| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 11300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5.82e-05, | |
| "loss": 10.446, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0001182, | |
| "loss": 3.652, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.00017819999999999997, | |
| "loss": 3.2102, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.0002382, | |
| "loss": 3.1322, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 0.0002982, | |
| "loss": 3.0798, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "eval_loss": 3.001023054122925, | |
| "eval_runtime": 56.1867, | |
| "eval_samples_per_second": 30.31, | |
| "eval_steps_per_second": 0.961, | |
| "eval_wer": 1.0011981566820276, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 0.00029730555555555554, | |
| "loss": 2.5544, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.0002945277777777777, | |
| "loss": 1.8568, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 0.00029174999999999996, | |
| "loss": 1.6575, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.0002889722222222222, | |
| "loss": 1.5125, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.85, | |
| "learning_rate": 0.0002861944444444444, | |
| "loss": 1.4336, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.85, | |
| "eval_loss": 0.8481376767158508, | |
| "eval_runtime": 54.9915, | |
| "eval_samples_per_second": 30.968, | |
| "eval_steps_per_second": 0.982, | |
| "eval_wer": 0.6910599078341014, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 0.0002834166666666666, | |
| "loss": 1.3709, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 0.00028063888888888886, | |
| "loss": 1.3058, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "learning_rate": 0.0002778611111111111, | |
| "loss": 1.2604, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.39, | |
| "learning_rate": 0.0002750833333333333, | |
| "loss": 1.2065, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.27, | |
| "learning_rate": 0.00027230555555555553, | |
| "loss": 1.2062, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.27, | |
| "eval_loss": 0.7312180995941162, | |
| "eval_runtime": 55.5977, | |
| "eval_samples_per_second": 30.631, | |
| "eval_steps_per_second": 0.971, | |
| "eval_wer": 0.6332718894009216, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "learning_rate": 0.00026952777777777777, | |
| "loss": 1.1712, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 0.00026674999999999995, | |
| "loss": 1.1348, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 0.0002639722222222222, | |
| "loss": 1.1077, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.81, | |
| "learning_rate": 0.00026119444444444443, | |
| "loss": 1.0821, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 0.0002584166666666666, | |
| "loss": 1.0481, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "eval_loss": 0.6849815845489502, | |
| "eval_runtime": 54.2403, | |
| "eval_samples_per_second": 31.397, | |
| "eval_steps_per_second": 0.996, | |
| "eval_wer": 0.5358525345622119, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.58, | |
| "learning_rate": 0.0002556388888888889, | |
| "loss": 1.0495, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 0.0002528611111111111, | |
| "loss": 1.0119, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 20.35, | |
| "learning_rate": 0.0002500833333333333, | |
| "loss": 1.0156, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.24, | |
| "learning_rate": 0.00024730555555555557, | |
| "loss": 0.9916, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 22.12, | |
| "learning_rate": 0.00024452777777777776, | |
| "loss": 0.9837, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 22.12, | |
| "eval_loss": 0.6336787343025208, | |
| "eval_runtime": 55.0498, | |
| "eval_samples_per_second": 30.936, | |
| "eval_steps_per_second": 0.981, | |
| "eval_wer": 0.5316129032258065, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 0.00024174999999999997, | |
| "loss": 0.9558, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 23.89, | |
| "learning_rate": 0.0002389722222222222, | |
| "loss": 0.9523, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 24.78, | |
| "learning_rate": 0.00023619444444444442, | |
| "loss": 0.946, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 25.66, | |
| "learning_rate": 0.00023341666666666663, | |
| "loss": 0.909, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 26.55, | |
| "learning_rate": 0.00023063888888888887, | |
| "loss": 0.9108, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 26.55, | |
| "eval_loss": 0.6257887482643127, | |
| "eval_runtime": 55.6823, | |
| "eval_samples_per_second": 30.584, | |
| "eval_steps_per_second": 0.97, | |
| "eval_wer": 0.507926267281106, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 27.43, | |
| "learning_rate": 0.00022786111111111108, | |
| "loss": 0.9005, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 28.32, | |
| "learning_rate": 0.00022508333333333332, | |
| "loss": 0.9069, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 29.2, | |
| "learning_rate": 0.00022230555555555553, | |
| "loss": 0.8605, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 30.09, | |
| "learning_rate": 0.00021952777777777774, | |
| "loss": 0.8815, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 0.00021674999999999998, | |
| "loss": 0.8439, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_loss": 0.6301265954971313, | |
| "eval_runtime": 54.5193, | |
| "eval_samples_per_second": 31.237, | |
| "eval_steps_per_second": 0.99, | |
| "eval_wer": 0.48884792626728113, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 31.86, | |
| "learning_rate": 0.0002139722222222222, | |
| "loss": 0.8522, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 32.74, | |
| "learning_rate": 0.0002111944444444444, | |
| "loss": 0.8477, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 33.63, | |
| "learning_rate": 0.00020841666666666665, | |
| "loss": 0.7978, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 34.51, | |
| "learning_rate": 0.00020563888888888886, | |
| "loss": 0.8127, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 35.4, | |
| "learning_rate": 0.0002028611111111111, | |
| "loss": 0.7901, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 35.4, | |
| "eval_loss": 0.6244927048683167, | |
| "eval_runtime": 53.6992, | |
| "eval_samples_per_second": 31.714, | |
| "eval_steps_per_second": 1.006, | |
| "eval_wer": 0.4976958525345622, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 36.28, | |
| "learning_rate": 0.0002000833333333333, | |
| "loss": 0.7978, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 37.17, | |
| "learning_rate": 0.00019730555555555552, | |
| "loss": 0.8046, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 38.05, | |
| "learning_rate": 0.00019452777777777776, | |
| "loss": 0.7892, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 38.94, | |
| "learning_rate": 0.00019174999999999997, | |
| "loss": 0.7657, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 39.82, | |
| "learning_rate": 0.00018899999999999999, | |
| "loss": 0.7669, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 39.82, | |
| "eval_loss": 0.6164370775222778, | |
| "eval_runtime": 54.5713, | |
| "eval_samples_per_second": 31.207, | |
| "eval_steps_per_second": 0.99, | |
| "eval_wer": 0.4671889400921659, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 40.71, | |
| "learning_rate": 0.0001862222222222222, | |
| "loss": 0.7389, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 41.59, | |
| "learning_rate": 0.00018344444444444444, | |
| "loss": 0.7595, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 42.48, | |
| "learning_rate": 0.00018066666666666665, | |
| "loss": 0.7264, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 43.36, | |
| "learning_rate": 0.00017788888888888886, | |
| "loss": 0.7251, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 44.25, | |
| "learning_rate": 0.0001751111111111111, | |
| "loss": 0.7196, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 44.25, | |
| "eval_loss": 0.6039230227470398, | |
| "eval_runtime": 55.8782, | |
| "eval_samples_per_second": 30.477, | |
| "eval_steps_per_second": 0.966, | |
| "eval_wer": 0.4688479262672811, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 45.13, | |
| "learning_rate": 0.0001723333333333333, | |
| "loss": 0.6932, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 46.02, | |
| "learning_rate": 0.00016955555555555555, | |
| "loss": 0.6983, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 46.9, | |
| "learning_rate": 0.00016677777777777776, | |
| "loss": 0.6769, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 47.79, | |
| "learning_rate": 0.00016399999999999997, | |
| "loss": 0.6928, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 48.67, | |
| "learning_rate": 0.00016122222222222221, | |
| "loss": 0.6715, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 48.67, | |
| "eval_loss": 0.5900057554244995, | |
| "eval_runtime": 53.576, | |
| "eval_samples_per_second": 31.787, | |
| "eval_steps_per_second": 1.008, | |
| "eval_wer": 0.45732718894009217, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 49.56, | |
| "learning_rate": 0.00015849999999999998, | |
| "loss": 0.6833, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 50.44, | |
| "learning_rate": 0.0001557222222222222, | |
| "loss": 0.6673, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 51.33, | |
| "learning_rate": 0.00015294444444444443, | |
| "loss": 0.6791, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 52.21, | |
| "learning_rate": 0.00015016666666666664, | |
| "loss": 0.6292, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 53.1, | |
| "learning_rate": 0.00014738888888888888, | |
| "loss": 0.6441, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 53.1, | |
| "eval_loss": 0.7002069354057312, | |
| "eval_runtime": 55.3769, | |
| "eval_samples_per_second": 30.753, | |
| "eval_steps_per_second": 0.975, | |
| "eval_wer": 0.479815668202765, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 53.98, | |
| "learning_rate": 0.0001446111111111111, | |
| "loss": 0.6217, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 54.87, | |
| "learning_rate": 0.00014183333333333333, | |
| "loss": 0.6131, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 55.75, | |
| "learning_rate": 0.00013905555555555554, | |
| "loss": 0.6211, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 56.64, | |
| "learning_rate": 0.00013627777777777775, | |
| "loss": 0.6104, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 57.52, | |
| "learning_rate": 0.0001335, | |
| "loss": 0.5938, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 57.52, | |
| "eval_loss": 0.6249451637268066, | |
| "eval_runtime": 55.668, | |
| "eval_samples_per_second": 30.592, | |
| "eval_steps_per_second": 0.97, | |
| "eval_wer": 0.4578801843317972, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 58.41, | |
| "learning_rate": 0.0001307222222222222, | |
| "loss": 0.6015, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 59.29, | |
| "learning_rate": 0.00012794444444444442, | |
| "loss": 0.591, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 60.18, | |
| "learning_rate": 0.00012516666666666666, | |
| "loss": 0.5734, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 61.06, | |
| "learning_rate": 0.0001223888888888889, | |
| "loss": 0.5685, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 61.95, | |
| "learning_rate": 0.0001196111111111111, | |
| "loss": 0.5541, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 61.95, | |
| "eval_loss": 0.6184473633766174, | |
| "eval_runtime": 55.2876, | |
| "eval_samples_per_second": 30.803, | |
| "eval_steps_per_second": 0.977, | |
| "eval_wer": 0.4424884792626728, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 62.83, | |
| "learning_rate": 0.00011683333333333332, | |
| "loss": 0.5546, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 63.72, | |
| "learning_rate": 0.00011405555555555554, | |
| "loss": 0.5473, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 64.6, | |
| "learning_rate": 0.00011127777777777777, | |
| "loss": 0.5592, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 65.49, | |
| "learning_rate": 0.0001085, | |
| "loss": 0.5349, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 66.37, | |
| "learning_rate": 0.0001057222222222222, | |
| "loss": 0.5506, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 66.37, | |
| "eval_loss": 0.6962713003158569, | |
| "eval_runtime": 55.4374, | |
| "eval_samples_per_second": 30.719, | |
| "eval_steps_per_second": 0.974, | |
| "eval_wer": 0.45852534562211983, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 67.26, | |
| "learning_rate": 0.00010294444444444443, | |
| "loss": 0.5313, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 68.14, | |
| "learning_rate": 0.00010016666666666666, | |
| "loss": 0.5267, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 69.03, | |
| "learning_rate": 9.738888888888888e-05, | |
| "loss": 0.5222, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 69.91, | |
| "learning_rate": 9.46111111111111e-05, | |
| "loss": 0.5101, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 70.8, | |
| "learning_rate": 9.183333333333332e-05, | |
| "loss": 0.4998, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 70.8, | |
| "eval_loss": 0.6778160333633423, | |
| "eval_runtime": 56.1738, | |
| "eval_samples_per_second": 30.317, | |
| "eval_steps_per_second": 0.961, | |
| "eval_wer": 0.44682027649769585, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 71.68, | |
| "learning_rate": 8.905555555555555e-05, | |
| "loss": 0.4941, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 72.57, | |
| "learning_rate": 8.627777777777776e-05, | |
| "loss": 0.492, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 73.45, | |
| "learning_rate": 8.349999999999998e-05, | |
| "loss": 0.4741, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 74.34, | |
| "learning_rate": 8.072222222222222e-05, | |
| "loss": 0.495, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 75.22, | |
| "learning_rate": 7.794444444444445e-05, | |
| "loss": 0.4729, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 75.22, | |
| "eval_loss": 0.6383044719696045, | |
| "eval_runtime": 53.9538, | |
| "eval_samples_per_second": 31.564, | |
| "eval_steps_per_second": 1.001, | |
| "eval_wer": 0.4392626728110599, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 76.11, | |
| "learning_rate": 7.516666666666665e-05, | |
| "loss": 0.4696, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 76.99, | |
| "learning_rate": 7.238888888888889e-05, | |
| "loss": 0.4581, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 77.88, | |
| "learning_rate": 6.961111111111111e-05, | |
| "loss": 0.4583, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 78.76, | |
| "learning_rate": 6.683333333333332e-05, | |
| "loss": 0.4451, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 79.65, | |
| "learning_rate": 6.405555555555555e-05, | |
| "loss": 0.4535, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 79.65, | |
| "eval_loss": 0.6592639684677124, | |
| "eval_runtime": 53.7527, | |
| "eval_samples_per_second": 31.682, | |
| "eval_steps_per_second": 1.005, | |
| "eval_wer": 0.4368663594470046, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 80.53, | |
| "learning_rate": 6.130555555555555e-05, | |
| "loss": 0.4324, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 81.42, | |
| "learning_rate": 5.8527777777777774e-05, | |
| "loss": 0.4546, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 82.3, | |
| "learning_rate": 5.574999999999999e-05, | |
| "loss": 0.4391, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 83.19, | |
| "learning_rate": 5.297222222222222e-05, | |
| "loss": 0.4306, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 84.07, | |
| "learning_rate": 5.019444444444444e-05, | |
| "loss": 0.4358, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 84.07, | |
| "eval_loss": 0.6913911700248718, | |
| "eval_runtime": 55.4673, | |
| "eval_samples_per_second": 30.703, | |
| "eval_steps_per_second": 0.974, | |
| "eval_wer": 0.4422119815668203, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 84.96, | |
| "learning_rate": 4.741666666666666e-05, | |
| "loss": 0.4095, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 85.84, | |
| "learning_rate": 4.463888888888888e-05, | |
| "loss": 0.4148, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 86.73, | |
| "learning_rate": 4.186111111111111e-05, | |
| "loss": 0.4113, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 87.61, | |
| "learning_rate": 3.9083333333333326e-05, | |
| "loss": 0.405, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 88.5, | |
| "learning_rate": 3.630555555555555e-05, | |
| "loss": 0.402, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 88.5, | |
| "eval_loss": 0.6743763089179993, | |
| "eval_runtime": 53.7113, | |
| "eval_samples_per_second": 31.707, | |
| "eval_steps_per_second": 1.005, | |
| "eval_wer": 0.4269124423963134, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 89.38, | |
| "learning_rate": 3.352777777777777e-05, | |
| "loss": 0.3915, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 90.27, | |
| "learning_rate": 3.0749999999999995e-05, | |
| "loss": 0.394, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 91.15, | |
| "learning_rate": 2.7972222222222217e-05, | |
| "loss": 0.392, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 92.04, | |
| "learning_rate": 2.519444444444444e-05, | |
| "loss": 0.3937, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 92.92, | |
| "learning_rate": 2.2416666666666665e-05, | |
| "loss": 0.3946, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 92.92, | |
| "eval_loss": 0.6894700527191162, | |
| "eval_runtime": 55.0145, | |
| "eval_samples_per_second": 30.955, | |
| "eval_steps_per_second": 0.982, | |
| "eval_wer": 0.42746543778801843, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 93.81, | |
| "learning_rate": 1.9638888888888887e-05, | |
| "loss": 0.3881, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 94.69, | |
| "learning_rate": 1.686111111111111e-05, | |
| "loss": 0.3778, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 95.58, | |
| "learning_rate": 1.4083333333333331e-05, | |
| "loss": 0.3783, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 96.46, | |
| "learning_rate": 1.1305555555555553e-05, | |
| "loss": 0.3612, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 97.35, | |
| "learning_rate": 8.527777777777777e-06, | |
| "loss": 0.3734, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 97.35, | |
| "eval_loss": 0.6888979077339172, | |
| "eval_runtime": 55.289, | |
| "eval_samples_per_second": 30.802, | |
| "eval_steps_per_second": 0.977, | |
| "eval_wer": 0.4319815668202765, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 98.23, | |
| "learning_rate": 5.749999999999999e-06, | |
| "loss": 0.3635, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 99.12, | |
| "learning_rate": 2.9999999999999997e-06, | |
| "loss": 0.3775, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 2.222222222222222e-07, | |
| "loss": 0.3737, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 11300, | |
| "total_flos": 4.438667922365353e+19, | |
| "train_loss": 0.9041624231254105, | |
| "train_runtime": 18573.3759, | |
| "train_samples_per_second": 19.388, | |
| "train_steps_per_second": 0.608 | |
| } | |
| ], | |
| "max_steps": 11300, | |
| "num_train_epochs": 100, | |
| "total_flos": 4.438667922365353e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |