| { | |
| "best_metric": 36.204844612672595, | |
| "best_model_checkpoint": "./checkpoint-5000", | |
| "epoch": 17.0089, | |
| "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": 4.4e-08, | |
| "loss": 2.8778, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.4e-08, | |
| "loss": 2.7378, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.44e-07, | |
| "loss": 2.5155, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.94e-07, | |
| "loss": 2.27, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.4399999999999996e-07, | |
| "loss": 2.0747, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.9399999999999996e-07, | |
| "loss": 1.9162, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.4399999999999996e-07, | |
| "loss": 1.8041, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.94e-07, | |
| "loss": 1.7071, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.44e-07, | |
| "loss": 1.6212, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.94e-07, | |
| "loss": 1.5404, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.44e-07, | |
| "loss": 1.4513, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5.939999999999999e-07, | |
| "loss": 1.3595, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.44e-07, | |
| "loss": 1.2568, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.939999999999999e-07, | |
| "loss": 1.1491, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.44e-07, | |
| "loss": 1.0415, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.94e-07, | |
| "loss": 0.951, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 8.439999999999999e-07, | |
| "loss": 0.8648, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 8.939999999999999e-07, | |
| "loss": 0.7923, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.439999999999999e-07, | |
| "loss": 0.7562, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.94e-07, | |
| "loss": 0.693, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_loss": 0.6936731338500977, | |
| "eval_runtime": 1973.4768, | |
| "eval_samples_per_second": 4.72, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 69.67821892924574, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.95111111111111e-07, | |
| "loss": 0.6595, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.895555555555554e-07, | |
| "loss": 0.6253, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.84e-07, | |
| "loss": 0.5966, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.784444444444444e-07, | |
| "loss": 0.5806, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.728888888888888e-07, | |
| "loss": 0.5642, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.673333333333332e-07, | |
| "loss": 0.5421, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.617777777777776e-07, | |
| "loss": 0.5176, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.562222222222223e-07, | |
| "loss": 0.5187, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.506666666666667e-07, | |
| "loss": 0.48, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.451111111111111e-07, | |
| "loss": 0.4798, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.395555555555556e-07, | |
| "loss": 0.4677, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.34e-07, | |
| "loss": 0.4496, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.284444444444444e-07, | |
| "loss": 0.4456, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.228888888888888e-07, | |
| "loss": 0.4341, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.173333333333333e-07, | |
| "loss": 0.4279, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 9.117777777777778e-07, | |
| "loss": 0.4274, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 9.062222222222222e-07, | |
| "loss": 0.43, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 9.006666666666666e-07, | |
| "loss": 0.4138, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.95111111111111e-07, | |
| "loss": 0.4109, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.895555555555555e-07, | |
| "loss": 0.3979, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 0.4168144762516022, | |
| "eval_runtime": 1968.8095, | |
| "eval_samples_per_second": 4.731, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 48.0702978890947, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.839999999999999e-07, | |
| "loss": 0.3922, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.784444444444444e-07, | |
| "loss": 0.3913, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.728888888888889e-07, | |
| "loss": 0.3782, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.673333333333332e-07, | |
| "loss": 0.3722, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.617777777777777e-07, | |
| "loss": 0.3735, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 8.562222222222222e-07, | |
| "loss": 0.3641, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.506666666666667e-07, | |
| "loss": 0.3701, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 8.451111111111111e-07, | |
| "loss": 0.3743, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 8.395555555555556e-07, | |
| "loss": 0.3638, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 8.34e-07, | |
| "loss": 0.3598, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 8.284444444444444e-07, | |
| "loss": 0.3623, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 8.228888888888889e-07, | |
| "loss": 0.3446, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 8.173333333333333e-07, | |
| "loss": 0.3464, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 8.117777777777778e-07, | |
| "loss": 0.3461, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 8.062222222222221e-07, | |
| "loss": 0.3358, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 8.006666666666666e-07, | |
| "loss": 0.3349, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.95111111111111e-07, | |
| "loss": 0.3318, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.895555555555555e-07, | |
| "loss": 0.3266, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.84e-07, | |
| "loss": 0.3386, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 7.784444444444444e-07, | |
| "loss": 0.3429, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "eval_loss": 0.3527003228664398, | |
| "eval_runtime": 1971.2193, | |
| "eval_samples_per_second": 4.725, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 42.810483834885865, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 7.728888888888888e-07, | |
| "loss": 0.3324, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 7.673333333333332e-07, | |
| "loss": 0.321, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 7.617777777777778e-07, | |
| "loss": 0.3334, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.562222222222222e-07, | |
| "loss": 0.3143, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.506666666666667e-07, | |
| "loss": 0.3185, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 7.451111111111111e-07, | |
| "loss": 0.317, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 7.395555555555555e-07, | |
| "loss": 0.3089, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 7.34e-07, | |
| "loss": 0.3118, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 7.284444444444444e-07, | |
| "loss": 0.3096, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.228888888888889e-07, | |
| "loss": 0.3069, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 7.173333333333333e-07, | |
| "loss": 0.3114, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 7.117777777777777e-07, | |
| "loss": 0.3183, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 7.062222222222222e-07, | |
| "loss": 0.3101, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 7.006666666666666e-07, | |
| "loss": 0.3047, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 6.951111111111111e-07, | |
| "loss": 0.3059, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 6.895555555555555e-07, | |
| "loss": 0.2954, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 6.84e-07, | |
| "loss": 0.3042, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 6.784444444444443e-07, | |
| "loss": 0.2969, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 6.728888888888888e-07, | |
| "loss": 0.2915, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 6.673333333333334e-07, | |
| "loss": 0.2907, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_loss": 0.3225308954715729, | |
| "eval_runtime": 1971.4065, | |
| "eval_samples_per_second": 4.725, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 40.42665521283545, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 6.617777777777778e-07, | |
| "loss": 0.2906, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 6.562222222222223e-07, | |
| "loss": 0.295, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 6.506666666666666e-07, | |
| "loss": 0.3019, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 6.451111111111111e-07, | |
| "loss": 0.2959, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 6.395555555555555e-07, | |
| "loss": 0.2901, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 6.34e-07, | |
| "loss": 0.2983, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 6.284444444444445e-07, | |
| "loss": 0.2824, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 6.228888888888889e-07, | |
| "loss": 0.2853, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 6.173333333333333e-07, | |
| "loss": 0.2909, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 6.117777777777777e-07, | |
| "loss": 0.2762, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 6.062222222222222e-07, | |
| "loss": 0.2782, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 6.006666666666666e-07, | |
| "loss": 0.2808, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 5.951111111111111e-07, | |
| "loss": 0.2797, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 5.895555555555555e-07, | |
| "loss": 0.2829, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 5.839999999999999e-07, | |
| "loss": 0.2908, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 5.784444444444444e-07, | |
| "loss": 0.2807, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 5.728888888888888e-07, | |
| "loss": 0.2754, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 5.673333333333334e-07, | |
| "loss": 0.2853, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 5.617777777777778e-07, | |
| "loss": 0.2702, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 5.562222222222222e-07, | |
| "loss": 0.2761, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "eval_loss": 0.3039405047893524, | |
| "eval_runtime": 1971.2169, | |
| "eval_samples_per_second": 4.725, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 38.8973897792453, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 5.506666666666666e-07, | |
| "loss": 0.275, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 5.451111111111111e-07, | |
| "loss": 0.2676, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 5.395555555555556e-07, | |
| "loss": 0.2709, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 5.34e-07, | |
| "loss": 0.2697, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 5.284444444444445e-07, | |
| "loss": 0.2688, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 5.228888888888888e-07, | |
| "loss": 0.2756, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 5.173333333333333e-07, | |
| "loss": 0.2786, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 5.117777777777777e-07, | |
| "loss": 0.2734, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 5.062222222222222e-07, | |
| "loss": 0.2732, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 5.006666666666667e-07, | |
| "loss": 0.2643, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.951111111111111e-07, | |
| "loss": 0.2644, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 4.895555555555555e-07, | |
| "loss": 0.2672, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 4.839999999999999e-07, | |
| "loss": 0.264, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 4.784444444444444e-07, | |
| "loss": 0.2598, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 4.728888888888889e-07, | |
| "loss": 0.263, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 4.673333333333333e-07, | |
| "loss": 0.2588, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4.6177777777777777e-07, | |
| "loss": 0.2697, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 4.5622222222222217e-07, | |
| "loss": 0.2698, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 4.506666666666666e-07, | |
| "loss": 0.2666, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 4.451111111111111e-07, | |
| "loss": 0.2637, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "eval_loss": 0.29210057854652405, | |
| "eval_runtime": 1966.1974, | |
| "eval_samples_per_second": 4.737, | |
| "eval_steps_per_second": 0.149, | |
| "eval_wer": 37.79272144819378, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 4.3955555555555554e-07, | |
| "loss": 0.2668, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 4.34e-07, | |
| "loss": 0.2561, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 4.2844444444444445e-07, | |
| "loss": 0.258, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 4.2288888888888886e-07, | |
| "loss": 0.2621, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 4.173333333333333e-07, | |
| "loss": 0.2516, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 4.1177777777777777e-07, | |
| "loss": 0.2584, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 4.0622222222222217e-07, | |
| "loss": 0.256, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 4.0066666666666663e-07, | |
| "loss": 0.2511, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9511111111111114e-07, | |
| "loss": 0.263, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 3.8955555555555554e-07, | |
| "loss": 0.264, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 3.84e-07, | |
| "loss": 0.2609, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 3.7844444444444445e-07, | |
| "loss": 0.2532, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 3.7288888888888886e-07, | |
| "loss": 0.2643, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 3.673333333333333e-07, | |
| "loss": 0.2477, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 3.617777777777777e-07, | |
| "loss": 0.2562, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 3.5622222222222223e-07, | |
| "loss": 0.2538, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 3.506666666666667e-07, | |
| "loss": 0.2461, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 3.451111111111111e-07, | |
| "loss": 0.2523, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 3.3955555555555554e-07, | |
| "loss": 0.2488, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.34e-07, | |
| "loss": 0.2507, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.2845877408981323, | |
| "eval_runtime": 1974.2354, | |
| "eval_samples_per_second": 4.718, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 37.073277674872266, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 3.284444444444444e-07, | |
| "loss": 0.2567, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 3.2288888888888886e-07, | |
| "loss": 0.2645, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 3.173333333333333e-07, | |
| "loss": 0.2523, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 3.1177777777777777e-07, | |
| "loss": 0.2509, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 3.0622222222222223e-07, | |
| "loss": 0.2493, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 3.006666666666667e-07, | |
| "loss": 0.2477, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 2.951111111111111e-07, | |
| "loss": 0.2513, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 2.8955555555555555e-07, | |
| "loss": 0.2474, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 2.8399999999999995e-07, | |
| "loss": 0.2433, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 2.784444444444444e-07, | |
| "loss": 0.2481, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 2.728888888888889e-07, | |
| "loss": 0.2436, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2.673333333333333e-07, | |
| "loss": 0.2522, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 2.617777777777778e-07, | |
| "loss": 0.2543, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 2.5622222222222223e-07, | |
| "loss": 0.2546, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 2.5066666666666663e-07, | |
| "loss": 0.2492, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 2.451111111111111e-07, | |
| "loss": 0.2502, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 2.3955555555555555e-07, | |
| "loss": 0.2427, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 2.34e-07, | |
| "loss": 0.2443, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 2.2844444444444443e-07, | |
| "loss": 0.2505, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 2.2288888888888886e-07, | |
| "loss": 0.2397, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "eval_loss": 0.27933579683303833, | |
| "eval_runtime": 1969.1014, | |
| "eval_samples_per_second": 4.73, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 36.60035972188666, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 2.1733333333333332e-07, | |
| "loss": 0.2447, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 2.1177777777777778e-07, | |
| "loss": 0.2398, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 2.062222222222222e-07, | |
| "loss": 0.2401, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 2.0066666666666666e-07, | |
| "loss": 0.2498, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 1.9511111111111112e-07, | |
| "loss": 0.2501, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 1.8955555555555555e-07, | |
| "loss": 0.2534, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 1.8399999999999998e-07, | |
| "loss": 0.2398, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 1.7844444444444444e-07, | |
| "loss": 0.253, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 1.728888888888889e-07, | |
| "loss": 0.2377, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 1.6733333333333332e-07, | |
| "loss": 0.2431, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 1.6177777777777775e-07, | |
| "loss": 0.2438, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 1.5622222222222224e-07, | |
| "loss": 0.2337, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 1.5066666666666667e-07, | |
| "loss": 0.2433, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 1.451111111111111e-07, | |
| "loss": 0.2398, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.3955555555555553e-07, | |
| "loss": 0.2428, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 1.34e-07, | |
| "loss": 0.2467, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 1.2844444444444444e-07, | |
| "loss": 0.255, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "learning_rate": 1.228888888888889e-07, | |
| "loss": 0.2407, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "learning_rate": 1.1733333333333333e-07, | |
| "loss": 0.2473, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 1.1177777777777778e-07, | |
| "loss": 0.243, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "eval_loss": 0.276321142911911, | |
| "eval_runtime": 1971.8605, | |
| "eval_samples_per_second": 4.723, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 36.35025457929541, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 1.0622222222222221e-07, | |
| "loss": 0.2404, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 1.0066666666666667e-07, | |
| "loss": 0.2417, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 9.51111111111111e-08, | |
| "loss": 0.2369, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 8.955555555555555e-08, | |
| "loss": 0.2388, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 8.4e-08, | |
| "loss": 0.2371, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 7.844444444444444e-08, | |
| "loss": 0.2371, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 7.288888888888888e-08, | |
| "loss": 0.2446, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 6.733333333333333e-08, | |
| "loss": 0.2473, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 6.177777777777777e-08, | |
| "loss": 0.251, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 5.622222222222222e-08, | |
| "loss": 0.2391, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 5.0666666666666664e-08, | |
| "loss": 0.248, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 4.511111111111111e-08, | |
| "loss": 0.2349, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 3.955555555555555e-08, | |
| "loss": 0.2406, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 3.4e-08, | |
| "loss": 0.2444, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 2.8444444444444443e-08, | |
| "loss": 0.2319, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 16.05, | |
| "learning_rate": 2.2888888888888887e-08, | |
| "loss": 0.2384, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 16.05, | |
| "learning_rate": 1.7333333333333333e-08, | |
| "loss": 0.2376, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "learning_rate": 1.1777777777777778e-08, | |
| "loss": 0.2379, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 6.222222222222222e-09, | |
| "loss": 0.2432, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 6.666666666666666e-10, | |
| "loss": 0.2501, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "eval_loss": 0.27535662055015564, | |
| "eval_runtime": 1985.1484, | |
| "eval_samples_per_second": 4.692, | |
| "eval_steps_per_second": 0.147, | |
| "eval_wer": 36.204844612672595, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "step": 5000, | |
| "total_flos": 2.072432541302784e+19, | |
| "train_loss": 0.4274742190361023, | |
| "train_runtime": 38107.6821, | |
| "train_samples_per_second": 8.397, | |
| "train_steps_per_second": 0.131 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 2.072432541302784e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |