| { | |
| "best_metric": 40.23026315789473, | |
| "best_model_checkpoint": "./checkpoint-1000", | |
| "epoch": 43.011, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.712113404111031e-06, | |
| "loss": 0.3226, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8.31378915840787e-06, | |
| "loss": 0.1591, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 9.256291743595376e-06, | |
| "loss": 0.0854, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 9.911356165197841e-06, | |
| "loss": 0.045, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 9.777777777777779e-06, | |
| "loss": 0.0313, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 9.511111111111112e-06, | |
| "loss": 0.0214, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 9.233333333333334e-06, | |
| "loss": 0.0142, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 8.955555555555555e-06, | |
| "loss": 0.0113, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_loss": 0.35009765625, | |
| "eval_runtime": 275.0727, | |
| "eval_samples_per_second": 1.443, | |
| "eval_steps_per_second": 0.091, | |
| "eval_wer": 43.7171052631579, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 8.677777777777779e-06, | |
| "loss": 0.0077, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 0.007, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 8.122222222222223e-06, | |
| "loss": 0.0081, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 7.844444444444446e-06, | |
| "loss": 0.005, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 7.566666666666667e-06, | |
| "loss": 0.0032, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 7.28888888888889e-06, | |
| "loss": 0.0009, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 7.011111111111112e-06, | |
| "loss": 0.0004, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 6.733333333333334e-06, | |
| "loss": 0.0003, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "eval_loss": 0.39892578125, | |
| "eval_runtime": 271.1212, | |
| "eval_samples_per_second": 1.464, | |
| "eval_steps_per_second": 0.092, | |
| "eval_wer": 40.78947368421053, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 6.455555555555556e-06, | |
| "loss": 0.0002, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 19.01, | |
| "learning_rate": 6.177777777777778e-06, | |
| "loss": 0.0002, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "learning_rate": 5.9e-06, | |
| "loss": 0.0001, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "learning_rate": 5.622222222222222e-06, | |
| "loss": 0.0001, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 5.344444444444446e-06, | |
| "loss": 0.0001, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "learning_rate": 5.0666666666666676e-06, | |
| "loss": 0.0001, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 4.7888888888888894e-06, | |
| "loss": 0.0001, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 4.511111111111111e-06, | |
| "loss": 0.0001, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.42822265625, | |
| "eval_runtime": 277.6471, | |
| "eval_samples_per_second": 1.43, | |
| "eval_steps_per_second": 0.09, | |
| "eval_wer": 40.46052631578947, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 4.233333333333334e-06, | |
| "loss": 0.0001, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 28.01, | |
| "learning_rate": 3.955555555555556e-06, | |
| "loss": 0.0001, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "learning_rate": 3.6777777777777778e-06, | |
| "loss": 0.0001, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 30.01, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 0.0001, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 31.01, | |
| "learning_rate": 3.1222222222222228e-06, | |
| "loss": 0.0001, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "learning_rate": 2.8444444444444446e-06, | |
| "loss": 0.0001, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "learning_rate": 2.566666666666667e-06, | |
| "loss": 0.0001, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 34.02, | |
| "learning_rate": 2.2888888888888892e-06, | |
| "loss": 0.0001, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 34.02, | |
| "eval_loss": 0.439208984375, | |
| "eval_runtime": 279.7337, | |
| "eval_samples_per_second": 1.419, | |
| "eval_steps_per_second": 0.089, | |
| "eval_wer": 40.26315789473684, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 35.02, | |
| "learning_rate": 2.011111111111111e-06, | |
| "loss": 0.0001, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 36.02, | |
| "learning_rate": 1.7333333333333336e-06, | |
| "loss": 0.0001, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 1.4555555555555557e-06, | |
| "loss": 0.0001, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 1.1777777777777778e-06, | |
| "loss": 0.0001, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 40.01, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 0.0001, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 41.01, | |
| "learning_rate": 6.222222222222223e-07, | |
| "loss": 0.0001, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 42.01, | |
| "learning_rate": 3.444444444444445e-07, | |
| "loss": 0.0001, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 6.666666666666668e-08, | |
| "loss": 0.0001, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "eval_loss": 0.44287109375, | |
| "eval_runtime": 276.1752, | |
| "eval_samples_per_second": 1.437, | |
| "eval_steps_per_second": 0.091, | |
| "eval_wer": 40.23026315789473, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "step": 1000, | |
| "total_flos": 6.575035801025485e+19, | |
| "train_loss": 0.018124743044376373, | |
| "train_runtime": 9931.7597, | |
| "train_samples_per_second": 3.222, | |
| "train_steps_per_second": 0.101 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 6.575035801025485e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |