| { | |
| "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": 0.1561, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.8694090760692135e-05, | |
| "loss": 0.1369, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9633558392524719, | |
| "eval_loss": 0.11954528838396072, | |
| "eval_runtime": 3.6034, | |
| "eval_samples_per_second": 2590.036, | |
| "eval_steps_per_second": 81.034, | |
| "step": 1021 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.80411361410382e-05, | |
| "loss": 0.0974, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.7388181521384265e-05, | |
| "loss": 0.1035, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9667845368385315, | |
| "eval_loss": 0.12783248722553253, | |
| "eval_runtime": 3.6648, | |
| "eval_samples_per_second": 2546.636, | |
| "eval_steps_per_second": 79.676, | |
| "step": 2042 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.673522690173033e-05, | |
| "loss": 0.0744, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.6082272282076395e-05, | |
| "loss": 0.0734, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9661416411399841, | |
| "eval_loss": 0.11725891381502151, | |
| "eval_runtime": 3.8091, | |
| "eval_samples_per_second": 2450.154, | |
| "eval_steps_per_second": 76.658, | |
| "step": 3063 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 1.5429317662422463e-05, | |
| "loss": 0.0586, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.4776363042768528e-05, | |
| "loss": 0.0448, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9683917164802551, | |
| "eval_loss": 0.16057069599628448, | |
| "eval_runtime": 3.7958, | |
| "eval_samples_per_second": 2458.759, | |
| "eval_steps_per_second": 76.927, | |
| "step": 4084 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 1.4123408423114595e-05, | |
| "loss": 0.031, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 1.3470453803460662e-05, | |
| "loss": 0.0321, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9644272923469543, | |
| "eval_loss": 0.16097016632556915, | |
| "eval_runtime": 3.8317, | |
| "eval_samples_per_second": 2435.729, | |
| "eval_steps_per_second": 76.206, | |
| "step": 5105 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 1.2817499183806728e-05, | |
| "loss": 0.0183, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 1.2164544564152792e-05, | |
| "loss": 0.0163, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9695703387260437, | |
| "eval_loss": 0.21039049327373505, | |
| "eval_runtime": 3.8392, | |
| "eval_samples_per_second": 2430.966, | |
| "eval_steps_per_second": 76.057, | |
| "step": 6126 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 1.1511589944498858e-05, | |
| "loss": 0.0123, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 1.0858635324844923e-05, | |
| "loss": 0.0107, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9709632396697998, | |
| "eval_loss": 0.2211793214082718, | |
| "eval_runtime": 3.818, | |
| "eval_samples_per_second": 2444.447, | |
| "eval_steps_per_second": 76.479, | |
| "step": 7147 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 1.020568070519099e-05, | |
| "loss": 0.009, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 9.552726085537057e-06, | |
| "loss": 0.005, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9671059846878052, | |
| "eval_loss": 0.26634445786476135, | |
| "eval_runtime": 3.8057, | |
| "eval_samples_per_second": 2452.35, | |
| "eval_steps_per_second": 76.726, | |
| "step": 8168 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 8.899771465883122e-06, | |
| "loss": 0.0047, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 8.246816846229188e-06, | |
| "loss": 0.0068, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.96924889087677, | |
| "eval_loss": 0.2791275084018707, | |
| "eval_runtime": 3.6479, | |
| "eval_samples_per_second": 2558.477, | |
| "eval_steps_per_second": 80.047, | |
| "step": 9189 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 7.593862226575253e-06, | |
| "loss": 0.0041, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 6.94090760692132e-06, | |
| "loss": 0.0042, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9676417112350464, | |
| "eval_loss": 0.2793237268924713, | |
| "eval_runtime": 3.6466, | |
| "eval_samples_per_second": 2559.346, | |
| "eval_steps_per_second": 80.074, | |
| "step": 10210 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 6.287952987267385e-06, | |
| "loss": 0.0028, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 5.634998367613451e-06, | |
| "loss": 0.0035, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9691417813301086, | |
| "eval_loss": 0.28977808356285095, | |
| "eval_runtime": 3.6274, | |
| "eval_samples_per_second": 2572.927, | |
| "eval_steps_per_second": 80.499, | |
| "step": 11231 | |
| } | |
| ], | |
| "max_steps": 15315, | |
| "num_train_epochs": 15, | |
| "total_flos": 2.363421365580595e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |