| { | |
| "best_metric": 1.1659743785858154, | |
| "best_model_checkpoint": "/tmp/model/checkpoint-3000", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 3.4301, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.1638, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.8374, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.7298, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.6427, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5828, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 1.5323, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 1.5066, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 1.4769, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.45, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 1.4125, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.3836, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 1.3856, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 1.3565, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.3546, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 1.3101, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 1.3605, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 1.3175, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 1.3235, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.3046, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 20.0, | |
| "eval_loss": 1.2609087228775024, | |
| "eval_rouge1": 13.4574, | |
| "eval_rouge2": 9.198, | |
| "eval_rougeL": 13.0881, | |
| "eval_rougeLsum": 13.3368, | |
| "eval_runtime": 191.5141, | |
| "eval_samples_per_second": 10.443, | |
| "eval_steps_per_second": 0.653, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 1.2635, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 1.2522, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 1.2465, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.2389, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 1.2321, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 1.2349, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 1.2442, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 1.2462, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 1.2083, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.2251, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 1.2295, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 1.2295, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.221, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 1.2059, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 1.1994, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 1.19, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 1.2009, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 1.1991, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 1.1965, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 1.1969, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 20.0, | |
| "eval_loss": 1.1954196691513062, | |
| "eval_rouge1": 13.7683, | |
| "eval_rouge2": 9.4853, | |
| "eval_rougeL": 13.3723, | |
| "eval_rougeLsum": 13.6358, | |
| "eval_runtime": 191.2997, | |
| "eval_samples_per_second": 10.455, | |
| "eval_steps_per_second": 0.653, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 1.1336, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.1388, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 1.1374, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 1.1564, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 1.1277, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 1.1429, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 1.1384, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 1.1301, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 1.1333, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 1.1182, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.1341, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 1.1139, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 1.1337, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 1.1241, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 1.1179, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 1.1187, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 1.1222, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 1.1274, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 1.1162, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.1146, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 20.0, | |
| "eval_loss": 1.1659743785858154, | |
| "eval_rouge1": 14.9893, | |
| "eval_rouge2": 10.2707, | |
| "eval_rougeL": 14.4389, | |
| "eval_rougeLsum": 14.7875, | |
| "eval_runtime": 191.6936, | |
| "eval_samples_per_second": 10.433, | |
| "eval_steps_per_second": 0.652, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 3000, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 6499688477884416.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |