| { | |
| "best_metric": 0.27871406078338623, | |
| "best_model_checkpoint": "/content/deberta-v3-small/checkpoint-230", | |
| "epoch": 10.0, | |
| "global_step": 2300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8921568627450981, | |
| "eval_combined_score": 0.9077509052401449, | |
| "eval_f1": 0.9233449477351917, | |
| "eval_loss": 0.27871406078338623, | |
| "eval_runtime": 0.9934, | |
| "eval_samples_per_second": 410.721, | |
| "eval_steps_per_second": 51.34, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.875, | |
| "eval_combined_score": 0.8943527918781726, | |
| "eval_f1": 0.9137055837563453, | |
| "eval_loss": 0.3650500774383545, | |
| "eval_runtime": 1.104, | |
| "eval_samples_per_second": 369.555, | |
| "eval_steps_per_second": 46.194, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8799019607843137, | |
| "eval_combined_score": 0.8989124544290077, | |
| "eval_f1": 0.9179229480737018, | |
| "eval_loss": 0.5237621068954468, | |
| "eval_runtime": 1.014, | |
| "eval_samples_per_second": 402.381, | |
| "eval_steps_per_second": 50.298, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8946078431372549, | |
| "eval_combined_score": 0.9084250788923165, | |
| "eval_f1": 0.9222423146473779, | |
| "eval_loss": 0.471233606338501, | |
| "eval_runtime": 1.021, | |
| "eval_samples_per_second": 399.621, | |
| "eval_steps_per_second": 49.953, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 1.6956521739130433e-05, | |
| "loss": 0.2147, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8946078431372549, | |
| "eval_combined_score": 0.9104257054451284, | |
| "eval_f1": 0.9262435677530019, | |
| "eval_loss": 0.570386528968811, | |
| "eval_runtime": 0.9794, | |
| "eval_samples_per_second": 416.598, | |
| "eval_steps_per_second": 52.075, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8995098039215687, | |
| "eval_combined_score": 0.9139782876501386, | |
| "eval_f1": 0.9284467713787086, | |
| "eval_loss": 0.5697466135025024, | |
| "eval_runtime": 0.9891, | |
| "eval_samples_per_second": 412.495, | |
| "eval_steps_per_second": 51.562, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8921568627450981, | |
| "eval_combined_score": 0.9067927170868348, | |
| "eval_f1": 0.9214285714285715, | |
| "eval_loss": 0.6650992631912231, | |
| "eval_runtime": 1.008, | |
| "eval_samples_per_second": 404.777, | |
| "eval_steps_per_second": 50.597, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8946078431372549, | |
| "eval_combined_score": 0.9092508242234947, | |
| "eval_f1": 0.9238938053097344, | |
| "eval_loss": 0.672588586807251, | |
| "eval_runtime": 1.0378, | |
| "eval_samples_per_second": 393.14, | |
| "eval_steps_per_second": 49.143, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 3.913043478260869e-06, | |
| "loss": 0.0183, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8848039215686274, | |
| "eval_combined_score": 0.9012460938841386, | |
| "eval_f1": 0.9176882661996497, | |
| "eval_loss": 0.7250252366065979, | |
| "eval_runtime": 1.0544, | |
| "eval_samples_per_second": 386.943, | |
| "eval_steps_per_second": 48.368, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8921568627450981, | |
| "eval_combined_score": 0.907209173422019, | |
| "eval_f1": 0.9222614840989399, | |
| "eval_loss": 0.7092880010604858, | |
| "eval_runtime": 1.0253, | |
| "eval_samples_per_second": 397.917, | |
| "eval_steps_per_second": 49.74, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2300, | |
| "total_flos": 1214769315164160.0, | |
| "train_loss": 0.1021328606294549, | |
| "train_runtime": 294.6784, | |
| "train_samples_per_second": 124.475, | |
| "train_steps_per_second": 7.805 | |
| } | |
| ], | |
| "max_steps": 2300, | |
| "num_train_epochs": 10, | |
| "total_flos": 1214769315164160.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |