| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.999161777032691, | |
| "global_step": 6705, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00029939999999999996, | |
| "loss": 3.0536, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.5348725914955139, | |
| "eval_runtime": 293.4912, | |
| "eval_samples_per_second": 18.958, | |
| "eval_steps_per_second": 4.739, | |
| "eval_wer": 0.43377988807868406, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0002758742949234488, | |
| "loss": 0.2543, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_loss": 0.38594746589660645, | |
| "eval_runtime": 294.551, | |
| "eval_samples_per_second": 18.89, | |
| "eval_steps_per_second": 4.722, | |
| "eval_wer": 0.3029082584364931, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 0.0002517002417405318, | |
| "loss": 0.1472, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_loss": 0.34706196188926697, | |
| "eval_runtime": 295.2527, | |
| "eval_samples_per_second": 18.845, | |
| "eval_steps_per_second": 4.711, | |
| "eval_wer": 0.28175343394946584, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.00022752618855761478, | |
| "loss": 0.1088, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "eval_loss": 0.34891438484191895, | |
| "eval_runtime": 293.508, | |
| "eval_samples_per_second": 18.957, | |
| "eval_steps_per_second": 4.739, | |
| "eval_wer": 0.27306257419026625, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.00020340048348106365, | |
| "loss": 0.0855, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "eval_loss": 0.3582000732421875, | |
| "eval_runtime": 293.4793, | |
| "eval_samples_per_second": 18.959, | |
| "eval_steps_per_second": 4.74, | |
| "eval_wer": 0.25578684076649144, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 0.00017922643029814666, | |
| "loss": 0.0721, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "eval_loss": 0.34574663639068604, | |
| "eval_runtime": 293.1368, | |
| "eval_samples_per_second": 18.981, | |
| "eval_steps_per_second": 4.745, | |
| "eval_wer": 0.24711717822621673, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 0.00015505237711522965, | |
| "loss": 0.0653, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "eval_loss": 0.3298647999763489, | |
| "eval_runtime": 293.814, | |
| "eval_samples_per_second": 18.937, | |
| "eval_steps_per_second": 4.734, | |
| "eval_wer": 0.23569187722570797, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 0.00013087832393231263, | |
| "loss": 0.0527, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "eval_loss": 0.3439774811267853, | |
| "eval_runtime": 293.0867, | |
| "eval_samples_per_second": 18.984, | |
| "eval_steps_per_second": 4.746, | |
| "eval_wer": 0.23338138036289638, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 0.00010670427074939564, | |
| "loss": 0.0444, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "eval_loss": 0.34170448780059814, | |
| "eval_runtime": 294.6348, | |
| "eval_samples_per_second": 18.884, | |
| "eval_steps_per_second": 4.721, | |
| "eval_wer": 0.22890876716974734, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 8.253021756647865e-05, | |
| "loss": 0.0404, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "eval_loss": 0.36909687519073486, | |
| "eval_runtime": 294.7932, | |
| "eval_samples_per_second": 18.874, | |
| "eval_steps_per_second": 4.719, | |
| "eval_wer": 0.2204298795997965, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.3, | |
| "learning_rate": 5.835616438356164e-05, | |
| "loss": 0.0345, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 12.3, | |
| "eval_loss": 0.3453255593776703, | |
| "eval_runtime": 293.4104, | |
| "eval_samples_per_second": 18.963, | |
| "eval_steps_per_second": 4.741, | |
| "eval_wer": 0.21021282007800576, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "learning_rate": 3.418211120064464e-05, | |
| "loss": 0.0288, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "eval_loss": 0.3634118437767029, | |
| "eval_runtime": 296.1491, | |
| "eval_samples_per_second": 18.788, | |
| "eval_steps_per_second": 4.697, | |
| "eval_wer": 0.20885619806681363, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 14.54, | |
| "learning_rate": 1.0008058017727637e-05, | |
| "loss": 0.027, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 14.54, | |
| "eval_loss": 0.35320568084716797, | |
| "eval_runtime": 290.8751, | |
| "eval_samples_per_second": 19.128, | |
| "eval_steps_per_second": 4.782, | |
| "eval_wer": 0.20438358487366456, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 6705, | |
| "total_flos": 2.897461391875666e+19, | |
| "train_loss": 0.3001849569908102, | |
| "train_runtime": 25737.7904, | |
| "train_samples_per_second": 8.343, | |
| "train_steps_per_second": 0.261 | |
| } | |
| ], | |
| "max_steps": 6705, | |
| "num_train_epochs": 15, | |
| "total_flos": 2.897461391875666e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |