| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 11.0, | |
| "global_step": 11231, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.9347045380346067e-05, | |
| "loss": 0.1574, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.8694090760692135e-05, | |
| "loss": 0.1384, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9631415605545044, | |
| "eval_loss": 0.12948766350746155, | |
| "eval_runtime": 3.6006, | |
| "eval_samples_per_second": 2592.076, | |
| "eval_steps_per_second": 81.098, | |
| "step": 1021 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.80411361410382e-05, | |
| "loss": 0.1024, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.7388181521384265e-05, | |
| "loss": 0.109, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9664630889892578, | |
| "eval_loss": 0.11389727890491486, | |
| "eval_runtime": 3.5512, | |
| "eval_samples_per_second": 2628.103, | |
| "eval_steps_per_second": 82.225, | |
| "step": 2042 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.673522690173033e-05, | |
| "loss": 0.0816, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.6082272282076395e-05, | |
| "loss": 0.0764, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9662488102912903, | |
| "eval_loss": 0.11298216134309769, | |
| "eval_runtime": 3.5558, | |
| "eval_samples_per_second": 2624.692, | |
| "eval_steps_per_second": 82.118, | |
| "step": 3063 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 1.5429317662422463e-05, | |
| "loss": 0.065, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.4776363042768528e-05, | |
| "loss": 0.0517, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9630343914031982, | |
| "eval_loss": 0.1652730107307434, | |
| "eval_runtime": 3.5703, | |
| "eval_samples_per_second": 2614.08, | |
| "eval_steps_per_second": 81.786, | |
| "step": 4084 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 1.4123408423114595e-05, | |
| "loss": 0.0348, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 1.3470453803460662e-05, | |
| "loss": 0.0385, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9632486701011658, | |
| "eval_loss": 0.18385031819343567, | |
| "eval_runtime": 3.5894, | |
| "eval_samples_per_second": 2600.184, | |
| "eval_steps_per_second": 81.352, | |
| "step": 5105 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 1.2817499183806728e-05, | |
| "loss": 0.0244, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 1.2164544564152792e-05, | |
| "loss": 0.0245, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9693560600280762, | |
| "eval_loss": 0.1997281014919281, | |
| "eval_runtime": 3.7802, | |
| "eval_samples_per_second": 2468.931, | |
| "eval_steps_per_second": 77.245, | |
| "step": 6126 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 1.1511589944498858e-05, | |
| "loss": 0.0175, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 1.0858635324844923e-05, | |
| "loss": 0.0171, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9687131643295288, | |
| "eval_loss": 0.1954147070646286, | |
| "eval_runtime": 3.5659, | |
| "eval_samples_per_second": 2617.281, | |
| "eval_steps_per_second": 81.886, | |
| "step": 7147 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 1.020568070519099e-05, | |
| "loss": 0.0127, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 9.552726085537057e-06, | |
| "loss": 0.0102, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9687131643295288, | |
| "eval_loss": 0.23564255237579346, | |
| "eval_runtime": 3.5772, | |
| "eval_samples_per_second": 2609.01, | |
| "eval_steps_per_second": 81.628, | |
| "step": 8168 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 8.899771465883122e-06, | |
| "loss": 0.0092, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 8.246816846229188e-06, | |
| "loss": 0.0078, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.966570258140564, | |
| "eval_loss": 0.2585618495941162, | |
| "eval_runtime": 3.6436, | |
| "eval_samples_per_second": 2561.473, | |
| "eval_steps_per_second": 80.14, | |
| "step": 9189 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 7.593862226575253e-06, | |
| "loss": 0.0061, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 6.94090760692132e-06, | |
| "loss": 0.0034, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9699989557266235, | |
| "eval_loss": 0.2755224406719208, | |
| "eval_runtime": 3.5689, | |
| "eval_samples_per_second": 2615.084, | |
| "eval_steps_per_second": 81.818, | |
| "step": 10210 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 6.287952987267385e-06, | |
| "loss": 0.0035, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 5.634998367613451e-06, | |
| "loss": 0.0029, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9706417918205261, | |
| "eval_loss": 0.2912203371524811, | |
| "eval_runtime": 3.5951, | |
| "eval_samples_per_second": 2596.026, | |
| "eval_steps_per_second": 81.221, | |
| "step": 11231 | |
| } | |
| ], | |
| "max_steps": 15315, | |
| "num_train_epochs": 15, | |
| "total_flos": 2.363421365580595e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |