| { | |
| "best_metric": 30.362728902323294, | |
| "best_model_checkpoint": "./checkpoint-5000", | |
| "epoch": 14.0368, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.5999999999999995e-08, | |
| "loss": 1.387, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.6e-08, | |
| "loss": 1.4556, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.4599999999999998e-07, | |
| "loss": 1.4467, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.96e-07, | |
| "loss": 1.4102, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.46e-07, | |
| "loss": 1.2113, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.9599999999999995e-07, | |
| "loss": 1.0881, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.4599999999999995e-07, | |
| "loss": 1.0059, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.96e-07, | |
| "loss": 0.7739, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.46e-07, | |
| "loss": 0.6612, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.96e-07, | |
| "loss": 0.5301, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.46e-07, | |
| "loss": 0.5301, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.96e-07, | |
| "loss": 0.5397, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.46e-07, | |
| "loss": 0.5302, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.959999999999999e-07, | |
| "loss": 0.4659, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 7.459999999999999e-07, | |
| "loss": 0.5914, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 7.96e-07, | |
| "loss": 0.5186, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.459999999999999e-07, | |
| "loss": 0.5277, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.96e-07, | |
| "loss": 0.4868, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.459999999999999e-07, | |
| "loss": 0.4496, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.959999999999999e-07, | |
| "loss": 0.4162, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 0.5339126586914062, | |
| "eval_runtime": 785.1652, | |
| "eval_samples_per_second": 14.054, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 35.4346068679484, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.948888888888888e-07, | |
| "loss": 0.4721, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.893333333333332e-07, | |
| "loss": 0.4764, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.837777777777778e-07, | |
| "loss": 0.4003, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.782222222222222e-07, | |
| "loss": 0.3526, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.726666666666666e-07, | |
| "loss": 0.3835, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.67111111111111e-07, | |
| "loss": 0.4236, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.615555555555556e-07, | |
| "loss": 0.4043, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.559999999999998e-07, | |
| "loss": 0.3823, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.504444444444443e-07, | |
| "loss": 0.4967, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.448888888888888e-07, | |
| "loss": 0.4285, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.393333333333334e-07, | |
| "loss": 0.4434, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.337777777777778e-07, | |
| "loss": 0.3761, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.282222222222222e-07, | |
| "loss": 0.3682, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.226666666666666e-07, | |
| "loss": 0.3629, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.171111111111111e-07, | |
| "loss": 0.4196, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.115555555555555e-07, | |
| "loss": 0.4124, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.06e-07, | |
| "loss": 0.3344, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.004444444444445e-07, | |
| "loss": 0.3146, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.948888888888888e-07, | |
| "loss": 0.3421, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.893333333333333e-07, | |
| "loss": 0.3776, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_loss": 0.4788414239883423, | |
| "eval_runtime": 776.5207, | |
| "eval_samples_per_second": 14.211, | |
| "eval_steps_per_second": 0.444, | |
| "eval_wer": 33.4891567958947, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 8.837777777777777e-07, | |
| "loss": 0.3498, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.782222222222222e-07, | |
| "loss": 0.3713, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.726666666666666e-07, | |
| "loss": 0.4487, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.671111111111111e-07, | |
| "loss": 0.3902, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.615555555555555e-07, | |
| "loss": 0.3955, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.559999999999999e-07, | |
| "loss": 0.3418, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.504444444444444e-07, | |
| "loss": 0.3173, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.448888888888888e-07, | |
| "loss": 0.3611, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.393333333333334e-07, | |
| "loss": 0.3956, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.337777777777777e-07, | |
| "loss": 0.3737, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.282222222222222e-07, | |
| "loss": 0.2894, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.226666666666666e-07, | |
| "loss": 0.2954, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 8.171111111111111e-07, | |
| "loss": 0.3201, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 8.115555555555556e-07, | |
| "loss": 0.359, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 8.06e-07, | |
| "loss": 0.3097, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.004444444444444e-07, | |
| "loss": 0.3752, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.948888888888888e-07, | |
| "loss": 0.4085, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.893333333333333e-07, | |
| "loss": 0.3599, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.837777777777777e-07, | |
| "loss": 0.3551, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.782222222222222e-07, | |
| "loss": 0.3238, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_loss": 0.456836998462677, | |
| "eval_runtime": 778.804, | |
| "eval_samples_per_second": 14.169, | |
| "eval_steps_per_second": 0.443, | |
| "eval_wer": 31.9496678499877, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.726666666666666e-07, | |
| "loss": 0.2951, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.67111111111111e-07, | |
| "loss": 0.3577, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.615555555555555e-07, | |
| "loss": 0.3706, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.559999999999999e-07, | |
| "loss": 0.3187, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.504444444444444e-07, | |
| "loss": 0.2742, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.448888888888889e-07, | |
| "loss": 0.2905, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.393333333333333e-07, | |
| "loss": 0.3154, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.337777777777778e-07, | |
| "loss": 0.3289, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.282222222222222e-07, | |
| "loss": 0.2943, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 7.226666666666667e-07, | |
| "loss": 0.389, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 7.171111111111111e-07, | |
| "loss": 0.3535, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 7.115555555555556e-07, | |
| "loss": 0.3661, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 7.059999999999999e-07, | |
| "loss": 0.3278, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.004444444444444e-07, | |
| "loss": 0.2889, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.948888888888889e-07, | |
| "loss": 0.2873, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.893333333333333e-07, | |
| "loss": 0.3464, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.837777777777778e-07, | |
| "loss": 0.3548, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 6.782222222222221e-07, | |
| "loss": 0.292, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 6.726666666666666e-07, | |
| "loss": 0.2577, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 6.67111111111111e-07, | |
| "loss": 0.2714, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "eval_loss": 0.44693753123283386, | |
| "eval_runtime": 778.4073, | |
| "eval_samples_per_second": 14.176, | |
| "eval_steps_per_second": 0.443, | |
| "eval_wer": 31.427717830656217, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 6.615555555555555e-07, | |
| "loss": 0.3127, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 6.56e-07, | |
| "loss": 0.3113, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.504444444444445e-07, | |
| "loss": 0.2835, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 6.448888888888889e-07, | |
| "loss": 0.3884, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 6.393333333333333e-07, | |
| "loss": 0.3317, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 6.34e-07, | |
| "loss": 0.3423, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 6.284444444444445e-07, | |
| "loss": 0.2979, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 6.228888888888889e-07, | |
| "loss": 0.2786, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 6.173333333333333e-07, | |
| "loss": 0.2819, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 6.117777777777777e-07, | |
| "loss": 0.3407, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 6.062222222222222e-07, | |
| "loss": 0.3473, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 6.006666666666666e-07, | |
| "loss": 0.25, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 5.951111111111111e-07, | |
| "loss": 0.2509, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.895555555555555e-07, | |
| "loss": 0.2712, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.839999999999999e-07, | |
| "loss": 0.3074, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 5.784444444444444e-07, | |
| "loss": 0.2795, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 5.728888888888888e-07, | |
| "loss": 0.2929, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.673333333333334e-07, | |
| "loss": 0.3763, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.617777777777778e-07, | |
| "loss": 0.3213, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.562222222222222e-07, | |
| "loss": 0.3232, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_loss": 0.43857714533805847, | |
| "eval_runtime": 776.2142, | |
| "eval_samples_per_second": 14.216, | |
| "eval_steps_per_second": 0.444, | |
| "eval_wer": 31.099082633299357, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.506666666666666e-07, | |
| "loss": 0.278, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.451111111111111e-07, | |
| "loss": 0.2533, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.395555555555556e-07, | |
| "loss": 0.2934, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 5.34e-07, | |
| "loss": 0.3414, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 5.284444444444445e-07, | |
| "loss": 0.3035, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 5.228888888888888e-07, | |
| "loss": 0.2427, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 5.173333333333333e-07, | |
| "loss": 0.2487, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 5.117777777777777e-07, | |
| "loss": 0.2633, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 5.062222222222222e-07, | |
| "loss": 0.2954, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 5.006666666666667e-07, | |
| "loss": 0.2622, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.951111111111111e-07, | |
| "loss": 0.3103, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.895555555555555e-07, | |
| "loss": 0.3497, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.839999999999999e-07, | |
| "loss": 0.3151, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.784444444444444e-07, | |
| "loss": 0.3004, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.728888888888889e-07, | |
| "loss": 0.2709, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.673333333333333e-07, | |
| "loss": 0.2458, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.6177777777777777e-07, | |
| "loss": 0.2998, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 4.5622222222222217e-07, | |
| "loss": 0.3245, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 4.506666666666666e-07, | |
| "loss": 0.2824, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.451111111111111e-07, | |
| "loss": 0.2324, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "eval_loss": 0.4352690875530243, | |
| "eval_runtime": 785.905, | |
| "eval_samples_per_second": 14.041, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 30.740571508910058, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.3955555555555554e-07, | |
| "loss": 0.2435, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 4.34e-07, | |
| "loss": 0.2745, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 4.2844444444444445e-07, | |
| "loss": 0.2749, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.2288888888888886e-07, | |
| "loss": 0.2531, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 4.173333333333333e-07, | |
| "loss": 0.3338, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 4.1177777777777777e-07, | |
| "loss": 0.3111, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 4.0622222222222217e-07, | |
| "loss": 0.3129, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 4.0066666666666663e-07, | |
| "loss": 0.2878, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.9511111111111114e-07, | |
| "loss": 0.2548, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.8955555555555554e-07, | |
| "loss": 0.2558, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.84e-07, | |
| "loss": 0.2963, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.7844444444444445e-07, | |
| "loss": 0.3112, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.7288888888888886e-07, | |
| "loss": 0.2602, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.673333333333333e-07, | |
| "loss": 0.2225, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.617777777777777e-07, | |
| "loss": 0.2441, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.5622222222222223e-07, | |
| "loss": 0.2739, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 3.506666666666667e-07, | |
| "loss": 0.2748, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.451111111111111e-07, | |
| "loss": 0.2493, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 3.3955555555555554e-07, | |
| "loss": 0.3451, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 3.34e-07, | |
| "loss": 0.2953, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "eval_loss": 0.43063539266586304, | |
| "eval_runtime": 775.735, | |
| "eval_samples_per_second": 14.225, | |
| "eval_steps_per_second": 0.445, | |
| "eval_wer": 30.603493726055326, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 3.284444444444444e-07, | |
| "loss": 0.3077, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 3.2288888888888886e-07, | |
| "loss": 0.2583, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 3.173333333333333e-07, | |
| "loss": 0.2527, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 3.1177777777777777e-07, | |
| "loss": 0.2498, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 3.0622222222222223e-07, | |
| "loss": 0.3043, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 3.006666666666667e-07, | |
| "loss": 0.3129, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.951111111111111e-07, | |
| "loss": 0.2311, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.8955555555555555e-07, | |
| "loss": 0.2212, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.8399999999999995e-07, | |
| "loss": 0.2403, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.784444444444444e-07, | |
| "loss": 0.2739, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 2.728888888888889e-07, | |
| "loss": 0.2567, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.673333333333333e-07, | |
| "loss": 0.2692, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.617777777777778e-07, | |
| "loss": 0.341, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.5622222222222223e-07, | |
| "loss": 0.288, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 2.5066666666666663e-07, | |
| "loss": 0.2934, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 2.451111111111111e-07, | |
| "loss": 0.2548, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.3955555555555555e-07, | |
| "loss": 0.2329, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.34e-07, | |
| "loss": 0.2731, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 2.2844444444444443e-07, | |
| "loss": 0.3028, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 2.2288888888888886e-07, | |
| "loss": 0.2878, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "eval_loss": 0.4291991889476776, | |
| "eval_runtime": 783.0355, | |
| "eval_samples_per_second": 14.093, | |
| "eval_steps_per_second": 0.441, | |
| "eval_wer": 30.427752978805668, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.1733333333333332e-07, | |
| "loss": 0.2194, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.1177777777777778e-07, | |
| "loss": 0.2206, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 2.062222222222222e-07, | |
| "loss": 0.2478, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 2.0066666666666666e-07, | |
| "loss": 0.267, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "learning_rate": 1.9511111111111112e-07, | |
| "loss": 0.2416, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1.8955555555555555e-07, | |
| "loss": 0.278, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.8399999999999998e-07, | |
| "loss": 0.3243, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.7844444444444444e-07, | |
| "loss": 0.294, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.728888888888889e-07, | |
| "loss": 0.2839, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.6733333333333332e-07, | |
| "loss": 0.2438, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.6177777777777775e-07, | |
| "loss": 0.2303, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.5622222222222224e-07, | |
| "loss": 0.2774, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.5066666666666667e-07, | |
| "loss": 0.3046, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.451111111111111e-07, | |
| "loss": 0.2681, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 1.3955555555555553e-07, | |
| "loss": 0.2079, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 1.34e-07, | |
| "loss": 0.2287, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1.2844444444444444e-07, | |
| "loss": 0.2503, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1.228888888888889e-07, | |
| "loss": 0.2646, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.1733333333333333e-07, | |
| "loss": 0.237, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1.1177777777777778e-07, | |
| "loss": 0.3077, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "eval_loss": 0.4285801947116852, | |
| "eval_runtime": 786.757, | |
| "eval_samples_per_second": 14.026, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 30.415451126498187, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1.0622222222222221e-07, | |
| "loss": 0.2994, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 1.0066666666666667e-07, | |
| "loss": 0.2908, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 9.51111111111111e-08, | |
| "loss": 0.2711, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 8.955555555555555e-08, | |
| "loss": 0.2408, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 8.4e-08, | |
| "loss": 0.2354, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 7.866666666666666e-08, | |
| "loss": 0.2875, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 7.311111111111111e-08, | |
| "loss": 0.2966, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 6.755555555555554e-08, | |
| "loss": 0.2451, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 6.2e-08, | |
| "loss": 0.2121, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 5.644444444444444e-08, | |
| "loss": 0.2249, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 5.0888888888888886e-08, | |
| "loss": 0.2615, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 4.5333333333333336e-08, | |
| "loss": 0.2547, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.977777777777778e-08, | |
| "loss": 0.2425, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 3.422222222222222e-08, | |
| "loss": 0.3237, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 2.8666666666666665e-08, | |
| "loss": 0.2843, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 2.311111111111111e-08, | |
| "loss": 0.2903, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 1.7555555555555555e-08, | |
| "loss": 0.2568, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 1.2e-08, | |
| "loss": 0.2378, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 6.444444444444444e-09, | |
| "loss": 0.2415, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 8.888888888888889e-10, | |
| "loss": 0.2914, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "eval_loss": 0.4279767870903015, | |
| "eval_runtime": 778.9506, | |
| "eval_samples_per_second": 14.166, | |
| "eval_steps_per_second": 0.443, | |
| "eval_wer": 30.362728902323294, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "step": 5000, | |
| "total_flos": 2.073340581249024e+19, | |
| "train_loss": 0.3546683967590332, | |
| "train_runtime": 22389.8509, | |
| "train_samples_per_second": 14.292, | |
| "train_steps_per_second": 0.223 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 2.073340581249024e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |