| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 9.99985909539242, |
| "global_step": 8870, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.56, |
| "learning_rate": 4.71815107102593e-05, |
| "loss": 0.2722, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.13, |
| "learning_rate": 4.43630214205186e-05, |
| "loss": 0.1261, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.13, |
| "eval_acc": 0.98, |
| "eval_loss": 0.09329384565353394, |
| "eval_num": 56776, |
| "eval_runtime": 5408.6044, |
| "eval_samples_per_second": 10.497, |
| "eval_steps_per_second": 2.624, |
| "eval_true_num": 55574, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.69, |
| "learning_rate": 4.1544532130777905e-05, |
| "loss": 0.1078, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.25, |
| "learning_rate": 3.872604284103721e-05, |
| "loss": 0.0951, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.25, |
| "eval_acc": 0.98, |
| "eval_loss": 0.06545884907245636, |
| "eval_num": 56776, |
| "eval_runtime": 5398.8946, |
| "eval_samples_per_second": 10.516, |
| "eval_steps_per_second": 2.629, |
| "eval_true_num": 55867, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.82, |
| "learning_rate": 3.590755355129651e-05, |
| "loss": 0.0832, |
| "step": 2500 |
| }, |
| { |
| "epoch": 3.38, |
| "learning_rate": 3.308906426155581e-05, |
| "loss": 0.0774, |
| "step": 3000 |
| }, |
| { |
| "epoch": 3.38, |
| "eval_acc": 0.99, |
| "eval_loss": 0.047995537519454956, |
| "eval_num": 56776, |
| "eval_runtime": 5407.5843, |
| "eval_samples_per_second": 10.499, |
| "eval_steps_per_second": 2.625, |
| "eval_true_num": 56047, |
| "step": 3000 |
| }, |
| { |
| "epoch": 3.95, |
| "learning_rate": 3.0270574971815107e-05, |
| "loss": 0.0689, |
| "step": 3500 |
| }, |
| { |
| "epoch": 4.51, |
| "learning_rate": 2.745208568207441e-05, |
| "loss": 0.0584, |
| "step": 4000 |
| }, |
| { |
| "epoch": 4.51, |
| "eval_acc": 0.99, |
| "eval_loss": 0.03338366374373436, |
| "eval_num": 56776, |
| "eval_runtime": 5517.5145, |
| "eval_samples_per_second": 10.29, |
| "eval_steps_per_second": 2.573, |
| "eval_true_num": 56252, |
| "step": 4000 |
| }, |
| { |
| "epoch": 5.07, |
| "learning_rate": 2.463359639233371e-05, |
| "loss": 0.0537, |
| "step": 4500 |
| }, |
| { |
| "epoch": 5.64, |
| "learning_rate": 2.181510710259301e-05, |
| "loss": 0.042, |
| "step": 5000 |
| }, |
| { |
| "epoch": 5.64, |
| "eval_acc": 0.99, |
| "eval_loss": 0.022190678864717484, |
| "eval_num": 56776, |
| "eval_runtime": 5470.5084, |
| "eval_samples_per_second": 10.379, |
| "eval_steps_per_second": 2.595, |
| "eval_true_num": 56411, |
| "step": 5000 |
| }, |
| { |
| "epoch": 6.2, |
| "learning_rate": 1.8996617812852312e-05, |
| "loss": 0.0431, |
| "step": 5500 |
| }, |
| { |
| "epoch": 6.76, |
| "learning_rate": 1.6178128523111614e-05, |
| "loss": 0.0329, |
| "step": 6000 |
| }, |
| { |
| "epoch": 6.76, |
| "eval_acc": 1.0, |
| "eval_loss": 0.013875513337552547, |
| "eval_num": 56776, |
| "eval_runtime": 5431.4557, |
| "eval_samples_per_second": 10.453, |
| "eval_steps_per_second": 2.613, |
| "eval_true_num": 56502, |
| "step": 6000 |
| }, |
| { |
| "epoch": 7.33, |
| "learning_rate": 1.3359639233370913e-05, |
| "loss": 0.0305, |
| "step": 6500 |
| }, |
| { |
| "epoch": 7.89, |
| "learning_rate": 1.0541149943630215e-05, |
| "loss": 0.0254, |
| "step": 7000 |
| }, |
| { |
| "epoch": 7.89, |
| "eval_acc": 1.0, |
| "eval_loss": 0.009383471682667732, |
| "eval_num": 56776, |
| "eval_runtime": 5439.4304, |
| "eval_samples_per_second": 10.438, |
| "eval_steps_per_second": 2.609, |
| "eval_true_num": 56626, |
| "step": 7000 |
| }, |
| { |
| "epoch": 8.46, |
| "learning_rate": 7.722660653889515e-06, |
| "loss": 0.0225, |
| "step": 7500 |
| }, |
| { |
| "epoch": 9.02, |
| "learning_rate": 4.904171364148816e-06, |
| "loss": 0.0214, |
| "step": 8000 |
| }, |
| { |
| "epoch": 9.02, |
| "eval_acc": 1.0, |
| "eval_loss": 0.007043090648949146, |
| "eval_num": 56776, |
| "eval_runtime": 5398.7111, |
| "eval_samples_per_second": 10.517, |
| "eval_steps_per_second": 2.629, |
| "eval_true_num": 56659, |
| "step": 8000 |
| }, |
| { |
| "epoch": 9.58, |
| "learning_rate": 2.0856820744081176e-06, |
| "loss": 0.017, |
| "step": 8500 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 8870, |
| "total_flos": 7.53120395931648e+17, |
| "train_loss": 0.06710476681774875, |
| "train_runtime": 130006.8001, |
| "train_samples_per_second": 4.367, |
| "train_steps_per_second": 0.068 |
| } |
| ], |
| "max_steps": 8870, |
| "num_train_epochs": 10, |
| "total_flos": 7.53120395931648e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
