| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 44.01, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.475000000000002e-06, | |
| "loss": 1.5129, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 8.85e-06, | |
| "loss": 0.4346, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "eval_loss": 0.3902237117290497, | |
| "eval_runtime": 91.4411, | |
| "eval_samples_per_second": 4.363, | |
| "eval_steps_per_second": 0.142, | |
| "eval_wer": 49.679733538303864, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 8.225e-06, | |
| "loss": 0.1652, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.0728, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.3811023235321045, | |
| "eval_runtime": 75.7771, | |
| "eval_samples_per_second": 5.265, | |
| "eval_steps_per_second": 0.172, | |
| "eval_wer": 40.73789392774788, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 6.975000000000001e-06, | |
| "loss": 0.0287, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 6.35e-06, | |
| "loss": 0.009, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "eval_loss": 0.4923767149448395, | |
| "eval_runtime": 76.4459, | |
| "eval_samples_per_second": 5.219, | |
| "eval_steps_per_second": 0.17, | |
| "eval_wer": 39.508070714834744, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 19.01, | |
| "learning_rate": 5.725e-06, | |
| "loss": 0.0061, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 5.1e-06, | |
| "loss": 0.0028, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.5309251546859741, | |
| "eval_runtime": 79.4543, | |
| "eval_samples_per_second": 5.022, | |
| "eval_steps_per_second": 0.164, | |
| "eval_wer": 38.73943120676403, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 4.475e-06, | |
| "loss": 0.0014, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "learning_rate": 3.85e-06, | |
| "loss": 0.0008, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "eval_loss": 0.5686793327331543, | |
| "eval_runtime": 84.4437, | |
| "eval_samples_per_second": 4.725, | |
| "eval_steps_per_second": 0.154, | |
| "eval_wer": 38.63694593902127, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 30.01, | |
| "learning_rate": 3.2250000000000005e-06, | |
| "loss": 0.0006, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 33.01, | |
| "learning_rate": 2.6e-06, | |
| "loss": 0.0006, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 33.01, | |
| "eval_loss": 0.5858641266822815, | |
| "eval_runtime": 74.0959, | |
| "eval_samples_per_second": 5.385, | |
| "eval_steps_per_second": 0.175, | |
| "eval_wer": 39.02126569305662, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 1.975e-06, | |
| "loss": 0.0005, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 38.02, | |
| "learning_rate": 1.3500000000000002e-06, | |
| "loss": 0.0005, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 38.02, | |
| "eval_loss": 0.5954424142837524, | |
| "eval_runtime": 274.4089, | |
| "eval_samples_per_second": 1.454, | |
| "eval_steps_per_second": 0.047, | |
| "eval_wer": 39.09812964386369, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 41.02, | |
| "learning_rate": 7.25e-07, | |
| "loss": 0.0005, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 44.01, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 0.0005, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 44.01, | |
| "eval_loss": 0.5990687608718872, | |
| "eval_runtime": 75.4625, | |
| "eval_samples_per_second": 5.287, | |
| "eval_steps_per_second": 0.172, | |
| "eval_wer": 39.04688700999232, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 44.01, | |
| "step": 400, | |
| "total_flos": 7.18462212636672e+18, | |
| "train_loss": 0.13985416618408636, | |
| "train_runtime": 4597.3293, | |
| "train_samples_per_second": 5.568, | |
| "train_steps_per_second": 0.087 | |
| } | |
| ], | |
| "max_steps": 400, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 7.18462212636672e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |