| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.99830220713073, | |
| "global_step": 14700, | |
| "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": 11.6851, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 7.3875e-06, | |
| "loss": 3.4316, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.1137499999999998e-05, | |
| "loss": 2.9224, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.48875e-05, | |
| "loss": 1.9859, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_loss": 1.1663333177566528, | |
| "eval_runtime": 129.56, | |
| "eval_samples_per_second": 16.17, | |
| "eval_steps_per_second": 1.011, | |
| "eval_wer": 0.7947625223889815, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 1.86375e-05, | |
| "loss": 1.625, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 2.23875e-05, | |
| "loss": 1.4408, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 2.6137499999999995e-05, | |
| "loss": 1.3534, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 2.9887499999999998e-05, | |
| "loss": 1.2969, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "eval_loss": 0.7725260853767395, | |
| "eval_runtime": 128.7049, | |
| "eval_samples_per_second": 16.278, | |
| "eval_steps_per_second": 1.018, | |
| "eval_wer": 0.656228769069236, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3.36375e-05, | |
| "loss": 1.2621, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 3.7387499999999994e-05, | |
| "loss": 1.2048, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 4.11375e-05, | |
| "loss": 1.1914, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 4.48875e-05, | |
| "loss": 1.1954, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "eval_loss": 0.5940413475036621, | |
| "eval_runtime": 129.7001, | |
| "eval_samples_per_second": 16.153, | |
| "eval_steps_per_second": 1.01, | |
| "eval_wer": 0.4903958989562102, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 4.8637499999999996e-05, | |
| "loss": 1.1675, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 5.23875e-05, | |
| "loss": 1.1484, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 5.61375e-05, | |
| "loss": 1.1564, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 5.988749999999999e-05, | |
| "loss": 1.164, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "eval_loss": 0.5338293313980103, | |
| "eval_runtime": 129.1415, | |
| "eval_samples_per_second": 16.223, | |
| "eval_steps_per_second": 1.014, | |
| "eval_wer": 0.4315978012476067, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 6.36375e-05, | |
| "loss": 1.1467, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 6.738749999999999e-05, | |
| "loss": 1.1664, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "learning_rate": 7.11375e-05, | |
| "loss": 1.1658, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 1.1464, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "eval_loss": 0.5432471632957458, | |
| "eval_runtime": 128.5422, | |
| "eval_samples_per_second": 16.298, | |
| "eval_steps_per_second": 1.019, | |
| "eval_wer": 0.42264220863442653, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 7.44271653543307e-05, | |
| "loss": 1.1256, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 7.383661417322834e-05, | |
| "loss": 1.1464, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 7.324606299212597e-05, | |
| "loss": 1.1545, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 16.33, | |
| "learning_rate": 7.265551181102362e-05, | |
| "loss": 1.1553, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.33, | |
| "eval_loss": 0.5470633506774902, | |
| "eval_runtime": 128.2784, | |
| "eval_samples_per_second": 16.332, | |
| "eval_steps_per_second": 1.021, | |
| "eval_wer": 0.42603915755666727, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 7.206496062992125e-05, | |
| "loss": 1.1332, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "learning_rate": 7.14744094488189e-05, | |
| "loss": 1.1317, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.37, | |
| "learning_rate": 7.088385826771653e-05, | |
| "loss": 1.1197, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 7.029330708661417e-05, | |
| "loss": 1.0985, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "eval_loss": 0.5289891958236694, | |
| "eval_runtime": 128.1965, | |
| "eval_samples_per_second": 16.342, | |
| "eval_steps_per_second": 1.022, | |
| "eval_wer": 0.407572107961213, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 6.970275590551181e-05, | |
| "loss": 1.0637, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.41, | |
| "learning_rate": 6.911220472440943e-05, | |
| "loss": 1.0786, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.09, | |
| "learning_rate": 6.852165354330708e-05, | |
| "loss": 1.0638, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "learning_rate": 6.793110236220472e-05, | |
| "loss": 1.0421, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "eval_loss": 0.5671822428703308, | |
| "eval_runtime": 128.6139, | |
| "eval_samples_per_second": 16.289, | |
| "eval_steps_per_second": 1.019, | |
| "eval_wer": 0.4181335309739979, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 22.45, | |
| "learning_rate": 6.734055118110236e-05, | |
| "loss": 1.0587, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 23.13, | |
| "learning_rate": 6.675e-05, | |
| "loss": 1.0431, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "learning_rate": 6.615944881889763e-05, | |
| "loss": 1.0282, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 24.49, | |
| "learning_rate": 6.556889763779528e-05, | |
| "loss": 0.9831, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 24.49, | |
| "eval_loss": 0.5740545392036438, | |
| "eval_runtime": 128.3992, | |
| "eval_samples_per_second": 16.316, | |
| "eval_steps_per_second": 1.02, | |
| "eval_wer": 0.41405719226730897, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 25.17, | |
| "learning_rate": 6.497834645669291e-05, | |
| "loss": 1.0409, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "learning_rate": 6.438779527559054e-05, | |
| "loss": 1.0023, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 26.53, | |
| "learning_rate": 6.379724409448818e-05, | |
| "loss": 0.9724, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 27.21, | |
| "learning_rate": 6.320669291338583e-05, | |
| "loss": 0.9827, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 27.21, | |
| "eval_loss": 0.5753942131996155, | |
| "eval_runtime": 129.7692, | |
| "eval_samples_per_second": 16.144, | |
| "eval_steps_per_second": 1.009, | |
| "eval_wer": 0.4178864801432895, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 27.89, | |
| "learning_rate": 6.261614173228346e-05, | |
| "loss": 0.9974, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 6.203149606299212e-05, | |
| "loss": 0.9666, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 6.144685039370079e-05, | |
| "loss": 0.9571, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 29.93, | |
| "learning_rate": 6.085629921259842e-05, | |
| "loss": 0.9669, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 29.93, | |
| "eval_loss": 0.5309818387031555, | |
| "eval_runtime": 127.6098, | |
| "eval_samples_per_second": 16.417, | |
| "eval_steps_per_second": 1.027, | |
| "eval_wer": 0.38885800753505034, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 30.61, | |
| "learning_rate": 6.026574803149606e-05, | |
| "loss": 0.9642, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 31.29, | |
| "learning_rate": 5.967519685039369e-05, | |
| "loss": 0.9843, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 31.97, | |
| "learning_rate": 5.908464566929133e-05, | |
| "loss": 0.9589, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 32.65, | |
| "learning_rate": 5.8494094488188974e-05, | |
| "loss": 0.9496, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 32.65, | |
| "eval_loss": 0.5648738145828247, | |
| "eval_runtime": 128.3031, | |
| "eval_samples_per_second": 16.329, | |
| "eval_steps_per_second": 1.021, | |
| "eval_wer": 0.4061515656846396, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 5.7903543307086614e-05, | |
| "loss": 0.9264, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "learning_rate": 5.731299212598425e-05, | |
| "loss": 0.9117, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 34.69, | |
| "learning_rate": 5.672244094488189e-05, | |
| "loss": 0.9176, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 35.37, | |
| "learning_rate": 5.613188976377952e-05, | |
| "loss": 0.9112, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 35.37, | |
| "eval_loss": 0.5737842321395874, | |
| "eval_runtime": 127.713, | |
| "eval_samples_per_second": 16.404, | |
| "eval_steps_per_second": 1.026, | |
| "eval_wer": 0.3925637699956766, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 36.05, | |
| "learning_rate": 5.5541338582677156e-05, | |
| "loss": 0.903, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 36.73, | |
| "learning_rate": 5.49507874015748e-05, | |
| "loss": 0.8615, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 37.41, | |
| "learning_rate": 5.436023622047244e-05, | |
| "loss": 0.904, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 38.1, | |
| "learning_rate": 5.376968503937008e-05, | |
| "loss": 0.8838, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 38.1, | |
| "eval_loss": 0.5232195258140564, | |
| "eval_runtime": 127.5056, | |
| "eval_samples_per_second": 16.431, | |
| "eval_steps_per_second": 1.027, | |
| "eval_wer": 0.37681427953801494, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 38.77, | |
| "learning_rate": 5.317913385826771e-05, | |
| "loss": 0.8693, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 39.46, | |
| "learning_rate": 5.258858267716535e-05, | |
| "loss": 0.8653, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 40.14, | |
| "learning_rate": 5.1998031496062986e-05, | |
| "loss": 0.8862, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 40.81, | |
| "learning_rate": 5.140748031496062e-05, | |
| "loss": 0.8666, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 40.81, | |
| "eval_loss": 0.5510496497154236, | |
| "eval_runtime": 128.5402, | |
| "eval_samples_per_second": 16.298, | |
| "eval_steps_per_second": 1.019, | |
| "eval_wer": 0.3852140077821012, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 41.5, | |
| "learning_rate": 5.081692913385826e-05, | |
| "loss": 0.8378, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 42.18, | |
| "learning_rate": 5.02263779527559e-05, | |
| "loss": 0.8636, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 42.86, | |
| "learning_rate": 4.963582677165354e-05, | |
| "loss": 0.8464, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 43.54, | |
| "learning_rate": 4.9045275590551175e-05, | |
| "loss": 0.8366, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 43.54, | |
| "eval_loss": 0.5436074733734131, | |
| "eval_runtime": 127.7975, | |
| "eval_samples_per_second": 16.393, | |
| "eval_steps_per_second": 1.025, | |
| "eval_wer": 0.3837317027978507, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 44.22, | |
| "learning_rate": 4.846062992125984e-05, | |
| "loss": 0.8298, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.9, | |
| "learning_rate": 4.7870078740157475e-05, | |
| "loss": 0.8385, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 45.58, | |
| "learning_rate": 4.727952755905511e-05, | |
| "loss": 0.8347, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 46.26, | |
| "learning_rate": 4.668897637795275e-05, | |
| "loss": 0.7957, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 46.26, | |
| "eval_loss": 0.5337450504302979, | |
| "eval_runtime": 128.0855, | |
| "eval_samples_per_second": 16.356, | |
| "eval_steps_per_second": 1.023, | |
| "eval_wer": 0.3774936693224631, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 46.94, | |
| "learning_rate": 4.609842519685039e-05, | |
| "loss": 0.8226, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 47.62, | |
| "learning_rate": 4.550787401574803e-05, | |
| "loss": 0.8228, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 48.3, | |
| "learning_rate": 4.4917322834645664e-05, | |
| "loss": 0.815, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 48.98, | |
| "learning_rate": 4.4326771653543305e-05, | |
| "loss": 0.7834, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 48.98, | |
| "eval_loss": 0.561106264591217, | |
| "eval_runtime": 127.7689, | |
| "eval_samples_per_second": 16.397, | |
| "eval_steps_per_second": 1.025, | |
| "eval_wer": 0.38441109258229883, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 49.66, | |
| "learning_rate": 4.3736220472440945e-05, | |
| "loss": 0.7998, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 50.34, | |
| "learning_rate": 4.314566929133857e-05, | |
| "loss": 0.7898, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 51.02, | |
| "learning_rate": 4.255511811023621e-05, | |
| "loss": 0.7567, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 51.7, | |
| "learning_rate": 4.1964566929133854e-05, | |
| "loss": 0.7685, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 51.7, | |
| "eval_loss": 0.571021556854248, | |
| "eval_runtime": 126.8927, | |
| "eval_samples_per_second": 16.51, | |
| "eval_steps_per_second": 1.032, | |
| "eval_wer": 0.40077821011673154, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 52.38, | |
| "learning_rate": 4.1374015748031494e-05, | |
| "loss": 0.7744, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 53.06, | |
| "learning_rate": 4.078346456692913e-05, | |
| "loss": 0.7483, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 53.74, | |
| "learning_rate": 4.019291338582677e-05, | |
| "loss": 0.76, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 54.42, | |
| "learning_rate": 3.960236220472441e-05, | |
| "loss": 0.7431, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 54.42, | |
| "eval_loss": 0.5636317729949951, | |
| "eval_runtime": 128.5647, | |
| "eval_samples_per_second": 16.295, | |
| "eval_steps_per_second": 1.019, | |
| "eval_wer": 0.37261441541597184, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 55.1, | |
| "learning_rate": 3.901181102362205e-05, | |
| "loss": 0.746, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 55.78, | |
| "learning_rate": 3.842125984251968e-05, | |
| "loss": 0.7162, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 56.46, | |
| "learning_rate": 3.783070866141732e-05, | |
| "loss": 0.7253, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 57.14, | |
| "learning_rate": 3.724015748031496e-05, | |
| "loss": 0.7353, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 57.14, | |
| "eval_loss": 0.5937429070472717, | |
| "eval_runtime": 127.7429, | |
| "eval_samples_per_second": 16.4, | |
| "eval_steps_per_second": 1.025, | |
| "eval_wer": 0.3836081773824965, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 57.82, | |
| "learning_rate": 3.664960629921259e-05, | |
| "loss": 0.6947, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 58.5, | |
| "learning_rate": 3.605905511811023e-05, | |
| "loss": 0.6965, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 59.18, | |
| "learning_rate": 3.546850393700787e-05, | |
| "loss": 0.6975, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 59.86, | |
| "learning_rate": 3.4877952755905506e-05, | |
| "loss": 0.7001, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 59.86, | |
| "eval_loss": 0.5815082788467407, | |
| "eval_runtime": 127.9156, | |
| "eval_samples_per_second": 16.378, | |
| "eval_steps_per_second": 1.024, | |
| "eval_wer": 0.38583163485887223, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 60.54, | |
| "learning_rate": 3.428740157480315e-05, | |
| "loss": 0.694, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 61.22, | |
| "learning_rate": 3.369685039370079e-05, | |
| "loss": 0.6867, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 61.9, | |
| "learning_rate": 3.310629921259842e-05, | |
| "loss": 0.7046, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 62.58, | |
| "learning_rate": 3.2515748031496055e-05, | |
| "loss": 0.6799, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 62.58, | |
| "eval_loss": 0.5861709713935852, | |
| "eval_runtime": 127.4436, | |
| "eval_samples_per_second": 16.439, | |
| "eval_steps_per_second": 1.028, | |
| "eval_wer": 0.36964980544747084, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 63.26, | |
| "learning_rate": 3.1925196850393696e-05, | |
| "loss": 0.6639, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 63.94, | |
| "learning_rate": 3.1334645669291336e-05, | |
| "loss": 0.6543, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 64.62, | |
| "learning_rate": 3.074409448818898e-05, | |
| "loss": 0.6746, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 65.31, | |
| "learning_rate": 3.015354330708661e-05, | |
| "loss": 0.6459, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 65.31, | |
| "eval_loss": 0.6180748343467712, | |
| "eval_runtime": 127.2163, | |
| "eval_samples_per_second": 16.468, | |
| "eval_steps_per_second": 1.03, | |
| "eval_wer": 0.3761966524612439, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 65.98, | |
| "learning_rate": 2.9562992125984248e-05, | |
| "loss": 0.625, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "learning_rate": 2.897244094488189e-05, | |
| "loss": 0.6256, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 67.35, | |
| "learning_rate": 2.8381889763779526e-05, | |
| "loss": 0.6284, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 68.03, | |
| "learning_rate": 2.7791338582677163e-05, | |
| "loss": 0.6121, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 68.03, | |
| "eval_loss": 0.5636932253837585, | |
| "eval_runtime": 128.1438, | |
| "eval_samples_per_second": 16.349, | |
| "eval_steps_per_second": 1.022, | |
| "eval_wer": 0.35896485701933173, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 68.71, | |
| "learning_rate": 2.7206692913385825e-05, | |
| "loss": 0.6123, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 69.39, | |
| "learning_rate": 2.6616141732283462e-05, | |
| "loss": 0.605, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 70.07, | |
| "learning_rate": 2.60255905511811e-05, | |
| "loss": 0.6029, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "learning_rate": 2.5435039370078737e-05, | |
| "loss": 0.5942, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "eval_loss": 0.6374208927154541, | |
| "eval_runtime": 127.5478, | |
| "eval_samples_per_second": 16.425, | |
| "eval_steps_per_second": 1.027, | |
| "eval_wer": 0.3881786177506022, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 71.43, | |
| "learning_rate": 2.4844488188976377e-05, | |
| "loss": 0.6081, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 72.11, | |
| "learning_rate": 2.4253937007874014e-05, | |
| "loss": 0.5865, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 72.79, | |
| "learning_rate": 2.3663385826771648e-05, | |
| "loss": 0.5695, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 73.47, | |
| "learning_rate": 2.307283464566929e-05, | |
| "loss": 0.5769, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 73.47, | |
| "eval_loss": 0.601510226726532, | |
| "eval_runtime": 127.2518, | |
| "eval_samples_per_second": 16.463, | |
| "eval_steps_per_second": 1.029, | |
| "eval_wer": 0.3639676363411772, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 74.15, | |
| "learning_rate": 2.2482283464566926e-05, | |
| "loss": 0.5827, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 74.83, | |
| "learning_rate": 2.1891732283464567e-05, | |
| "loss": 0.5751, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 75.51, | |
| "learning_rate": 2.1307086614173225e-05, | |
| "loss": 0.5448, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 76.19, | |
| "learning_rate": 2.0716535433070866e-05, | |
| "loss": 0.5689, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 76.19, | |
| "eval_loss": 0.5669254064559937, | |
| "eval_runtime": 127.9372, | |
| "eval_samples_per_second": 16.375, | |
| "eval_steps_per_second": 1.024, | |
| "eval_wer": 0.3507504168982768, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 76.87, | |
| "learning_rate": 2.0125984251968503e-05, | |
| "loss": 0.5606, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 77.55, | |
| "learning_rate": 1.9535433070866137e-05, | |
| "loss": 0.544, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 78.23, | |
| "learning_rate": 1.8944881889763778e-05, | |
| "loss": 0.5373, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 78.91, | |
| "learning_rate": 1.8354330708661415e-05, | |
| "loss": 0.5461, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 78.91, | |
| "eval_loss": 0.5967420339584351, | |
| "eval_runtime": 127.7784, | |
| "eval_samples_per_second": 16.396, | |
| "eval_steps_per_second": 1.025, | |
| "eval_wer": 0.36205299240318695, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 79.59, | |
| "learning_rate": 1.7763779527559052e-05, | |
| "loss": 0.5342, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 80.27, | |
| "learning_rate": 1.7173228346456693e-05, | |
| "loss": 0.5105, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 80.95, | |
| "learning_rate": 1.658267716535433e-05, | |
| "loss": 0.5289, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 81.63, | |
| "learning_rate": 1.5992125984251967e-05, | |
| "loss": 0.5286, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 81.63, | |
| "eval_loss": 0.5839831829071045, | |
| "eval_runtime": 127.0078, | |
| "eval_samples_per_second": 16.495, | |
| "eval_steps_per_second": 1.031, | |
| "eval_wer": 0.36050892471125934, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 82.31, | |
| "learning_rate": 1.5401574803149604e-05, | |
| "loss": 0.5179, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 82.99, | |
| "learning_rate": 1.4811023622047243e-05, | |
| "loss": 0.502, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 83.67, | |
| "learning_rate": 1.422047244094488e-05, | |
| "loss": 0.5153, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 84.35, | |
| "learning_rate": 1.3629921259842519e-05, | |
| "loss": 0.5057, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 84.35, | |
| "eval_loss": 0.5848367214202881, | |
| "eval_runtime": 128.5953, | |
| "eval_samples_per_second": 16.291, | |
| "eval_steps_per_second": 1.019, | |
| "eval_wer": 0.34889753566796367, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 85.03, | |
| "learning_rate": 1.3039370078740156e-05, | |
| "loss": 0.5021, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 85.71, | |
| "learning_rate": 1.2448818897637795e-05, | |
| "loss": 0.5002, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 86.39, | |
| "learning_rate": 1.1858267716535432e-05, | |
| "loss": 0.495, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 87.07, | |
| "learning_rate": 1.1267716535433071e-05, | |
| "loss": 0.482, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 87.07, | |
| "eval_loss": 0.5860108733177185, | |
| "eval_runtime": 129.85, | |
| "eval_samples_per_second": 16.134, | |
| "eval_steps_per_second": 1.009, | |
| "eval_wer": 0.34877401025260946, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 87.75, | |
| "learning_rate": 1.0677165354330707e-05, | |
| "loss": 0.4651, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 88.43, | |
| "learning_rate": 1.0086614173228345e-05, | |
| "loss": 0.4894, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 89.12, | |
| "learning_rate": 9.496062992125983e-06, | |
| "loss": 0.49, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 89.79, | |
| "learning_rate": 8.905511811023621e-06, | |
| "loss": 0.4655, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 89.79, | |
| "eval_loss": 0.5780399441719055, | |
| "eval_runtime": 126.603, | |
| "eval_samples_per_second": 16.548, | |
| "eval_steps_per_second": 1.035, | |
| "eval_wer": 0.3453152986226916, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 90.48, | |
| "learning_rate": 8.31496062992126e-06, | |
| "loss": 0.4661, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 91.16, | |
| "learning_rate": 7.724409448818898e-06, | |
| "loss": 0.4776, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 91.84, | |
| "learning_rate": 7.133858267716534e-06, | |
| "loss": 0.4697, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 92.52, | |
| "learning_rate": 6.543307086614172e-06, | |
| "loss": 0.4523, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 92.52, | |
| "eval_loss": 0.6150020956993103, | |
| "eval_runtime": 127.4399, | |
| "eval_samples_per_second": 16.439, | |
| "eval_steps_per_second": 1.028, | |
| "eval_wer": 0.353220925205361, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 93.2, | |
| "learning_rate": 5.95275590551181e-06, | |
| "loss": 0.4522, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 93.88, | |
| "learning_rate": 5.362204724409448e-06, | |
| "loss": 0.4595, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 94.56, | |
| "learning_rate": 4.77755905511811e-06, | |
| "loss": 0.4518, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 95.24, | |
| "learning_rate": 4.1870078740157475e-06, | |
| "loss": 0.4422, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 95.24, | |
| "eval_loss": 0.5930356383323669, | |
| "eval_runtime": 128.0708, | |
| "eval_samples_per_second": 16.358, | |
| "eval_steps_per_second": 1.023, | |
| "eval_wer": 0.3451917732073374, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 95.92, | |
| "learning_rate": 3.5964566929133855e-06, | |
| "loss": 0.4558, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 96.6, | |
| "learning_rate": 3.0059055118110235e-06, | |
| "loss": 0.4427, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 97.28, | |
| "learning_rate": 2.415354330708661e-06, | |
| "loss": 0.4263, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 97.96, | |
| "learning_rate": 1.8248031496062992e-06, | |
| "loss": 0.4436, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 97.96, | |
| "eval_loss": 0.5867109894752502, | |
| "eval_runtime": 127.0812, | |
| "eval_samples_per_second": 16.486, | |
| "eval_steps_per_second": 1.031, | |
| "eval_wer": 0.34278302760793034, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 98.64, | |
| "learning_rate": 1.234251968503937e-06, | |
| "loss": 0.447, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 99.32, | |
| "learning_rate": 6.437007874015748e-07, | |
| "loss": 0.4325, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 5.31496062992126e-08, | |
| "loss": 0.421, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 14700, | |
| "total_flos": 1.7835239032674828e+20, | |
| "train_loss": 0.9074438333835732, | |
| "train_runtime": 47510.6321, | |
| "train_samples_per_second": 9.916, | |
| "train_steps_per_second": 0.309 | |
| } | |
| ], | |
| "max_steps": 14700, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.7835239032674828e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |