| { | |
| "best_metric": 63.10532687651331, | |
| "best_model_checkpoint": "./checkpoint-5100", | |
| "epoch": 742.8571428571429, | |
| "global_step": 5200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 1.32e-07, | |
| "loss": 2.8512, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 2.8199999999999996e-07, | |
| "loss": 2.7082, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 2.9305263157894735e-07, | |
| "loss": 2.3515, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 2.851578947368421e-07, | |
| "loss": 2.0871, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "eval_loss": 2.0101583003997803, | |
| "eval_runtime": 720.7636, | |
| "eval_samples_per_second": 0.71, | |
| "eval_steps_per_second": 0.044, | |
| "eval_wer": 230.27391041162227, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 2.7726315789473684e-07, | |
| "loss": 1.8622, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 21.43, | |
| "learning_rate": 2.693684210526316e-07, | |
| "loss": 1.7104, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 2.614736842105263e-07, | |
| "loss": 1.5736, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 2.53578947368421e-07, | |
| "loss": 1.465, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "eval_loss": 1.4968725442886353, | |
| "eval_runtime": 628.5425, | |
| "eval_samples_per_second": 0.815, | |
| "eval_steps_per_second": 0.051, | |
| "eval_wer": 137.24273607748182, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 32.14, | |
| "learning_rate": 2.4568421052631577e-07, | |
| "loss": 1.3669, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 35.71, | |
| "learning_rate": 2.3778947368421054e-07, | |
| "loss": 1.2898, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 39.29, | |
| "learning_rate": 2.2989473684210523e-07, | |
| "loss": 1.2205, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 42.86, | |
| "learning_rate": 2.2199999999999998e-07, | |
| "loss": 1.1617, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 42.86, | |
| "eval_loss": 1.2715740203857422, | |
| "eval_runtime": 499.3378, | |
| "eval_samples_per_second": 1.025, | |
| "eval_steps_per_second": 0.064, | |
| "eval_wer": 76.32415254237289, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 46.43, | |
| "learning_rate": 2.1410526315789472e-07, | |
| "loss": 1.1091, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 2.0621052631578947e-07, | |
| "loss": 1.0738, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 53.57, | |
| "learning_rate": 1.9831578947368419e-07, | |
| "loss": 1.033, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 57.14, | |
| "learning_rate": 1.9042105263157893e-07, | |
| "loss": 1.0019, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 57.14, | |
| "eval_loss": 1.16450834274292, | |
| "eval_runtime": 480.3701, | |
| "eval_samples_per_second": 1.066, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 71.37560532687651, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 60.71, | |
| "learning_rate": 1.8252631578947368e-07, | |
| "loss": 0.9712, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 64.29, | |
| "learning_rate": 1.7463157894736842e-07, | |
| "loss": 0.9437, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 67.86, | |
| "learning_rate": 1.6673684210526314e-07, | |
| "loss": 0.9215, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 71.43, | |
| "learning_rate": 1.588421052631579e-07, | |
| "loss": 0.9052, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 71.43, | |
| "eval_loss": 1.1051170825958252, | |
| "eval_runtime": 486.3374, | |
| "eval_samples_per_second": 1.053, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 69.78662227602905, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 1.5094736842105263e-07, | |
| "loss": 0.8773, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 78.57, | |
| "learning_rate": 1.4305263157894735e-07, | |
| "loss": 0.8643, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 82.14, | |
| "learning_rate": 1.351578947368421e-07, | |
| "loss": 0.8449, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 85.71, | |
| "learning_rate": 1.2726315789473684e-07, | |
| "loss": 0.8334, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 85.71, | |
| "eval_loss": 1.0691100358963013, | |
| "eval_runtime": 478.3463, | |
| "eval_samples_per_second": 1.07, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 68.26573849878935, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 89.29, | |
| "learning_rate": 1.1936842105263156e-07, | |
| "loss": 0.8132, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 92.86, | |
| "learning_rate": 1.1147368421052631e-07, | |
| "loss": 0.8058, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 96.43, | |
| "learning_rate": 1.0357894736842104e-07, | |
| "loss": 0.7913, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 9.568421052631579e-08, | |
| "loss": 0.7838, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 1.0482958555221558, | |
| "eval_runtime": 478.3861, | |
| "eval_samples_per_second": 1.07, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 67.16858353510897, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 103.57, | |
| "learning_rate": 8.778947368421052e-08, | |
| "loss": 0.7768, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 107.14, | |
| "learning_rate": 7.989473684210526e-08, | |
| "loss": 0.7673, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 110.71, | |
| "learning_rate": 7.2e-08, | |
| "loss": 0.7643, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 114.29, | |
| "learning_rate": 6.410526315789473e-08, | |
| "loss": 0.7539, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 114.29, | |
| "eval_loss": 1.0362622737884521, | |
| "eval_runtime": 484.8883, | |
| "eval_samples_per_second": 1.056, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 66.41949152542372, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 117.86, | |
| "learning_rate": 5.621052631578947e-08, | |
| "loss": 0.7527, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 121.43, | |
| "learning_rate": 4.8315789473684206e-08, | |
| "loss": 0.7441, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "learning_rate": 4.0421052631578945e-08, | |
| "loss": 0.7417, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 128.57, | |
| "learning_rate": 3.2526315789473684e-08, | |
| "loss": 0.7377, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 128.57, | |
| "eval_loss": 1.0297424793243408, | |
| "eval_runtime": 471.9816, | |
| "eval_samples_per_second": 1.085, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 66.20006053268766, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 132.14, | |
| "learning_rate": 2.463157894736842e-08, | |
| "loss": 0.7387, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 135.71, | |
| "learning_rate": 1.673684210526316e-08, | |
| "loss": 0.7329, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 139.29, | |
| "learning_rate": 8.842105263157895e-09, | |
| "loss": 0.7312, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 142.86, | |
| "learning_rate": 9.473684210526316e-10, | |
| "loss": 0.7325, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 142.86, | |
| "eval_loss": 1.0276601314544678, | |
| "eval_runtime": 477.2948, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 66.00332929782083, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 146.43, | |
| "learning_rate": 1.5046153846153844e-07, | |
| "loss": 0.7238, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 1.466153846153846e-07, | |
| "loss": 0.7163, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 153.57, | |
| "learning_rate": 1.4276923076923076e-07, | |
| "loss": 0.7028, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 157.14, | |
| "learning_rate": 1.389230769230769e-07, | |
| "loss": 0.6952, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 157.14, | |
| "eval_loss": 1.0121634006500244, | |
| "eval_runtime": 471.3724, | |
| "eval_samples_per_second": 1.086, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 65.05750605326877, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 160.71, | |
| "learning_rate": 1.3507692307692308e-07, | |
| "loss": 0.6843, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 164.29, | |
| "learning_rate": 1.3123076923076923e-07, | |
| "loss": 0.6699, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 167.86, | |
| "learning_rate": 1.2738461538461538e-07, | |
| "loss": 0.6671, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 171.43, | |
| "learning_rate": 1.2353846153846153e-07, | |
| "loss": 0.6531, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 171.43, | |
| "eval_loss": 1.0014406442642212, | |
| "eval_runtime": 475.4519, | |
| "eval_samples_per_second": 1.077, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 64.42191283292978, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 175.0, | |
| "learning_rate": 1.1969230769230767e-07, | |
| "loss": 0.6487, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 178.57, | |
| "learning_rate": 1.1584615384615385e-07, | |
| "loss": 0.6369, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 182.14, | |
| "learning_rate": 1.12e-07, | |
| "loss": 0.6336, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 185.71, | |
| "learning_rate": 1.0815384615384614e-07, | |
| "loss": 0.6189, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 185.71, | |
| "eval_loss": 0.9944669008255005, | |
| "eval_runtime": 470.4039, | |
| "eval_samples_per_second": 1.088, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 63.79388619854721, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 189.29, | |
| "learning_rate": 1.043076923076923e-07, | |
| "loss": 0.6213, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 192.86, | |
| "learning_rate": 1.0046153846153845e-07, | |
| "loss": 0.608, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 196.43, | |
| "learning_rate": 9.66153846153846e-08, | |
| "loss": 0.6029, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 9.276923076923078e-08, | |
| "loss": 0.5993, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "eval_loss": 0.9895604252815247, | |
| "eval_runtime": 473.2431, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 63.35502421307506, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 203.57, | |
| "learning_rate": 8.892307692307692e-08, | |
| "loss": 0.593, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 207.14, | |
| "learning_rate": 8.507692307692307e-08, | |
| "loss": 0.5817, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 210.71, | |
| "learning_rate": 8.123076923076922e-08, | |
| "loss": 0.5782, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 214.29, | |
| "learning_rate": 7.738461538461538e-08, | |
| "loss": 0.5757, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 214.29, | |
| "eval_loss": 0.9864457845687866, | |
| "eval_runtime": 474.414, | |
| "eval_samples_per_second": 1.079, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 63.22639225181598, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 217.86, | |
| "learning_rate": 7.353846153846153e-08, | |
| "loss": 0.5706, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 221.43, | |
| "learning_rate": 6.969230769230769e-08, | |
| "loss": 0.5624, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 225.0, | |
| "learning_rate": 6.584615384615385e-08, | |
| "loss": 0.5638, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 228.57, | |
| "learning_rate": 6.2e-08, | |
| "loss": 0.5601, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 228.57, | |
| "eval_loss": 0.9844600558280945, | |
| "eval_runtime": 478.7212, | |
| "eval_samples_per_second": 1.07, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 62.916162227602904, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 232.14, | |
| "learning_rate": 5.815384615384615e-08, | |
| "loss": 0.5537, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 235.71, | |
| "learning_rate": 5.430769230769231e-08, | |
| "loss": 0.5488, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 239.29, | |
| "learning_rate": 5.0461538461538456e-08, | |
| "loss": 0.5479, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 242.86, | |
| "learning_rate": 4.661538461538461e-08, | |
| "loss": 0.5482, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 242.86, | |
| "eval_loss": 0.9833234548568726, | |
| "eval_runtime": 488.3079, | |
| "eval_samples_per_second": 1.049, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 62.817796610169495, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 246.43, | |
| "learning_rate": 4.2769230769230765e-08, | |
| "loss": 0.5441, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 250.0, | |
| "learning_rate": 3.892307692307692e-08, | |
| "loss": 0.54, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 253.57, | |
| "learning_rate": 3.5076923076923074e-08, | |
| "loss": 0.538, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 257.14, | |
| "learning_rate": 3.123076923076923e-08, | |
| "loss": 0.5382, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 257.14, | |
| "eval_loss": 0.9826769828796387, | |
| "eval_runtime": 486.3686, | |
| "eval_samples_per_second": 1.053, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 62.84049636803874, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 260.71, | |
| "learning_rate": 2.7384615384615387e-08, | |
| "loss": 0.5343, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 264.29, | |
| "learning_rate": 2.3538461538461535e-08, | |
| "loss": 0.5313, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 267.86, | |
| "learning_rate": 1.9692307692307693e-08, | |
| "loss": 0.5318, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 271.43, | |
| "learning_rate": 1.5846153846153844e-08, | |
| "loss": 0.5325, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 271.43, | |
| "eval_loss": 0.9823360443115234, | |
| "eval_runtime": 483.6609, | |
| "eval_samples_per_second": 1.059, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 62.76483050847458, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 275.0, | |
| "learning_rate": 1.2e-08, | |
| "loss": 0.529, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 278.57, | |
| "learning_rate": 8.153846153846154e-09, | |
| "loss": 0.5294, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 282.14, | |
| "learning_rate": 4.307692307692307e-09, | |
| "loss": 0.525, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 285.71, | |
| "learning_rate": 4.615384615384615e-10, | |
| "loss": 0.5287, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 285.71, | |
| "eval_loss": 0.9822061061859131, | |
| "eval_runtime": 484.044, | |
| "eval_samples_per_second": 1.058, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 62.817796610169495, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 289.29, | |
| "learning_rate": 1.9853333333333334e-07, | |
| "loss": 0.523, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 292.86, | |
| "learning_rate": 1.9686666666666667e-07, | |
| "loss": 0.5133, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 296.43, | |
| "learning_rate": 1.9519999999999997e-07, | |
| "loss": 0.4999, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 300.0, | |
| "learning_rate": 1.935333333333333e-07, | |
| "loss": 0.4924, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 303.57, | |
| "learning_rate": 1.9186666666666666e-07, | |
| "loss": 0.4855, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 307.14, | |
| "learning_rate": 1.902e-07, | |
| "loss": 0.4701, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 310.71, | |
| "learning_rate": 1.8853333333333333e-07, | |
| "loss": 0.4601, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 314.29, | |
| "learning_rate": 1.8686666666666669e-07, | |
| "loss": 0.4525, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 317.86, | |
| "learning_rate": 1.852e-07, | |
| "loss": 0.4448, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 321.43, | |
| "learning_rate": 1.8353333333333332e-07, | |
| "loss": 0.4364, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 325.0, | |
| "learning_rate": 1.8186666666666665e-07, | |
| "loss": 0.4232, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 328.57, | |
| "learning_rate": 1.8019999999999999e-07, | |
| "loss": 0.4163, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 332.14, | |
| "learning_rate": 1.7853333333333334e-07, | |
| "loss": 0.4089, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 335.71, | |
| "learning_rate": 1.7686666666666668e-07, | |
| "loss": 0.4031, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 339.29, | |
| "learning_rate": 1.7519999999999998e-07, | |
| "loss": 0.3887, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 342.86, | |
| "learning_rate": 1.735333333333333e-07, | |
| "loss": 0.3826, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 346.43, | |
| "learning_rate": 1.7186666666666667e-07, | |
| "loss": 0.3766, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 350.0, | |
| "learning_rate": 1.702e-07, | |
| "loss": 0.3647, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 353.57, | |
| "learning_rate": 1.6853333333333333e-07, | |
| "loss": 0.3601, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 357.14, | |
| "learning_rate": 1.6686666666666664e-07, | |
| "loss": 0.3494, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 357.14, | |
| "eval_loss": 1.0025562047958374, | |
| "eval_runtime": 489.174, | |
| "eval_samples_per_second": 1.047, | |
| "eval_steps_per_second": 0.065, | |
| "eval_wer": 61.61470944309927, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 360.71, | |
| "learning_rate": 1.652e-07, | |
| "loss": 0.3448, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 364.29, | |
| "learning_rate": 1.6353333333333333e-07, | |
| "loss": 0.3367, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 367.86, | |
| "learning_rate": 1.6186666666666666e-07, | |
| "loss": 0.3308, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 371.43, | |
| "learning_rate": 1.602e-07, | |
| "loss": 0.3226, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 375.0, | |
| "learning_rate": 1.5853333333333335e-07, | |
| "loss": 0.3165, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 378.57, | |
| "learning_rate": 1.5686666666666666e-07, | |
| "loss": 0.3099, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 382.14, | |
| "learning_rate": 1.552e-07, | |
| "loss": 0.3021, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 385.71, | |
| "learning_rate": 1.5353333333333332e-07, | |
| "loss": 0.2964, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 389.29, | |
| "learning_rate": 1.5186666666666668e-07, | |
| "loss": 0.2901, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 392.86, | |
| "learning_rate": 1.502e-07, | |
| "loss": 0.284, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 396.43, | |
| "learning_rate": 1.4853333333333334e-07, | |
| "loss": 0.279, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "learning_rate": 1.4686666666666667e-07, | |
| "loss": 0.2715, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 403.57, | |
| "learning_rate": 1.4519999999999998e-07, | |
| "loss": 0.2646, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 407.14, | |
| "learning_rate": 1.4353333333333333e-07, | |
| "loss": 0.2606, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 410.71, | |
| "learning_rate": 1.4186666666666667e-07, | |
| "loss": 0.2564, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 414.29, | |
| "learning_rate": 1.402e-07, | |
| "loss": 0.2486, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 417.86, | |
| "learning_rate": 1.3853333333333333e-07, | |
| "loss": 0.2463, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 421.43, | |
| "learning_rate": 1.3686666666666666e-07, | |
| "loss": 0.239, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 425.0, | |
| "learning_rate": 1.352e-07, | |
| "loss": 0.2341, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 428.57, | |
| "learning_rate": 1.3353333333333332e-07, | |
| "loss": 0.2287, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 428.57, | |
| "eval_loss": 1.0533033609390259, | |
| "eval_runtime": 465.8233, | |
| "eval_samples_per_second": 1.099, | |
| "eval_steps_per_second": 0.069, | |
| "eval_wer": 61.516343825665864, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 432.14, | |
| "learning_rate": 1.3186666666666666e-07, | |
| "loss": 0.2231, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 435.71, | |
| "learning_rate": 1.302e-07, | |
| "loss": 0.2225, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 439.29, | |
| "learning_rate": 1.2853333333333332e-07, | |
| "loss": 0.216, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 442.86, | |
| "learning_rate": 1.2686666666666665e-07, | |
| "loss": 0.2099, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 446.43, | |
| "learning_rate": 1.2519999999999998e-07, | |
| "loss": 0.2053, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 450.0, | |
| "learning_rate": 1.2353333333333334e-07, | |
| "loss": 0.2034, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 453.57, | |
| "learning_rate": 1.2186666666666665e-07, | |
| "loss": 0.1979, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 457.14, | |
| "learning_rate": 1.202e-07, | |
| "loss": 0.1938, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 460.71, | |
| "learning_rate": 1.1853333333333334e-07, | |
| "loss": 0.19, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 464.29, | |
| "learning_rate": 1.1686666666666665e-07, | |
| "loss": 0.1877, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 467.86, | |
| "learning_rate": 1.152e-07, | |
| "loss": 0.1814, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 471.43, | |
| "learning_rate": 1.1353333333333333e-07, | |
| "loss": 0.1785, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 475.0, | |
| "learning_rate": 1.1186666666666666e-07, | |
| "loss": 0.1738, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 478.57, | |
| "learning_rate": 1.102e-07, | |
| "loss": 0.1716, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 482.14, | |
| "learning_rate": 1.0853333333333333e-07, | |
| "loss": 0.1666, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 485.71, | |
| "learning_rate": 1.0686666666666666e-07, | |
| "loss": 0.1663, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 489.29, | |
| "learning_rate": 1.052e-07, | |
| "loss": 0.1611, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 492.86, | |
| "learning_rate": 1.0353333333333332e-07, | |
| "loss": 0.158, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 496.43, | |
| "learning_rate": 1.0186666666666667e-07, | |
| "loss": 0.1556, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "learning_rate": 1.002e-07, | |
| "loss": 0.1525, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "eval_loss": 1.1040765047073364, | |
| "eval_runtime": 467.1311, | |
| "eval_samples_per_second": 1.096, | |
| "eval_steps_per_second": 0.069, | |
| "eval_wer": 62.05357142857143, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 503.57, | |
| "learning_rate": 9.853333333333333e-08, | |
| "loss": 0.1499, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 507.14, | |
| "learning_rate": 9.686666666666666e-08, | |
| "loss": 0.1474, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 510.71, | |
| "learning_rate": 9.52e-08, | |
| "loss": 0.1429, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 514.29, | |
| "learning_rate": 9.353333333333332e-08, | |
| "loss": 0.1407, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 517.86, | |
| "learning_rate": 9.186666666666667e-08, | |
| "loss": 0.1394, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 521.43, | |
| "learning_rate": 9.02e-08, | |
| "loss": 0.136, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 525.0, | |
| "learning_rate": 8.853333333333333e-08, | |
| "loss": 0.1347, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 528.57, | |
| "learning_rate": 8.686666666666666e-08, | |
| "loss": 0.1307, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 532.14, | |
| "learning_rate": 8.519999999999998e-08, | |
| "loss": 0.1297, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 535.71, | |
| "learning_rate": 8.353333333333333e-08, | |
| "loss": 0.1282, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 539.29, | |
| "learning_rate": 8.186666666666667e-08, | |
| "loss": 0.1247, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 542.86, | |
| "learning_rate": 8.019999999999999e-08, | |
| "loss": 0.1225, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 546.43, | |
| "learning_rate": 7.853333333333334e-08, | |
| "loss": 0.1206, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 550.0, | |
| "learning_rate": 7.686666666666667e-08, | |
| "loss": 0.1197, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 553.57, | |
| "learning_rate": 7.519999999999999e-08, | |
| "loss": 0.117, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 557.14, | |
| "learning_rate": 7.353333333333333e-08, | |
| "loss": 0.1154, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 560.71, | |
| "learning_rate": 7.186666666666666e-08, | |
| "loss": 0.1128, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 564.29, | |
| "learning_rate": 7.02e-08, | |
| "loss": 0.1133, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 567.86, | |
| "learning_rate": 6.853333333333334e-08, | |
| "loss": 0.1091, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 571.43, | |
| "learning_rate": 6.686666666666666e-08, | |
| "loss": 0.1089, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 571.43, | |
| "eval_loss": 1.1451234817504883, | |
| "eval_runtime": 481.8401, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 62.50756658595642, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 575.0, | |
| "learning_rate": 6.519999999999999e-08, | |
| "loss": 0.1069, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 578.57, | |
| "learning_rate": 6.353333333333333e-08, | |
| "loss": 0.1055, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 582.14, | |
| "learning_rate": 6.186666666666666e-08, | |
| "loss": 0.1043, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 585.71, | |
| "learning_rate": 6.02e-08, | |
| "loss": 0.1031, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 589.29, | |
| "learning_rate": 5.8533333333333335e-08, | |
| "loss": 0.1021, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 592.86, | |
| "learning_rate": 5.6866666666666666e-08, | |
| "loss": 0.1, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 596.43, | |
| "learning_rate": 5.52e-08, | |
| "loss": 0.0987, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 600.0, | |
| "learning_rate": 5.353333333333333e-08, | |
| "loss": 0.0984, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 603.57, | |
| "learning_rate": 5.186666666666667e-08, | |
| "loss": 0.0959, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 607.14, | |
| "learning_rate": 5.02e-08, | |
| "loss": 0.0957, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 610.71, | |
| "learning_rate": 4.853333333333333e-08, | |
| "loss": 0.094, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 614.29, | |
| "learning_rate": 4.686666666666667e-08, | |
| "loss": 0.0931, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 617.86, | |
| "learning_rate": 4.52e-08, | |
| "loss": 0.0945, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 621.43, | |
| "learning_rate": 4.353333333333333e-08, | |
| "loss": 0.0909, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 625.0, | |
| "learning_rate": 4.1866666666666664e-08, | |
| "loss": 0.0909, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 628.57, | |
| "learning_rate": 4.02e-08, | |
| "loss": 0.0885, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 632.14, | |
| "learning_rate": 3.8533333333333334e-08, | |
| "loss": 0.0893, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 635.71, | |
| "learning_rate": 3.6866666666666666e-08, | |
| "loss": 0.089, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 639.29, | |
| "learning_rate": 3.52e-08, | |
| "loss": 0.0875, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 642.86, | |
| "learning_rate": 3.3533333333333336e-08, | |
| "loss": 0.0871, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 642.86, | |
| "eval_loss": 1.1704117059707642, | |
| "eval_runtime": 470.4344, | |
| "eval_samples_per_second": 1.088, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 62.93129539951574, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 646.43, | |
| "learning_rate": 3.186666666666666e-08, | |
| "loss": 0.0866, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 650.0, | |
| "learning_rate": 3.02e-08, | |
| "loss": 0.0849, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 653.57, | |
| "learning_rate": 2.853333333333333e-08, | |
| "loss": 0.0851, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 657.14, | |
| "learning_rate": 2.6866666666666666e-08, | |
| "loss": 0.0837, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 660.71, | |
| "learning_rate": 2.52e-08, | |
| "loss": 0.0841, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 664.29, | |
| "learning_rate": 2.3533333333333332e-08, | |
| "loss": 0.0836, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 667.86, | |
| "learning_rate": 2.1866666666666667e-08, | |
| "loss": 0.0826, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 671.43, | |
| "learning_rate": 2.02e-08, | |
| "loss": 0.0826, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 675.0, | |
| "learning_rate": 1.8533333333333334e-08, | |
| "loss": 0.0822, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 678.57, | |
| "learning_rate": 1.6866666666666665e-08, | |
| "loss": 0.0823, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 682.14, | |
| "learning_rate": 1.52e-08, | |
| "loss": 0.0818, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 685.71, | |
| "learning_rate": 1.3533333333333332e-08, | |
| "loss": 0.0804, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 689.29, | |
| "learning_rate": 1.1866666666666665e-08, | |
| "loss": 0.0808, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 692.86, | |
| "learning_rate": 1.02e-08, | |
| "loss": 0.0806, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 696.43, | |
| "learning_rate": 8.533333333333334e-09, | |
| "loss": 0.0817, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 700.0, | |
| "learning_rate": 6.866666666666666e-09, | |
| "loss": 0.0805, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 703.57, | |
| "learning_rate": 5.1999999999999994e-09, | |
| "loss": 0.0797, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 707.14, | |
| "learning_rate": 3.533333333333333e-09, | |
| "loss": 0.0802, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 710.71, | |
| "learning_rate": 1.8666666666666664e-09, | |
| "loss": 0.08, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 714.29, | |
| "learning_rate": 1.9999999999999998e-10, | |
| "loss": 0.0797, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 714.29, | |
| "eval_loss": 1.1790930032730103, | |
| "eval_runtime": 463.678, | |
| "eval_samples_per_second": 1.104, | |
| "eval_steps_per_second": 0.069, | |
| "eval_wer": 63.16585956416465, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 717.86, | |
| "learning_rate": 1.0289017341040463e-08, | |
| "loss": 0.0796, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 721.43, | |
| "learning_rate": 8.84393063583815e-09, | |
| "loss": 0.0804, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 725.0, | |
| "learning_rate": 7.398843930635838e-09, | |
| "loss": 0.0799, | |
| "step": 5075 | |
| }, | |
| { | |
| "epoch": 728.57, | |
| "learning_rate": 5.953757225433526e-09, | |
| "loss": 0.0799, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 728.57, | |
| "eval_loss": 1.1799688339233398, | |
| "eval_runtime": 463.8166, | |
| "eval_samples_per_second": 1.104, | |
| "eval_steps_per_second": 0.069, | |
| "eval_wer": 63.10532687651331, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 732.14, | |
| "learning_rate": 4.508670520231213e-09, | |
| "loss": 0.079, | |
| "step": 5125 | |
| }, | |
| { | |
| "epoch": 735.71, | |
| "learning_rate": 3.0635838150289017e-09, | |
| "loss": 0.0791, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 739.29, | |
| "learning_rate": 1.6184971098265895e-09, | |
| "loss": 0.0789, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 742.86, | |
| "learning_rate": 1.7341040462427745e-10, | |
| "loss": 0.0791, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 742.86, | |
| "eval_loss": 1.1803033351898193, | |
| "eval_runtime": 463.7133, | |
| "eval_samples_per_second": 1.104, | |
| "eval_steps_per_second": 0.069, | |
| "eval_wer": 63.112893462469735, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 742.86, | |
| "step": 5200, | |
| "total_flos": 9.240100498685952e+19, | |
| "train_loss": 0.0030569748695080095, | |
| "train_runtime": 1690.7628, | |
| "train_samples_per_second": 196.834, | |
| "train_steps_per_second": 3.076 | |
| } | |
| ], | |
| "max_steps": 5200, | |
| "num_train_epochs": 743, | |
| "total_flos": 9.240100498685952e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |