| { | |
| "best_metric": 2.608954906463623, | |
| "best_model_checkpoint": "./models/final_bart/checkpoint-2000", | |
| "epoch": 5.0, | |
| "global_step": 3310, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.063444108761328e-06, | |
| "loss": 5.5998, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.8126888217522657e-05, | |
| "loss": 2.7748, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.719033232628399e-05, | |
| "loss": 2.5362, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.9305135951661632e-05, | |
| "loss": 2.4214, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 2.8298086606243708e-05, | |
| "loss": 2.3583, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 2.729103726082578e-05, | |
| "loss": 2.3157, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.6283987915407858e-05, | |
| "loss": 2.2639, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 2.527693856998993e-05, | |
| "loss": 2.1867, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.4269889224572005e-05, | |
| "loss": 2.155, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.3262839879154077e-05, | |
| "loss": 2.1622, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_bleu1": 29.4018, | |
| "eval_bleu2": 17.2004, | |
| "eval_bleu3": 10.3744, | |
| "eval_bleu4": 6.052, | |
| "eval_gen_len": 49.4266, | |
| "eval_loss": 2.6686596870422363, | |
| "eval_rdass": 0.6378999948501587, | |
| "eval_rouge1": 35.4366, | |
| "eval_rouge2": 12.8631, | |
| "eval_rougeL": 23.1588, | |
| "eval_runtime": 119.7752, | |
| "eval_samples_per_second": 3.582, | |
| "eval_steps_per_second": 0.058, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.2255790533736152e-05, | |
| "loss": 2.1548, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.124874118831823e-05, | |
| "loss": 2.1442, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 2.0241691842900302e-05, | |
| "loss": 2.1273, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1.9234642497482378e-05, | |
| "loss": 2.0555, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.822759315206445e-05, | |
| "loss": 2.0243, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 1.7220543806646525e-05, | |
| "loss": 2.0054, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 1.6213494461228603e-05, | |
| "loss": 2.0066, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.5206445115810675e-05, | |
| "loss": 2.009, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 1.419939577039275e-05, | |
| "loss": 2.0079, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1.3192346424974824e-05, | |
| "loss": 2.0114, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_bleu1": 28.8917, | |
| "eval_bleu2": 17.0965, | |
| "eval_bleu3": 10.1873, | |
| "eval_bleu4": 5.896, | |
| "eval_gen_len": 46.1096, | |
| "eval_loss": 2.608954906463623, | |
| "eval_rdass": 0.6388999819755554, | |
| "eval_rouge1": 35.1436, | |
| "eval_rouge2": 13.0347, | |
| "eval_rougeL": 23.4682, | |
| "eval_runtime": 105.5108, | |
| "eval_samples_per_second": 4.066, | |
| "eval_steps_per_second": 0.066, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 1.2185297079556899e-05, | |
| "loss": 1.9198, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 1.1178247734138972e-05, | |
| "loss": 1.9285, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 1.0171198388721048e-05, | |
| "loss": 1.9271, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 9.164149043303123e-06, | |
| "loss": 1.9449, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 8.157099697885196e-06, | |
| "loss": 1.9189, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 7.1500503524672715e-06, | |
| "loss": 1.9269, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 6.143001007049345e-06, | |
| "loss": 1.8913, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 5.13595166163142e-06, | |
| "loss": 1.8721, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 4.1289023162134946e-06, | |
| "loss": 1.8779, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 3.121852970795569e-06, | |
| "loss": 1.8758, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_bleu1": 29.5206, | |
| "eval_bleu2": 17.3914, | |
| "eval_bleu3": 10.5577, | |
| "eval_bleu4": 6.1502, | |
| "eval_gen_len": 49.7389, | |
| "eval_loss": 2.6099660396575928, | |
| "eval_rdass": 0.6449000239372253, | |
| "eval_rouge1": 35.5593, | |
| "eval_rouge2": 13.0497, | |
| "eval_rougeL": 23.5672, | |
| "eval_runtime": 115.4788, | |
| "eval_samples_per_second": 3.715, | |
| "eval_steps_per_second": 0.061, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 2.1148036253776437e-06, | |
| "loss": 1.8709, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 1.107754279959718e-06, | |
| "loss": 1.8656, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 1.0070493454179255e-07, | |
| "loss": 1.8666, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 3310, | |
| "total_flos": 5.527100997669888e+16, | |
| "train_loss": 2.181030597859639, | |
| "train_runtime": 4138.7205, | |
| "train_samples_per_second": 51.184, | |
| "train_steps_per_second": 0.8 | |
| } | |
| ], | |
| "max_steps": 3310, | |
| "num_train_epochs": 5, | |
| "total_flos": 5.527100997669888e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |