| { | |
| "best_metric": 0.4198276698589325, | |
| "best_model_checkpoint": "hBERTv1_qnli/checkpoint-1230", | |
| "epoch": 8.0, | |
| "global_step": 3280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6667, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6873512721947648, | |
| "eval_loss": 0.5954663157463074, | |
| "eval_runtime": 6.817, | |
| "eval_samples_per_second": 801.373, | |
| "eval_steps_per_second": 3.227, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.4998, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.794801391177009, | |
| "eval_loss": 0.44860708713531494, | |
| "eval_runtime": 6.81, | |
| "eval_samples_per_second": 802.199, | |
| "eval_steps_per_second": 3.231, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.3985, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8112758557569101, | |
| "eval_loss": 0.4198276698589325, | |
| "eval_runtime": 6.8165, | |
| "eval_samples_per_second": 801.434, | |
| "eval_steps_per_second": 3.227, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.3106, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7865641588870584, | |
| "eval_loss": 0.48407182097435, | |
| "eval_runtime": 6.8451, | |
| "eval_samples_per_second": 798.094, | |
| "eval_steps_per_second": 3.214, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.2286, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.790591250228812, | |
| "eval_loss": 0.5340340733528137, | |
| "eval_runtime": 6.828, | |
| "eval_samples_per_second": 800.093, | |
| "eval_steps_per_second": 3.222, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.1662, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7728354384038074, | |
| "eval_loss": 0.6281628608703613, | |
| "eval_runtime": 6.835, | |
| "eval_samples_per_second": 799.265, | |
| "eval_steps_per_second": 3.219, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.1237, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7752150832875709, | |
| "eval_loss": 0.6678447723388672, | |
| "eval_runtime": 6.8476, | |
| "eval_samples_per_second": 797.793, | |
| "eval_steps_per_second": 3.213, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0945, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7752150832875709, | |
| "eval_loss": 0.7668098211288452, | |
| "eval_runtime": 6.8357, | |
| "eval_samples_per_second": 799.183, | |
| "eval_steps_per_second": 3.218, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 3280, | |
| "total_flos": 1.0584157486002995e+17, | |
| "train_loss": 0.3110694268854653, | |
| "train_runtime": 3241.717, | |
| "train_samples_per_second": 1615.548, | |
| "train_steps_per_second": 6.324 | |
| } | |
| ], | |
| "max_steps": 20500, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.0584157486002995e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |