| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.880770180202418, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003949642063687978, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 25.19717788696289, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.9992594421130585, | |
| "eval_gen_len": 16.6, | |
| "eval_loss": 13.611336708068848, | |
| "eval_rouge1": 9.8775, | |
| "eval_rouge2": 2.958, | |
| "eval_rougeL": 8.7847, | |
| "eval_rougeLsum": 8.871, | |
| "eval_runtime": 220.704, | |
| "eval_samples_per_second": 4.585, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 1.9755615897309307, | |
| "learning_rate": 4.176954732510288e-05, | |
| "loss": 13.314938470691382, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9992594421130585, | |
| "eval_gen_len": 13.0, | |
| "eval_loss": 7.852785110473633, | |
| "eval_rouge1": 10.5151, | |
| "eval_rouge2": 3.2488, | |
| "eval_rougeL": 9.5196, | |
| "eval_rougeLsum": 9.5659, | |
| "eval_runtime": 203.1626, | |
| "eval_samples_per_second": 4.981, | |
| "step": 506 | |
| }, | |
| { | |
| "epoch": 2.9992594421130585, | |
| "eval_gen_len": 5.3, | |
| "eval_loss": 5.0542707443237305, | |
| "eval_rouge1": 3.3057, | |
| "eval_rouge2": 1.0908, | |
| "eval_rougeL": 2.9248, | |
| "eval_rougeLsum": 2.906, | |
| "eval_runtime": 191.681, | |
| "eval_samples_per_second": 5.28, | |
| "step": 759 | |
| }, | |
| { | |
| "epoch": 3.951863737348803, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 6.2216669921875, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.9992594421130585, | |
| "eval_gen_len": 14.8, | |
| "eval_loss": 4.541400909423828, | |
| "eval_rouge1": 15.0004, | |
| "eval_rouge2": 5.7011, | |
| "eval_rougeL": 13.2283, | |
| "eval_rougeLsum": 13.2086, | |
| "eval_runtime": 197.5356, | |
| "eval_samples_per_second": 5.123, | |
| "step": 1012 | |
| }, | |
| { | |
| "epoch": 4.9992594421130585, | |
| "eval_gen_len": 16.3, | |
| "eval_loss": 4.135443687438965, | |
| "eval_rouge1": 17.3311, | |
| "eval_rouge2": 6.3922, | |
| "eval_rougeL": 15.3676, | |
| "eval_rougeLsum": 15.3563, | |
| "eval_runtime": 200.4736, | |
| "eval_samples_per_second": 5.048, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 5.928165884966675, | |
| "learning_rate": 2.1193415637860082e-05, | |
| "loss": 4.99077294921875, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.9992594421130585, | |
| "eval_gen_len": 17.1, | |
| "eval_loss": 3.8587934970855713, | |
| "eval_rouge1": 18.2338, | |
| "eval_rouge2": 6.8747, | |
| "eval_rougeL": 16.1719, | |
| "eval_rougeLsum": 16.1563, | |
| "eval_runtime": 201.2411, | |
| "eval_samples_per_second": 5.029, | |
| "step": 1518 | |
| }, | |
| { | |
| "epoch": 6.9992594421130585, | |
| "eval_gen_len": 17.4, | |
| "eval_loss": 3.6997532844543457, | |
| "eval_rouge1": 19.4383, | |
| "eval_rouge2": 7.6786, | |
| "eval_rougeL": 17.1778, | |
| "eval_rougeLsum": 17.1963, | |
| "eval_runtime": 200.0236, | |
| "eval_samples_per_second": 5.059, | |
| "step": 1771 | |
| }, | |
| { | |
| "epoch": 7.904468032584547, | |
| "learning_rate": 1.0905349794238684e-05, | |
| "loss": 4.552568359375, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.9992594421130585, | |
| "eval_gen_len": 17.6, | |
| "eval_loss": 3.6233103275299072, | |
| "eval_rouge1": 20.0385, | |
| "eval_rouge2": 7.9683, | |
| "eval_rougeL": 17.6542, | |
| "eval_rougeLsum": 17.6752, | |
| "eval_runtime": 201.3719, | |
| "eval_samples_per_second": 5.026, | |
| "step": 2024 | |
| }, | |
| { | |
| "epoch": 8.999259442113058, | |
| "eval_gen_len": 17.7, | |
| "eval_loss": 3.5823326110839844, | |
| "eval_rouge1": 20.504, | |
| "eval_rouge2": 8.2176, | |
| "eval_rougeL": 18.0877, | |
| "eval_rougeLsum": 18.082, | |
| "eval_runtime": 199.3661, | |
| "eval_samples_per_second": 5.076, | |
| "step": 2277 | |
| }, | |
| { | |
| "epoch": 9.880770180202418, | |
| "learning_rate": 6.17283950617284e-07, | |
| "loss": 4.4207666015625, | |
| "step": 2500 | |
| } | |
| ], | |
| "max_steps": 2530, | |
| "num_train_epochs": 10, | |
| "total_flos": 63945902037307392, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |