| { | |
| "best_metric": 55.04238164151334, | |
| "best_model_checkpoint": "./whisper-large-somali/checkpoint-1000", | |
| "epoch": 76.88888888888889, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 6.611096473669596e-06, | |
| "loss": 2.191, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 8.31378915840787e-06, | |
| "loss": 1.015, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 9.28666248215634e-06, | |
| "loss": 0.5048, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 9.933858671331224e-06, | |
| "loss": 0.2122, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 9.755555555555556e-06, | |
| "loss": 0.0748, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 9.466666666666667e-06, | |
| "loss": 0.0338, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 9.17777777777778e-06, | |
| "loss": 0.0206, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 15.37, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.0137, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 17.3, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.0101, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 19.22, | |
| "learning_rate": 8.311111111111111e-06, | |
| "loss": 0.0066, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "learning_rate": 8.022222222222222e-06, | |
| "loss": 0.0057, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 23.07, | |
| "learning_rate": 7.733333333333334e-06, | |
| "loss": 0.0055, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "learning_rate": 7.444444444444445e-06, | |
| "loss": 0.0058, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 26.89, | |
| "learning_rate": 7.155555555555556e-06, | |
| "loss": 0.0047, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 28.81, | |
| "learning_rate": 6.866666666666667e-06, | |
| "loss": 0.0031, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 30.74, | |
| "learning_rate": 6.577777777777779e-06, | |
| "loss": 0.0026, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 32.67, | |
| "learning_rate": 6.28888888888889e-06, | |
| "loss": 0.0037, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 34.59, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0026, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 36.52, | |
| "learning_rate": 5.711111111111112e-06, | |
| "loss": 0.0018, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 38.44, | |
| "learning_rate": 5.422222222222223e-06, | |
| "loss": 0.0012, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 38.44, | |
| "eval_loss": 1.890625, | |
| "eval_runtime": 421.0342, | |
| "eval_samples_per_second": 2.42, | |
| "eval_steps_per_second": 0.038, | |
| "eval_wer": 55.41451312797189, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 40.37, | |
| "learning_rate": 5.133333333333334e-06, | |
| "loss": 0.0009, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 42.3, | |
| "learning_rate": 4.8444444444444446e-06, | |
| "loss": 0.0008, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 44.22, | |
| "learning_rate": 4.555555555555556e-06, | |
| "loss": 0.0007, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 4.266666666666668e-06, | |
| "loss": 0.0006, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 48.07, | |
| "learning_rate": 3.977777777777778e-06, | |
| "loss": 0.0006, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 49.96, | |
| "learning_rate": 3.7e-06, | |
| "loss": 0.0006, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 51.89, | |
| "learning_rate": 3.4111111111111113e-06, | |
| "loss": 0.0006, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 53.81, | |
| "learning_rate": 3.1222222222222228e-06, | |
| "loss": 0.0005, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 55.74, | |
| "learning_rate": 2.8333333333333335e-06, | |
| "loss": 0.0005, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 57.67, | |
| "learning_rate": 2.5444444444444446e-06, | |
| "loss": 0.0005, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 59.59, | |
| "learning_rate": 2.2555555555555557e-06, | |
| "loss": 0.0005, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 61.52, | |
| "learning_rate": 1.9666666666666668e-06, | |
| "loss": 0.0005, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 63.44, | |
| "learning_rate": 1.6777777777777779e-06, | |
| "loss": 0.0005, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 65.37, | |
| "learning_rate": 1.3888888888888892e-06, | |
| "loss": 0.0005, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 67.3, | |
| "learning_rate": 1.1e-06, | |
| "loss": 0.0005, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 69.22, | |
| "learning_rate": 8.111111111111112e-07, | |
| "loss": 0.0005, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 71.15, | |
| "learning_rate": 5.222222222222223e-07, | |
| "loss": 0.0004, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 73.07, | |
| "learning_rate": 2.3333333333333336e-07, | |
| "loss": 0.0004, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 74.96, | |
| "learning_rate": 0.0, | |
| "loss": 0.0004, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 76.89, | |
| "learning_rate": 0.0, | |
| "loss": 0.0004, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 76.89, | |
| "eval_loss": 2.00390625, | |
| "eval_runtime": 421.4824, | |
| "eval_samples_per_second": 2.418, | |
| "eval_steps_per_second": 0.038, | |
| "eval_wer": 55.04238164151334, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 76.89, | |
| "step": 1000, | |
| "total_flos": 1.384647263524592e+20, | |
| "train_loss": 0.1032530371248722, | |
| "train_runtime": 9059.3866, | |
| "train_samples_per_second": 7.064, | |
| "train_steps_per_second": 0.11 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 77, | |
| "total_flos": 1.384647263524592e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |