| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "global_step": 29940, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.500000000000001e-07, | |
| "loss": 10.0397, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.9500000000000004e-06, | |
| "loss": 3.4906, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.9400000000000002e-06, | |
| "loss": 3.0663, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3.94e-06, | |
| "loss": 2.9734, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.94e-06, | |
| "loss": 2.8979, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.9413259029388428, | |
| "eval_runtime": 78.7489, | |
| "eval_samples_per_second": 10.413, | |
| "eval_steps_per_second": 0.444, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5.94e-06, | |
| "loss": 2.7525, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 6.9400000000000005e-06, | |
| "loss": 2.2756, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 7.94e-06, | |
| "loss": 1.7591, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 8.94e-06, | |
| "loss": 1.3666, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 1.2224, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.035875678062439, | |
| "eval_runtime": 78.5392, | |
| "eval_samples_per_second": 10.441, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.7802082338332219, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.0940000000000002e-05, | |
| "loss": 1.0768, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.1940000000000001e-05, | |
| "loss": 0.9695, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1.2940000000000001e-05, | |
| "loss": 0.9245, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.394e-05, | |
| "loss": 0.928, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 1.4940000000000001e-05, | |
| "loss": 0.8643, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_loss": 0.7745792865753174, | |
| "eval_runtime": 78.3017, | |
| "eval_samples_per_second": 10.472, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.5969051485337663, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 1.5940000000000003e-05, | |
| "loss": 0.7038, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 1.694e-05, | |
| "loss": 0.8009, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.794e-05, | |
| "loss": 0.7845, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 1.894e-05, | |
| "loss": 0.6791, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 1.9940000000000002e-05, | |
| "loss": 0.8211, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "eval_loss": 0.4882005751132965, | |
| "eval_runtime": 78.45, | |
| "eval_samples_per_second": 10.453, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.371000095520107, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 1.9932712956335006e-05, | |
| "loss": 0.5981, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 1.9861130994989265e-05, | |
| "loss": 0.766, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 1.9789549033643523e-05, | |
| "loss": 0.6564, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 1.9717967072297782e-05, | |
| "loss": 0.7201, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 1.964638511095204e-05, | |
| "loss": 0.5287, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "eval_loss": 0.4059687852859497, | |
| "eval_runtime": 78.4782, | |
| "eval_samples_per_second": 10.449, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.308529945553539, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 1.95748031496063e-05, | |
| "loss": 0.767, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 1.950322118826056e-05, | |
| "loss": 0.542, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 1.943163922691482e-05, | |
| "loss": 0.4444, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 1.9360057265569078e-05, | |
| "loss": 0.4599, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 1.9288475304223337e-05, | |
| "loss": 0.4724, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "eval_loss": 0.3296636939048767, | |
| "eval_runtime": 78.2325, | |
| "eval_samples_per_second": 10.482, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.2516954818989397, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 1.9216893342877595e-05, | |
| "loss": 0.4485, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 1.9145311381531854e-05, | |
| "loss": 0.4176, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 1.9073729420186116e-05, | |
| "loss": 0.4659, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 1.9002147458840374e-05, | |
| "loss": 0.5221, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 1.8930565497494633e-05, | |
| "loss": 0.4357, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "eval_loss": 0.31057485938072205, | |
| "eval_runtime": 78.313, | |
| "eval_samples_per_second": 10.471, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.2342153023211386, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 1.885898353614889e-05, | |
| "loss": 0.3869, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 1.878740157480315e-05, | |
| "loss": 0.4398, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 1.871581961345741e-05, | |
| "loss": 0.4936, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 1.864423765211167e-05, | |
| "loss": 0.45, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 1.857265569076593e-05, | |
| "loss": 0.376, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_loss": 0.277599960565567, | |
| "eval_runtime": 78.2545, | |
| "eval_samples_per_second": 10.479, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.2071831120450855, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 1.8501073729420187e-05, | |
| "loss": 0.322, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 1.842949176807445e-05, | |
| "loss": 0.339, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 1.8357909806728708e-05, | |
| "loss": 0.4353, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 1.8286327845382966e-05, | |
| "loss": 0.3572, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 1.8214745884037225e-05, | |
| "loss": 0.3286, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "eval_loss": 0.2887697219848633, | |
| "eval_runtime": 78.0501, | |
| "eval_samples_per_second": 10.506, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.20317126755181966, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 1.8143163922691483e-05, | |
| "loss": 0.3499, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 1.8071581961345742e-05, | |
| "loss": 0.3484, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.4217, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 1.792841803865426e-05, | |
| "loss": 0.3234, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 1.7856836077308518e-05, | |
| "loss": 0.3731, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "eval_loss": 0.2690582573413849, | |
| "eval_runtime": 78.1013, | |
| "eval_samples_per_second": 10.499, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.18349412551342056, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 1.778525411596278e-05, | |
| "loss": 0.2826, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "learning_rate": 1.7713672154617038e-05, | |
| "loss": 0.3204, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 1.7642090193271297e-05, | |
| "loss": 0.2912, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 1.7570508231925555e-05, | |
| "loss": 0.3001, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 1.7498926270579814e-05, | |
| "loss": 0.306, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "eval_loss": 0.2535899877548218, | |
| "eval_runtime": 78.0097, | |
| "eval_samples_per_second": 10.512, | |
| "eval_steps_per_second": 0.449, | |
| "eval_wer": 0.18349412551342056, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 1.7427344309234076e-05, | |
| "loss": 0.3393, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 11.42, | |
| "learning_rate": 1.7355762347888334e-05, | |
| "loss": 0.2619, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 1.7284180386542593e-05, | |
| "loss": 0.3324, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "learning_rate": 1.721259842519685e-05, | |
| "loss": 0.2864, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.714101646385111e-05, | |
| "loss": 0.3025, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "eval_loss": 0.2758459448814392, | |
| "eval_runtime": 78.012, | |
| "eval_samples_per_second": 10.511, | |
| "eval_steps_per_second": 0.449, | |
| "eval_wer": 0.18091508262489253, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 1.706943450250537e-05, | |
| "loss": 0.2557, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "learning_rate": 1.699785254115963e-05, | |
| "loss": 0.3373, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "learning_rate": 1.692627057981389e-05, | |
| "loss": 0.2631, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 1.6854688618468147e-05, | |
| "loss": 0.2559, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 1.6783106657122406e-05, | |
| "loss": 0.3413, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "eval_loss": 0.2791474759578705, | |
| "eval_runtime": 77.9246, | |
| "eval_samples_per_second": 10.523, | |
| "eval_steps_per_second": 0.449, | |
| "eval_wer": 0.18234788422963033, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 1.6711524695776668e-05, | |
| "loss": 0.2453, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 13.43, | |
| "learning_rate": 1.6639942734430926e-05, | |
| "loss": 0.2544, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 13.63, | |
| "learning_rate": 1.6568360773085185e-05, | |
| "loss": 0.3139, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 1.6496778811739443e-05, | |
| "loss": 0.251, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 1.6425196850393702e-05, | |
| "loss": 0.2601, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "eval_loss": 0.29117700457572937, | |
| "eval_runtime": 78.1768, | |
| "eval_samples_per_second": 10.489, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.1759480370618015, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 14.23, | |
| "learning_rate": 1.635361488904796e-05, | |
| "loss": 0.2767, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 14.43, | |
| "learning_rate": 1.628203292770222e-05, | |
| "loss": 0.2755, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 1.6210450966356477e-05, | |
| "loss": 0.2512, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 1.613886900501074e-05, | |
| "loss": 0.219, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 1.6067287043664998e-05, | |
| "loss": 0.2332, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "eval_loss": 0.25818365812301636, | |
| "eval_runtime": 78.301, | |
| "eval_samples_per_second": 10.472, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.16935714968000765, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 15.23, | |
| "learning_rate": 1.5995705082319256e-05, | |
| "loss": 0.257, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "learning_rate": 1.5924123120973515e-05, | |
| "loss": 0.2611, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 15.63, | |
| "learning_rate": 1.5852541159627774e-05, | |
| "loss": 0.2659, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 15.83, | |
| "learning_rate": 1.5780959198282032e-05, | |
| "loss": 0.237, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 1.5709377236936294e-05, | |
| "loss": 0.2108, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "eval_loss": 0.27170413732528687, | |
| "eval_runtime": 78.0575, | |
| "eval_samples_per_second": 10.505, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.16601394593561944, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 16.23, | |
| "learning_rate": 1.5637795275590553e-05, | |
| "loss": 0.1965, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 16.43, | |
| "learning_rate": 1.556621331424481e-05, | |
| "loss": 0.1961, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 16.63, | |
| "learning_rate": 1.549463135289907e-05, | |
| "loss": 0.2383, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 1.542304939155333e-05, | |
| "loss": 0.2734, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 1.535146743020759e-05, | |
| "loss": 0.2122, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "eval_loss": 0.2848033905029297, | |
| "eval_runtime": 78.1507, | |
| "eval_samples_per_second": 10.493, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.16467666443786416, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 17.23, | |
| "learning_rate": 1.527988546886185e-05, | |
| "loss": 0.186, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 17.43, | |
| "learning_rate": 1.5208303507516107e-05, | |
| "loss": 0.2194, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "learning_rate": 1.5136721546170366e-05, | |
| "loss": 0.1889, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 17.84, | |
| "learning_rate": 1.5065139584824626e-05, | |
| "loss": 0.2185, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 18.04, | |
| "learning_rate": 1.4993557623478884e-05, | |
| "loss": 0.2369, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 18.04, | |
| "eval_loss": 0.25476953387260437, | |
| "eval_runtime": 78.1145, | |
| "eval_samples_per_second": 10.497, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.16458114433088164, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 1.4921975662133143e-05, | |
| "loss": 0.2145, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 1.4850393700787402e-05, | |
| "loss": 0.1801, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 1.477881173944166e-05, | |
| "loss": 0.1962, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 1.4707945597709379e-05, | |
| "loss": 0.1993, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 1.4636363636363637e-05, | |
| "loss": 0.1906, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "eval_loss": 0.2667234539985657, | |
| "eval_runtime": 78.3018, | |
| "eval_samples_per_second": 10.472, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.16267074219123126, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 19.24, | |
| "learning_rate": 1.4564781675017896e-05, | |
| "loss": 0.1563, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 19.44, | |
| "learning_rate": 1.4493199713672154e-05, | |
| "loss": 0.1914, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 1.4421617752326416e-05, | |
| "loss": 0.1778, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "learning_rate": 1.4350035790980675e-05, | |
| "loss": 0.1918, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 20.04, | |
| "learning_rate": 1.4278453829634933e-05, | |
| "loss": 0.1943, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 20.04, | |
| "eval_loss": 0.2662443220615387, | |
| "eval_runtime": 78.3134, | |
| "eval_samples_per_second": 10.471, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.16228866176330117, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 20.24, | |
| "learning_rate": 1.4206871868289192e-05, | |
| "loss": 0.1673, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 20.44, | |
| "learning_rate": 1.4135289906943452e-05, | |
| "loss": 0.1945, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 20.64, | |
| "learning_rate": 1.406370794559771e-05, | |
| "loss": 0.1977, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 20.84, | |
| "learning_rate": 1.3992125984251969e-05, | |
| "loss": 0.1727, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 21.04, | |
| "learning_rate": 1.3920544022906228e-05, | |
| "loss": 0.18, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 21.04, | |
| "eval_loss": 0.2768990099430084, | |
| "eval_runtime": 78.5924, | |
| "eval_samples_per_second": 10.434, | |
| "eval_steps_per_second": 0.445, | |
| "eval_wer": 0.1560798548094374, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 21.24, | |
| "learning_rate": 1.3848962061560486e-05, | |
| "loss": 0.2274, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 21.44, | |
| "learning_rate": 1.3777380100214748e-05, | |
| "loss": 0.1886, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 21.64, | |
| "learning_rate": 1.3705798138869007e-05, | |
| "loss": 0.2035, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 21.84, | |
| "learning_rate": 1.3634216177523265e-05, | |
| "loss": 0.2019, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 22.04, | |
| "learning_rate": 1.3562634216177524e-05, | |
| "loss": 0.1654, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 22.04, | |
| "eval_loss": 0.2661066949367523, | |
| "eval_runtime": 78.0693, | |
| "eval_samples_per_second": 10.503, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.15579329448848983, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 22.24, | |
| "learning_rate": 1.3491052254831784e-05, | |
| "loss": 0.1539, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 22.44, | |
| "learning_rate": 1.3419470293486042e-05, | |
| "loss": 0.1496, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 22.65, | |
| "learning_rate": 1.3347888332140303e-05, | |
| "loss": 0.1517, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 22.85, | |
| "learning_rate": 1.3276306370794561e-05, | |
| "loss": 0.171, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 23.05, | |
| "learning_rate": 1.320472440944882e-05, | |
| "loss": 0.1515, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 23.05, | |
| "eval_loss": 0.2870410680770874, | |
| "eval_runtime": 78.1611, | |
| "eval_samples_per_second": 10.491, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.15970961887477314, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 23.25, | |
| "learning_rate": 1.313314244810308e-05, | |
| "loss": 0.1429, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 23.45, | |
| "learning_rate": 1.3061560486757339e-05, | |
| "loss": 0.1803, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 23.65, | |
| "learning_rate": 1.2989978525411597e-05, | |
| "loss": 0.1645, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 23.85, | |
| "learning_rate": 1.2918396564065856e-05, | |
| "loss": 0.1711, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 24.05, | |
| "learning_rate": 1.2846814602720114e-05, | |
| "loss": 0.147, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 24.05, | |
| "eval_loss": 0.27776607871055603, | |
| "eval_runtime": 78.2174, | |
| "eval_samples_per_second": 10.484, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.15512465373961218, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 24.25, | |
| "learning_rate": 1.2775232641374376e-05, | |
| "loss": 0.1507, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 24.45, | |
| "learning_rate": 1.2703650680028635e-05, | |
| "loss": 0.1681, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 24.65, | |
| "learning_rate": 1.2632068718682893e-05, | |
| "loss": 0.1581, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 24.85, | |
| "learning_rate": 1.2560486757337152e-05, | |
| "loss": 0.1455, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 25.05, | |
| "learning_rate": 1.2488904795991412e-05, | |
| "loss": 0.1622, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 25.05, | |
| "eval_loss": 0.2752845883369446, | |
| "eval_runtime": 78.4264, | |
| "eval_samples_per_second": 10.456, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.15407393256280447, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 25.25, | |
| "learning_rate": 1.241732283464567e-05, | |
| "loss": 0.1399, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 25.45, | |
| "learning_rate": 1.2345740873299929e-05, | |
| "loss": 0.1565, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 25.65, | |
| "learning_rate": 1.2274158911954188e-05, | |
| "loss": 0.1464, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "learning_rate": 1.2202576950608446e-05, | |
| "loss": 0.1562, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 26.05, | |
| "learning_rate": 1.2130994989262708e-05, | |
| "loss": 0.1522, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 26.05, | |
| "eval_loss": 0.2932421565055847, | |
| "eval_runtime": 78.5851, | |
| "eval_samples_per_second": 10.435, | |
| "eval_steps_per_second": 0.445, | |
| "eval_wer": 0.15206801031617156, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "learning_rate": 1.2059413027916967e-05, | |
| "loss": 0.1532, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 26.45, | |
| "learning_rate": 1.1987831066571225e-05, | |
| "loss": 0.1303, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 26.65, | |
| "learning_rate": 1.1916249105225484e-05, | |
| "loss": 0.1393, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 26.85, | |
| "learning_rate": 1.1844667143879742e-05, | |
| "loss": 0.1278, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 27.05, | |
| "learning_rate": 1.1773085182534002e-05, | |
| "loss": 0.1522, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 27.05, | |
| "eval_loss": 0.2547505497932434, | |
| "eval_runtime": 78.6033, | |
| "eval_samples_per_second": 10.432, | |
| "eval_steps_per_second": 0.445, | |
| "eval_wer": 0.1513038494603114, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 27.25, | |
| "learning_rate": 1.1701503221188261e-05, | |
| "loss": 0.1283, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 27.45, | |
| "learning_rate": 1.1629921259842521e-05, | |
| "loss": 0.1501, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 27.66, | |
| "learning_rate": 1.155833929849678e-05, | |
| "loss": 0.1195, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 27.86, | |
| "learning_rate": 1.148675733715104e-05, | |
| "loss": 0.1314, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 28.06, | |
| "learning_rate": 1.1415175375805298e-05, | |
| "loss": 0.1319, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 28.06, | |
| "eval_loss": 0.28105294704437256, | |
| "eval_runtime": 78.1505, | |
| "eval_samples_per_second": 10.493, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.1532142515999618, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 28.26, | |
| "learning_rate": 1.1343593414459557e-05, | |
| "loss": 0.1283, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 28.46, | |
| "learning_rate": 1.1272011453113816e-05, | |
| "loss": 0.1301, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 28.66, | |
| "learning_rate": 1.1201145311381534e-05, | |
| "loss": 0.1196, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 28.86, | |
| "learning_rate": 1.1129563350035793e-05, | |
| "loss": 0.1375, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 29.06, | |
| "learning_rate": 1.1057981388690051e-05, | |
| "loss": 0.1261, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 29.06, | |
| "eval_loss": 0.27864107489585876, | |
| "eval_runtime": 78.2482, | |
| "eval_samples_per_second": 10.479, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.15206801031617156, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 29.26, | |
| "learning_rate": 1.098639942734431e-05, | |
| "loss": 0.1302, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 29.46, | |
| "learning_rate": 1.0914817465998568e-05, | |
| "loss": 0.1362, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 29.66, | |
| "learning_rate": 1.0843235504652828e-05, | |
| "loss": 0.1396, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 29.86, | |
| "learning_rate": 1.0771653543307089e-05, | |
| "loss": 0.1213, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 30.06, | |
| "learning_rate": 1.0700071581961347e-05, | |
| "loss": 0.1391, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 30.06, | |
| "eval_loss": 0.2651180028915405, | |
| "eval_runtime": 78.2395, | |
| "eval_samples_per_second": 10.481, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.14614576368325533, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 30.26, | |
| "learning_rate": 1.0628489620615606e-05, | |
| "loss": 0.1208, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 30.46, | |
| "learning_rate": 1.0556907659269864e-05, | |
| "loss": 0.1154, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 30.66, | |
| "learning_rate": 1.0485325697924125e-05, | |
| "loss": 0.1456, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "learning_rate": 1.0413743736578383e-05, | |
| "loss": 0.1355, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 31.06, | |
| "learning_rate": 1.0342161775232642e-05, | |
| "loss": 0.1486, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 31.06, | |
| "eval_loss": 0.28657251596450806, | |
| "eval_runtime": 78.4786, | |
| "eval_samples_per_second": 10.449, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.149393447320661, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 31.26, | |
| "learning_rate": 1.02705798138869e-05, | |
| "loss": 0.1324, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 31.46, | |
| "learning_rate": 1.0198997852541162e-05, | |
| "loss": 0.1181, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 31.66, | |
| "learning_rate": 1.012741589119542e-05, | |
| "loss": 0.118, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 31.86, | |
| "learning_rate": 1.005583392984968e-05, | |
| "loss": 0.1299, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 32.06, | |
| "learning_rate": 9.984251968503938e-06, | |
| "loss": 0.1121, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 32.06, | |
| "eval_loss": 0.264145165681839, | |
| "eval_runtime": 78.3489, | |
| "eval_samples_per_second": 10.466, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.14776960550195817, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 9.913385826771654e-06, | |
| "loss": 0.1105, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 32.46, | |
| "learning_rate": 9.841803865425913e-06, | |
| "loss": 0.1257, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 32.67, | |
| "learning_rate": 9.770221904080173e-06, | |
| "loss": 0.1085, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 32.87, | |
| "learning_rate": 9.698639942734432e-06, | |
| "loss": 0.1217, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 33.07, | |
| "learning_rate": 9.62705798138869e-06, | |
| "loss": 0.1114, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 33.07, | |
| "eval_loss": 0.2910135090351105, | |
| "eval_runtime": 78.4013, | |
| "eval_samples_per_second": 10.459, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.14776960550195817, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 33.27, | |
| "learning_rate": 9.555476020042949e-06, | |
| "loss": 0.1288, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 33.47, | |
| "learning_rate": 9.483894058697209e-06, | |
| "loss": 0.1036, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 33.67, | |
| "learning_rate": 9.412312097351468e-06, | |
| "loss": 0.1192, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 33.87, | |
| "learning_rate": 9.340730136005728e-06, | |
| "loss": 0.1153, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 34.07, | |
| "learning_rate": 9.269148174659986e-06, | |
| "loss": 0.101, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 34.07, | |
| "eval_loss": 0.28841477632522583, | |
| "eval_runtime": 78.2927, | |
| "eval_samples_per_second": 10.474, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.14433088165058744, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 34.27, | |
| "learning_rate": 9.197566213314245e-06, | |
| "loss": 0.1141, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 34.47, | |
| "learning_rate": 9.125984251968505e-06, | |
| "loss": 0.1116, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 34.67, | |
| "learning_rate": 9.054402290622764e-06, | |
| "loss": 0.1003, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 34.87, | |
| "learning_rate": 8.982820329277024e-06, | |
| "loss": 0.1073, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 35.07, | |
| "learning_rate": 8.911238367931283e-06, | |
| "loss": 0.1135, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 35.07, | |
| "eval_loss": 0.3028663396835327, | |
| "eval_runtime": 78.2285, | |
| "eval_samples_per_second": 10.482, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.14690992453911547, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 35.27, | |
| "learning_rate": 8.839656406585541e-06, | |
| "loss": 0.1006, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 35.47, | |
| "learning_rate": 8.7680744452398e-06, | |
| "loss": 0.0951, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 35.67, | |
| "learning_rate": 8.69649248389406e-06, | |
| "loss": 0.1052, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 35.87, | |
| "learning_rate": 8.624910522548318e-06, | |
| "loss": 0.0925, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 36.07, | |
| "learning_rate": 8.553328561202577e-06, | |
| "loss": 0.0972, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 36.07, | |
| "eval_loss": 0.28704050183296204, | |
| "eval_runtime": 78.3292, | |
| "eval_samples_per_second": 10.469, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.14671888432515046, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 36.27, | |
| "learning_rate": 8.481746599856837e-06, | |
| "loss": 0.1159, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 36.47, | |
| "learning_rate": 8.410164638511096e-06, | |
| "loss": 0.1044, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 36.67, | |
| "learning_rate": 8.338582677165356e-06, | |
| "loss": 0.1029, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 36.87, | |
| "learning_rate": 8.267000715819614e-06, | |
| "loss": 0.1022, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 37.07, | |
| "learning_rate": 8.195418754473873e-06, | |
| "loss": 0.1178, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 37.07, | |
| "eval_loss": 0.2744777500629425, | |
| "eval_runtime": 78.3098, | |
| "eval_samples_per_second": 10.471, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.1449995223994651, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 37.27, | |
| "learning_rate": 8.123836793128133e-06, | |
| "loss": 0.1053, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 37.47, | |
| "learning_rate": 8.052254831782392e-06, | |
| "loss": 0.0995, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 37.68, | |
| "learning_rate": 7.98067287043665e-06, | |
| "loss": 0.099, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 37.88, | |
| "learning_rate": 7.909090909090909e-06, | |
| "loss": 0.1069, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 38.08, | |
| "learning_rate": 7.837508947745169e-06, | |
| "loss": 0.0885, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 38.08, | |
| "eval_loss": 0.2836405038833618, | |
| "eval_runtime": 78.1038, | |
| "eval_samples_per_second": 10.499, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.1440443213296399, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 38.28, | |
| "learning_rate": 7.765926986399428e-06, | |
| "loss": 0.0973, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 38.48, | |
| "learning_rate": 7.694345025053686e-06, | |
| "loss": 0.0968, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "learning_rate": 7.622763063707946e-06, | |
| "loss": 0.0997, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 38.88, | |
| "learning_rate": 7.551181102362205e-06, | |
| "loss": 0.0904, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 39.08, | |
| "learning_rate": 7.479599141016465e-06, | |
| "loss": 0.1144, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 39.08, | |
| "eval_loss": 0.27612900733947754, | |
| "eval_runtime": 78.4941, | |
| "eval_samples_per_second": 10.447, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.144617441971535, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 39.28, | |
| "learning_rate": 7.408017179670724e-06, | |
| "loss": 0.1003, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 39.48, | |
| "learning_rate": 7.336435218324983e-06, | |
| "loss": 0.1035, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 39.68, | |
| "learning_rate": 7.264853256979242e-06, | |
| "loss": 0.1005, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 39.88, | |
| "learning_rate": 7.1932712956335e-06, | |
| "loss": 0.0937, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 40.08, | |
| "learning_rate": 7.12168933428776e-06, | |
| "loss": 0.0997, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 40.08, | |
| "eval_loss": 0.28064894676208496, | |
| "eval_runtime": 78.1984, | |
| "eval_samples_per_second": 10.486, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.14385328111567486, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 40.28, | |
| "learning_rate": 7.050823192555477e-06, | |
| "loss": 0.1024, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 40.48, | |
| "learning_rate": 6.979241231209736e-06, | |
| "loss": 0.0928, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 40.68, | |
| "learning_rate": 6.907659269863994e-06, | |
| "loss": 0.0991, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 40.88, | |
| "learning_rate": 6.8360773085182545e-06, | |
| "loss": 0.0982, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 41.08, | |
| "learning_rate": 6.764495347172513e-06, | |
| "loss": 0.1012, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 41.08, | |
| "eval_loss": 0.2878414988517761, | |
| "eval_runtime": 78.3493, | |
| "eval_samples_per_second": 10.466, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.14127423822714683, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 41.28, | |
| "learning_rate": 6.692913385826772e-06, | |
| "loss": 0.086, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 41.48, | |
| "learning_rate": 6.621331424481031e-06, | |
| "loss": 0.093, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 41.68, | |
| "learning_rate": 6.550465282748748e-06, | |
| "loss": 0.0884, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 41.88, | |
| "learning_rate": 6.478883321403006e-06, | |
| "loss": 0.0949, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 42.08, | |
| "learning_rate": 6.4073013600572665e-06, | |
| "loss": 0.0902, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 42.08, | |
| "eval_loss": 0.2832191586494446, | |
| "eval_runtime": 78.1254, | |
| "eval_samples_per_second": 10.496, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.14519056261343014, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 42.28, | |
| "learning_rate": 6.335719398711525e-06, | |
| "loss": 0.0786, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 42.48, | |
| "learning_rate": 6.2641374373657844e-06, | |
| "loss": 0.0794, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 42.69, | |
| "learning_rate": 6.192555476020044e-06, | |
| "loss": 0.0843, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 42.89, | |
| "learning_rate": 6.120973514674302e-06, | |
| "loss": 0.0913, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 43.09, | |
| "learning_rate": 6.049391553328562e-06, | |
| "loss": 0.0804, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 43.09, | |
| "eval_loss": 0.2910655438899994, | |
| "eval_runtime": 78.1303, | |
| "eval_samples_per_second": 10.495, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.14576368325532524, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 43.29, | |
| "learning_rate": 5.97780959198282e-06, | |
| "loss": 0.0815, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 43.49, | |
| "learning_rate": 5.9062276306370805e-06, | |
| "loss": 0.0924, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 43.69, | |
| "learning_rate": 5.834645669291339e-06, | |
| "loss": 0.0916, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 43.89, | |
| "learning_rate": 5.7630637079455985e-06, | |
| "loss": 0.0859, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 44.09, | |
| "learning_rate": 5.691481746599857e-06, | |
| "loss": 0.0762, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 44.09, | |
| "eval_loss": 0.2707769572734833, | |
| "eval_runtime": 78.5003, | |
| "eval_samples_per_second": 10.446, | |
| "eval_steps_per_second": 0.446, | |
| "eval_wer": 0.1441398414366224, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 44.29, | |
| "learning_rate": 5.6198997852541155e-06, | |
| "loss": 0.0991, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 44.49, | |
| "learning_rate": 5.548317823908376e-06, | |
| "loss": 0.0824, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 44.69, | |
| "learning_rate": 5.476735862562634e-06, | |
| "loss": 0.1, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 44.89, | |
| "learning_rate": 5.405153901216894e-06, | |
| "loss": 0.0689, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 45.09, | |
| "learning_rate": 5.333571939871153e-06, | |
| "loss": 0.0758, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 45.09, | |
| "eval_loss": 0.2803756594657898, | |
| "eval_runtime": 78.2094, | |
| "eval_samples_per_second": 10.485, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.14337568058076225, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 45.29, | |
| "learning_rate": 5.2619899785254125e-06, | |
| "loss": 0.0776, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 45.49, | |
| "learning_rate": 5.190408017179671e-06, | |
| "loss": 0.079, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 45.69, | |
| "learning_rate": 5.1188260558339295e-06, | |
| "loss": 0.069, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 45.89, | |
| "learning_rate": 5.04724409448819e-06, | |
| "loss": 0.0835, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 46.09, | |
| "learning_rate": 4.975662133142448e-06, | |
| "loss": 0.0874, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 46.09, | |
| "eval_loss": 0.28313755989074707, | |
| "eval_runtime": 78.3395, | |
| "eval_samples_per_second": 10.467, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.1407011175852517, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 46.29, | |
| "learning_rate": 4.904080171796708e-06, | |
| "loss": 0.0767, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 46.49, | |
| "learning_rate": 4.832498210450966e-06, | |
| "loss": 0.0795, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 46.69, | |
| "learning_rate": 4.760916249105226e-06, | |
| "loss": 0.0866, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 46.89, | |
| "learning_rate": 4.689334287759485e-06, | |
| "loss": 0.0884, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 47.09, | |
| "learning_rate": 4.617752326413744e-06, | |
| "loss": 0.0895, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 47.09, | |
| "eval_loss": 0.2913013994693756, | |
| "eval_runtime": 78.135, | |
| "eval_samples_per_second": 10.495, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.13955487630146146, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 47.29, | |
| "learning_rate": 4.546170365068003e-06, | |
| "loss": 0.0746, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 47.49, | |
| "learning_rate": 4.474588403722262e-06, | |
| "loss": 0.0588, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 47.7, | |
| "learning_rate": 4.403006442376522e-06, | |
| "loss": 0.079, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 47.9, | |
| "learning_rate": 4.33142448103078e-06, | |
| "loss": 0.0971, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 48.1, | |
| "learning_rate": 4.260558339298497e-06, | |
| "loss": 0.0975, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 48.1, | |
| "eval_loss": 0.2956128418445587, | |
| "eval_runtime": 78.2552, | |
| "eval_samples_per_second": 10.479, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.14108319801318178, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 48.3, | |
| "learning_rate": 4.1889763779527564e-06, | |
| "loss": 0.0898, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 48.5, | |
| "learning_rate": 4.117394416607015e-06, | |
| "loss": 0.0827, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 48.7, | |
| "learning_rate": 4.045812455261274e-06, | |
| "loss": 0.084, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 48.9, | |
| "learning_rate": 3.974230493915534e-06, | |
| "loss": 0.0747, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 49.1, | |
| "learning_rate": 3.902648532569793e-06, | |
| "loss": 0.0758, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 49.1, | |
| "eval_loss": 0.292031854391098, | |
| "eval_runtime": 78.3103, | |
| "eval_samples_per_second": 10.471, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.13850415512465375, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 49.3, | |
| "learning_rate": 3.831066571224052e-06, | |
| "loss": 0.0654, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "learning_rate": 3.759484609878311e-06, | |
| "loss": 0.0727, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 49.7, | |
| "learning_rate": 3.6879026485325704e-06, | |
| "loss": 0.0897, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 49.9, | |
| "learning_rate": 3.616320687186829e-06, | |
| "loss": 0.0673, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 50.1, | |
| "learning_rate": 3.544738725841088e-06, | |
| "loss": 0.0704, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 50.1, | |
| "eval_loss": 0.2788284718990326, | |
| "eval_runtime": 78.6013, | |
| "eval_samples_per_second": 10.432, | |
| "eval_steps_per_second": 0.445, | |
| "eval_wer": 0.1383131149106887, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 50.3, | |
| "learning_rate": 3.4731567644953473e-06, | |
| "loss": 0.0701, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 50.5, | |
| "learning_rate": 3.4015748031496067e-06, | |
| "loss": 0.0721, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 50.7, | |
| "learning_rate": 3.3299928418038657e-06, | |
| "loss": 0.0608, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 50.9, | |
| "learning_rate": 3.258410880458125e-06, | |
| "loss": 0.0722, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 51.1, | |
| "learning_rate": 3.186828919112384e-06, | |
| "loss": 0.0707, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 51.1, | |
| "eval_loss": 0.28223100304603577, | |
| "eval_runtime": 78.2491, | |
| "eval_samples_per_second": 10.479, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.1387907154456013, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 51.3, | |
| "learning_rate": 3.1152469577666434e-06, | |
| "loss": 0.0762, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 51.5, | |
| "learning_rate": 3.043664996420902e-06, | |
| "loss": 0.0688, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 51.7, | |
| "learning_rate": 2.9720830350751613e-06, | |
| "loss": 0.07, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 51.9, | |
| "learning_rate": 2.9005010737294203e-06, | |
| "loss": 0.0758, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 52.1, | |
| "learning_rate": 2.8289191123836797e-06, | |
| "loss": 0.0664, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 52.1, | |
| "eval_loss": 0.2876470386981964, | |
| "eval_runtime": 78.2924, | |
| "eval_samples_per_second": 10.474, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.13707135351991595, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 52.3, | |
| "learning_rate": 2.7573371510379386e-06, | |
| "loss": 0.0676, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 52.51, | |
| "learning_rate": 2.685755189692198e-06, | |
| "loss": 0.0645, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 52.71, | |
| "learning_rate": 2.614173228346457e-06, | |
| "loss": 0.0742, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 52.91, | |
| "learning_rate": 2.542591267000716e-06, | |
| "loss": 0.0846, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 53.11, | |
| "learning_rate": 2.4710093056549753e-06, | |
| "loss": 0.0692, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 53.11, | |
| "eval_loss": 0.28145480155944824, | |
| "eval_runtime": 77.9062, | |
| "eval_samples_per_second": 10.525, | |
| "eval_steps_per_second": 0.449, | |
| "eval_wer": 0.13773999426879358, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 53.31, | |
| "learning_rate": 2.3994273443092343e-06, | |
| "loss": 0.0829, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 53.51, | |
| "learning_rate": 2.3278453829634933e-06, | |
| "loss": 0.0688, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 53.71, | |
| "learning_rate": 2.2562634216177526e-06, | |
| "loss": 0.0838, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 53.91, | |
| "learning_rate": 2.1846814602720116e-06, | |
| "loss": 0.0733, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 54.11, | |
| "learning_rate": 2.1130994989262706e-06, | |
| "loss": 0.0799, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 54.11, | |
| "eval_loss": 0.28059905767440796, | |
| "eval_runtime": 78.1345, | |
| "eval_samples_per_second": 10.495, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.13630719266405578, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 54.31, | |
| "learning_rate": 2.04151753758053e-06, | |
| "loss": 0.063, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 54.51, | |
| "learning_rate": 1.9706513958482463e-06, | |
| "loss": 0.0858, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 54.71, | |
| "learning_rate": 1.8990694345025055e-06, | |
| "loss": 0.0664, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 54.91, | |
| "learning_rate": 1.8274874731567647e-06, | |
| "loss": 0.0706, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 55.11, | |
| "learning_rate": 1.7559055118110239e-06, | |
| "loss": 0.0611, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 55.11, | |
| "eval_loss": 0.2878268361091614, | |
| "eval_runtime": 78.1003, | |
| "eval_samples_per_second": 10.499, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 0.13630719266405578, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 55.31, | |
| "learning_rate": 1.684323550465283e-06, | |
| "loss": 0.0912, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 55.51, | |
| "learning_rate": 1.612741589119542e-06, | |
| "loss": 0.0597, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 55.71, | |
| "learning_rate": 1.5411596277738012e-06, | |
| "loss": 0.0683, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 55.91, | |
| "learning_rate": 1.4695776664280603e-06, | |
| "loss": 0.0633, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 56.11, | |
| "learning_rate": 1.3979957050823193e-06, | |
| "loss": 0.0759, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 56.11, | |
| "eval_loss": 0.2899620532989502, | |
| "eval_runtime": 78.8347, | |
| "eval_samples_per_second": 10.402, | |
| "eval_steps_per_second": 0.444, | |
| "eval_wer": 0.13649823287802082, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 56.31, | |
| "learning_rate": 1.3264137437365785e-06, | |
| "loss": 0.07, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 56.51, | |
| "learning_rate": 1.2548317823908377e-06, | |
| "loss": 0.0648, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 56.71, | |
| "learning_rate": 1.1832498210450966e-06, | |
| "loss": 0.0739, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 56.91, | |
| "learning_rate": 1.111667859699356e-06, | |
| "loss": 0.0707, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 57.11, | |
| "learning_rate": 1.040085898353615e-06, | |
| "loss": 0.0801, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 57.11, | |
| "eval_loss": 0.28810349106788635, | |
| "eval_runtime": 78.3272, | |
| "eval_samples_per_second": 10.469, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.13754895405482853, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 57.31, | |
| "learning_rate": 9.685039370078741e-07, | |
| "loss": 0.0555, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 57.52, | |
| "learning_rate": 8.969219756621332e-07, | |
| "loss": 0.0552, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 57.72, | |
| "learning_rate": 8.253400143163923e-07, | |
| "loss": 0.0604, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 57.92, | |
| "learning_rate": 7.537580529706514e-07, | |
| "loss": 0.0668, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 58.12, | |
| "learning_rate": 6.821760916249105e-07, | |
| "loss": 0.0644, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 58.12, | |
| "eval_loss": 0.28975263237953186, | |
| "eval_runtime": 78.3272, | |
| "eval_samples_per_second": 10.469, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.13621167255707325, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 58.32, | |
| "learning_rate": 6.105941302791697e-07, | |
| "loss": 0.0651, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 58.52, | |
| "learning_rate": 5.390121689334289e-07, | |
| "loss": 0.0632, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 58.72, | |
| "learning_rate": 4.6743020758768793e-07, | |
| "loss": 0.0678, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 58.92, | |
| "learning_rate": 3.9584824624194705e-07, | |
| "loss": 0.0584, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 59.12, | |
| "learning_rate": 3.2426628489620617e-07, | |
| "loss": 0.068, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 59.12, | |
| "eval_loss": 0.2913411557674408, | |
| "eval_runtime": 78.2895, | |
| "eval_samples_per_second": 10.474, | |
| "eval_steps_per_second": 0.447, | |
| "eval_wer": 0.1368803133059509, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 59.32, | |
| "learning_rate": 2.526843235504653e-07, | |
| "loss": 0.0588, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 59.52, | |
| "learning_rate": 1.8110236220472441e-07, | |
| "loss": 0.0662, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 59.72, | |
| "learning_rate": 1.0952040085898355e-07, | |
| "loss": 0.0619, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 59.92, | |
| "learning_rate": 3.7938439513242663e-08, | |
| "loss": 0.067, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 29940, | |
| "total_flos": 3.460380972676701e+20, | |
| "train_loss": 0.29242472694807237, | |
| "train_runtime": 84446.646, | |
| "train_samples_per_second": 8.498, | |
| "train_steps_per_second": 0.355 | |
| } | |
| ], | |
| "max_steps": 29940, | |
| "num_train_epochs": 60, | |
| "total_flos": 3.460380972676701e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |