| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.999682034976153, | |
| "global_step": 23580, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.950000000000001e-06, | |
| "loss": 9.2024, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.950000000000001e-06, | |
| "loss": 5.1397, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.4950000000000001e-05, | |
| "loss": 3.6109, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.995e-05, | |
| "loss": 3.2035, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 2.495e-05, | |
| "loss": 3.0307, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 3.0025861263275146, | |
| "eval_runtime": 463.575, | |
| "eval_samples_per_second": 11.841, | |
| "eval_steps_per_second": 0.742, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.995e-05, | |
| "loss": 2.9725, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.495e-05, | |
| "loss": 2.9483, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.995e-05, | |
| "loss": 2.9349, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.495e-05, | |
| "loss": 2.9042, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.995e-05, | |
| "loss": 2.7865, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 2.4848814010620117, | |
| "eval_runtime": 473.8576, | |
| "eval_samples_per_second": 11.584, | |
| "eval_steps_per_second": 0.726, | |
| "eval_wer": 0.9926163412027592, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 5.495e-05, | |
| "loss": 2.0237, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 5.99e-05, | |
| "loss": 1.2269, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 6.49e-05, | |
| "loss": 1.0577, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 6.99e-05, | |
| "loss": 0.8144, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 7.49e-05, | |
| "loss": 0.7522, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.4566805362701416, | |
| "eval_runtime": 465.7671, | |
| "eval_samples_per_second": 11.785, | |
| "eval_steps_per_second": 0.739, | |
| "eval_wer": 0.3594287379772661, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.99e-05, | |
| "loss": 0.6976, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.49e-05, | |
| "loss": 0.6571, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 8.99e-05, | |
| "loss": 0.604, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 9.49e-05, | |
| "loss": 0.5644, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 9.99e-05, | |
| "loss": 0.5703, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.34396621584892273, | |
| "eval_runtime": 464.1797, | |
| "eval_samples_per_second": 11.825, | |
| "eval_steps_per_second": 0.741, | |
| "eval_wer": 0.2586223647138832, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 9.954587581093605e-05, | |
| "loss": 0.5726, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 9.908248378127896e-05, | |
| "loss": 0.5471, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 9.861909175162187e-05, | |
| "loss": 0.5291, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 9.816033364226136e-05, | |
| "loss": 0.5082, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 9.769694161260426e-05, | |
| "loss": 0.4762, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_loss": 0.2925226390361786, | |
| "eval_runtime": 464.1083, | |
| "eval_samples_per_second": 11.827, | |
| "eval_steps_per_second": 0.741, | |
| "eval_wer": 0.21781793451860487, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 9.723354958294718e-05, | |
| "loss": 0.4993, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 9.677015755329009e-05, | |
| "loss": 0.4715, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 9.6306765523633e-05, | |
| "loss": 0.4497, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 9.584337349397591e-05, | |
| "loss": 0.45, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 9.537998146431882e-05, | |
| "loss": 0.4585, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_loss": 0.24418394267559052, | |
| "eval_runtime": 462.5404, | |
| "eval_samples_per_second": 11.867, | |
| "eval_steps_per_second": 0.744, | |
| "eval_wer": 0.19812493927912173, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 9.491658943466173e-05, | |
| "loss": 0.4531, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.445319740500464e-05, | |
| "loss": 0.5227, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.398980537534755e-05, | |
| "loss": 0.4052, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.352641334569045e-05, | |
| "loss": 0.3941, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 9.306302131603337e-05, | |
| "loss": 0.4013, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_loss": 0.24954961240291595, | |
| "eval_runtime": 463.9078, | |
| "eval_samples_per_second": 11.832, | |
| "eval_steps_per_second": 0.742, | |
| "eval_wer": 0.18175459049839696, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 9.259962928637628e-05, | |
| "loss": 0.3812, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 9.21362372567192e-05, | |
| "loss": 0.3919, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 9.16728452270621e-05, | |
| "loss": 0.389, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 9.1209453197405e-05, | |
| "loss": 0.3597, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 9.074606116774792e-05, | |
| "loss": 0.449, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_loss": 0.21517333388328552, | |
| "eval_runtime": 464.2336, | |
| "eval_samples_per_second": 11.824, | |
| "eval_steps_per_second": 0.741, | |
| "eval_wer": 0.18082191780821918, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 9.028266913809083e-05, | |
| "loss": 0.383, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 8.981927710843374e-05, | |
| "loss": 0.3625, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 8.935588507877664e-05, | |
| "loss": 0.3481, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 8.889712696941613e-05, | |
| "loss": 0.3706, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 8.843373493975903e-05, | |
| "loss": 0.355, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_loss": 0.21792450547218323, | |
| "eval_runtime": 460.2784, | |
| "eval_samples_per_second": 11.925, | |
| "eval_steps_per_second": 0.747, | |
| "eval_wer": 0.16701641892548333, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 8.797034291010195e-05, | |
| "loss": 0.3509, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 8.750695088044486e-05, | |
| "loss": 0.3432, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.704355885078776e-05, | |
| "loss": 0.3184, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 8.658016682113068e-05, | |
| "loss": 0.3203, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 8.611677479147359e-05, | |
| "loss": 0.3142, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "eval_loss": 0.19529223442077637, | |
| "eval_runtime": 460.5041, | |
| "eval_samples_per_second": 11.92, | |
| "eval_steps_per_second": 0.747, | |
| "eval_wer": 0.15418245409501602, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 8.56533827618165e-05, | |
| "loss": 0.3081, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 8.518999073215941e-05, | |
| "loss": 0.3916, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 8.472659870250233e-05, | |
| "loss": 0.3128, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 8.426320667284524e-05, | |
| "loss": 0.3095, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 8.379981464318814e-05, | |
| "loss": 0.3242, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_loss": 0.21034082770347595, | |
| "eval_runtime": 458.718, | |
| "eval_samples_per_second": 11.966, | |
| "eval_steps_per_second": 0.75, | |
| "eval_wer": 0.15256970756825028, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 8.333642261353105e-05, | |
| "loss": 0.3029, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 8.287303058387395e-05, | |
| "loss": 0.3136, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 8.240963855421687e-05, | |
| "loss": 0.3069, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 8.194624652455978e-05, | |
| "loss": 0.3158, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 8.14828544949027e-05, | |
| "loss": 0.3016, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "eval_loss": 0.19107381999492645, | |
| "eval_runtime": 460.0828, | |
| "eval_samples_per_second": 11.93, | |
| "eval_steps_per_second": 0.748, | |
| "eval_wer": 0.1477217526474303, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 8.10194624652456e-05, | |
| "loss": 0.3116, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 8.055607043558852e-05, | |
| "loss": 0.302, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 8.009731232622799e-05, | |
| "loss": 0.4194, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 7.963392029657091e-05, | |
| "loss": 0.2762, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 7.917052826691382e-05, | |
| "loss": 0.2713, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "eval_loss": 0.18356408178806305, | |
| "eval_runtime": 459.9096, | |
| "eval_samples_per_second": 11.935, | |
| "eval_steps_per_second": 0.748, | |
| "eval_wer": 0.14217429320897698, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 7.870713623725672e-05, | |
| "loss": 0.2812, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 7.824374420759963e-05, | |
| "loss": 0.2775, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 7.778035217794253e-05, | |
| "loss": 0.263, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 7.731696014828545e-05, | |
| "loss": 0.2734, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 7.685820203892493e-05, | |
| "loss": 0.2807, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "eval_loss": 0.1924305111169815, | |
| "eval_runtime": 455.2841, | |
| "eval_samples_per_second": 12.056, | |
| "eval_steps_per_second": 0.756, | |
| "eval_wer": 0.14470999708539783, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 7.639481000926784e-05, | |
| "loss": 0.2794, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 7.593141797961075e-05, | |
| "loss": 0.2705, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 7.546802594995367e-05, | |
| "loss": 0.2698, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 7.500463392029657e-05, | |
| "loss": 0.2761, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 7.454124189063948e-05, | |
| "loss": 0.2929, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "eval_loss": 0.1848202645778656, | |
| "eval_runtime": 454.4795, | |
| "eval_samples_per_second": 12.078, | |
| "eval_steps_per_second": 0.757, | |
| "eval_wer": 0.14020207908287186, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 7.40778498609824e-05, | |
| "loss": 0.2729, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 7.36144578313253e-05, | |
| "loss": 0.2654, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7.315106580166822e-05, | |
| "loss": 0.2627, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.268767377201113e-05, | |
| "loss": 0.255, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 7.222428174235403e-05, | |
| "loss": 0.2595, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "eval_loss": 0.17834417521953583, | |
| "eval_runtime": 456.3292, | |
| "eval_samples_per_second": 12.029, | |
| "eval_steps_per_second": 0.754, | |
| "eval_wer": 0.13299329641503935, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 7.176088971269694e-05, | |
| "loss": 0.2411, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 7.129749768303986e-05, | |
| "loss": 0.256, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 7.083410565338276e-05, | |
| "loss": 0.2605, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 7.037071362372567e-05, | |
| "loss": 0.2468, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 6.990732159406859e-05, | |
| "loss": 0.2289, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "eval_loss": 0.19014818966388702, | |
| "eval_runtime": 455.481, | |
| "eval_samples_per_second": 12.051, | |
| "eval_steps_per_second": 0.755, | |
| "eval_wer": 0.13133197318566014, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 6.94439295644115e-05, | |
| "loss": 0.2556, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 6.898517145505098e-05, | |
| "loss": 0.2458, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 6.852177942539389e-05, | |
| "loss": 0.2389, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 6.80583873957368e-05, | |
| "loss": 0.251, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 6.759499536607971e-05, | |
| "loss": 0.2567, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "eval_loss": 0.17836444079875946, | |
| "eval_runtime": 456.7005, | |
| "eval_samples_per_second": 12.019, | |
| "eval_steps_per_second": 0.753, | |
| "eval_wer": 0.12981638006412125, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 6.713160333642262e-05, | |
| "loss": 0.2503, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 6.666821130676552e-05, | |
| "loss": 0.2525, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 6.620481927710843e-05, | |
| "loss": 0.2416, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 6.574142724745135e-05, | |
| "loss": 0.3107, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 6.527803521779425e-05, | |
| "loss": 0.2401, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "eval_loss": 0.19564008712768555, | |
| "eval_runtime": 455.4461, | |
| "eval_samples_per_second": 12.052, | |
| "eval_steps_per_second": 0.755, | |
| "eval_wer": 0.1297775187020305, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 6.481464318813717e-05, | |
| "loss": 0.2336, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 6.435125115848008e-05, | |
| "loss": 0.2305, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 6.3887859128823e-05, | |
| "loss": 0.2109, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 6.34244670991659e-05, | |
| "loss": 0.2321, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 6.296570898980539e-05, | |
| "loss": 0.2098, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "eval_loss": 0.17475813627243042, | |
| "eval_runtime": 454.9716, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 0.756, | |
| "eval_wer": 0.12771786651122122, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "learning_rate": 6.250231696014829e-05, | |
| "loss": 0.2226, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 6.20389249304912e-05, | |
| "loss": 0.2165, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 6.157553290083412e-05, | |
| "loss": 0.2228, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 6.111214087117701e-05, | |
| "loss": 0.224, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 6.0648748841519934e-05, | |
| "loss": 0.2246, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "eval_loss": 0.1776755452156067, | |
| "eval_runtime": 454.2959, | |
| "eval_samples_per_second": 12.082, | |
| "eval_steps_per_second": 0.757, | |
| "eval_wer": 0.12535703876420867, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 6.018535681186284e-05, | |
| "loss": 0.2872, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 5.972196478220575e-05, | |
| "loss": 0.2305, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 5.925857275254866e-05, | |
| "loss": 0.2252, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 5.8795180722891576e-05, | |
| "loss": 0.2218, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 5.833178869323448e-05, | |
| "loss": 0.2197, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.17029327154159546, | |
| "eval_runtime": 456.3521, | |
| "eval_samples_per_second": 12.028, | |
| "eval_steps_per_second": 0.754, | |
| "eval_wer": 0.12218983775381327, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 5.786839666357739e-05, | |
| "loss": 0.1959, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 5.74050046339203e-05, | |
| "loss": 0.2026, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 5.6941612604263204e-05, | |
| "loss": 0.1998, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 5.6478220574606123e-05, | |
| "loss": 0.2016, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 5.601482854494903e-05, | |
| "loss": 0.2122, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "eval_loss": 0.1917443722486496, | |
| "eval_runtime": 453.8398, | |
| "eval_samples_per_second": 12.095, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.12212183037015448, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 5.555143651529194e-05, | |
| "loss": 0.2096, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 5.5088044485634847e-05, | |
| "loss": 0.2012, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 5.4624652455977766e-05, | |
| "loss": 0.2071, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "learning_rate": 5.416126042632067e-05, | |
| "loss": 0.1978, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 5.3697868396663576e-05, | |
| "loss": 0.2746, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "eval_loss": 0.1768909990787506, | |
| "eval_runtime": 454.1012, | |
| "eval_samples_per_second": 12.088, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.12147090255513456, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 5.323447636700649e-05, | |
| "loss": 0.2089, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 7.76, | |
| "learning_rate": 5.277571825764597e-05, | |
| "loss": 0.2055, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 5.231232622798888e-05, | |
| "loss": 0.197, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 5.1848934198331786e-05, | |
| "loss": 0.1963, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 5.1385542168674705e-05, | |
| "loss": 0.2148, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "eval_loss": 0.1736288219690323, | |
| "eval_runtime": 454.4442, | |
| "eval_samples_per_second": 12.078, | |
| "eval_steps_per_second": 0.757, | |
| "eval_wer": 0.11932381229962111, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 5.092215013901761e-05, | |
| "loss": 0.2049, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "learning_rate": 5.045875810936053e-05, | |
| "loss": 0.1886, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 4.9995366079703435e-05, | |
| "loss": 0.1955, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 4.953197405004634e-05, | |
| "loss": 0.1903, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "learning_rate": 4.906858202038925e-05, | |
| "loss": 0.1915, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "eval_loss": 0.18138590455055237, | |
| "eval_runtime": 453.4078, | |
| "eval_samples_per_second": 12.106, | |
| "eval_steps_per_second": 0.759, | |
| "eval_wer": 0.11608860390556688, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 4.860982391102873e-05, | |
| "loss": 0.1925, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 4.8146431881371644e-05, | |
| "loss": 0.201, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 4.768303985171455e-05, | |
| "loss": 0.2033, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 4.721964782205746e-05, | |
| "loss": 0.1893, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "learning_rate": 4.6756255792400374e-05, | |
| "loss": 0.2462, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "eval_loss": 0.1748134046792984, | |
| "eval_runtime": 459.3573, | |
| "eval_samples_per_second": 11.949, | |
| "eval_steps_per_second": 0.749, | |
| "eval_wer": 0.11656465559117847, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 4.629286376274328e-05, | |
| "loss": 0.1884, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 4.582947173308619e-05, | |
| "loss": 0.1893, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 4.5366079703429104e-05, | |
| "loss": 0.1927, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 4.4902687673772016e-05, | |
| "loss": 0.1889, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "learning_rate": 4.443929564411493e-05, | |
| "loss": 0.1872, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "eval_loss": 0.1768641322851181, | |
| "eval_runtime": 453.6375, | |
| "eval_samples_per_second": 12.1, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.11334887787816963, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 4.3975903614457834e-05, | |
| "loss": 0.1842, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.351251158480074e-05, | |
| "loss": 0.1824, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 4.304911955514365e-05, | |
| "loss": 0.1811, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 4.2585727525486564e-05, | |
| "loss": 0.1686, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 4.2122335495829476e-05, | |
| "loss": 0.1886, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "eval_loss": 0.18517228960990906, | |
| "eval_runtime": 454.3062, | |
| "eval_samples_per_second": 12.082, | |
| "eval_steps_per_second": 0.757, | |
| "eval_wer": 0.11428155056834742, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 4.1663577386468956e-05, | |
| "loss": 0.166, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 4.120018535681187e-05, | |
| "loss": 0.1796, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 4.0736793327154773e-05, | |
| "loss": 0.1744, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 4.0273401297497686e-05, | |
| "loss": 0.1798, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "learning_rate": 3.98100092678406e-05, | |
| "loss": 0.1789, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "eval_loss": 0.1695661097764969, | |
| "eval_runtime": 453.7028, | |
| "eval_samples_per_second": 12.098, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.11260079665792286, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 3.93466172381835e-05, | |
| "loss": 0.2102, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 3.8883225208526416e-05, | |
| "loss": 0.1897, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 3.841983317886933e-05, | |
| "loss": 0.1768, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 3.795644114921223e-05, | |
| "loss": 0.1751, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 3.7493049119555145e-05, | |
| "loss": 0.1692, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "eval_loss": 0.18170228600502014, | |
| "eval_runtime": 455.8222, | |
| "eval_samples_per_second": 12.042, | |
| "eval_steps_per_second": 0.755, | |
| "eval_wer": 0.11220246769649277, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "learning_rate": 3.702965708989806e-05, | |
| "loss": 0.1833, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 3.656626506024097e-05, | |
| "loss": 0.179, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 3.6102873030583875e-05, | |
| "loss": 0.1645, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 3.563948100092678e-05, | |
| "loss": 0.1699, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "learning_rate": 3.517608897126969e-05, | |
| "loss": 0.1765, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "eval_loss": 0.17690393328666687, | |
| "eval_runtime": 456.2471, | |
| "eval_samples_per_second": 12.031, | |
| "eval_steps_per_second": 0.754, | |
| "eval_wer": 0.10927815019916448, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 3.4712696941612605e-05, | |
| "loss": 0.1586, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 10.31, | |
| "learning_rate": 3.424930491195552e-05, | |
| "loss": 0.1762, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 10.37, | |
| "learning_rate": 3.378591288229842e-05, | |
| "loss": 0.1766, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 3.3322520852641335e-05, | |
| "loss": 0.1698, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 3.285912882298425e-05, | |
| "loss": 0.1699, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "eval_loss": 0.16039559245109558, | |
| "eval_runtime": 453.5742, | |
| "eval_samples_per_second": 12.102, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.10837462353055474, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "learning_rate": 3.239573679332716e-05, | |
| "loss": 0.1649, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 3.193234476367007e-05, | |
| "loss": 0.1714, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 3.146895273401298e-05, | |
| "loss": 0.1609, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 10.75, | |
| "learning_rate": 3.100556070435588e-05, | |
| "loss": 0.1678, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 3.054680259499537e-05, | |
| "loss": 0.1591, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "eval_loss": 0.17774704098701477, | |
| "eval_runtime": 452.7751, | |
| "eval_samples_per_second": 12.123, | |
| "eval_steps_per_second": 0.76, | |
| "eval_wer": 0.10795686388807928, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 3.0083410565338278e-05, | |
| "loss": 0.1589, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "learning_rate": 2.9620018535681187e-05, | |
| "loss": 0.2177, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.91566265060241e-05, | |
| "loss": 0.171, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "learning_rate": 2.8693234476367008e-05, | |
| "loss": 0.1625, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 2.822984244670992e-05, | |
| "loss": 0.1499, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "eval_loss": 0.16445568203926086, | |
| "eval_runtime": 453.5993, | |
| "eval_samples_per_second": 12.101, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.10737394345671816, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 2.7766450417052832e-05, | |
| "loss": 0.1796, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 11.26, | |
| "learning_rate": 2.7303058387395734e-05, | |
| "loss": 0.1524, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 2.6839666357738646e-05, | |
| "loss": 0.148, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 2.637627432808156e-05, | |
| "loss": 0.1566, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 2.5912882298424467e-05, | |
| "loss": 0.163, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "eval_loss": 0.17037604749202728, | |
| "eval_runtime": 453.4998, | |
| "eval_samples_per_second": 12.104, | |
| "eval_steps_per_second": 0.759, | |
| "eval_wer": 0.10646070144758574, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 2.544949026876738e-05, | |
| "loss": 0.1622, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "learning_rate": 2.498609823911029e-05, | |
| "loss": 0.1528, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 2.4522706209453197e-05, | |
| "loss": 0.1699, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 11.7, | |
| "learning_rate": 2.405931417979611e-05, | |
| "loss": 0.1587, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 11.77, | |
| "learning_rate": 2.3595922150139018e-05, | |
| "loss": 0.1597, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 11.77, | |
| "eval_loss": 0.15760080516338348, | |
| "eval_runtime": 452.5827, | |
| "eval_samples_per_second": 12.128, | |
| "eval_steps_per_second": 0.76, | |
| "eval_wer": 0.10640240940444963, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 2.31371640407785e-05, | |
| "loss": 0.1507, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 11.9, | |
| "learning_rate": 2.267377201112141e-05, | |
| "loss": 0.1675, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 11.96, | |
| "learning_rate": 2.221037998146432e-05, | |
| "loss": 0.1515, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 2.1746987951807228e-05, | |
| "loss": 0.1398, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "learning_rate": 2.128359592215014e-05, | |
| "loss": 0.1484, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "eval_loss": 0.16369372606277466, | |
| "eval_runtime": 452.8951, | |
| "eval_samples_per_second": 12.12, | |
| "eval_steps_per_second": 0.76, | |
| "eval_wer": 0.10407072767900515, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 2.0820203892493052e-05, | |
| "loss": 0.1459, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 2.0361445783132532e-05, | |
| "loss": 0.155, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 1.989805375347544e-05, | |
| "loss": 0.1655, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 1.943466172381835e-05, | |
| "loss": 0.137, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 1.8971269694161262e-05, | |
| "loss": 0.1464, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "eval_loss": 0.16305148601531982, | |
| "eval_runtime": 454.6084, | |
| "eval_samples_per_second": 12.074, | |
| "eval_steps_per_second": 0.757, | |
| "eval_wer": 0.10473137083454775, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.47, | |
| "learning_rate": 1.8507877664504174e-05, | |
| "loss": 0.1514, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 1.804448563484708e-05, | |
| "loss": 0.1485, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 1.7581093605189992e-05, | |
| "loss": 0.1524, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "learning_rate": 1.71177015755329e-05, | |
| "loss": 0.1488, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 12.72, | |
| "learning_rate": 1.6654309545875813e-05, | |
| "loss": 0.156, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.72, | |
| "eval_loss": 0.16862976551055908, | |
| "eval_runtime": 454.3682, | |
| "eval_samples_per_second": 12.081, | |
| "eval_steps_per_second": 0.757, | |
| "eval_wer": 0.10287574079471486, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.79, | |
| "learning_rate": 1.619091751621872e-05, | |
| "loss": 0.1517, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 1.572752548656163e-05, | |
| "loss": 0.1482, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 1.5264133456904543e-05, | |
| "loss": 0.1451, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "learning_rate": 1.4800741427247453e-05, | |
| "loss": 0.1506, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 1.4337349397590364e-05, | |
| "loss": 0.1625, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "eval_loss": 0.16482920944690704, | |
| "eval_runtime": 453.7585, | |
| "eval_samples_per_second": 12.097, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.10226367434178568, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 1.387395736793327e-05, | |
| "loss": 0.1317, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 1.3410565338276181e-05, | |
| "loss": 0.1543, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 1.2947173308619093e-05, | |
| "loss": 0.1401, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "learning_rate": 1.2483781278962002e-05, | |
| "loss": 0.1445, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 13.36, | |
| "learning_rate": 1.2020389249304913e-05, | |
| "loss": 0.1395, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 13.36, | |
| "eval_loss": 0.16876418888568878, | |
| "eval_runtime": 457.4225, | |
| "eval_samples_per_second": 12.0, | |
| "eval_steps_per_second": 0.752, | |
| "eval_wer": 0.10274944136791994, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "learning_rate": 1.1556997219647822e-05, | |
| "loss": 0.1385, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 1.1093605189990732e-05, | |
| "loss": 0.1528, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 1.0634847080630214e-05, | |
| "loss": 0.14, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 1.0171455050973124e-05, | |
| "loss": 0.1438, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "learning_rate": 9.708063021316035e-06, | |
| "loss": 0.1387, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "eval_loss": 0.16695508360862732, | |
| "eval_runtime": 453.4195, | |
| "eval_samples_per_second": 12.106, | |
| "eval_steps_per_second": 0.759, | |
| "eval_wer": 0.10130185563003984, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "learning_rate": 9.244670991658943e-06, | |
| "loss": 0.151, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 13.8, | |
| "learning_rate": 8.781278962001854e-06, | |
| "loss": 0.1425, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "learning_rate": 8.317886932344763e-06, | |
| "loss": 0.1429, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 7.854494902687675e-06, | |
| "loss": 0.139, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 7.391102873030584e-06, | |
| "loss": 0.1434, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_loss": 0.16770704090595245, | |
| "eval_runtime": 452.4411, | |
| "eval_samples_per_second": 12.132, | |
| "eval_steps_per_second": 0.76, | |
| "eval_wer": 0.10168075391042455, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 6.927710843373494e-06, | |
| "loss": 0.1225, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "learning_rate": 6.464318813716404e-06, | |
| "loss": 0.1387, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 6.0009267840593145e-06, | |
| "loss": 0.135, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 5.537534754402224e-06, | |
| "loss": 0.1362, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 5.074142724745135e-06, | |
| "loss": 0.1442, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "eval_loss": 0.16875232756137848, | |
| "eval_runtime": 455.5341, | |
| "eval_samples_per_second": 12.05, | |
| "eval_steps_per_second": 0.755, | |
| "eval_wer": 0.10080637326338288, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.1383, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "learning_rate": 4.151992585727526e-06, | |
| "loss": 0.1407, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "learning_rate": 3.688600556070436e-06, | |
| "loss": 0.1412, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "learning_rate": 3.225208526413346e-06, | |
| "loss": 0.1467, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 2.761816496756256e-06, | |
| "loss": 0.1439, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "eval_loss": 0.16467291116714478, | |
| "eval_runtime": 451.7915, | |
| "eval_samples_per_second": 12.149, | |
| "eval_steps_per_second": 0.761, | |
| "eval_wer": 0.10040804430195278, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 2.298424467099166e-06, | |
| "loss": 0.1439, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 1.835032437442076e-06, | |
| "loss": 0.1346, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 14.82, | |
| "learning_rate": 1.3716404077849862e-06, | |
| "loss": 0.1441, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "learning_rate": 9.082483781278962e-07, | |
| "loss": 0.1386, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 4.448563484708063e-07, | |
| "loss": 0.137, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "eval_loss": 0.1636081039905548, | |
| "eval_runtime": 454.2899, | |
| "eval_samples_per_second": 12.083, | |
| "eval_steps_per_second": 0.757, | |
| "eval_wer": 0.10055377440979306, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 23580, | |
| "total_flos": 1.645470128774266e+20, | |
| "train_loss": 0.41333935497371294, | |
| "train_runtime": 85566.3736, | |
| "train_samples_per_second": 8.82, | |
| "train_steps_per_second": 0.276 | |
| } | |
| ], | |
| "max_steps": 23580, | |
| "num_train_epochs": 15, | |
| "total_flos": 1.645470128774266e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |