| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.582800284292822, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 5.2679, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 4.0457, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.5e-05, | |
| "loss": 3.7746, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 3.4735970497131348, | |
| "eval_rouge2_fmeasure": 0.0608, | |
| "eval_rouge2_precision": 0.0752, | |
| "eval_rouge2_recall": 0.0676, | |
| "eval_runtime": 10444.2518, | |
| "eval_samples_per_second": 0.09, | |
| "eval_steps_per_second": 0.09, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 3.6193, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 3.5271, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 5e-05, | |
| "loss": 3.4793, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 3.331468105316162, | |
| "eval_rouge2_fmeasure": 0.0653, | |
| "eval_rouge2_precision": 0.0796, | |
| "eval_rouge2_recall": 0.0735, | |
| "eval_runtime": 10901.7035, | |
| "eval_samples_per_second": 0.086, | |
| "eval_steps_per_second": 0.086, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 4.8529411764705885e-05, | |
| "loss": 3.4251, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 3.4313, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 4.558823529411765e-05, | |
| "loss": 3.3459, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 3.2606232166290283, | |
| "eval_rouge2_fmeasure": 0.0685, | |
| "eval_rouge2_precision": 0.0825, | |
| "eval_rouge2_recall": 0.077, | |
| "eval_runtime": 10929.9339, | |
| "eval_samples_per_second": 0.086, | |
| "eval_steps_per_second": 0.086, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.411764705882353e-05, | |
| "loss": 3.291, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 4.2647058823529415e-05, | |
| "loss": 3.2803, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 3.2948, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 3.227094888687134, | |
| "eval_rouge2_fmeasure": 0.0684, | |
| "eval_rouge2_precision": 0.0815, | |
| "eval_rouge2_recall": 0.078, | |
| "eval_runtime": 11536.812, | |
| "eval_samples_per_second": 0.081, | |
| "eval_steps_per_second": 0.081, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.970588235294117e-05, | |
| "loss": 3.2314, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.8235294117647055e-05, | |
| "loss": 3.2134, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 3.6764705882352945e-05, | |
| "loss": 3.1879, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "eval_loss": 3.2067530155181885, | |
| "eval_rouge2_fmeasure": 0.0704, | |
| "eval_rouge2_precision": 0.083, | |
| "eval_rouge2_recall": 0.0809, | |
| "eval_runtime": 11358.1561, | |
| "eval_samples_per_second": 0.083, | |
| "eval_steps_per_second": 0.083, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 3.1617, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 3.382352941176471e-05, | |
| "loss": 3.172, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 3.235294117647059e-05, | |
| "loss": 3.1038, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "eval_loss": 3.1938178539276123, | |
| "eval_rouge2_fmeasure": 0.0716, | |
| "eval_rouge2_precision": 0.0858, | |
| "eval_rouge2_recall": 0.0813, | |
| "eval_runtime": 10958.7861, | |
| "eval_samples_per_second": 0.086, | |
| "eval_steps_per_second": 0.086, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 3.0882352941176475e-05, | |
| "loss": 3.1153, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 3.1037, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 2.7941176470588236e-05, | |
| "loss": 3.1155, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_loss": 3.187068462371826, | |
| "eval_rouge2_fmeasure": 0.0725, | |
| "eval_rouge2_precision": 0.0856, | |
| "eval_rouge2_recall": 0.0839, | |
| "eval_runtime": 11370.9481, | |
| "eval_samples_per_second": 0.083, | |
| "eval_steps_per_second": 0.083, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 2.647058823529412e-05, | |
| "loss": 3.048, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 2.5e-05, | |
| "loss": 3.0603, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 3.0534, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "eval_loss": 3.181117057800293, | |
| "eval_rouge2_fmeasure": 0.0735, | |
| "eval_rouge2_precision": 0.0865, | |
| "eval_rouge2_recall": 0.0848, | |
| "eval_runtime": 11629.6122, | |
| "eval_samples_per_second": 0.081, | |
| "eval_steps_per_second": 0.081, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 2.2058823529411766e-05, | |
| "loss": 3.0664, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 2.058823529411765e-05, | |
| "loss": 3.0255, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 1.9117647058823528e-05, | |
| "loss": 3.0338, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 3.1781060695648193, | |
| "eval_rouge2_fmeasure": 0.0747, | |
| "eval_rouge2_precision": 0.0879, | |
| "eval_rouge2_recall": 0.0865, | |
| "eval_runtime": 11275.9298, | |
| "eval_samples_per_second": 0.083, | |
| "eval_steps_per_second": 0.083, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 2.9872, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 1.6176470588235296e-05, | |
| "loss": 3.0128, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.4705882352941177e-05, | |
| "loss": 3.0266, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "eval_loss": 3.1731793880462646, | |
| "eval_rouge2_fmeasure": 0.074, | |
| "eval_rouge2_precision": 0.0886, | |
| "eval_rouge2_recall": 0.0836, | |
| "eval_runtime": 11197.1963, | |
| "eval_samples_per_second": 0.084, | |
| "eval_steps_per_second": 0.084, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 1.323529411764706e-05, | |
| "loss": 2.9676, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 2.9834, | |
| "step": 8000 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.248414149407408e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |