| { | |
| "best_metric": 0.8575587868690491, | |
| "best_model_checkpoint": "./electra-base-discriminator/fine_tuned_models/checkpoint-3729", | |
| "epoch": 10.0, | |
| "global_step": 12430, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.9998212370396856e-05, | |
| "loss": 0.5211, | |
| "step": 1243 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8279955387115479, | |
| "eval_loss": 0.3951820731163025, | |
| "eval_runtime": 7.0965, | |
| "eval_samples_per_second": 629.183, | |
| "eval_steps_per_second": 78.771, | |
| "step": 1243 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.7776188773686092e-05, | |
| "loss": 0.3301, | |
| "step": 2486 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8539753556251526, | |
| "eval_loss": 0.356336385011673, | |
| "eval_runtime": 7.2692, | |
| "eval_samples_per_second": 614.237, | |
| "eval_steps_per_second": 76.9, | |
| "step": 2486 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.5554165176975332e-05, | |
| "loss": 0.2267, | |
| "step": 3729 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8575587868690491, | |
| "eval_loss": 0.3921225965023041, | |
| "eval_runtime": 7.1896, | |
| "eval_samples_per_second": 621.038, | |
| "eval_steps_per_second": 77.752, | |
| "step": 3729 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.333214158026457e-05, | |
| "loss": 0.1557, | |
| "step": 4972 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8546472787857056, | |
| "eval_loss": 0.46923860907554626, | |
| "eval_runtime": 7.2458, | |
| "eval_samples_per_second": 616.217, | |
| "eval_steps_per_second": 77.148, | |
| "step": 4972 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.1110117983553808e-05, | |
| "loss": 0.1155, | |
| "step": 6215 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8559910655021667, | |
| "eval_loss": 0.5102965831756592, | |
| "eval_runtime": 7.4396, | |
| "eval_samples_per_second": 600.169, | |
| "eval_steps_per_second": 75.139, | |
| "step": 6215 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.888094386843046e-06, | |
| "loss": 0.0894, | |
| "step": 7458 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.856662929058075, | |
| "eval_loss": 0.6104263663291931, | |
| "eval_runtime": 7.2873, | |
| "eval_samples_per_second": 612.707, | |
| "eval_steps_per_second": 76.708, | |
| "step": 7458 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 6.666070790132285e-06, | |
| "loss": 0.0702, | |
| "step": 8701 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8533034920692444, | |
| "eval_loss": 0.6788500547409058, | |
| "eval_runtime": 7.0995, | |
| "eval_samples_per_second": 628.92, | |
| "eval_steps_per_second": 78.738, | |
| "step": 8701 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.444047193421523e-06, | |
| "loss": 0.0594, | |
| "step": 9944 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8524076342582703, | |
| "eval_loss": 0.7277155518531799, | |
| "eval_runtime": 7.218, | |
| "eval_samples_per_second": 618.59, | |
| "eval_steps_per_second": 77.445, | |
| "step": 9944 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.2220235967107616e-06, | |
| "loss": 0.0472, | |
| "step": 11187 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8546472787857056, | |
| "eval_loss": 0.8223575949668884, | |
| "eval_runtime": 7.1336, | |
| "eval_samples_per_second": 625.908, | |
| "eval_steps_per_second": 78.361, | |
| "step": 11187 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0379, | |
| "step": 12430 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8521836400032043, | |
| "eval_loss": 0.8311443328857422, | |
| "eval_runtime": 7.1227, | |
| "eval_samples_per_second": 626.866, | |
| "eval_steps_per_second": 78.481, | |
| "step": 12430 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 12430, | |
| "total_flos": 1.671959181036288e+16, | |
| "train_loss": 0.16532029376716645, | |
| "train_runtime": 1732.3817, | |
| "train_samples_per_second": 229.563, | |
| "train_steps_per_second": 7.175 | |
| } | |
| ], | |
| "max_steps": 12430, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.671959181036288e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |