| { | |
| "best_metric": 14.884437596302003, | |
| "best_model_checkpoint": "whisper-small-pt-cv11-v4/checkpoint-5000", | |
| "epoch": 11.970534069981584, | |
| "global_step": 6500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 2.2e-07, | |
| "loss": 6.7515, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.7000000000000005e-07, | |
| "loss": 5.8564, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.2e-07, | |
| "loss": 5.5525, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.7e-07, | |
| "loss": 5.4545, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.2200000000000002e-06, | |
| "loss": 4.2057, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.4700000000000001e-06, | |
| "loss": 3.82, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.72e-06, | |
| "loss": 3.5564, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.97e-06, | |
| "loss": 3.2007, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.2200000000000003e-06, | |
| "loss": 3.0785, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.47e-06, | |
| "loss": 2.9379, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_cer": 7.057198121027907, | |
| "eval_loss": 0.47825536131858826, | |
| "eval_runtime": 193.582, | |
| "eval_samples_per_second": 5.166, | |
| "eval_steps_per_second": 0.325, | |
| "eval_wer": 17.380585516178737, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.7200000000000002e-06, | |
| "loss": 2.83, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.97e-06, | |
| "loss": 2.8714, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.2200000000000005e-06, | |
| "loss": 2.5881, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.4700000000000002e-06, | |
| "loss": 2.2805, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.7200000000000004e-06, | |
| "loss": 2.5818, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 3.97e-06, | |
| "loss": 2.9589, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 4.22e-06, | |
| "loss": 2.5782, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 4.47e-06, | |
| "loss": 2.3677, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.7200000000000005e-06, | |
| "loss": 2.5862, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.970000000000001e-06, | |
| "loss": 2.1727, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_cer": 6.797457861287648, | |
| "eval_loss": 0.3721262812614441, | |
| "eval_runtime": 195.8834, | |
| "eval_samples_per_second": 5.105, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 17.272727272727273, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 4.9755555555555554e-06, | |
| "loss": 2.0283, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.947777777777778e-06, | |
| "loss": 2.1114, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 4.92e-06, | |
| "loss": 2.0833, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 4.892222222222222e-06, | |
| "loss": 2.0559, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 4.864444444444445e-06, | |
| "loss": 1.9963, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 4.836666666666667e-06, | |
| "loss": 2.1794, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 4.808888888888889e-06, | |
| "loss": 2.1072, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.781111111111111e-06, | |
| "loss": 1.8907, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 4.753333333333333e-06, | |
| "loss": 1.8439, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 4.725555555555556e-06, | |
| "loss": 1.7856, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_cer": 6.402321083172147, | |
| "eval_loss": 0.3466031551361084, | |
| "eval_runtime": 198.2589, | |
| "eval_samples_per_second": 5.044, | |
| "eval_steps_per_second": 0.318, | |
| "eval_wer": 16.37904468412943, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 4.697777777777778e-06, | |
| "loss": 1.9889, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 4.670000000000001e-06, | |
| "loss": 1.9626, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.642222222222222e-06, | |
| "loss": 2.2689, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.614444444444445e-06, | |
| "loss": 1.5772, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 4.586666666666667e-06, | |
| "loss": 1.8441, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 4.558888888888889e-06, | |
| "loss": 1.6514, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 4.531111111111112e-06, | |
| "loss": 1.9828, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 4.503333333333333e-06, | |
| "loss": 1.8731, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 4.475555555555556e-06, | |
| "loss": 1.6879, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 4.447777777777778e-06, | |
| "loss": 1.7803, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_cer": 6.20889748549323, | |
| "eval_loss": 0.33716297149658203, | |
| "eval_runtime": 196.0701, | |
| "eval_samples_per_second": 5.1, | |
| "eval_steps_per_second": 0.321, | |
| "eval_wer": 15.901386748844375, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.42e-06, | |
| "loss": 1.8693, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.392222222222223e-06, | |
| "loss": 1.7551, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 4.364444444444445e-06, | |
| "loss": 2.2832, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 4.336666666666667e-06, | |
| "loss": 1.9009, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 4.308888888888889e-06, | |
| "loss": 1.5534, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 4.281111111111112e-06, | |
| "loss": 1.5566, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 4.253333333333334e-06, | |
| "loss": 1.5025, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 4.225555555555556e-06, | |
| "loss": 1.4498, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 4.197777777777779e-06, | |
| "loss": 1.4484, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 4.17e-06, | |
| "loss": 1.8312, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "eval_cer": 6.090080132633324, | |
| "eval_loss": 0.3303045630455017, | |
| "eval_runtime": 197.9104, | |
| "eval_samples_per_second": 5.053, | |
| "eval_steps_per_second": 0.318, | |
| "eval_wer": 15.747303543913713, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 4.142222222222223e-06, | |
| "loss": 1.5531, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 4.114444444444445e-06, | |
| "loss": 1.6399, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 4.086666666666667e-06, | |
| "loss": 1.68, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 4.05888888888889e-06, | |
| "loss": 1.5905, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 4.031111111111111e-06, | |
| "loss": 1.4962, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.003333333333334e-06, | |
| "loss": 1.283, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 3.975555555555556e-06, | |
| "loss": 1.5967, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 3.9477777777777786e-06, | |
| "loss": 1.627, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 1.6596, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 3.892222222222223e-06, | |
| "loss": 1.6403, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "eval_cer": 6.189555125725339, | |
| "eval_loss": 0.3256286084651947, | |
| "eval_runtime": 196.2193, | |
| "eval_samples_per_second": 5.096, | |
| "eval_steps_per_second": 0.321, | |
| "eval_wer": 15.947611710323576, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 3.864444444444445e-06, | |
| "loss": 1.6811, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 3.836666666666667e-06, | |
| "loss": 1.3881, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 3.808888888888889e-06, | |
| "loss": 1.9165, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 3.7811111111111115e-06, | |
| "loss": 1.7782, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 3.753333333333334e-06, | |
| "loss": 1.6915, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 3.7255555555555556e-06, | |
| "loss": 1.6901, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 3.697777777777778e-06, | |
| "loss": 1.8462, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 3.6700000000000004e-06, | |
| "loss": 1.3781, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 3.6422222222222225e-06, | |
| "loss": 1.4356, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 3.614444444444445e-06, | |
| "loss": 1.536, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_cer": 6.09284332688588, | |
| "eval_loss": 0.32352736592292786, | |
| "eval_runtime": 196.5479, | |
| "eval_samples_per_second": 5.088, | |
| "eval_steps_per_second": 0.321, | |
| "eval_wer": 15.500770416024654, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 3.5866666666666673e-06, | |
| "loss": 1.4201, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 3.558888888888889e-06, | |
| "loss": 1.4551, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 3.5311111111111114e-06, | |
| "loss": 1.5927, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3.5033333333333334e-06, | |
| "loss": 1.357, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 3.475555555555556e-06, | |
| "loss": 1.5396, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.4477777777777783e-06, | |
| "loss": 1.5673, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 1.3435, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 3.3922222222222223e-06, | |
| "loss": 1.4499, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 3.3644444444444447e-06, | |
| "loss": 1.4198, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 3.3366666666666668e-06, | |
| "loss": 1.4223, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "eval_cer": 6.0735009671179885, | |
| "eval_loss": 0.32085078954696655, | |
| "eval_runtime": 197.2684, | |
| "eval_samples_per_second": 5.069, | |
| "eval_steps_per_second": 0.319, | |
| "eval_wer": 15.362095531587059, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 3.308888888888889e-06, | |
| "loss": 1.5456, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 3.2811111111111116e-06, | |
| "loss": 1.2197, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 3.2533333333333332e-06, | |
| "loss": 1.4893, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 3.2255555555555557e-06, | |
| "loss": 1.4071, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 3.197777777777778e-06, | |
| "loss": 1.5669, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 3.17e-06, | |
| "loss": 1.3333, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 3.1422222222222226e-06, | |
| "loss": 1.4102, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 3.114444444444445e-06, | |
| "loss": 1.525, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 3.0866666666666666e-06, | |
| "loss": 1.6757, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 3.058888888888889e-06, | |
| "loss": 1.4652, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "eval_cer": 5.932578060237635, | |
| "eval_loss": 0.3209191560745239, | |
| "eval_runtime": 200.0634, | |
| "eval_samples_per_second": 4.998, | |
| "eval_steps_per_second": 0.315, | |
| "eval_wer": 15.26964560862866, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 3.0311111111111115e-06, | |
| "loss": 1.6026, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 3.0033333333333335e-06, | |
| "loss": 1.425, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 2.975555555555556e-06, | |
| "loss": 1.548, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 2.9477777777777784e-06, | |
| "loss": 1.1192, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 2.92e-06, | |
| "loss": 1.5373, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 2.8922222222222224e-06, | |
| "loss": 1.4464, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 2.864444444444445e-06, | |
| "loss": 1.0499, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 2.836666666666667e-06, | |
| "loss": 1.2791, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 2.8088888888888893e-06, | |
| "loss": 1.2165, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 2.781111111111111e-06, | |
| "loss": 1.2572, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "eval_cer": 5.74468085106383, | |
| "eval_loss": 0.3191336393356323, | |
| "eval_runtime": 195.9428, | |
| "eval_samples_per_second": 5.104, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 14.884437596302003, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 2.7533333333333334e-06, | |
| "loss": 1.4301, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 2.725555555555556e-06, | |
| "loss": 1.4087, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.6977777777777783e-06, | |
| "loss": 1.6344, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 2.6700000000000003e-06, | |
| "loss": 1.4237, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 2.6422222222222227e-06, | |
| "loss": 1.5428, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 2.6144444444444443e-06, | |
| "loss": 1.2484, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 2.5866666666666667e-06, | |
| "loss": 1.424, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "learning_rate": 2.558888888888889e-06, | |
| "loss": 1.4562, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.531111111111111e-06, | |
| "loss": 1.2812, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "learning_rate": 2.5033333333333336e-06, | |
| "loss": 1.7142, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "eval_cer": 5.8469190384084, | |
| "eval_loss": 0.3182465732097626, | |
| "eval_runtime": 195.9115, | |
| "eval_samples_per_second": 5.104, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 15.007704160246533, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 2.4755555555555557e-06, | |
| "loss": 1.238, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 10.31, | |
| "learning_rate": 2.447777777777778e-06, | |
| "loss": 1.178, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 2.42e-06, | |
| "loss": 1.3038, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 2.3922222222222226e-06, | |
| "loss": 1.3901, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 2.3644444444444446e-06, | |
| "loss": 1.4346, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 2.3366666666666666e-06, | |
| "loss": 1.3559, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 2.308888888888889e-06, | |
| "loss": 1.2192, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 2.281111111111111e-06, | |
| "loss": 1.0437, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 2.2533333333333335e-06, | |
| "loss": 1.8052, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.225555555555556e-06, | |
| "loss": 1.4195, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "eval_cer": 5.885603757944184, | |
| "eval_loss": 0.3170944154262543, | |
| "eval_runtime": 198.0232, | |
| "eval_samples_per_second": 5.05, | |
| "eval_steps_per_second": 0.318, | |
| "eval_wer": 15.069337442218798, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 2.197777777777778e-06, | |
| "loss": 1.1544, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "learning_rate": 2.17e-06, | |
| "loss": 1.3013, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 11.33, | |
| "learning_rate": 2.1422222222222224e-06, | |
| "loss": 1.4788, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 11.42, | |
| "learning_rate": 2.1144444444444444e-06, | |
| "loss": 1.345, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 2.086666666666667e-06, | |
| "loss": 1.4488, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 2.0588888888888893e-06, | |
| "loss": 1.0227, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 11.69, | |
| "learning_rate": 2.0311111111111113e-06, | |
| "loss": 1.4288, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 2.0033333333333334e-06, | |
| "loss": 1.0756, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 1.975555555555556e-06, | |
| "loss": 1.3951, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "learning_rate": 1.947777777777778e-06, | |
| "loss": 1.3965, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "eval_cer": 5.857971815418623, | |
| "eval_loss": 0.3166500926017761, | |
| "eval_runtime": 195.9821, | |
| "eval_samples_per_second": 5.103, | |
| "eval_steps_per_second": 0.321, | |
| "eval_wer": 15.053929121725732, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "step": 6500, | |
| "total_flos": 6.002576326656e+19, | |
| "train_loss": 1.8605577674278846, | |
| "train_runtime": 13282.9377, | |
| "train_samples_per_second": 24.091, | |
| "train_steps_per_second": 0.753 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 19, | |
| "total_flos": 6.002576326656e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |