| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.72454448017149, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0002394, | |
| "loss": 3.2993, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 828.7984, | |
| "eval_samples_per_second": 9.006, | |
| "eval_steps_per_second": 1.126, | |
| "eval_wer": 0.36457931084030065, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.0002723048327137546, | |
| "loss": 0.3837, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 820.745, | |
| "eval_samples_per_second": 9.094, | |
| "eval_steps_per_second": 1.137, | |
| "eval_wer": 0.25638285452569753, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.00023522304832713752, | |
| "loss": 0.2576, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 816.7249, | |
| "eval_samples_per_second": 9.139, | |
| "eval_steps_per_second": 1.142, | |
| "eval_wer": 0.17959103032697818, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 0.0001981412639405204, | |
| "loss": 0.1963, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 819.1322, | |
| "eval_samples_per_second": 9.112, | |
| "eval_steps_per_second": 1.139, | |
| "eval_wer": 0.1695059854696326, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.00016096654275092934, | |
| "loss": 0.158, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 815.3058, | |
| "eval_samples_per_second": 9.155, | |
| "eval_steps_per_second": 1.144, | |
| "eval_wer": 0.16685675284904763, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.00012388475836431226, | |
| "loss": 0.125, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 815.712, | |
| "eval_samples_per_second": 9.15, | |
| "eval_steps_per_second": 1.144, | |
| "eval_wer": 0.15157202769570643, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 8.680297397769517e-05, | |
| "loss": 0.101, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 815.5191, | |
| "eval_samples_per_second": 9.152, | |
| "eval_steps_per_second": 1.144, | |
| "eval_wer": 0.1440015087155263, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 4.972118959107806e-05, | |
| "loss": 0.0793, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 816.2609, | |
| "eval_samples_per_second": 9.144, | |
| "eval_steps_per_second": 1.143, | |
| "eval_wer": 0.139223909548912, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 1.2639405204460965e-05, | |
| "loss": 0.0642, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 815.052, | |
| "eval_samples_per_second": 9.158, | |
| "eval_steps_per_second": 1.145, | |
| "eval_wer": 0.1354611011827252, | |
| "step": 3600 | |
| } | |
| ], | |
| "max_steps": 3728, | |
| "num_train_epochs": 8, | |
| "total_flos": 4.913815144448316e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |