| { | |
| "best_metric": 39.89950864725314, | |
| "best_model_checkpoint": "./whisper-small-Persian/checkpoint-5000", | |
| "epoch": 15.57542768273717, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.4e-08, | |
| "loss": 1.4232, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.4e-08, | |
| "loss": 1.4027, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.44e-07, | |
| "loss": 1.3731, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.94e-07, | |
| "loss": 1.2969, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.4399999999999996e-07, | |
| "loss": 1.1964, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.9399999999999996e-07, | |
| "loss": 1.0995, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.4399999999999996e-07, | |
| "loss": 1.0068, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.94e-07, | |
| "loss": 0.8949, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.44e-07, | |
| "loss": 0.8521, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.94e-07, | |
| "loss": 0.7694, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 5.44e-07, | |
| "loss": 0.7196, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 5.939999999999999e-07, | |
| "loss": 0.6902, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.44e-07, | |
| "loss": 0.6748, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.939999999999999e-07, | |
| "loss": 0.6247, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 7.44e-07, | |
| "loss": 0.63, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 7.94e-07, | |
| "loss": 0.5899, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 8.439999999999999e-07, | |
| "loss": 0.5628, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 8.939999999999999e-07, | |
| "loss": 0.5634, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 9.439999999999999e-07, | |
| "loss": 0.5549, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.94e-07, | |
| "loss": 0.5533, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 0.7044046521186829, | |
| "eval_runtime": 1265.1903, | |
| "eval_samples_per_second": 8.132, | |
| "eval_steps_per_second": 0.127, | |
| "eval_wer": 54.54987091580379, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.95111111111111e-07, | |
| "loss": 0.5329, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 9.895555555555554e-07, | |
| "loss": 0.5116, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 9.84e-07, | |
| "loss": 0.511, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 9.784444444444444e-07, | |
| "loss": 0.4963, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 9.728888888888888e-07, | |
| "loss": 0.5177, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.673333333333332e-07, | |
| "loss": 0.4947, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.617777777777776e-07, | |
| "loss": 0.4529, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 9.562222222222223e-07, | |
| "loss": 0.4543, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 9.506666666666667e-07, | |
| "loss": 0.4478, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 9.451111111111111e-07, | |
| "loss": 0.4421, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 9.395555555555556e-07, | |
| "loss": 0.4465, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 9.34e-07, | |
| "loss": 0.4164, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 9.284444444444444e-07, | |
| "loss": 0.4291, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 9.228888888888888e-07, | |
| "loss": 0.4149, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 9.173333333333333e-07, | |
| "loss": 0.4242, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 9.117777777777778e-07, | |
| "loss": 0.4329, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 9.062222222222222e-07, | |
| "loss": 0.4165, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 9.006666666666666e-07, | |
| "loss": 0.4166, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.95111111111111e-07, | |
| "loss": 0.409, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 8.895555555555555e-07, | |
| "loss": 0.3951, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_loss": 0.5893104076385498, | |
| "eval_runtime": 1326.5175, | |
| "eval_samples_per_second": 7.756, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 47.521028231962916, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 8.839999999999999e-07, | |
| "loss": 0.3808, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 8.784444444444444e-07, | |
| "loss": 0.3786, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 8.728888888888889e-07, | |
| "loss": 0.3811, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 8.673333333333332e-07, | |
| "loss": 0.3644, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 8.617777777777777e-07, | |
| "loss": 0.3737, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 8.562222222222222e-07, | |
| "loss": 0.3698, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 8.506666666666667e-07, | |
| "loss": 0.3601, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 8.451111111111111e-07, | |
| "loss": 0.3675, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 8.395555555555556e-07, | |
| "loss": 0.3588, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 8.34e-07, | |
| "loss": 0.3541, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 8.284444444444444e-07, | |
| "loss": 0.3537, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 8.228888888888889e-07, | |
| "loss": 0.3666, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 8.173333333333333e-07, | |
| "loss": 0.3242, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 8.117777777777778e-07, | |
| "loss": 0.3212, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 8.062222222222221e-07, | |
| "loss": 0.3374, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 8.006666666666666e-07, | |
| "loss": 0.3259, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 7.95111111111111e-07, | |
| "loss": 0.3361, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 7.895555555555555e-07, | |
| "loss": 0.3355, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 7.84e-07, | |
| "loss": 0.3254, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 7.784444444444444e-07, | |
| "loss": 0.3296, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_loss": 0.5428555011749268, | |
| "eval_runtime": 1299.6224, | |
| "eval_samples_per_second": 7.916, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 44.64508786053355, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 7.728888888888888e-07, | |
| "loss": 0.3229, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 7.673333333333332e-07, | |
| "loss": 0.3372, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 7.617777777777778e-07, | |
| "loss": 0.3297, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 7.562222222222222e-07, | |
| "loss": 0.3239, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 7.506666666666667e-07, | |
| "loss": 0.3113, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 7.451111111111111e-07, | |
| "loss": 0.3142, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 7.395555555555555e-07, | |
| "loss": 0.303, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 7.34e-07, | |
| "loss": 0.2973, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 7.284444444444444e-07, | |
| "loss": 0.2998, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 7.228888888888889e-07, | |
| "loss": 0.2962, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 7.173333333333333e-07, | |
| "loss": 0.3007, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 7.117777777777777e-07, | |
| "loss": 0.2906, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "learning_rate": 7.062222222222222e-07, | |
| "loss": 0.295, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 7.006666666666666e-07, | |
| "loss": 0.3083, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 6.951111111111111e-07, | |
| "loss": 0.2869, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 6.895555555555555e-07, | |
| "loss": 0.2936, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.84e-07, | |
| "loss": 0.2982, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 6.784444444444443e-07, | |
| "loss": 0.2885, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 6.728888888888888e-07, | |
| "loss": 0.2719, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 6.673333333333334e-07, | |
| "loss": 0.2662, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "eval_loss": 0.5223153829574585, | |
| "eval_runtime": 1347.3336, | |
| "eval_samples_per_second": 7.636, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 42.664408849901456, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 6.617777777777778e-07, | |
| "loss": 0.2739, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 6.562222222222223e-07, | |
| "loss": 0.2694, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 6.506666666666666e-07, | |
| "loss": 0.2678, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 6.451111111111111e-07, | |
| "loss": 0.2716, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 6.395555555555555e-07, | |
| "loss": 0.27, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 6.34e-07, | |
| "loss": 0.2715, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 6.284444444444445e-07, | |
| "loss": 0.2705, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 6.228888888888889e-07, | |
| "loss": 0.2771, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 6.173333333333333e-07, | |
| "loss": 0.2814, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 6.117777777777777e-07, | |
| "loss": 0.302, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 6.062222222222222e-07, | |
| "loss": 0.2619, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 6.006666666666666e-07, | |
| "loss": 0.2561, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 5.951111111111111e-07, | |
| "loss": 0.2559, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 5.895555555555555e-07, | |
| "loss": 0.2524, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 5.839999999999999e-07, | |
| "loss": 0.2583, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 5.784444444444444e-07, | |
| "loss": 0.2434, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 5.728888888888888e-07, | |
| "loss": 0.2546, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 5.673333333333334e-07, | |
| "loss": 0.2492, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "learning_rate": 5.617777777777778e-07, | |
| "loss": 0.2544, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 5.562222222222222e-07, | |
| "loss": 0.2535, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "eval_loss": 0.5044606924057007, | |
| "eval_runtime": 1321.646, | |
| "eval_samples_per_second": 7.784, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 41.53041112622491, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 5.506666666666666e-07, | |
| "loss": 0.2591, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 5.451111111111111e-07, | |
| "loss": 0.2583, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 5.395555555555556e-07, | |
| "loss": 0.2566, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 5.34e-07, | |
| "loss": 0.2384, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 5.284444444444445e-07, | |
| "loss": 0.2447, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 5.228888888888888e-07, | |
| "loss": 0.236, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 5.173333333333333e-07, | |
| "loss": 0.2419, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 5.117777777777777e-07, | |
| "loss": 0.2426, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 5.062222222222222e-07, | |
| "loss": 0.2431, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 5.006666666666667e-07, | |
| "loss": 0.242, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 4.951111111111111e-07, | |
| "loss": 0.2384, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 4.895555555555555e-07, | |
| "loss": 0.2436, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 4.839999999999999e-07, | |
| "loss": 0.2453, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 4.784444444444444e-07, | |
| "loss": 0.2286, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 4.728888888888889e-07, | |
| "loss": 0.236, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.673333333333333e-07, | |
| "loss": 0.2303, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 4.6177777777777777e-07, | |
| "loss": 0.2184, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 4.5622222222222217e-07, | |
| "loss": 0.2256, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 4.506666666666666e-07, | |
| "loss": 0.2316, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 4.451111111111111e-07, | |
| "loss": 0.224, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "eval_loss": 0.5001869201660156, | |
| "eval_runtime": 1349.9905, | |
| "eval_samples_per_second": 7.621, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 40.882214140965495, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 4.3955555555555554e-07, | |
| "loss": 0.2199, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 4.34e-07, | |
| "loss": 0.2358, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 4.2844444444444445e-07, | |
| "loss": 0.2392, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 4.2288888888888886e-07, | |
| "loss": 0.2226, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 4.173333333333333e-07, | |
| "loss": 0.2223, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "learning_rate": 4.1177777777777777e-07, | |
| "loss": 0.2359, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 4.0622222222222217e-07, | |
| "loss": 0.2255, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "learning_rate": 4.0066666666666663e-07, | |
| "loss": 0.2312, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 3.9511111111111114e-07, | |
| "loss": 0.2126, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 3.8955555555555554e-07, | |
| "loss": 0.2168, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 3.84e-07, | |
| "loss": 0.2014, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 3.7844444444444445e-07, | |
| "loss": 0.2123, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "learning_rate": 3.7288888888888886e-07, | |
| "loss": 0.2253, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 10.44, | |
| "learning_rate": 3.673333333333333e-07, | |
| "loss": 0.2134, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 10.51, | |
| "learning_rate": 3.617777777777777e-07, | |
| "loss": 0.212, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 3.5622222222222223e-07, | |
| "loss": 0.211, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "learning_rate": 3.506666666666667e-07, | |
| "loss": 0.2247, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 10.75, | |
| "learning_rate": 3.451111111111111e-07, | |
| "loss": 0.2118, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 3.3955555555555554e-07, | |
| "loss": 0.217, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "learning_rate": 3.34e-07, | |
| "loss": 0.2204, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "eval_loss": 0.49665194749832153, | |
| "eval_runtime": 1350.5042, | |
| "eval_samples_per_second": 7.618, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 40.30758126752353, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "learning_rate": 3.284444444444444e-07, | |
| "loss": 0.2276, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 3.2288888888888886e-07, | |
| "loss": 0.215, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 3.173333333333333e-07, | |
| "loss": 0.2065, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "learning_rate": 3.1177777777777777e-07, | |
| "loss": 0.2141, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 3.0622222222222223e-07, | |
| "loss": 0.2064, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 11.37, | |
| "learning_rate": 3.006666666666667e-07, | |
| "loss": 0.2039, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 2.951111111111111e-07, | |
| "loss": 0.2146, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 11.53, | |
| "learning_rate": 2.8955555555555555e-07, | |
| "loss": 0.2031, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 2.8399999999999995e-07, | |
| "loss": 0.2091, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 11.68, | |
| "learning_rate": 2.784444444444444e-07, | |
| "loss": 0.2105, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 2.728888888888889e-07, | |
| "loss": 0.204, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 11.84, | |
| "learning_rate": 2.673333333333333e-07, | |
| "loss": 0.2096, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 11.91, | |
| "learning_rate": 2.617777777777778e-07, | |
| "loss": 0.2051, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 2.5622222222222223e-07, | |
| "loss": 0.2063, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 2.5066666666666663e-07, | |
| "loss": 0.205, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 2.451111111111111e-07, | |
| "loss": 0.2041, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 2.3955555555555555e-07, | |
| "loss": 0.1955, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 12.3, | |
| "learning_rate": 2.34e-07, | |
| "loss": 0.2116, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 2.2844444444444443e-07, | |
| "loss": 0.1994, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 12.46, | |
| "learning_rate": 2.2288888888888886e-07, | |
| "loss": 0.2024, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.46, | |
| "eval_loss": 0.4951171875, | |
| "eval_runtime": 1349.6224, | |
| "eval_samples_per_second": 7.623, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 39.98834078227799, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.54, | |
| "learning_rate": 2.1755555555555554e-07, | |
| "loss": 0.2007, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 2.12e-07, | |
| "loss": 0.1959, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 2.0644444444444445e-07, | |
| "loss": 0.2056, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "learning_rate": 2.0088888888888888e-07, | |
| "loss": 0.2017, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 1.953333333333333e-07, | |
| "loss": 0.1976, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 12.93, | |
| "learning_rate": 1.8977777777777777e-07, | |
| "loss": 0.201, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1.8422222222222222e-07, | |
| "loss": 0.202, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "learning_rate": 1.7866666666666665e-07, | |
| "loss": 0.19, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 1.731111111111111e-07, | |
| "loss": 0.1999, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 1.6755555555555556e-07, | |
| "loss": 0.1981, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 13.32, | |
| "learning_rate": 1.62e-07, | |
| "loss": 0.1914, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 1.5644444444444442e-07, | |
| "loss": 0.2056, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 13.47, | |
| "learning_rate": 1.5088888888888888e-07, | |
| "loss": 0.1925, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 1.4533333333333334e-07, | |
| "loss": 0.1945, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 13.63, | |
| "learning_rate": 1.3977777777777777e-07, | |
| "loss": 0.1997, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 1.342222222222222e-07, | |
| "loss": 0.1856, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 13.78, | |
| "learning_rate": 1.2866666666666668e-07, | |
| "loss": 0.2028, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 1.231111111111111e-07, | |
| "loss": 0.1924, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 1.1755555555555554e-07, | |
| "loss": 0.194, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 1.12e-07, | |
| "loss": 0.2099, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "eval_loss": 0.49205756187438965, | |
| "eval_runtime": 1375.2125, | |
| "eval_samples_per_second": 7.481, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 39.9841767759487, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 1.0644444444444444e-07, | |
| "loss": 0.1899, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "learning_rate": 1.0088888888888888e-07, | |
| "loss": 0.1927, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 9.533333333333334e-08, | |
| "loss": 0.1888, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 14.33, | |
| "learning_rate": 8.977777777777777e-08, | |
| "loss": 0.1938, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 8.422222222222223e-08, | |
| "loss": 0.1912, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 7.866666666666666e-08, | |
| "loss": 0.1874, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 14.56, | |
| "learning_rate": 7.311111111111111e-08, | |
| "loss": 0.1858, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 14.64, | |
| "learning_rate": 6.755555555555554e-08, | |
| "loss": 0.1903, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 6.2e-08, | |
| "loss": 0.1843, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "learning_rate": 5.644444444444444e-08, | |
| "loss": 0.197, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 14.87, | |
| "learning_rate": 5.0888888888888886e-08, | |
| "loss": 0.2016, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 4.5333333333333336e-08, | |
| "loss": 0.2038, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 3.977777777777778e-08, | |
| "loss": 0.1978, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 3.422222222222222e-08, | |
| "loss": 0.1945, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 15.19, | |
| "learning_rate": 2.8666666666666665e-08, | |
| "loss": 0.1972, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 15.26, | |
| "learning_rate": 2.311111111111111e-08, | |
| "loss": 0.1834, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 15.34, | |
| "learning_rate": 1.7555555555555555e-08, | |
| "loss": 0.2, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 15.42, | |
| "learning_rate": 1.2e-08, | |
| "loss": 0.1897, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 6.444444444444444e-09, | |
| "loss": 0.1903, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "learning_rate": 8.888888888888889e-10, | |
| "loss": 0.1836, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "eval_loss": 0.4923795163631439, | |
| "eval_runtime": 1373.8353, | |
| "eval_samples_per_second": 7.489, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 39.89950864725314, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 16, | |
| "total_flos": 9.248584972351046e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |