| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.3640661938534278, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.892434988179669e-05, | |
| "loss": 9.0256, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.774231678486998e-05, | |
| "loss": 5.1429, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.656028368794326e-05, | |
| "loss": 4.6719, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.537825059101655e-05, | |
| "loss": 4.3761, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.419621749408984e-05, | |
| "loss": 4.1596, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.301418439716312e-05, | |
| "loss": 3.9129, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.1832151300236414e-05, | |
| "loss": 3.7629, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.0650118203309695e-05, | |
| "loss": 3.5985, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 3.946808510638298e-05, | |
| "loss": 3.3403, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3.828605200945627e-05, | |
| "loss": 3.103, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_cer": 11.597362920083615, | |
| "eval_loss": 8.085211753845215, | |
| "eval_runtime": 146.0243, | |
| "eval_samples_per_second": 5.143, | |
| "eval_steps_per_second": 0.322, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.710401891252955e-05, | |
| "loss": 2.9059, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 3.592198581560284e-05, | |
| "loss": 2.7383, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 3.473995271867612e-05, | |
| "loss": 2.5918, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.3557919621749415e-05, | |
| "loss": 2.4015, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.2375886524822696e-05, | |
| "loss": 2.2971, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3.119385342789598e-05, | |
| "loss": 2.0906, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 3.0011820330969272e-05, | |
| "loss": 1.7917, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 2.8829787234042553e-05, | |
| "loss": 1.4973, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 2.7647754137115837e-05, | |
| "loss": 1.3962, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.646572104018913e-05, | |
| "loss": 1.2535, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_cer": 0.49043254542530956, | |
| "eval_loss": 2.039987325668335, | |
| "eval_runtime": 35.6188, | |
| "eval_samples_per_second": 21.084, | |
| "eval_steps_per_second": 1.32, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 4230, | |
| "num_train_epochs": 5, | |
| "total_flos": 5.790864009434628e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |