": 92.5358, | |
| "eval_samples_per_second": 22.64, | |
| "eval_steps_per_second": 0.713, | |
| "eval_wer": 1.0197976483456386, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 95.27, | |
| "learning_rate": 4.1484375e-06, | |
| "loss": 0.7499, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 95.95, | |
| "learning_rate": 3.5624999999999998e-06, | |
| "loss": 0.7511, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 96.62, | |
| "learning_rate": 2.9765625e-06, | |
| "loss": 0.7652, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 97.3, | |
| "learning_rate": 2.3906249999999997e-06, | |
| "loss": 0.7596, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 97.97, | |
| "learning_rate": 1.8046874999999998e-06, | |
| "loss": 0.7502, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 97.97, | |
| "eval_loss": 0.5409459471702576, | |
| "eval_runtime": 94.0294, | |
| "eval_samples_per_second": 22.28, | |
| "eval_steps_per_second": 0.702, | |
| "eval_wer": 1.0191413727098715, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 98.65, | |
| "learning_rate": 1.21875e-06, | |
| "loss": 0.7587, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 99.32, | |
| "learning_rate": 6.328125e-07, | |
| "loss": 0.7614, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 4.6874999999999995e-08, | |
| "loss": 0.7354, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 14800, | |
| "total_flos": 6.011583745907916e+19, | |
| "train_loss": 1.4159005551724821, | |
| "train_runtime": 24218.0743, | |
| "train_samples_per_second": 19.452, | |
| "train_steps_per_second": 0.611 | |
| } | |
| ], | |
| "max_steps": 14800, | |
| "num_train_epochs": 100, | |
| "total_flos": 6.011583745907916e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |