| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 12520, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.5e-05, | |
| "loss": 7.7286, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5e-05, | |
| "loss": 4.7251, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.782986111111111e-05, | |
| "loss": 4.02, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.5659722222222224e-05, | |
| "loss": 3.7877, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.3489583333333334e-05, | |
| "loss": 3.7233, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu1_score": 0.1167, | |
| "eval_bleu2_score": 0.0543, | |
| "eval_bleu3_score": 0.0297, | |
| "eval_bleu4_score": 0.0163, | |
| "eval_loss": 3.503084421157837, | |
| "eval_rougeL": 0.0994, | |
| "eval_runtime": 340.0803, | |
| "eval_samples_per_second": 4.911, | |
| "eval_steps_per_second": 0.82, | |
| "step": 2504 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.1319444444444445e-05, | |
| "loss": 3.0028, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 3.914930555555556e-05, | |
| "loss": 2.9884, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.697916666666667e-05, | |
| "loss": 3.0136, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 3.4809027777777784e-05, | |
| "loss": 2.9996, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.263888888888889e-05, | |
| "loss": 2.9889, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu1_score": 0.1443, | |
| "eval_bleu2_score": 0.0736, | |
| "eval_bleu3_score": 0.0435, | |
| "eval_bleu4_score": 0.0277, | |
| "eval_loss": 3.3509583473205566, | |
| "eval_rougeL": 0.1289, | |
| "eval_runtime": 338.6709, | |
| "eval_samples_per_second": 4.931, | |
| "eval_steps_per_second": 0.824, | |
| "step": 5008 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 3.0468750000000002e-05, | |
| "loss": 2.0035, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.8298611111111113e-05, | |
| "loss": 2.015, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 2.6128472222222223e-05, | |
| "loss": 2.0214, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.3958333333333334e-05, | |
| "loss": 2.0377, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.1788194444444445e-05, | |
| "loss": 2.01, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu1_score": 0.1648, | |
| "eval_bleu2_score": 0.0926, | |
| "eval_bleu3_score": 0.0606, | |
| "eval_bleu4_score": 0.0442, | |
| "eval_loss": 3.430915117263794, | |
| "eval_rougeL": 0.1547, | |
| "eval_runtime": 334.0631, | |
| "eval_samples_per_second": 4.999, | |
| "eval_steps_per_second": 0.835, | |
| "step": 7512 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 1.9618055555555555e-05, | |
| "loss": 1.2721, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 1.7447916666666666e-05, | |
| "loss": 1.2343, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 1.2384, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.3107638888888891e-05, | |
| "loss": 1.2651, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 1.09375e-05, | |
| "loss": 1.2431, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu1_score": 0.1802, | |
| "eval_bleu2_score": 0.1133, | |
| "eval_bleu3_score": 0.0848, | |
| "eval_bleu4_score": 0.0701, | |
| "eval_loss": 3.5641274452209473, | |
| "eval_rougeL": 0.1702, | |
| "eval_runtime": 335.4185, | |
| "eval_samples_per_second": 4.979, | |
| "eval_steps_per_second": 0.832, | |
| "step": 10016 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 8.76736111111111e-06, | |
| "loss": 0.7971, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 6.597222222222223e-06, | |
| "loss": 0.7706, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 4.427083333333334e-06, | |
| "loss": 0.76, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 2.2569444444444443e-06, | |
| "loss": 0.7854, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 8.680555555555556e-08, | |
| "loss": 0.7708, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu1_score": 0.1834, | |
| "eval_bleu2_score": 0.1191, | |
| "eval_bleu3_score": 0.0914, | |
| "eval_bleu4_score": 0.0772, | |
| "eval_loss": 3.666842460632324, | |
| "eval_rougeL": 0.1742, | |
| "eval_runtime": 359.6469, | |
| "eval_samples_per_second": 4.643, | |
| "eval_steps_per_second": 0.776, | |
| "step": 12520 | |
| } | |
| ], | |
| "max_steps": 12520, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.6262752763904e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |