| { | |
| "best_metric": 38.5992, | |
| "best_model_checkpoint": "pegasus-newsroom-cnn_full-adafactor-bs6/checkpoint-897", | |
| "epoch": 0.6356132321902493, | |
| "global_step": 1901, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.839464882943145e-05, | |
| "loss": 3.3516, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_gen_len": 75.2876, | |
| "eval_loss": 3.0071940422058105, | |
| "eval_rouge1": 37.4459, | |
| "eval_rouge2": 16.5527, | |
| "eval_rougeL": 26.0178, | |
| "eval_rougeLsum": 34.4955, | |
| "eval_runtime": 1262.9542, | |
| "eval_samples_per_second": 1.979, | |
| "eval_steps_per_second": 0.33, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00015812709030100337, | |
| "loss": 3.1391, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_gen_len": 72.434, | |
| "eval_loss": 2.974590539932251, | |
| "eval_rouge1": 37.9369, | |
| "eval_rouge2": 16.9058, | |
| "eval_rougeL": 26.566, | |
| "eval_rougeLsum": 35.0327, | |
| "eval_runtime": 1247.518, | |
| "eval_samples_per_second": 2.004, | |
| "eval_steps_per_second": 0.334, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0001405351170568562, | |
| "loss": 3.1086, | |
| "step": 897 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_gen_len": 66.2268, | |
| "eval_loss": 2.9494450092315674, | |
| "eval_rouge1": 38.5992, | |
| "eval_rouge2": 17.5271, | |
| "eval_rougeL": 27.3099, | |
| "eval_rougeLsum": 35.6531, | |
| "eval_runtime": 1158.9422, | |
| "eval_samples_per_second": 2.157, | |
| "eval_steps_per_second": 0.36, | |
| "step": 897 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00012060200668896322, | |
| "loss": 3.1037, | |
| "step": 1196 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_gen_len": 70.6692, | |
| "eval_loss": 2.95255184173584, | |
| "eval_rouge1": 38.3805, | |
| "eval_rouge2": 17.5267, | |
| "eval_rougeL": 27.1742, | |
| "eval_rougeLsum": 35.5035, | |
| "eval_runtime": 1183.4051, | |
| "eval_samples_per_second": 2.113, | |
| "eval_steps_per_second": 0.352, | |
| "step": 1196 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00010100334448160536, | |
| "loss": 3.2294, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_gen_len": 56.5128, | |
| "eval_loss": 3.183689832687378, | |
| "eval_rouge1": 36.2619, | |
| "eval_rouge2": 15.599, | |
| "eval_rougeL": 25.8333, | |
| "eval_rougeLsum": 33.3692, | |
| "eval_runtime": 997.3694, | |
| "eval_samples_per_second": 2.507, | |
| "eval_steps_per_second": 0.418, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.127090301003345e-05, | |
| "loss": 3.695, | |
| "step": 1794 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_gen_len": 47.1888, | |
| "eval_loss": 3.4275190830230713, | |
| "eval_rouge1": 33.0969, | |
| "eval_rouge2": 13.6577, | |
| "eval_rougeL": 23.9839, | |
| "eval_rougeLsum": 30.6005, | |
| "eval_runtime": 927.662, | |
| "eval_samples_per_second": 2.695, | |
| "eval_steps_per_second": 0.45, | |
| "step": 1794 | |
| } | |
| ], | |
| "max_steps": 2990, | |
| "num_train_epochs": 1, | |
| "total_flos": 2.4881742419263488e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |