| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 18300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.950000000000001e-06, | |
| "loss": 13.5246, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 9.950000000000001e-06, | |
| "loss": 6.1441, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1.4950000000000001e-05, | |
| "loss": 4.2189, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.995e-05, | |
| "loss": 3.6075, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 2.495e-05, | |
| "loss": 3.3043, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_loss": 3.241492748260498, | |
| "eval_runtime": 169.6485, | |
| "eval_samples_per_second": 15.444, | |
| "eval_steps_per_second": 1.933, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 2.995e-05, | |
| "loss": 3.2087, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 3.495e-05, | |
| "loss": 3.1835, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 3.995e-05, | |
| "loss": 3.1414, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 4.495e-05, | |
| "loss": 3.0995, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 4.995e-05, | |
| "loss": 3.0482, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "eval_loss": 2.9591026306152344, | |
| "eval_runtime": 128.3332, | |
| "eval_samples_per_second": 20.416, | |
| "eval_steps_per_second": 2.556, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 5.495e-05, | |
| "loss": 2.7368, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 5.995000000000001e-05, | |
| "loss": 2.0079, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 6.494999999999999e-05, | |
| "loss": 1.6588, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 6.995e-05, | |
| "loss": 1.5644, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 7.495e-05, | |
| "loss": 1.4767, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "eval_loss": 0.47794264554977417, | |
| "eval_runtime": 128.1544, | |
| "eval_samples_per_second": 20.444, | |
| "eval_steps_per_second": 2.559, | |
| "eval_wer": 0.5776572037593256, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 7.995e-05, | |
| "loss": 1.4421, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 8.495e-05, | |
| "loss": 1.4009, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "learning_rate": 8.995e-05, | |
| "loss": 1.3682, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 9.495e-05, | |
| "loss": 1.3377, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 9.995e-05, | |
| "loss": 1.3152, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "eval_loss": 0.36967846751213074, | |
| "eval_runtime": 127.926, | |
| "eval_samples_per_second": 20.481, | |
| "eval_steps_per_second": 2.564, | |
| "eval_wer": 0.49384749539773276, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "learning_rate": 9.939877300613497e-05, | |
| "loss": 1.3009, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 9.878527607361964e-05, | |
| "loss": 1.2799, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.57, | |
| "learning_rate": 9.81717791411043e-05, | |
| "loss": 1.2559, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 13.11, | |
| "learning_rate": 9.756441717791411e-05, | |
| "loss": 1.2379, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 9.695092024539878e-05, | |
| "loss": 1.2246, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "eval_loss": 0.3084094822406769, | |
| "eval_runtime": 127.6781, | |
| "eval_samples_per_second": 20.52, | |
| "eval_steps_per_second": 2.569, | |
| "eval_wer": 0.44593547136905337, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 9.633742331288344e-05, | |
| "loss": 1.2209, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.75, | |
| "learning_rate": 9.57239263803681e-05, | |
| "loss": 1.1972, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 15.3, | |
| "learning_rate": 9.511042944785277e-05, | |
| "loss": 1.1885, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 15.85, | |
| "learning_rate": 9.449693251533743e-05, | |
| "loss": 1.1807, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 16.39, | |
| "learning_rate": 9.388343558282209e-05, | |
| "loss": 1.1781, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.39, | |
| "eval_loss": 0.2842142581939697, | |
| "eval_runtime": 128.3875, | |
| "eval_samples_per_second": 20.407, | |
| "eval_steps_per_second": 2.555, | |
| "eval_wer": 0.4154151729483577, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 9.326993865030675e-05, | |
| "loss": 1.1757, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 9.265644171779141e-05, | |
| "loss": 1.155, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 18.03, | |
| "learning_rate": 9.204294478527608e-05, | |
| "loss": 1.1455, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 18.58, | |
| "learning_rate": 9.142944785276074e-05, | |
| "loss": 1.1376, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 9.081595092024541e-05, | |
| "loss": 1.1351, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "eval_loss": 0.26151829957962036, | |
| "eval_runtime": 127.9176, | |
| "eval_samples_per_second": 20.482, | |
| "eval_steps_per_second": 2.564, | |
| "eval_wer": 0.3929367309369247, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 19.67, | |
| "learning_rate": 9.020245398773006e-05, | |
| "loss": 1.1262, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 20.22, | |
| "learning_rate": 8.958895705521472e-05, | |
| "loss": 1.1265, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 20.77, | |
| "learning_rate": 8.897546012269939e-05, | |
| "loss": 1.1033, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 21.31, | |
| "learning_rate": 8.836196319018405e-05, | |
| "loss": 1.1016, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "learning_rate": 8.774846625766872e-05, | |
| "loss": 1.1052, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "eval_loss": 0.24618586897850037, | |
| "eval_runtime": 128.2681, | |
| "eval_samples_per_second": 20.426, | |
| "eval_steps_per_second": 2.557, | |
| "eval_wer": 0.3746729968026354, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 8.714110429447854e-05, | |
| "loss": 1.0964, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 22.95, | |
| "learning_rate": 8.652760736196319e-05, | |
| "loss": 1.0848, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 23.5, | |
| "learning_rate": 8.591411042944786e-05, | |
| "loss": 1.0714, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 24.04, | |
| "learning_rate": 8.530061349693252e-05, | |
| "loss": 1.0696, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 24.59, | |
| "learning_rate": 8.468711656441717e-05, | |
| "loss": 1.0711, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.59, | |
| "eval_loss": 0.23661433160305023, | |
| "eval_runtime": 128.1823, | |
| "eval_samples_per_second": 20.44, | |
| "eval_steps_per_second": 2.559, | |
| "eval_wer": 0.36522623776765817, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 25.14, | |
| "learning_rate": 8.407361963190185e-05, | |
| "loss": 1.0519, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 25.68, | |
| "learning_rate": 8.346012269938652e-05, | |
| "loss": 1.0673, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 26.23, | |
| "learning_rate": 8.284662576687117e-05, | |
| "loss": 1.0606, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 26.78, | |
| "learning_rate": 8.223312883435583e-05, | |
| "loss": 1.0711, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 27.32, | |
| "learning_rate": 8.16196319018405e-05, | |
| "loss": 1.035, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 27.32, | |
| "eval_loss": 0.22680768370628357, | |
| "eval_runtime": 127.7363, | |
| "eval_samples_per_second": 20.511, | |
| "eval_steps_per_second": 2.568, | |
| "eval_wer": 0.3556825888964248, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "learning_rate": 8.100613496932515e-05, | |
| "loss": 1.0428, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 28.42, | |
| "learning_rate": 8.039877300613497e-05, | |
| "loss": 1.0322, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "learning_rate": 7.978527607361964e-05, | |
| "loss": 1.0291, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 29.51, | |
| "learning_rate": 7.91717791411043e-05, | |
| "loss": 1.0258, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 30.05, | |
| "learning_rate": 7.855828220858897e-05, | |
| "loss": 1.0277, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 30.05, | |
| "eval_loss": 0.22434431314468384, | |
| "eval_runtime": 127.9511, | |
| "eval_samples_per_second": 20.477, | |
| "eval_steps_per_second": 2.563, | |
| "eval_wer": 0.3449762619901172, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 30.6, | |
| "learning_rate": 7.794478527607363e-05, | |
| "loss": 1.0208, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 31.15, | |
| "learning_rate": 7.733128834355828e-05, | |
| "loss": 1.0252, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 31.69, | |
| "learning_rate": 7.671779141104295e-05, | |
| "loss": 1.0157, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 32.24, | |
| "learning_rate": 7.610429447852761e-05, | |
| "loss": 1.0094, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 32.79, | |
| "learning_rate": 7.549079754601228e-05, | |
| "loss": 1.002, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.79, | |
| "eval_loss": 0.22044695913791656, | |
| "eval_runtime": 127.6026, | |
| "eval_samples_per_second": 20.532, | |
| "eval_steps_per_second": 2.57, | |
| "eval_wer": 0.3388722023059781, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 7.487730061349694e-05, | |
| "loss": 0.996, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 33.88, | |
| "learning_rate": 7.42638036809816e-05, | |
| "loss": 1.002, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 34.43, | |
| "learning_rate": 7.365030674846626e-05, | |
| "loss": 0.9845, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "learning_rate": 7.303680981595092e-05, | |
| "loss": 0.9958, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 35.52, | |
| "learning_rate": 7.242331288343559e-05, | |
| "loss": 0.9837, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 35.52, | |
| "eval_loss": 0.2156379520893097, | |
| "eval_runtime": 127.7813, | |
| "eval_samples_per_second": 20.504, | |
| "eval_steps_per_second": 2.567, | |
| "eval_wer": 0.33489971901947485, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 36.07, | |
| "learning_rate": 7.180981595092025e-05, | |
| "loss": 1.0028, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 36.61, | |
| "learning_rate": 7.119631901840491e-05, | |
| "loss": 0.9722, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "learning_rate": 7.058282208588958e-05, | |
| "loss": 0.9784, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 37.7, | |
| "learning_rate": 6.996932515337423e-05, | |
| "loss": 0.9822, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 38.25, | |
| "learning_rate": 6.93558282208589e-05, | |
| "loss": 0.9773, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 38.25, | |
| "eval_loss": 0.21265123784542084, | |
| "eval_runtime": 128.3357, | |
| "eval_samples_per_second": 20.415, | |
| "eval_steps_per_second": 2.556, | |
| "eval_wer": 0.3288925491715919, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 38.8, | |
| "learning_rate": 6.874233128834356e-05, | |
| "loss": 0.9649, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 39.34, | |
| "learning_rate": 6.812883435582822e-05, | |
| "loss": 0.9728, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 39.89, | |
| "learning_rate": 6.751533742331289e-05, | |
| "loss": 0.9663, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 40.44, | |
| "learning_rate": 6.690184049079755e-05, | |
| "loss": 0.9762, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 40.98, | |
| "learning_rate": 6.629447852760736e-05, | |
| "loss": 0.9807, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 40.98, | |
| "eval_loss": 0.21417580544948578, | |
| "eval_runtime": 128.3216, | |
| "eval_samples_per_second": 20.417, | |
| "eval_steps_per_second": 2.556, | |
| "eval_wer": 0.32743920162774925, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 41.53, | |
| "learning_rate": 6.568098159509203e-05, | |
| "loss": 0.9647, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 42.08, | |
| "learning_rate": 6.506748466257669e-05, | |
| "loss": 0.9748, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 42.62, | |
| "learning_rate": 6.445398773006134e-05, | |
| "loss": 0.9484, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 43.17, | |
| "learning_rate": 6.384049079754602e-05, | |
| "loss": 0.9558, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 43.72, | |
| "learning_rate": 6.322699386503069e-05, | |
| "loss": 0.9582, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 43.72, | |
| "eval_loss": 0.20038354396820068, | |
| "eval_runtime": 127.7709, | |
| "eval_samples_per_second": 20.505, | |
| "eval_steps_per_second": 2.567, | |
| "eval_wer": 0.314165294060653, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 44.26, | |
| "learning_rate": 6.261349693251534e-05, | |
| "loss": 0.949, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 44.81, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.9491, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 45.36, | |
| "learning_rate": 6.138650306748467e-05, | |
| "loss": 0.9447, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 45.9, | |
| "learning_rate": 6.0773006134969325e-05, | |
| "loss": 0.9368, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "learning_rate": 6.015950920245399e-05, | |
| "loss": 0.9548, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "eval_loss": 0.20219053328037262, | |
| "eval_runtime": 127.6869, | |
| "eval_samples_per_second": 20.519, | |
| "eval_steps_per_second": 2.569, | |
| "eval_wer": 0.30500920453444436, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 46.99, | |
| "learning_rate": 5.9546012269938655e-05, | |
| "loss": 0.9407, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 47.54, | |
| "learning_rate": 5.893251533742331e-05, | |
| "loss": 0.9312, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 48.09, | |
| "learning_rate": 5.831901840490798e-05, | |
| "loss": 0.9446, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 48.63, | |
| "learning_rate": 5.770552147239264e-05, | |
| "loss": 0.9225, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 49.18, | |
| "learning_rate": 5.70920245398773e-05, | |
| "loss": 0.9251, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 49.18, | |
| "eval_loss": 0.2018980085849762, | |
| "eval_runtime": 130.6299, | |
| "eval_samples_per_second": 20.057, | |
| "eval_steps_per_second": 2.511, | |
| "eval_wer": 0.3035074120724736, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 49.73, | |
| "learning_rate": 5.6478527607361965e-05, | |
| "loss": 0.9251, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 50.27, | |
| "learning_rate": 5.586503067484663e-05, | |
| "loss": 0.9172, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 50.82, | |
| "learning_rate": 5.5251533742331294e-05, | |
| "loss": 0.9103, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 51.37, | |
| "learning_rate": 5.463803680981595e-05, | |
| "loss": 0.9133, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 51.91, | |
| "learning_rate": 5.402453987730062e-05, | |
| "loss": 0.9103, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 51.91, | |
| "eval_loss": 0.1963759958744049, | |
| "eval_runtime": 127.8377, | |
| "eval_samples_per_second": 20.495, | |
| "eval_steps_per_second": 2.566, | |
| "eval_wer": 0.30210250944675904, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 52.46, | |
| "learning_rate": 5.341104294478528e-05, | |
| "loss": 0.9109, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 53.01, | |
| "learning_rate": 5.279754601226994e-05, | |
| "loss": 0.9153, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 53.55, | |
| "learning_rate": 5.2184049079754604e-05, | |
| "loss": 0.9113, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 54.1, | |
| "learning_rate": 5.157055214723927e-05, | |
| "loss": 0.9181, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 54.64, | |
| "learning_rate": 5.095705521472393e-05, | |
| "loss": 0.915, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 54.64, | |
| "eval_loss": 0.19702854752540588, | |
| "eval_runtime": 128.314, | |
| "eval_samples_per_second": 20.419, | |
| "eval_steps_per_second": 2.556, | |
| "eval_wer": 0.30316829764557696, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 55.19, | |
| "learning_rate": 5.034969325153375e-05, | |
| "loss": 0.8971, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 55.74, | |
| "learning_rate": 4.973619631901841e-05, | |
| "loss": 0.9066, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 56.28, | |
| "learning_rate": 4.9122699386503065e-05, | |
| "loss": 0.8993, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 56.83, | |
| "learning_rate": 4.850920245398774e-05, | |
| "loss": 0.8933, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 57.38, | |
| "learning_rate": 4.7895705521472395e-05, | |
| "loss": 0.8962, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 57.38, | |
| "eval_loss": 0.2006961703300476, | |
| "eval_runtime": 128.3801, | |
| "eval_samples_per_second": 20.408, | |
| "eval_steps_per_second": 2.555, | |
| "eval_wer": 0.30462164518941964, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 57.92, | |
| "learning_rate": 4.728220858895705e-05, | |
| "loss": 0.8829, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 58.47, | |
| "learning_rate": 4.6668711656441724e-05, | |
| "loss": 0.8932, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 59.02, | |
| "learning_rate": 4.605521472392638e-05, | |
| "loss": 0.8986, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 59.56, | |
| "learning_rate": 4.544171779141104e-05, | |
| "loss": 0.8892, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 60.11, | |
| "learning_rate": 4.482822085889571e-05, | |
| "loss": 0.8729, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 60.11, | |
| "eval_loss": 0.19668185710906982, | |
| "eval_runtime": 128.5612, | |
| "eval_samples_per_second": 20.379, | |
| "eval_steps_per_second": 2.551, | |
| "eval_wer": 0.2942059877918806, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 60.66, | |
| "learning_rate": 4.421472392638037e-05, | |
| "loss": 0.876, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 61.2, | |
| "learning_rate": 4.3601226993865034e-05, | |
| "loss": 0.8759, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 61.75, | |
| "learning_rate": 4.29877300613497e-05, | |
| "loss": 0.8813, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 62.3, | |
| "learning_rate": 4.237423312883436e-05, | |
| "loss": 0.8684, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 62.84, | |
| "learning_rate": 4.176073619631902e-05, | |
| "loss": 0.8744, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 62.84, | |
| "eval_loss": 0.19520752131938934, | |
| "eval_runtime": 127.705, | |
| "eval_samples_per_second": 20.516, | |
| "eval_steps_per_second": 2.568, | |
| "eval_wer": 0.2885379323708943, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 63.39, | |
| "learning_rate": 4.1147239263803686e-05, | |
| "loss": 0.8665, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 63.93, | |
| "learning_rate": 4.0533742331288344e-05, | |
| "loss": 0.8757, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 64.48, | |
| "learning_rate": 3.992024539877301e-05, | |
| "loss": 0.8694, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 65.03, | |
| "learning_rate": 3.930674846625767e-05, | |
| "loss": 0.8578, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 65.57, | |
| "learning_rate": 3.869938650306748e-05, | |
| "loss": 0.874, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 65.57, | |
| "eval_loss": 0.18939977884292603, | |
| "eval_runtime": 128.0037, | |
| "eval_samples_per_second": 20.468, | |
| "eval_steps_per_second": 2.562, | |
| "eval_wer": 0.28950683073345607, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 66.12, | |
| "learning_rate": 3.808588957055215e-05, | |
| "loss": 0.8628, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "learning_rate": 3.747239263803681e-05, | |
| "loss": 0.8564, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 67.21, | |
| "learning_rate": 3.685889570552147e-05, | |
| "loss": 0.8502, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 67.76, | |
| "learning_rate": 3.6245398773006135e-05, | |
| "loss": 0.8521, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 68.31, | |
| "learning_rate": 3.56319018404908e-05, | |
| "loss": 0.8457, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 68.31, | |
| "eval_loss": 0.18946239352226257, | |
| "eval_runtime": 128.0623, | |
| "eval_samples_per_second": 20.459, | |
| "eval_steps_per_second": 2.561, | |
| "eval_wer": 0.28282143203177984, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 68.85, | |
| "learning_rate": 3.501840490797546e-05, | |
| "loss": 0.8624, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 69.4, | |
| "learning_rate": 3.440490797546013e-05, | |
| "loss": 0.8394, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 69.95, | |
| "learning_rate": 3.379141104294479e-05, | |
| "loss": 0.8381, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 70.49, | |
| "learning_rate": 3.3177914110429445e-05, | |
| "loss": 0.8431, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 71.04, | |
| "learning_rate": 3.2564417177914117e-05, | |
| "loss": 0.8519, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 71.04, | |
| "eval_loss": 0.19119836390018463, | |
| "eval_runtime": 127.8936, | |
| "eval_samples_per_second": 20.486, | |
| "eval_steps_per_second": 2.565, | |
| "eval_wer": 0.28747214417207634, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 71.58, | |
| "learning_rate": 3.1950920245398774e-05, | |
| "loss": 0.8463, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 72.13, | |
| "learning_rate": 3.133742331288343e-05, | |
| "loss": 0.8492, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 72.68, | |
| "learning_rate": 3.0723926380368104e-05, | |
| "loss": 0.8354, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 73.22, | |
| "learning_rate": 3.0110429447852762e-05, | |
| "loss": 0.8408, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 73.77, | |
| "learning_rate": 2.9496932515337423e-05, | |
| "loss": 0.8301, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 73.77, | |
| "eval_loss": 0.18781304359436035, | |
| "eval_runtime": 127.7845, | |
| "eval_samples_per_second": 20.503, | |
| "eval_steps_per_second": 2.567, | |
| "eval_wer": 0.2760391434938475, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 74.32, | |
| "learning_rate": 2.8883435582822088e-05, | |
| "loss": 0.8425, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 74.86, | |
| "learning_rate": 2.826993865030675e-05, | |
| "loss": 0.8416, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 75.41, | |
| "learning_rate": 2.7656441717791414e-05, | |
| "loss": 0.8269, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 75.96, | |
| "learning_rate": 2.7042944785276075e-05, | |
| "loss": 0.8212, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 76.5, | |
| "learning_rate": 2.6429447852760736e-05, | |
| "loss": 0.8226, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 76.5, | |
| "eval_loss": 0.18075355887413025, | |
| "eval_runtime": 127.7671, | |
| "eval_samples_per_second": 20.506, | |
| "eval_steps_per_second": 2.567, | |
| "eval_wer": 0.2701288634822207, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 77.05, | |
| "learning_rate": 2.58159509202454e-05, | |
| "loss": 0.8265, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 77.6, | |
| "learning_rate": 2.5202453987730063e-05, | |
| "loss": 0.8279, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 78.14, | |
| "learning_rate": 2.4588957055214727e-05, | |
| "loss": 0.8273, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 78.69, | |
| "learning_rate": 2.3975460122699385e-05, | |
| "loss": 0.8067, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 79.23, | |
| "learning_rate": 2.33680981595092e-05, | |
| "loss": 0.8071, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 79.23, | |
| "eval_loss": 0.18486912548542023, | |
| "eval_runtime": 127.9551, | |
| "eval_samples_per_second": 20.476, | |
| "eval_steps_per_second": 2.563, | |
| "eval_wer": 0.27410134676872394, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 79.78, | |
| "learning_rate": 2.2754601226993866e-05, | |
| "loss": 0.8231, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 80.33, | |
| "learning_rate": 2.214110429447853e-05, | |
| "loss": 0.808, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 80.87, | |
| "learning_rate": 2.1527607361963192e-05, | |
| "loss": 0.8107, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 81.42, | |
| "learning_rate": 2.0914110429447853e-05, | |
| "loss": 0.7966, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 81.97, | |
| "learning_rate": 2.0300613496932515e-05, | |
| "loss": 0.7999, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 81.97, | |
| "eval_loss": 0.18083913624286652, | |
| "eval_runtime": 127.4398, | |
| "eval_samples_per_second": 20.559, | |
| "eval_steps_per_second": 2.574, | |
| "eval_wer": 0.2717275457804476, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 82.51, | |
| "learning_rate": 1.968711656441718e-05, | |
| "loss": 0.8171, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 83.06, | |
| "learning_rate": 1.907361963190184e-05, | |
| "loss": 0.8034, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 83.61, | |
| "learning_rate": 1.8460122699386502e-05, | |
| "loss": 0.8086, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 84.15, | |
| "learning_rate": 1.7846625766871167e-05, | |
| "loss": 0.8005, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 84.7, | |
| "learning_rate": 1.723312883435583e-05, | |
| "loss": 0.7947, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 84.7, | |
| "eval_loss": 0.1820572018623352, | |
| "eval_runtime": 127.1268, | |
| "eval_samples_per_second": 20.609, | |
| "eval_steps_per_second": 2.58, | |
| "eval_wer": 0.2715822110260634, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 85.25, | |
| "learning_rate": 1.661963190184049e-05, | |
| "loss": 0.7947, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 85.79, | |
| "learning_rate": 1.6006134969325154e-05, | |
| "loss": 0.7892, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 86.34, | |
| "learning_rate": 1.539263803680982e-05, | |
| "loss": 0.7969, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 86.89, | |
| "learning_rate": 1.477914110429448e-05, | |
| "loss": 0.801, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 87.43, | |
| "learning_rate": 1.4165644171779141e-05, | |
| "loss": 0.7783, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 87.43, | |
| "eval_loss": 0.18241995573043823, | |
| "eval_runtime": 127.0573, | |
| "eval_samples_per_second": 20.621, | |
| "eval_steps_per_second": 2.582, | |
| "eval_wer": 0.26610793527758936, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 87.98, | |
| "learning_rate": 1.3552147239263804e-05, | |
| "loss": 0.7915, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 88.52, | |
| "learning_rate": 1.2938650306748467e-05, | |
| "loss": 0.7862, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 89.07, | |
| "learning_rate": 1.232515337423313e-05, | |
| "loss": 0.7816, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 89.62, | |
| "learning_rate": 1.1711656441717792e-05, | |
| "loss": 0.7816, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 90.16, | |
| "learning_rate": 1.1098159509202455e-05, | |
| "loss": 0.7729, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 90.16, | |
| "eval_loss": 0.17727895081043243, | |
| "eval_runtime": 128.0807, | |
| "eval_samples_per_second": 20.456, | |
| "eval_steps_per_second": 2.561, | |
| "eval_wer": 0.2638794690436973, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 90.71, | |
| "learning_rate": 1.0484662576687116e-05, | |
| "loss": 0.7778, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 91.26, | |
| "learning_rate": 9.87116564417178e-06, | |
| "loss": 0.7844, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 91.8, | |
| "learning_rate": 9.257668711656442e-06, | |
| "loss": 0.7842, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 92.35, | |
| "learning_rate": 8.644171779141105e-06, | |
| "loss": 0.7755, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 92.9, | |
| "learning_rate": 8.030674846625766e-06, | |
| "loss": 0.7759, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 92.9, | |
| "eval_loss": 0.17666833102703094, | |
| "eval_runtime": 128.2582, | |
| "eval_samples_per_second": 20.428, | |
| "eval_steps_per_second": 2.557, | |
| "eval_wer": 0.26291057068113555, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 93.44, | |
| "learning_rate": 7.417177914110429e-06, | |
| "loss": 0.7672, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 93.99, | |
| "learning_rate": 6.8036809815950924e-06, | |
| "loss": 0.7813, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 94.54, | |
| "learning_rate": 6.1963190184049085e-06, | |
| "loss": 0.7781, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 95.08, | |
| "learning_rate": 5.582822085889571e-06, | |
| "loss": 0.7711, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 95.63, | |
| "learning_rate": 4.969325153374233e-06, | |
| "loss": 0.7713, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 95.63, | |
| "eval_loss": 0.17804710566997528, | |
| "eval_runtime": 127.8273, | |
| "eval_samples_per_second": 20.496, | |
| "eval_steps_per_second": 2.566, | |
| "eval_wer": 0.26208700707295807, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 96.17, | |
| "learning_rate": 4.355828220858896e-06, | |
| "loss": 0.7789, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 96.72, | |
| "learning_rate": 3.7423312883435584e-06, | |
| "loss": 0.7732, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 97.27, | |
| "learning_rate": 3.128834355828221e-06, | |
| "loss": 0.7688, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 97.81, | |
| "learning_rate": 2.5153374233128836e-06, | |
| "loss": 0.7724, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 98.36, | |
| "learning_rate": 1.9018404907975462e-06, | |
| "loss": 0.7628, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 98.36, | |
| "eval_loss": 0.17734766006469727, | |
| "eval_runtime": 128.0092, | |
| "eval_samples_per_second": 20.467, | |
| "eval_steps_per_second": 2.562, | |
| "eval_wer": 0.2594225365759132, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 98.91, | |
| "learning_rate": 1.2883435582822088e-06, | |
| "loss": 0.7634, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 99.45, | |
| "learning_rate": 6.748466257668713e-07, | |
| "loss": 0.7638, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 6.134969325153375e-08, | |
| "loss": 0.7746, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 18300, | |
| "total_flos": 1.0288191185677785e+20, | |
| "train_loss": 1.1684310275218526, | |
| "train_runtime": 46836.9916, | |
| "train_samples_per_second": 12.501, | |
| "train_steps_per_second": 0.391 | |
| } | |
| ], | |
| "max_steps": 18300, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.0288191185677785e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |