| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.59504132231405, | |
| "global_step": 3200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.000237, | |
| "loss": 6.8607, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_cer": 0.9999129564346956, | |
| "eval_loss": 3.4867048263549805, | |
| "eval_runtime": 258.7779, | |
| "eval_samples_per_second": 8.455, | |
| "eval_steps_per_second": 1.059, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.0002840252707581227, | |
| "loss": 2.7103, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_cer": 0.3108760934847891, | |
| "eval_loss": 1.061523675918579, | |
| "eval_runtime": 261.8355, | |
| "eval_samples_per_second": 8.356, | |
| "eval_steps_per_second": 1.046, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 0.0002623646209386281, | |
| "loss": 1.0054, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_cer": 0.22384703544123835, | |
| "eval_loss": 0.6996724605560303, | |
| "eval_runtime": 261.6376, | |
| "eval_samples_per_second": 8.363, | |
| "eval_steps_per_second": 1.047, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 0.00024070397111913356, | |
| "loss": 0.6836, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "eval_cer": 0.19226472849661255, | |
| "eval_loss": 0.6144178509712219, | |
| "eval_runtime": 259.8219, | |
| "eval_samples_per_second": 8.421, | |
| "eval_steps_per_second": 1.055, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.00021904332129963898, | |
| "loss": 0.5372, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "eval_cer": 0.1836474155314735, | |
| "eval_loss": 0.60425865650177, | |
| "eval_runtime": 259.9127, | |
| "eval_samples_per_second": 8.418, | |
| "eval_steps_per_second": 1.054, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 0.0001973826714801444, | |
| "loss": 0.4427, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "eval_cer": 0.16931424177800988, | |
| "eval_loss": 0.5793657898902893, | |
| "eval_runtime": 260.1302, | |
| "eval_samples_per_second": 8.411, | |
| "eval_steps_per_second": 1.053, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 0.0001757220216606498, | |
| "loss": 0.3773, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "eval_cer": 0.1633082357720039, | |
| "eval_loss": 0.6081105470657349, | |
| "eval_runtime": 260.6021, | |
| "eval_samples_per_second": 8.396, | |
| "eval_steps_per_second": 1.051, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 0.0001540613718411552, | |
| "loss": 0.3241, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "eval_cer": 0.15936226081153618, | |
| "eval_loss": 0.6171871423721313, | |
| "eval_runtime": 260.2528, | |
| "eval_samples_per_second": 8.407, | |
| "eval_steps_per_second": 1.053, | |
| "step": 3200 | |
| } | |
| ], | |
| "max_steps": 6040, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.610023346999888e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |