| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 139.66480446927375, | |
| "global_step": 25000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.000194, | |
| "loss": 3.3358, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00039400000000000004, | |
| "loss": 0.9471, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.000594, | |
| "loss": 0.9236, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.0007940000000000001, | |
| "loss": 0.996, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.000994, | |
| "loss": 1.0296, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_loss": 0.8106316328048706, | |
| "eval_runtime": 194.0347, | |
| "eval_samples_per_second": 13.467, | |
| "eval_steps_per_second": 0.423, | |
| "eval_wer": 0.8029419223941161, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0009960408163265306, | |
| "loss": 1.0459, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.0009919591836734694, | |
| "loss": 0.9901, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.0009878775510204082, | |
| "loss": 0.9455, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 0.000983795918367347, | |
| "loss": 0.9814, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.0009797142857142858, | |
| "loss": 0.9339, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "eval_loss": 0.7419061064720154, | |
| "eval_runtime": 190.1458, | |
| "eval_samples_per_second": 13.742, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 0.7931777834136444, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.0009756326530612245, | |
| "loss": 0.9541, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 0.0009715510204081632, | |
| "loss": 0.8954, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.0009674693877551021, | |
| "loss": 0.9209, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 0.0009633877551020408, | |
| "loss": 0.9078, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.0009593061224489796, | |
| "loss": 0.8925, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "eval_loss": 0.7136574983596802, | |
| "eval_runtime": 190.5136, | |
| "eval_samples_per_second": 13.716, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.7706061374587877, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 0.0009552244897959184, | |
| "loss": 0.8785, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 0.0009511428571428572, | |
| "loss": 0.8793, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.0009470612244897959, | |
| "loss": 0.8766, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.61, | |
| "learning_rate": 0.0009429795918367347, | |
| "loss": 0.8601, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 0.0009388979591836735, | |
| "loss": 0.8484, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "eval_loss": 0.7019568681716919, | |
| "eval_runtime": 193.4802, | |
| "eval_samples_per_second": 13.505, | |
| "eval_steps_per_second": 0.424, | |
| "eval_wer": 0.7676895764646209, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.73, | |
| "learning_rate": 0.0009348163265306122, | |
| "loss": 0.8123, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 0.000930734693877551, | |
| "loss": 0.7892, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 0.0009266530612244898, | |
| "loss": 0.7836, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 13.41, | |
| "learning_rate": 0.0009225714285714286, | |
| "loss": 0.7694, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.97, | |
| "learning_rate": 0.0009184897959183673, | |
| "loss": 0.7521, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.97, | |
| "eval_loss": 0.7042976021766663, | |
| "eval_runtime": 193.425, | |
| "eval_samples_per_second": 13.509, | |
| "eval_steps_per_second": 0.424, | |
| "eval_wer": 0.7375095105249809, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 14.53, | |
| "learning_rate": 0.0009144081632653061, | |
| "loss": 0.7472, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 15.08, | |
| "learning_rate": 0.0009103265306122449, | |
| "loss": 0.7254, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 15.64, | |
| "learning_rate": 0.0009062448979591837, | |
| "loss": 0.7541, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 16.2, | |
| "learning_rate": 0.0009021632653061224, | |
| "loss": 0.7361, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 0.0008980816326530612, | |
| "loss": 0.719, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "eval_loss": 0.6617498397827148, | |
| "eval_runtime": 191.0945, | |
| "eval_samples_per_second": 13.674, | |
| "eval_steps_per_second": 0.429, | |
| "eval_wer": 0.742772001014456, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 17.32, | |
| "learning_rate": 0.000894, | |
| "loss": 0.6862, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 17.88, | |
| "learning_rate": 0.0008899183673469387, | |
| "loss": 0.6599, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 0.0008858367346938775, | |
| "loss": 0.6363, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 0.0008817551020408164, | |
| "loss": 0.6861, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 19.55, | |
| "learning_rate": 0.0008776734693877552, | |
| "loss": 0.656, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 19.55, | |
| "eval_loss": 0.6388168334960938, | |
| "eval_runtime": 190.9254, | |
| "eval_samples_per_second": 13.686, | |
| "eval_steps_per_second": 0.429, | |
| "eval_wer": 0.7201686533096627, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 20.11, | |
| "learning_rate": 0.0008735918367346939, | |
| "loss": 0.6348, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 20.67, | |
| "learning_rate": 0.0008695102040816327, | |
| "loss": 0.5911, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 21.23, | |
| "learning_rate": 0.0008654285714285715, | |
| "loss": 0.605, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 21.79, | |
| "learning_rate": 0.0008613469387755102, | |
| "loss": 0.6051, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 22.35, | |
| "learning_rate": 0.0008572653061224491, | |
| "loss": 0.6085, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.35, | |
| "eval_loss": 0.6211119890213013, | |
| "eval_runtime": 188.9632, | |
| "eval_samples_per_second": 13.828, | |
| "eval_steps_per_second": 0.434, | |
| "eval_wer": 0.6960119198579762, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.91, | |
| "learning_rate": 0.0008531836734693878, | |
| "loss": 0.5777, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 23.46, | |
| "learning_rate": 0.0008491020408163266, | |
| "loss": 0.5582, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 0.0008450204081632653, | |
| "loss": 0.5859, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 24.58, | |
| "learning_rate": 0.0008409387755102042, | |
| "loss": 0.5368, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 25.14, | |
| "learning_rate": 0.0008368571428571429, | |
| "loss": 0.5598, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 25.14, | |
| "eval_loss": 0.6131592988967896, | |
| "eval_runtime": 191.2607, | |
| "eval_samples_per_second": 13.662, | |
| "eval_steps_per_second": 0.429, | |
| "eval_wer": 0.664436976921126, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 25.7, | |
| "learning_rate": 0.0008327755102040816, | |
| "loss": 0.5139, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 26.26, | |
| "learning_rate": 0.0008286938775510205, | |
| "loss": 0.5455, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 26.82, | |
| "learning_rate": 0.0008246122448979592, | |
| "loss": 0.5448, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 27.37, | |
| "learning_rate": 0.000820530612244898, | |
| "loss": 0.5316, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 27.93, | |
| "learning_rate": 0.0008164489795918367, | |
| "loss": 0.4969, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 27.93, | |
| "eval_loss": 0.6065218448638916, | |
| "eval_runtime": 194.8518, | |
| "eval_samples_per_second": 13.41, | |
| "eval_steps_per_second": 0.421, | |
| "eval_wer": 0.65210499619579, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 28.49, | |
| "learning_rate": 0.0008123673469387756, | |
| "loss": 0.48, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 29.05, | |
| "learning_rate": 0.0008082857142857143, | |
| "loss": 0.5028, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 29.61, | |
| "learning_rate": 0.0008042040816326531, | |
| "loss": 0.4577, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 30.17, | |
| "learning_rate": 0.0008001224489795919, | |
| "loss": 0.4636, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 30.73, | |
| "learning_rate": 0.0007960408163265306, | |
| "loss": 0.4638, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 30.73, | |
| "eval_loss": 0.697826623916626, | |
| "eval_runtime": 189.0268, | |
| "eval_samples_per_second": 13.823, | |
| "eval_steps_per_second": 0.434, | |
| "eval_wer": 0.6577479076845042, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 31.28, | |
| "learning_rate": 0.0007919591836734694, | |
| "loss": 0.46, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 31.84, | |
| "learning_rate": 0.0007878775510204081, | |
| "loss": 0.4577, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 32.4, | |
| "learning_rate": 0.000783795918367347, | |
| "loss": 0.4502, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "learning_rate": 0.0007797142857142857, | |
| "loss": 0.4395, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 33.52, | |
| "learning_rate": 0.0007756326530612245, | |
| "loss": 0.4385, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 33.52, | |
| "eval_loss": 0.5994424223899841, | |
| "eval_runtime": 190.5342, | |
| "eval_samples_per_second": 13.714, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.6565115394369769, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 34.08, | |
| "learning_rate": 0.0007715510204081633, | |
| "loss": 0.4173, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 34.64, | |
| "learning_rate": 0.0007674693877551021, | |
| "loss": 0.4144, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "learning_rate": 0.0007633877551020408, | |
| "loss": 0.3906, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 35.75, | |
| "learning_rate": 0.0007593061224489795, | |
| "loss": 0.4085, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 36.31, | |
| "learning_rate": 0.0007552244897959184, | |
| "loss": 0.396, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 36.31, | |
| "eval_loss": 0.6170300841331482, | |
| "eval_runtime": 190.0522, | |
| "eval_samples_per_second": 13.749, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 0.6257608419984784, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 36.87, | |
| "learning_rate": 0.0007511428571428571, | |
| "loss": 0.3864, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 37.43, | |
| "learning_rate": 0.0007470612244897959, | |
| "loss": 0.3864, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 37.99, | |
| "learning_rate": 0.0007429795918367347, | |
| "loss": 0.3975, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 38.55, | |
| "learning_rate": 0.0007388979591836735, | |
| "loss": 0.3873, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 39.11, | |
| "learning_rate": 0.0007348163265306122, | |
| "loss": 0.3861, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 39.11, | |
| "eval_loss": 0.6486256718635559, | |
| "eval_runtime": 191.5175, | |
| "eval_samples_per_second": 13.644, | |
| "eval_steps_per_second": 0.428, | |
| "eval_wer": 0.6216713162566574, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 39.66, | |
| "learning_rate": 0.0007307346938775509, | |
| "loss": 0.3629, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 40.22, | |
| "learning_rate": 0.0007266530612244898, | |
| "loss": 0.3545, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 40.78, | |
| "learning_rate": 0.0007225714285714285, | |
| "loss": 0.381, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 41.34, | |
| "learning_rate": 0.0007184897959183673, | |
| "loss": 0.3679, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 41.9, | |
| "learning_rate": 0.0007144081632653061, | |
| "loss": 0.3602, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 41.9, | |
| "eval_loss": 0.6508038640022278, | |
| "eval_runtime": 188.3071, | |
| "eval_samples_per_second": 13.876, | |
| "eval_steps_per_second": 0.435, | |
| "eval_wer": 0.6114633527770733, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 42.46, | |
| "learning_rate": 0.000710326530612245, | |
| "loss": 0.3431, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "learning_rate": 0.0007062448979591836, | |
| "loss": 0.3555, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 43.58, | |
| "learning_rate": 0.0007021632653061226, | |
| "loss": 0.3327, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 44.13, | |
| "learning_rate": 0.0006980816326530613, | |
| "loss": 0.3327, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 44.69, | |
| "learning_rate": 0.000694, | |
| "loss": 0.3251, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 44.69, | |
| "eval_loss": 0.7021898031234741, | |
| "eval_runtime": 189.1301, | |
| "eval_samples_per_second": 13.816, | |
| "eval_steps_per_second": 0.434, | |
| "eval_wer": 0.625317017499366, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 45.25, | |
| "learning_rate": 0.0006899183673469388, | |
| "loss": 0.3403, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 45.81, | |
| "learning_rate": 0.0006858367346938776, | |
| "loss": 0.3242, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 46.37, | |
| "learning_rate": 0.0006817551020408164, | |
| "loss": 0.316, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 46.93, | |
| "learning_rate": 0.0006776734693877551, | |
| "loss": 0.3113, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 47.49, | |
| "learning_rate": 0.000673591836734694, | |
| "loss": 0.3197, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 47.49, | |
| "eval_loss": 0.7705556750297546, | |
| "eval_runtime": 188.475, | |
| "eval_samples_per_second": 13.864, | |
| "eval_steps_per_second": 0.435, | |
| "eval_wer": 0.621544509256911, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 48.04, | |
| "learning_rate": 0.0006695102040816327, | |
| "loss": 0.3188, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 48.6, | |
| "learning_rate": 0.0006654285714285715, | |
| "loss": 0.3134, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 49.16, | |
| "learning_rate": 0.0006613469387755102, | |
| "loss": 0.3093, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 49.72, | |
| "learning_rate": 0.000657265306122449, | |
| "loss": 0.2872, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 50.28, | |
| "learning_rate": 0.0006531836734693878, | |
| "loss": 0.3013, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 50.28, | |
| "eval_loss": 0.6418728232383728, | |
| "eval_runtime": 189.3407, | |
| "eval_samples_per_second": 13.801, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 0.5998922140502155, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 50.84, | |
| "learning_rate": 0.0006491020408163265, | |
| "loss": 0.2866, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 51.4, | |
| "learning_rate": 0.0006450204081632654, | |
| "loss": 0.2794, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 51.96, | |
| "learning_rate": 0.0006409387755102041, | |
| "loss": 0.2842, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 52.51, | |
| "learning_rate": 0.0006368571428571429, | |
| "loss": 0.2894, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 53.07, | |
| "learning_rate": 0.0006327755102040816, | |
| "loss": 0.2813, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 53.07, | |
| "eval_loss": 0.6907714009284973, | |
| "eval_runtime": 188.5038, | |
| "eval_samples_per_second": 13.862, | |
| "eval_steps_per_second": 0.435, | |
| "eval_wer": 0.595929495308141, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 53.63, | |
| "learning_rate": 0.0006286938775510205, | |
| "loss": 0.2846, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 54.19, | |
| "learning_rate": 0.0006246122448979592, | |
| "loss": 0.2784, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 54.75, | |
| "learning_rate": 0.0006205306122448979, | |
| "loss": 0.2841, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 55.31, | |
| "learning_rate": 0.0006164489795918368, | |
| "loss": 0.2838, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 55.87, | |
| "learning_rate": 0.0006123673469387755, | |
| "loss": 0.286, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 55.87, | |
| "eval_loss": 0.7150660157203674, | |
| "eval_runtime": 188.1197, | |
| "eval_samples_per_second": 13.89, | |
| "eval_steps_per_second": 0.436, | |
| "eval_wer": 0.5916180573167639, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 56.42, | |
| "learning_rate": 0.0006082857142857143, | |
| "loss": 0.2831, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 56.98, | |
| "learning_rate": 0.000604204081632653, | |
| "loss": 0.2738, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 57.54, | |
| "learning_rate": 0.0006001224489795919, | |
| "loss": 0.2613, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 58.1, | |
| "learning_rate": 0.0005960816326530613, | |
| "loss": 0.2645, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 58.66, | |
| "learning_rate": 0.000592, | |
| "loss": 0.2645, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 58.66, | |
| "eval_loss": 0.7181155681610107, | |
| "eval_runtime": 192.5668, | |
| "eval_samples_per_second": 13.569, | |
| "eval_steps_per_second": 0.426, | |
| "eval_wer": 0.5860068475779863, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 59.22, | |
| "learning_rate": 0.0005879183673469388, | |
| "loss": 0.2501, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 59.78, | |
| "learning_rate": 0.0005838367346938776, | |
| "loss": 0.2612, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 60.34, | |
| "learning_rate": 0.0005797551020408164, | |
| "loss": 0.2541, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 60.89, | |
| "learning_rate": 0.0005756734693877551, | |
| "loss": 0.2496, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 61.45, | |
| "learning_rate": 0.0005715918367346939, | |
| "loss": 0.2535, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 61.45, | |
| "eval_loss": 0.7877444624900818, | |
| "eval_runtime": 190.8548, | |
| "eval_samples_per_second": 13.691, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.5979267055541466, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "learning_rate": 0.0005675102040816327, | |
| "loss": 0.2619, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 62.57, | |
| "learning_rate": 0.0005634285714285714, | |
| "loss": 0.2495, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 63.13, | |
| "learning_rate": 0.0005593469387755102, | |
| "loss": 0.2521, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 63.69, | |
| "learning_rate": 0.000555265306122449, | |
| "loss": 0.2326, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 64.25, | |
| "learning_rate": 0.0005511836734693878, | |
| "loss": 0.247, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 64.25, | |
| "eval_loss": 0.8198513984680176, | |
| "eval_runtime": 187.6839, | |
| "eval_samples_per_second": 13.922, | |
| "eval_steps_per_second": 0.437, | |
| "eval_wer": 0.6128899315242201, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 64.8, | |
| "learning_rate": 0.0005471020408163265, | |
| "loss": 0.249, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 65.36, | |
| "learning_rate": 0.0005430204081632654, | |
| "loss": 0.2463, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 65.92, | |
| "learning_rate": 0.0005389387755102041, | |
| "loss": 0.2333, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 66.48, | |
| "learning_rate": 0.0005348571428571428, | |
| "loss": 0.2381, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 67.04, | |
| "learning_rate": 0.0005307755102040816, | |
| "loss": 0.2412, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 67.04, | |
| "eval_loss": 0.767884373664856, | |
| "eval_runtime": 188.3627, | |
| "eval_samples_per_second": 13.872, | |
| "eval_steps_per_second": 0.435, | |
| "eval_wer": 0.5884161805731677, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 67.6, | |
| "learning_rate": 0.0005266938775510204, | |
| "loss": 0.2285, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 68.16, | |
| "learning_rate": 0.0005226122448979592, | |
| "loss": 0.241, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 68.72, | |
| "learning_rate": 0.0005185306122448979, | |
| "loss": 0.229, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 69.27, | |
| "learning_rate": 0.0005144489795918368, | |
| "loss": 0.2371, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 69.83, | |
| "learning_rate": 0.0005103673469387755, | |
| "loss": 0.2404, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 69.83, | |
| "eval_loss": 0.7266025543212891, | |
| "eval_runtime": 190.7393, | |
| "eval_samples_per_second": 13.699, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.581632006086736, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 70.39, | |
| "learning_rate": 0.0005062857142857143, | |
| "loss": 0.2315, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 70.95, | |
| "learning_rate": 0.0005022040816326531, | |
| "loss": 0.2305, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 71.51, | |
| "learning_rate": 0.0004981224489795918, | |
| "loss": 0.2307, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 72.07, | |
| "learning_rate": 0.0004940408163265306, | |
| "loss": 0.2204, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 72.63, | |
| "learning_rate": 0.00049, | |
| "loss": 0.2293, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 72.63, | |
| "eval_loss": 0.792820394039154, | |
| "eval_runtime": 188.6653, | |
| "eval_samples_per_second": 13.85, | |
| "eval_steps_per_second": 0.435, | |
| "eval_wer": 0.5794762870910474, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 73.18, | |
| "learning_rate": 0.0004859183673469388, | |
| "loss": 0.2185, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 73.74, | |
| "learning_rate": 0.00048183673469387754, | |
| "loss": 0.2126, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 74.3, | |
| "learning_rate": 0.00047775510204081634, | |
| "loss": 0.2177, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 74.86, | |
| "learning_rate": 0.00047367346938775515, | |
| "loss": 0.2204, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 75.42, | |
| "learning_rate": 0.0004695918367346939, | |
| "loss": 0.2176, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 75.42, | |
| "eval_loss": 0.7916468381881714, | |
| "eval_runtime": 189.1525, | |
| "eval_samples_per_second": 13.814, | |
| "eval_steps_per_second": 0.434, | |
| "eval_wer": 0.5845802688308395, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 75.98, | |
| "learning_rate": 0.00046551020408163265, | |
| "loss": 0.2119, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 76.54, | |
| "learning_rate": 0.0004614285714285714, | |
| "loss": 0.2149, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 77.09, | |
| "learning_rate": 0.0004573469387755102, | |
| "loss": 0.214, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 77.65, | |
| "learning_rate": 0.000453265306122449, | |
| "loss": 0.2095, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 78.21, | |
| "learning_rate": 0.00044918367346938776, | |
| "loss": 0.2143, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 78.21, | |
| "eval_loss": 0.7954298853874207, | |
| "eval_runtime": 187.5182, | |
| "eval_samples_per_second": 13.935, | |
| "eval_steps_per_second": 0.437, | |
| "eval_wer": 0.5764646208470707, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 78.77, | |
| "learning_rate": 0.00044510204081632656, | |
| "loss": 0.2182, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 79.33, | |
| "learning_rate": 0.0004410204081632653, | |
| "loss": 0.2042, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 79.89, | |
| "learning_rate": 0.0004369387755102041, | |
| "loss": 0.2184, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 80.45, | |
| "learning_rate": 0.00043285714285714287, | |
| "loss": 0.2104, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 81.01, | |
| "learning_rate": 0.0004287755102040816, | |
| "loss": 0.2185, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 81.01, | |
| "eval_loss": 0.8317196369171143, | |
| "eval_runtime": 188.0902, | |
| "eval_samples_per_second": 13.892, | |
| "eval_steps_per_second": 0.436, | |
| "eval_wer": 0.5906987065686026, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 81.56, | |
| "learning_rate": 0.0004246938775510204, | |
| "loss": 0.2026, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 82.12, | |
| "learning_rate": 0.0004206122448979592, | |
| "loss": 0.1972, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 82.68, | |
| "learning_rate": 0.000416530612244898, | |
| "loss": 0.2083, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 83.24, | |
| "learning_rate": 0.00041244897959183673, | |
| "loss": 0.2132, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 83.8, | |
| "learning_rate": 0.00040836734693877553, | |
| "loss": 0.2057, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 83.8, | |
| "eval_loss": 0.8015716075897217, | |
| "eval_runtime": 188.5358, | |
| "eval_samples_per_second": 13.859, | |
| "eval_steps_per_second": 0.435, | |
| "eval_wer": 0.5850557950798884, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 84.36, | |
| "learning_rate": 0.0004042857142857143, | |
| "loss": 0.2027, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 84.92, | |
| "learning_rate": 0.0004002040816326531, | |
| "loss": 0.2016, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 85.47, | |
| "learning_rate": 0.00039612244897959184, | |
| "loss": 0.1935, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 86.03, | |
| "learning_rate": 0.0003920408163265306, | |
| "loss": 0.1948, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 86.59, | |
| "learning_rate": 0.0003879591836734694, | |
| "loss": 0.1895, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 86.59, | |
| "eval_loss": 0.8080111145973206, | |
| "eval_runtime": 192.0878, | |
| "eval_samples_per_second": 13.603, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.5679368501141263, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 87.15, | |
| "learning_rate": 0.00038387755102040815, | |
| "loss": 0.1894, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 87.71, | |
| "learning_rate": 0.00037979591836734695, | |
| "loss": 0.1798, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 88.27, | |
| "learning_rate": 0.00037571428571428575, | |
| "loss": 0.1891, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 88.83, | |
| "learning_rate": 0.0003716326530612245, | |
| "loss": 0.1916, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 89.39, | |
| "learning_rate": 0.0003675510204081633, | |
| "loss": 0.1883, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 89.39, | |
| "eval_loss": 0.8103044033050537, | |
| "eval_runtime": 192.0932, | |
| "eval_samples_per_second": 13.603, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.5712338321075323, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 89.94, | |
| "learning_rate": 0.000363469387755102, | |
| "loss": 0.1774, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 90.5, | |
| "learning_rate": 0.0003593877551020408, | |
| "loss": 0.1831, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 91.06, | |
| "learning_rate": 0.00035530612244897956, | |
| "loss": 0.1858, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 91.62, | |
| "learning_rate": 0.00035122448979591837, | |
| "loss": 0.1863, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 92.18, | |
| "learning_rate": 0.00034714285714285717, | |
| "loss": 0.1802, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 92.18, | |
| "eval_loss": 0.8383206129074097, | |
| "eval_runtime": 187.0696, | |
| "eval_samples_per_second": 13.968, | |
| "eval_steps_per_second": 0.438, | |
| "eval_wer": 0.5644496576211007, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 92.74, | |
| "learning_rate": 0.0003430612244897959, | |
| "loss": 0.1818, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 93.3, | |
| "learning_rate": 0.0003389795918367347, | |
| "loss": 0.1805, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 93.85, | |
| "learning_rate": 0.0003348979591836735, | |
| "loss": 0.1784, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 94.41, | |
| "learning_rate": 0.0003308163265306123, | |
| "loss": 0.1731, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 94.97, | |
| "learning_rate": 0.000326734693877551, | |
| "loss": 0.1826, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 94.97, | |
| "eval_loss": 0.8798549771308899, | |
| "eval_runtime": 188.9431, | |
| "eval_samples_per_second": 13.83, | |
| "eval_steps_per_second": 0.434, | |
| "eval_wer": 0.5657494293685011, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 95.53, | |
| "learning_rate": 0.0003226530612244898, | |
| "loss": 0.1733, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 96.09, | |
| "learning_rate": 0.0003185714285714286, | |
| "loss": 0.1723, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 96.65, | |
| "learning_rate": 0.00031448979591836734, | |
| "loss": 0.1693, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 97.21, | |
| "learning_rate": 0.00031044897959183674, | |
| "loss": 0.1705, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 97.77, | |
| "learning_rate": 0.0003063673469387755, | |
| "loss": 0.1717, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 97.77, | |
| "eval_loss": 0.8619566559791565, | |
| "eval_runtime": 188.0319, | |
| "eval_samples_per_second": 13.897, | |
| "eval_steps_per_second": 0.436, | |
| "eval_wer": 0.5709168146081663, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 98.32, | |
| "learning_rate": 0.0003022857142857143, | |
| "loss": 0.1616, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 98.88, | |
| "learning_rate": 0.00029820408163265305, | |
| "loss": 0.1653, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 99.44, | |
| "learning_rate": 0.00029412244897959185, | |
| "loss": 0.1708, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 0.0002900408163265306, | |
| "loss": 0.1681, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 100.56, | |
| "learning_rate": 0.00028599999999999996, | |
| "loss": 0.1701, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 100.56, | |
| "eval_loss": 0.8717033267021179, | |
| "eval_runtime": 192.2586, | |
| "eval_samples_per_second": 13.591, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.5661932538676135, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 101.12, | |
| "learning_rate": 0.00028191836734693876, | |
| "loss": 0.1614, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 101.68, | |
| "learning_rate": 0.00027783673469387757, | |
| "loss": 0.1641, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 102.23, | |
| "learning_rate": 0.0002737551020408163, | |
| "loss": 0.1615, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 102.79, | |
| "learning_rate": 0.0002696734693877551, | |
| "loss": 0.1588, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 103.35, | |
| "learning_rate": 0.0002655918367346939, | |
| "loss": 0.1623, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 103.35, | |
| "eval_loss": 0.8533861041069031, | |
| "eval_runtime": 187.659, | |
| "eval_samples_per_second": 13.924, | |
| "eval_steps_per_second": 0.437, | |
| "eval_wer": 0.5594090793811819, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 103.91, | |
| "learning_rate": 0.0002615102040816327, | |
| "loss": 0.1566, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 104.47, | |
| "learning_rate": 0.0002574285714285715, | |
| "loss": 0.1528, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 105.03, | |
| "learning_rate": 0.0002533469387755102, | |
| "loss": 0.1538, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 105.59, | |
| "learning_rate": 0.000249265306122449, | |
| "loss": 0.1557, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 106.15, | |
| "learning_rate": 0.00024518367346938773, | |
| "loss": 0.158, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 106.15, | |
| "eval_loss": 0.8595470786094666, | |
| "eval_runtime": 194.975, | |
| "eval_samples_per_second": 13.402, | |
| "eval_steps_per_second": 0.421, | |
| "eval_wer": 0.5546221151407558, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 106.7, | |
| "learning_rate": 0.00024110204081632654, | |
| "loss": 0.1459, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 107.26, | |
| "learning_rate": 0.00023702040816326532, | |
| "loss": 0.152, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 107.82, | |
| "learning_rate": 0.0002329387755102041, | |
| "loss": 0.1533, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 108.38, | |
| "learning_rate": 0.00022885714285714287, | |
| "loss": 0.146, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 108.94, | |
| "learning_rate": 0.00022477551020408162, | |
| "loss": 0.1508, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 108.94, | |
| "eval_loss": 0.8573695421218872, | |
| "eval_runtime": 191.3527, | |
| "eval_samples_per_second": 13.655, | |
| "eval_steps_per_second": 0.429, | |
| "eval_wer": 0.5544636063910728, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 109.5, | |
| "learning_rate": 0.0002206938775510204, | |
| "loss": 0.1454, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 110.06, | |
| "learning_rate": 0.0002166122448979592, | |
| "loss": 0.1466, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 110.61, | |
| "learning_rate": 0.00021253061224489798, | |
| "loss": 0.1416, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 111.17, | |
| "learning_rate": 0.00020844897959183673, | |
| "loss": 0.1444, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 111.73, | |
| "learning_rate": 0.0002043673469387755, | |
| "loss": 0.142, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 111.73, | |
| "eval_loss": 0.867087721824646, | |
| "eval_runtime": 187.948, | |
| "eval_samples_per_second": 13.903, | |
| "eval_steps_per_second": 0.436, | |
| "eval_wer": 0.5536710626426579, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 112.29, | |
| "learning_rate": 0.0002002857142857143, | |
| "loss": 0.1485, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 112.85, | |
| "learning_rate": 0.00019620408163265306, | |
| "loss": 0.1466, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 113.41, | |
| "learning_rate": 0.00019212244897959184, | |
| "loss": 0.1364, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 113.97, | |
| "learning_rate": 0.00018804081632653062, | |
| "loss": 0.1421, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 114.53, | |
| "learning_rate": 0.0001839591836734694, | |
| "loss": 0.1395, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 114.53, | |
| "eval_loss": 0.843588650226593, | |
| "eval_runtime": 191.7789, | |
| "eval_samples_per_second": 13.625, | |
| "eval_steps_per_second": 0.428, | |
| "eval_wer": 0.5524663961450672, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 115.08, | |
| "learning_rate": 0.00017987755102040817, | |
| "loss": 0.1409, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 115.64, | |
| "learning_rate": 0.00017579591836734692, | |
| "loss": 0.1356, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 116.2, | |
| "learning_rate": 0.0001717142857142857, | |
| "loss": 0.1364, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 116.76, | |
| "learning_rate": 0.0001676326530612245, | |
| "loss": 0.1328, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 117.32, | |
| "learning_rate": 0.0001635918367346939, | |
| "loss": 0.1373, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 117.32, | |
| "eval_loss": 0.8807795643806458, | |
| "eval_runtime": 186.8391, | |
| "eval_samples_per_second": 13.985, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 0.5481866599036267, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 117.88, | |
| "learning_rate": 0.00015951020408163267, | |
| "loss": 0.1316, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 118.44, | |
| "learning_rate": 0.00015542857142857142, | |
| "loss": 0.133, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 118.99, | |
| "learning_rate": 0.0001513469387755102, | |
| "loss": 0.1348, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 119.55, | |
| "learning_rate": 0.00014726530612244897, | |
| "loss": 0.1321, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 120.11, | |
| "learning_rate": 0.00014318367346938778, | |
| "loss": 0.1338, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 120.11, | |
| "eval_loss": 0.902350127696991, | |
| "eval_runtime": 188.1097, | |
| "eval_samples_per_second": 13.891, | |
| "eval_steps_per_second": 0.436, | |
| "eval_wer": 0.5418146081663708, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 120.67, | |
| "learning_rate": 0.00013910204081632655, | |
| "loss": 0.1248, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 121.23, | |
| "learning_rate": 0.0001350204081632653, | |
| "loss": 0.1267, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 121.79, | |
| "learning_rate": 0.00013093877551020408, | |
| "loss": 0.1282, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 122.35, | |
| "learning_rate": 0.00012685714285714286, | |
| "loss": 0.1284, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 122.91, | |
| "learning_rate": 0.00012277551020408164, | |
| "loss": 0.1278, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 122.91, | |
| "eval_loss": 0.9142583608627319, | |
| "eval_runtime": 187.6194, | |
| "eval_samples_per_second": 13.927, | |
| "eval_steps_per_second": 0.437, | |
| "eval_wer": 0.5408952574182094, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 123.46, | |
| "learning_rate": 0.00011869387755102041, | |
| "loss": 0.1292, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 124.02, | |
| "learning_rate": 0.00011461224489795918, | |
| "loss": 0.1201, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 124.58, | |
| "learning_rate": 0.00011053061224489797, | |
| "loss": 0.1249, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 125.14, | |
| "learning_rate": 0.00010644897959183673, | |
| "loss": 0.1251, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 125.7, | |
| "learning_rate": 0.00010236734693877551, | |
| "loss": 0.1207, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 125.7, | |
| "eval_loss": 0.8916703462600708, | |
| "eval_runtime": 187.4806, | |
| "eval_samples_per_second": 13.937, | |
| "eval_steps_per_second": 0.437, | |
| "eval_wer": 0.5357912756784174, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 126.26, | |
| "learning_rate": 9.828571428571429e-05, | |
| "loss": 0.1233, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 126.82, | |
| "learning_rate": 9.420408163265307e-05, | |
| "loss": 0.1217, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 127.37, | |
| "learning_rate": 9.012244897959183e-05, | |
| "loss": 0.1197, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 127.93, | |
| "learning_rate": 8.604081632653062e-05, | |
| "loss": 0.1167, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 128.49, | |
| "learning_rate": 8.195918367346938e-05, | |
| "loss": 0.1203, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 128.49, | |
| "eval_loss": 0.9041045904159546, | |
| "eval_runtime": 186.9588, | |
| "eval_samples_per_second": 13.976, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 0.5341110829317778, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 129.05, | |
| "learning_rate": 7.791836734693878e-05, | |
| "loss": 0.1199, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 129.61, | |
| "learning_rate": 7.383673469387756e-05, | |
| "loss": 0.1171, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 130.17, | |
| "learning_rate": 6.975510204081632e-05, | |
| "loss": 0.1136, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 130.73, | |
| "learning_rate": 6.567346938775511e-05, | |
| "loss": 0.1133, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 131.28, | |
| "learning_rate": 6.159183673469388e-05, | |
| "loss": 0.1083, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 131.28, | |
| "eval_loss": 0.8883697986602783, | |
| "eval_runtime": 186.6245, | |
| "eval_samples_per_second": 14.001, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 0.5340793811818413, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 131.84, | |
| "learning_rate": 5.751020408163265e-05, | |
| "loss": 0.115, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 132.4, | |
| "learning_rate": 5.342857142857143e-05, | |
| "loss": 0.1132, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 132.96, | |
| "learning_rate": 4.934693877551021e-05, | |
| "loss": 0.1111, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 133.52, | |
| "learning_rate": 4.526530612244898e-05, | |
| "loss": 0.1079, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 134.08, | |
| "learning_rate": 4.1183673469387756e-05, | |
| "loss": 0.1147, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 134.08, | |
| "eval_loss": 0.891002357006073, | |
| "eval_runtime": 186.8507, | |
| "eval_samples_per_second": 13.984, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 0.525456505199087, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 134.64, | |
| "learning_rate": 3.7102040816326533e-05, | |
| "loss": 0.1085, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 135.2, | |
| "learning_rate": 3.3020408163265304e-05, | |
| "loss": 0.1104, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 135.75, | |
| "learning_rate": 2.8938775510204082e-05, | |
| "loss": 0.1096, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 136.31, | |
| "learning_rate": 2.4857142857142856e-05, | |
| "loss": 0.1064, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 136.87, | |
| "learning_rate": 2.0775510204081633e-05, | |
| "loss": 0.1129, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 136.87, | |
| "eval_loss": 0.8826149106025696, | |
| "eval_runtime": 187.6626, | |
| "eval_samples_per_second": 13.924, | |
| "eval_steps_per_second": 0.437, | |
| "eval_wer": 0.52412503170175, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 137.43, | |
| "learning_rate": 1.669387755102041e-05, | |
| "loss": 0.1086, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 137.99, | |
| "learning_rate": 1.2612244897959185e-05, | |
| "loss": 0.1061, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 138.55, | |
| "learning_rate": 8.53061224489796e-06, | |
| "loss": 0.1104, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 139.11, | |
| "learning_rate": 4.448979591836735e-06, | |
| "loss": 0.1095, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 139.66, | |
| "learning_rate": 3.673469387755102e-07, | |
| "loss": 0.1029, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 139.66, | |
| "eval_loss": 0.882430911064148, | |
| "eval_runtime": 187.8168, | |
| "eval_samples_per_second": 13.912, | |
| "eval_steps_per_second": 0.437, | |
| "eval_wer": 0.5246322597007355, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 139.66, | |
| "step": 25000, | |
| "total_flos": 4.393390203049772e+20, | |
| "train_loss": 0.3391121254348755, | |
| "train_runtime": 108834.0383, | |
| "train_samples_per_second": 7.351, | |
| "train_steps_per_second": 0.23 | |
| } | |
| ], | |
| "max_steps": 25000, | |
| "num_train_epochs": 140, | |
| "total_flos": 4.393390203049772e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |