| { | |
| "best_metric": 0.28330111503601074, | |
| "best_model_checkpoint": "/content/best_model/checkpoint-65000", | |
| "epoch": 2.0, | |
| "global_step": 69122, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.638320650444142e-05, | |
| "loss": 0.3604, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.8821028470993042, | |
| "eval_loss": 0.3162487745285034, | |
| "eval_runtime": 142.0782, | |
| "eval_samples_per_second": 203.634, | |
| "eval_steps_per_second": 3.188, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.276641300888285e-05, | |
| "loss": 0.3326, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.8842803835868835, | |
| "eval_loss": 0.311200350522995, | |
| "eval_runtime": 142.1706, | |
| "eval_samples_per_second": 203.502, | |
| "eval_steps_per_second": 3.186, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.914961951332427e-05, | |
| "loss": 0.3293, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.8869763612747192, | |
| "eval_loss": 0.3043750524520874, | |
| "eval_runtime": 142.2145, | |
| "eval_samples_per_second": 203.439, | |
| "eval_steps_per_second": 3.185, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3.5532826017765694e-05, | |
| "loss": 0.3246, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.8871491551399231, | |
| "eval_loss": 0.30401167273521423, | |
| "eval_runtime": 142.2635, | |
| "eval_samples_per_second": 203.369, | |
| "eval_steps_per_second": 3.184, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.191603252220711e-05, | |
| "loss": 0.32, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.8887736797332764, | |
| "eval_loss": 0.29694026708602905, | |
| "eval_runtime": 142.2955, | |
| "eval_samples_per_second": 203.323, | |
| "eval_steps_per_second": 3.184, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.829923902664854e-05, | |
| "loss": 0.3143, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.8903290629386902, | |
| "eval_loss": 0.2928813695907593, | |
| "eval_runtime": 142.3659, | |
| "eval_samples_per_second": 203.223, | |
| "eval_steps_per_second": 3.182, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.468244553108996e-05, | |
| "loss": 0.3095, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.8899142742156982, | |
| "eval_loss": 0.29173970222473145, | |
| "eval_runtime": 142.5262, | |
| "eval_samples_per_second": 202.994, | |
| "eval_steps_per_second": 3.178, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.106565203553138e-05, | |
| "loss": 0.2844, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_accuracy": 0.8886008858680725, | |
| "eval_loss": 0.29569417238235474, | |
| "eval_runtime": 142.5524, | |
| "eval_samples_per_second": 202.957, | |
| "eval_steps_per_second": 3.178, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.7448858539972804e-05, | |
| "loss": 0.2778, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.890640139579773, | |
| "eval_loss": 0.2942551076412201, | |
| "eval_runtime": 142.6301, | |
| "eval_samples_per_second": 202.846, | |
| "eval_steps_per_second": 3.176, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.3832065044414225e-05, | |
| "loss": 0.2779, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 0.8934743404388428, | |
| "eval_loss": 0.28896576166152954, | |
| "eval_runtime": 142.7675, | |
| "eval_samples_per_second": 202.651, | |
| "eval_steps_per_second": 3.173, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.0215271548855646e-05, | |
| "loss": 0.2752, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_accuracy": 0.891884446144104, | |
| "eval_loss": 0.28808724880218506, | |
| "eval_runtime": 142.5778, | |
| "eval_samples_per_second": 202.921, | |
| "eval_steps_per_second": 3.177, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 6.598478053297069e-06, | |
| "loss": 0.2736, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.8943729996681213, | |
| "eval_loss": 0.28354716300964355, | |
| "eval_runtime": 142.8817, | |
| "eval_samples_per_second": 202.489, | |
| "eval_steps_per_second": 3.17, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.9816845577384916e-06, | |
| "loss": 0.2725, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_accuracy": 0.8941656351089478, | |
| "eval_loss": 0.28330111503601074, | |
| "eval_runtime": 142.8534, | |
| "eval_samples_per_second": 202.529, | |
| "eval_steps_per_second": 3.171, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 69122, | |
| "total_flos": 3.410043502198626e+17, | |
| "train_loss": 0.30218412260715205, | |
| "train_runtime": 63905.8286, | |
| "train_samples_per_second": 69.223, | |
| "train_steps_per_second": 1.082 | |
| } | |
| ], | |
| "max_steps": 69122, | |
| "num_train_epochs": 2, | |
| "total_flos": 3.410043502198626e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |