| { | |
| "best_metric": 28.648953267516852, | |
| "best_model_checkpoint": "./checkpoint-5000", | |
| "epoch": 65.012, | |
| "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": 4.8e-08, | |
| "loss": 1.6929, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.8e-08, | |
| "loss": 1.6161, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.4799999999999998e-07, | |
| "loss": 1.5665, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.98e-07, | |
| "loss": 1.5875, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.48e-07, | |
| "loss": 1.4633, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.98e-07, | |
| "loss": 1.3517, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.4799999999999994e-07, | |
| "loss": 1.3473, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 3.98e-07, | |
| "loss": 1.2353, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.48e-07, | |
| "loss": 1.1232, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.979999999999999e-07, | |
| "loss": 1.0898, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 5.48e-07, | |
| "loss": 0.9945, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 5.979999999999999e-07, | |
| "loss": 0.8935, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6.48e-07, | |
| "loss": 0.8898, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 6.979999999999999e-07, | |
| "loss": 0.8355, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.48e-07, | |
| "loss": 0.7488, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.98e-07, | |
| "loss": 0.7638, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 8.48e-07, | |
| "loss": 0.7228, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 8.98e-07, | |
| "loss": 0.6542, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 9.479999999999999e-07, | |
| "loss": 0.6631, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 9.98e-07, | |
| "loss": 0.6425, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "eval_loss": 0.702507495880127, | |
| "eval_runtime": 521.2801, | |
| "eval_samples_per_second": 5.96, | |
| "eval_steps_per_second": 0.188, | |
| "eval_wer": 41.44768424787627, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 9.946666666666666e-07, | |
| "loss": 0.5748, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 9.891111111111112e-07, | |
| "loss": 0.5951, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 9.835555555555554e-07, | |
| "loss": 0.5775, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 9.78e-07, | |
| "loss": 0.5234, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9.724444444444444e-07, | |
| "loss": 0.5375, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 9.668888888888888e-07, | |
| "loss": 0.5398, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 9.613333333333334e-07, | |
| "loss": 0.479, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 9.557777777777776e-07, | |
| "loss": 0.4986, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 9.502222222222221e-07, | |
| "loss": 0.5009, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 9.446666666666666e-07, | |
| "loss": 0.4526, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 9.39111111111111e-07, | |
| "loss": 0.4678, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 9.335555555555555e-07, | |
| "loss": 0.4692, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 9.28e-07, | |
| "loss": 0.4286, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 9.224444444444443e-07, | |
| "loss": 0.4415, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 9.168888888888889e-07, | |
| "loss": 0.4503, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 9.113333333333333e-07, | |
| "loss": 0.4035, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 9.057777777777778e-07, | |
| "loss": 0.4175, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 9.002222222222222e-07, | |
| "loss": 0.4333, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 8.946666666666667e-07, | |
| "loss": 0.3904, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 8.89111111111111e-07, | |
| "loss": 0.3973, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.5367385149002075, | |
| "eval_runtime": 333.3595, | |
| "eval_samples_per_second": 9.32, | |
| "eval_steps_per_second": 0.294, | |
| "eval_wer": 33.96923462252927, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 8.835555555555555e-07, | |
| "loss": 0.4134, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 8.78e-07, | |
| "loss": 0.3698, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 8.724444444444444e-07, | |
| "loss": 0.3784, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 8.668888888888889e-07, | |
| "loss": 0.4007, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 8.613333333333332e-07, | |
| "loss": 0.3591, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 8.557777777777777e-07, | |
| "loss": 0.3615, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 8.502222222222221e-07, | |
| "loss": 0.3856, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 8.446666666666666e-07, | |
| "loss": 0.3527, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 8.391111111111111e-07, | |
| "loss": 0.3435, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 8.335555555555555e-07, | |
| "loss": 0.3771, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 8.28e-07, | |
| "loss": 0.3343, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 8.224444444444444e-07, | |
| "loss": 0.3335, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 8.168888888888889e-07, | |
| "loss": 0.363, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 8.113333333333333e-07, | |
| "loss": 0.3262, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 8.057777777777778e-07, | |
| "loss": 0.3186, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 8.002222222222222e-07, | |
| "loss": 0.3554, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 7.946666666666666e-07, | |
| "loss": 0.3174, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 7.891111111111111e-07, | |
| "loss": 0.3104, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 19.01, | |
| "learning_rate": 7.835555555555555e-07, | |
| "loss": 0.3405, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 19.01, | |
| "learning_rate": 7.78e-07, | |
| "loss": 0.3125, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 19.01, | |
| "eval_loss": 0.4926978647708893, | |
| "eval_runtime": 338.2106, | |
| "eval_samples_per_second": 9.187, | |
| "eval_steps_per_second": 0.29, | |
| "eval_wer": 31.44580576485567, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 7.724444444444444e-07, | |
| "loss": 0.2975, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 20.01, | |
| "learning_rate": 7.668888888888888e-07, | |
| "loss": 0.335, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 20.01, | |
| "learning_rate": 7.613333333333333e-07, | |
| "loss": 0.3028, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 7.557777777777777e-07, | |
| "loss": 0.291, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 7.502222222222222e-07, | |
| "loss": 0.3219, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 7.446666666666666e-07, | |
| "loss": 0.2992, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 7.391111111111111e-07, | |
| "loss": 0.2821, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "learning_rate": 7.335555555555554e-07, | |
| "loss": 0.3121, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "learning_rate": 7.28e-07, | |
| "loss": 0.2933, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 7.224444444444445e-07, | |
| "loss": 0.2704, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 7.168888888888889e-07, | |
| "loss": 0.3079, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 7.113333333333334e-07, | |
| "loss": 0.284, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 7.057777777777777e-07, | |
| "loss": 0.2646, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "learning_rate": 7.002222222222222e-07, | |
| "loss": 0.2982, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "learning_rate": 6.946666666666666e-07, | |
| "loss": 0.2794, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 6.891111111111111e-07, | |
| "loss": 0.26, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 6.835555555555555e-07, | |
| "loss": 0.294, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 25.01, | |
| "learning_rate": 6.78e-07, | |
| "loss": 0.2774, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "learning_rate": 6.724444444444444e-07, | |
| "loss": 0.2473, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 6.668888888888888e-07, | |
| "loss": 0.2848, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.47389376163482666, | |
| "eval_runtime": 332.7237, | |
| "eval_samples_per_second": 9.338, | |
| "eval_steps_per_second": 0.295, | |
| "eval_wer": 30.103734006804284, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 26.01, | |
| "learning_rate": 6.613333333333333e-07, | |
| "loss": 0.2707, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 26.01, | |
| "learning_rate": 6.557777777777777e-07, | |
| "loss": 0.2461, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 6.502222222222222e-07, | |
| "loss": 0.2753, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "learning_rate": 6.446666666666666e-07, | |
| "loss": 0.2666, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "learning_rate": 6.39111111111111e-07, | |
| "loss": 0.2428, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 6.335555555555556e-07, | |
| "loss": 0.2672, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 28.01, | |
| "learning_rate": 6.28e-07, | |
| "loss": 0.2628, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 28.01, | |
| "learning_rate": 6.224444444444445e-07, | |
| "loss": 0.2378, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 6.168888888888889e-07, | |
| "loss": 0.2615, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "learning_rate": 6.113333333333333e-07, | |
| "loss": 0.2566, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "learning_rate": 6.057777777777777e-07, | |
| "loss": 0.2334, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 6.002222222222222e-07, | |
| "loss": 0.2507, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 30.01, | |
| "learning_rate": 5.946666666666667e-07, | |
| "loss": 0.2576, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 30.01, | |
| "learning_rate": 5.891111111111111e-07, | |
| "loss": 0.2266, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 5.835555555555556e-07, | |
| "loss": 0.2519, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 31.01, | |
| "learning_rate": 5.779999999999999e-07, | |
| "loss": 0.2478, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 31.01, | |
| "learning_rate": 5.724444444444444e-07, | |
| "loss": 0.2257, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 5.668888888888888e-07, | |
| "loss": 0.2446, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "learning_rate": 5.613333333333333e-07, | |
| "loss": 0.2452, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "learning_rate": 5.557777777777778e-07, | |
| "loss": 0.2201, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "eval_loss": 0.46750521659851074, | |
| "eval_runtime": 335.6468, | |
| "eval_samples_per_second": 9.257, | |
| "eval_steps_per_second": 0.292, | |
| "eval_wer": 29.485921813362275, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "learning_rate": 5.502222222222221e-07, | |
| "loss": 0.2385, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 33.01, | |
| "learning_rate": 5.446666666666666e-07, | |
| "loss": 0.2409, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 33.01, | |
| "learning_rate": 5.39111111111111e-07, | |
| "loss": 0.2203, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 5.335555555555556e-07, | |
| "loss": 0.2328, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "learning_rate": 5.28e-07, | |
| "loss": 0.24, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "learning_rate": 5.224444444444445e-07, | |
| "loss": 0.2136, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 5.168888888888888e-07, | |
| "loss": 0.2286, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 35.01, | |
| "learning_rate": 5.113333333333333e-07, | |
| "loss": 0.2334, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 35.01, | |
| "learning_rate": 5.057777777777778e-07, | |
| "loss": 0.2142, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 5.002222222222222e-07, | |
| "loss": 0.2218, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 36.01, | |
| "learning_rate": 4.946666666666666e-07, | |
| "loss": 0.2342, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 36.01, | |
| "learning_rate": 4.891111111111111e-07, | |
| "loss": 0.2074, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "learning_rate": 4.835555555555555e-07, | |
| "loss": 0.2169, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "learning_rate": 4.779999999999999e-07, | |
| "loss": 0.2273, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "learning_rate": 4.724444444444444e-07, | |
| "loss": 0.2087, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 4.668888888888889e-07, | |
| "loss": 0.2139, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 38.01, | |
| "learning_rate": 4.613333333333333e-07, | |
| "loss": 0.2256, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 38.01, | |
| "learning_rate": 4.5577777777777774e-07, | |
| "loss": 0.2044, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 4.5022222222222225e-07, | |
| "loss": 0.2059, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 39.01, | |
| "learning_rate": 4.4466666666666665e-07, | |
| "loss": 0.2257, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 39.01, | |
| "eval_loss": 0.46374234557151794, | |
| "eval_runtime": 331.7281, | |
| "eval_samples_per_second": 9.366, | |
| "eval_steps_per_second": 0.295, | |
| "eval_wer": 28.99334182129365, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 39.01, | |
| "learning_rate": 4.391111111111111e-07, | |
| "loss": 0.2031, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 4.3355555555555556e-07, | |
| "loss": 0.205, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 40.01, | |
| "learning_rate": 4.2799999999999997e-07, | |
| "loss": 0.2182, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 40.01, | |
| "learning_rate": 4.224444444444444e-07, | |
| "loss": 0.2037, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "learning_rate": 4.1688888888888883e-07, | |
| "loss": 0.1995, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 41.01, | |
| "learning_rate": 4.113333333333333e-07, | |
| "loss": 0.2205, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 41.01, | |
| "learning_rate": 4.057777777777778e-07, | |
| "loss": 0.1982, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 4.002222222222222e-07, | |
| "loss": 0.1954, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 42.01, | |
| "learning_rate": 3.9466666666666665e-07, | |
| "loss": 0.2145, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 42.01, | |
| "learning_rate": 3.891111111111111e-07, | |
| "loss": 0.1975, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 3.835555555555555e-07, | |
| "loss": 0.1911, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 3.7799999999999997e-07, | |
| "loss": 0.2149, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 3.7244444444444443e-07, | |
| "loss": 0.1959, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 3.668888888888889e-07, | |
| "loss": 0.1898, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 44.01, | |
| "learning_rate": 3.6133333333333334e-07, | |
| "loss": 0.2104, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 44.01, | |
| "learning_rate": 3.557777777777778e-07, | |
| "loss": 0.1947, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 3.502222222222222e-07, | |
| "loss": 0.1863, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 45.01, | |
| "learning_rate": 3.4466666666666666e-07, | |
| "loss": 0.2093, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 45.01, | |
| "learning_rate": 3.3911111111111106e-07, | |
| "loss": 0.1914, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 3.335555555555555e-07, | |
| "loss": 0.1837, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 0.46572384238243103, | |
| "eval_runtime": 334.5411, | |
| "eval_samples_per_second": 9.287, | |
| "eval_steps_per_second": 0.293, | |
| "eval_wer": 28.914028093757178, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 46.01, | |
| "learning_rate": 3.28e-07, | |
| "loss": 0.2041, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 46.01, | |
| "learning_rate": 3.2244444444444443e-07, | |
| "loss": 0.1936, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 3.168888888888889e-07, | |
| "loss": 0.1808, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 47.01, | |
| "learning_rate": 3.1133333333333334e-07, | |
| "loss": 0.2032, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 47.01, | |
| "learning_rate": 3.0577777777777775e-07, | |
| "loss": 0.1924, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 3.002222222222222e-07, | |
| "loss": 0.1767, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 48.01, | |
| "learning_rate": 2.9466666666666666e-07, | |
| "loss": 0.2025, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 48.01, | |
| "learning_rate": 2.8911111111111106e-07, | |
| "loss": 0.1902, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 2.8355555555555557e-07, | |
| "loss": 0.174, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "learning_rate": 2.7800000000000003e-07, | |
| "loss": 0.2009, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "learning_rate": 2.7244444444444443e-07, | |
| "loss": 0.1884, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "learning_rate": 2.668888888888889e-07, | |
| "loss": 0.1728, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 50.01, | |
| "learning_rate": 2.613333333333333e-07, | |
| "loss": 0.1984, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 50.01, | |
| "learning_rate": 2.5577777777777775e-07, | |
| "loss": 0.1884, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 50.02, | |
| "learning_rate": 2.502222222222222e-07, | |
| "loss": 0.1717, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "learning_rate": 2.4466666666666666e-07, | |
| "loss": 0.1927, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 51.01, | |
| "learning_rate": 2.391111111111111e-07, | |
| "loss": 0.1909, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 51.01, | |
| "learning_rate": 2.3355555555555552e-07, | |
| "loss": 0.1698, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "learning_rate": 2.28e-07, | |
| "loss": 0.1901, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 52.01, | |
| "learning_rate": 2.2244444444444443e-07, | |
| "loss": 0.1897, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 52.01, | |
| "eval_loss": 0.4658238887786865, | |
| "eval_runtime": 335.2394, | |
| "eval_samples_per_second": 9.268, | |
| "eval_steps_per_second": 0.292, | |
| "eval_wer": 28.74496462190311, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 52.01, | |
| "learning_rate": 2.1688888888888886e-07, | |
| "loss": 0.1677, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "learning_rate": 2.1133333333333335e-07, | |
| "loss": 0.1899, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 53.01, | |
| "learning_rate": 2.0577777777777778e-07, | |
| "loss": 0.1883, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 53.01, | |
| "learning_rate": 2.002222222222222e-07, | |
| "loss": 0.1679, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "learning_rate": 1.9466666666666664e-07, | |
| "loss": 0.1869, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 54.01, | |
| "learning_rate": 1.8911111111111112e-07, | |
| "loss": 0.1881, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 54.01, | |
| "learning_rate": 1.8355555555555555e-07, | |
| "loss": 0.1673, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "learning_rate": 1.7799999999999998e-07, | |
| "loss": 0.1847, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 55.01, | |
| "learning_rate": 1.7244444444444444e-07, | |
| "loss": 0.1876, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 55.01, | |
| "learning_rate": 1.668888888888889e-07, | |
| "loss": 0.1667, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 1.6133333333333332e-07, | |
| "loss": 0.1807, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 56.01, | |
| "learning_rate": 1.5577777777777775e-07, | |
| "loss": 0.1851, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 56.01, | |
| "learning_rate": 1.5022222222222224e-07, | |
| "loss": 0.1705, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "learning_rate": 1.4466666666666667e-07, | |
| "loss": 0.1773, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 57.01, | |
| "learning_rate": 1.391111111111111e-07, | |
| "loss": 0.1905, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 57.01, | |
| "learning_rate": 1.3355555555555555e-07, | |
| "loss": 0.1629, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "learning_rate": 1.28e-07, | |
| "loss": 0.1774, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 58.01, | |
| "learning_rate": 1.2244444444444444e-07, | |
| "loss": 0.1853, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 58.01, | |
| "learning_rate": 1.1688888888888888e-07, | |
| "loss": 0.1659, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "learning_rate": 1.1133333333333334e-07, | |
| "loss": 0.1764, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_loss": 0.4676055610179901, | |
| "eval_runtime": 335.034, | |
| "eval_samples_per_second": 9.274, | |
| "eval_steps_per_second": 0.293, | |
| "eval_wer": 28.717830978272215, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 59.01, | |
| "learning_rate": 1.0577777777777777e-07, | |
| "loss": 0.1844, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 59.01, | |
| "learning_rate": 1.0022222222222222e-07, | |
| "loss": 0.1682, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 9.466666666666665e-08, | |
| "loss": 0.174, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 60.01, | |
| "learning_rate": 8.911111111111111e-08, | |
| "loss": 0.1835, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 60.01, | |
| "learning_rate": 8.355555555555554e-08, | |
| "loss": 0.166, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "learning_rate": 7.8e-08, | |
| "loss": 0.1747, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 61.01, | |
| "learning_rate": 7.244444444444445e-08, | |
| "loss": 0.1837, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 61.01, | |
| "learning_rate": 6.688888888888888e-08, | |
| "loss": 0.1673, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "learning_rate": 6.133333333333333e-08, | |
| "loss": 0.1741, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "learning_rate": 5.5777777777777777e-08, | |
| "loss": 0.1816, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "learning_rate": 5.022222222222222e-08, | |
| "loss": 0.166, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "learning_rate": 4.466666666666666e-08, | |
| "loss": 0.1725, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 63.01, | |
| "learning_rate": 3.9111111111111106e-08, | |
| "loss": 0.1849, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 63.01, | |
| "learning_rate": 3.355555555555555e-08, | |
| "loss": 0.1655, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "learning_rate": 2.8e-08, | |
| "loss": 0.1708, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 64.01, | |
| "learning_rate": 2.2444444444444442e-08, | |
| "loss": 0.1835, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 64.01, | |
| "learning_rate": 1.6888888888888886e-08, | |
| "loss": 0.1664, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 1.1333333333333334e-08, | |
| "loss": 0.1673, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 65.01, | |
| "learning_rate": 5.777777777777777e-09, | |
| "loss": 0.1839, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 65.01, | |
| "learning_rate": 2.2222222222222221e-10, | |
| "loss": 0.1681, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 65.01, | |
| "eval_loss": 0.46787169575691223, | |
| "eval_runtime": 334.6695, | |
| "eval_samples_per_second": 9.284, | |
| "eval_steps_per_second": 0.293, | |
| "eval_wer": 28.648953267516852, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 65.01, | |
| "step": 5000, | |
| "total_flos": 2.06793125756928e+19, | |
| "train_loss": 0.34305178251266477, | |
| "train_runtime": 22591.6172, | |
| "train_samples_per_second": 14.165, | |
| "train_steps_per_second": 0.221 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 2.06793125756928e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |