| { | |
| "best_metric": 96.06429012981661, | |
| "best_model_checkpoint": "./another_local/checkpoint-1500", | |
| "epoch": 12.779552715654953, | |
| "eval_steps": 500, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 3.5849, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 2.9636, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.42e-06, | |
| "loss": 2.182, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.45, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.42e-06, | |
| "loss": 0.9854, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.92e-06, | |
| "loss": 0.889, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.8304, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.7477, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.6921, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.5867, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.4866, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.3247, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.2519, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.1937, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.2031, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.2097, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.1868, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.1966, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.1949, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.1972, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.426580011844635, | |
| "eval_runtime": 2503.3967, | |
| "eval_samples_per_second": 0.799, | |
| "eval_steps_per_second": 0.1, | |
| "eval_wer": 105.19266433134145, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 0.1983, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 9.86857142857143e-06, | |
| "loss": 0.2248, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 9.797142857142858e-06, | |
| "loss": 0.1991, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 9.725714285714287e-06, | |
| "loss": 0.2179, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.654285714285716e-06, | |
| "loss": 0.2157, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.582857142857143e-06, | |
| "loss": 0.0822, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.511428571428572e-06, | |
| "loss": 0.0778, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 9.440000000000001e-06, | |
| "loss": 0.0879, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 9.368571428571428e-06, | |
| "loss": 0.0868, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 9.297142857142857e-06, | |
| "loss": 0.0785, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 9.225714285714286e-06, | |
| "loss": 0.075, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 9.154285714285715e-06, | |
| "loss": 0.0804, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 9.082857142857143e-06, | |
| "loss": 0.0794, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 9.011428571428572e-06, | |
| "loss": 0.0797, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 8.94e-06, | |
| "loss": 0.076, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 8.86857142857143e-06, | |
| "loss": 0.0802, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 8.797142857142857e-06, | |
| "loss": 0.0755, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.725714285714286e-06, | |
| "loss": 0.0619, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 8.654285714285715e-06, | |
| "loss": 0.0327, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 8.582857142857144e-06, | |
| "loss": 0.033, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "eval_loss": 0.45066991448402405, | |
| "eval_runtime": 2606.2897, | |
| "eval_samples_per_second": 0.767, | |
| "eval_steps_per_second": 0.096, | |
| "eval_wer": 209.08201112713786, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 8.511428571428571e-06, | |
| "loss": 0.0358, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 8.44e-06, | |
| "loss": 0.0293, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 8.36857142857143e-06, | |
| "loss": 0.0332, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 8.297142857142859e-06, | |
| "loss": 0.0354, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 8.225714285714288e-06, | |
| "loss": 0.036, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 8.154285714285715e-06, | |
| "loss": 0.035, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 8.082857142857144e-06, | |
| "loss": 0.0284, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.011428571428573e-06, | |
| "loss": 0.0366, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 7.94e-06, | |
| "loss": 0.0306, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 7.86857142857143e-06, | |
| "loss": 0.0313, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 7.797142857142858e-06, | |
| "loss": 0.0132, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 7.725714285714286e-06, | |
| "loss": 0.0193, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 7.654285714285715e-06, | |
| "loss": 0.0147, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 7.5828571428571444e-06, | |
| "loss": 0.0202, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 7.511428571428572e-06, | |
| "loss": 0.0145, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 7.440000000000001e-06, | |
| "loss": 0.0148, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 7.36857142857143e-06, | |
| "loss": 0.0191, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 7.297142857142858e-06, | |
| "loss": 0.016, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 7.225714285714286e-06, | |
| "loss": 0.0166, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 7.154285714285715e-06, | |
| "loss": 0.0167, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "eval_loss": 0.4738132357597351, | |
| "eval_runtime": 2193.3956, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.114, | |
| "eval_wer": 96.06429012981661, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 7.082857142857143e-06, | |
| "loss": 0.0154, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 7.011428571428572e-06, | |
| "loss": 0.0162, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.9400000000000005e-06, | |
| "loss": 0.0115, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 6.868571428571429e-06, | |
| "loss": 0.0091, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 6.797142857142858e-06, | |
| "loss": 0.0092, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 6.725714285714287e-06, | |
| "loss": 0.0115, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 6.654285714285716e-06, | |
| "loss": 0.0076, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 6.582857142857143e-06, | |
| "loss": 0.0089, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 6.511428571428572e-06, | |
| "loss": 0.0115, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 6.440000000000001e-06, | |
| "loss": 0.0086, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 6.368571428571429e-06, | |
| "loss": 0.0062, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 6.297142857142857e-06, | |
| "loss": 0.0074, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 6.225714285714286e-06, | |
| "loss": 0.0084, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 6.1542857142857145e-06, | |
| "loss": 0.0061, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 6.0828571428571435e-06, | |
| "loss": 0.0068, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 6.011428571428572e-06, | |
| "loss": 0.0093, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 5.94e-06, | |
| "loss": 0.0068, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 5.868571428571429e-06, | |
| "loss": 0.0067, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 5.797142857142858e-06, | |
| "loss": 0.0049, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 5.725714285714287e-06, | |
| "loss": 0.0052, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "eval_loss": 0.4947805106639862, | |
| "eval_runtime": 2469.5913, | |
| "eval_samples_per_second": 0.81, | |
| "eval_steps_per_second": 0.101, | |
| "eval_wer": 143.76159076859673, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 5.654285714285714e-06, | |
| "loss": 0.0045, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 5.582857142857143e-06, | |
| "loss": 0.0033, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 5.511428571428572e-06, | |
| "loss": 0.0054, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 5.4400000000000004e-06, | |
| "loss": 0.0034, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 5.368571428571429e-06, | |
| "loss": 0.0059, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 5.297142857142858e-06, | |
| "loss": 0.0043, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 5.225714285714286e-06, | |
| "loss": 0.0051, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.154285714285715e-06, | |
| "loss": 0.0062, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 5.082857142857144e-06, | |
| "loss": 0.0038, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 5.011428571428571e-06, | |
| "loss": 0.0033, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 4.94e-06, | |
| "loss": 0.0039, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 4.868571428571429e-06, | |
| "loss": 0.0027, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 4.797142857142857e-06, | |
| "loss": 0.0032, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 4.725714285714286e-06, | |
| "loss": 0.0032, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 4.6542857142857145e-06, | |
| "loss": 0.0027, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 4.5828571428571435e-06, | |
| "loss": 0.0028, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 4.511428571428572e-06, | |
| "loss": 0.0046, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 4.440000000000001e-06, | |
| "loss": 0.0034, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 4.368571428571429e-06, | |
| "loss": 0.0061, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 4.297142857142858e-06, | |
| "loss": 0.0035, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_loss": 0.5143901109695435, | |
| "eval_runtime": 2403.8841, | |
| "eval_samples_per_second": 0.832, | |
| "eval_steps_per_second": 0.104, | |
| "eval_wer": 126.81331135380178, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 4.225714285714286e-06, | |
| "loss": 0.0036, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 4.154285714285714e-06, | |
| "loss": 0.0037, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 4.082857142857143e-06, | |
| "loss": 0.0025, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "learning_rate": 4.011428571428571e-06, | |
| "loss": 0.0031, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 3.94e-06, | |
| "loss": 0.0022, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 3.8685714285714286e-06, | |
| "loss": 0.0034, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 3.7971428571428576e-06, | |
| "loss": 0.0021, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 3.7257142857142857e-06, | |
| "loss": 0.0032, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 3.6542857142857148e-06, | |
| "loss": 0.0028, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 3.582857142857143e-06, | |
| "loss": 0.0032, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 3.511428571428572e-06, | |
| "loss": 0.0035, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 3.44e-06, | |
| "loss": 0.0023, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.3685714285714287e-06, | |
| "loss": 0.0029, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 3.2971428571428577e-06, | |
| "loss": 0.0021, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 3.225714285714286e-06, | |
| "loss": 0.0022, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 3.154285714285715e-06, | |
| "loss": 0.0057, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 3.082857142857143e-06, | |
| "loss": 0.0026, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 3.0114285714285716e-06, | |
| "loss": 0.0019, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 2.9400000000000002e-06, | |
| "loss": 0.0017, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 2.868571428571429e-06, | |
| "loss": 0.0047, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "eval_loss": 0.5272812247276306, | |
| "eval_runtime": 2446.9046, | |
| "eval_samples_per_second": 0.817, | |
| "eval_steps_per_second": 0.102, | |
| "eval_wer": 133.89655882958996, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 2.797142857142857e-06, | |
| "loss": 0.0027, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 2.725714285714286e-06, | |
| "loss": 0.0014, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 2.654285714285714e-06, | |
| "loss": 0.0026, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 2.582857142857143e-06, | |
| "loss": 0.0017, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "learning_rate": 2.5114285714285718e-06, | |
| "loss": 0.0016, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.4400000000000004e-06, | |
| "loss": 0.0022, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 2.3685714285714285e-06, | |
| "loss": 0.0023, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 2.297142857142857e-06, | |
| "loss": 0.0021, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.3, | |
| "learning_rate": 2.2257142857142857e-06, | |
| "loss": 0.0016, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 2.1542857142857147e-06, | |
| "loss": 0.0024, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 10.46, | |
| "learning_rate": 2.0828571428571433e-06, | |
| "loss": 0.0035, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 2.0114285714285715e-06, | |
| "loss": 0.0029, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 1.94e-06, | |
| "loss": 0.0013, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 1.8685714285714289e-06, | |
| "loss": 0.0034, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "learning_rate": 1.7971428571428572e-06, | |
| "loss": 0.0021, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "learning_rate": 1.7257142857142858e-06, | |
| "loss": 0.0017, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "learning_rate": 1.6542857142857144e-06, | |
| "loss": 0.0023, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 1.582857142857143e-06, | |
| "loss": 0.0016, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 1.5114285714285714e-06, | |
| "loss": 0.0018, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 1.44e-06, | |
| "loss": 0.0033, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "eval_loss": 0.5348898768424988, | |
| "eval_runtime": 2303.5501, | |
| "eval_samples_per_second": 0.868, | |
| "eval_steps_per_second": 0.109, | |
| "eval_wer": 137.85802596332167, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.26, | |
| "learning_rate": 1.3685714285714286e-06, | |
| "loss": 0.0018, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 11.34, | |
| "learning_rate": 1.2971428571428574e-06, | |
| "loss": 0.0012, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 11.42, | |
| "learning_rate": 1.2257142857142857e-06, | |
| "loss": 0.0018, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "learning_rate": 1.1542857142857143e-06, | |
| "loss": 0.0025, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "learning_rate": 1.082857142857143e-06, | |
| "loss": 0.0024, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 11.66, | |
| "learning_rate": 1.0114285714285715e-06, | |
| "loss": 0.003, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 11.74, | |
| "learning_rate": 9.400000000000001e-07, | |
| "loss": 0.001, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "learning_rate": 8.685714285714286e-07, | |
| "loss": 0.0023, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 11.9, | |
| "learning_rate": 7.971428571428572e-07, | |
| "loss": 0.0024, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 11.98, | |
| "learning_rate": 7.257142857142857e-07, | |
| "loss": 0.0034, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 6.542857142857144e-07, | |
| "loss": 0.003, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 5.82857142857143e-07, | |
| "loss": 0.0012, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 5.114285714285714e-07, | |
| "loss": 0.0023, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 12.3, | |
| "learning_rate": 4.4e-07, | |
| "loss": 0.0017, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 3.685714285714286e-07, | |
| "loss": 0.0022, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 12.46, | |
| "learning_rate": 2.9714285714285715e-07, | |
| "loss": 0.0033, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 12.54, | |
| "learning_rate": 2.2571428571428574e-07, | |
| "loss": 0.0019, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 1.542857142857143e-07, | |
| "loss": 0.001, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 8.285714285714285e-08, | |
| "loss": 0.0019, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "learning_rate": 1.142857142857143e-08, | |
| "loss": 0.0026, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "eval_loss": 0.5372153520584106, | |
| "eval_runtime": 2322.1625, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.108, | |
| "eval_wer": 130.3266021017927, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 13, | |
| "save_steps": 500, | |
| "total_flos": 1.844176142204928e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |