| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 38.2043935052531, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00016, | |
| "loss": 9.8321, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00032, | |
| "loss": 4.0687, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.00048, | |
| "loss": 2.9707, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.00064, | |
| "loss": 2.5203, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0008, | |
| "loss": 2.2282, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.0007822222222222222, | |
| "loss": 2.0162, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 0.0007644444444444445, | |
| "loss": 1.8634, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.0007466666666666667, | |
| "loss": 1.7508, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.0007288888888888889, | |
| "loss": 1.6579, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.0007111111111111111, | |
| "loss": 1.5774, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "eval_bleu": 3.3312, | |
| "eval_gen_len": 18.8762, | |
| "eval_loss": 1.546088695526123, | |
| "eval_runtime": 341.773, | |
| "eval_samples_per_second": 223.804, | |
| "eval_steps_per_second": 1.75, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.0006933333333333333, | |
| "loss": 1.5111, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 0.0006755555555555555, | |
| "loss": 1.4553, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 0.0006577777777777777, | |
| "loss": 1.4026, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 0.00064, | |
| "loss": 1.3509, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "learning_rate": 0.0006222222222222223, | |
| "loss": 1.3105, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 0.0006044444444444445, | |
| "loss": 1.2714, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 0.0005866666666666667, | |
| "loss": 1.2358, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 0.000568888888888889, | |
| "loss": 1.1962, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 0.0005511111111111112, | |
| "loss": 1.1651, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.28, | |
| "learning_rate": 0.0005333333333333334, | |
| "loss": 1.137, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.28, | |
| "eval_bleu": 3.8148, | |
| "eval_gen_len": 18.8755, | |
| "eval_loss": 1.4426332712173462, | |
| "eval_runtime": 303.4706, | |
| "eval_samples_per_second": 252.051, | |
| "eval_steps_per_second": 1.971, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.05, | |
| "learning_rate": 0.0005155555555555557, | |
| "loss": 1.1102, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.81, | |
| "learning_rate": 0.0004977777777777778, | |
| "loss": 1.0793, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.57, | |
| "learning_rate": 0.00048, | |
| "loss": 1.053, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 0.0004622222222222222, | |
| "loss": 1.0295, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 0.00044444444444444447, | |
| "loss": 1.01, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 19.87, | |
| "learning_rate": 0.00042666666666666667, | |
| "loss": 0.986, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.63, | |
| "learning_rate": 0.00040888888888888887, | |
| "loss": 0.9645, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 21.39, | |
| "learning_rate": 0.0003911111111111111, | |
| "loss": 0.9447, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 22.16, | |
| "learning_rate": 0.0003733333333333334, | |
| "loss": 0.9289, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 22.92, | |
| "learning_rate": 0.00035555555555555557, | |
| "loss": 0.9109, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.92, | |
| "eval_bleu": 3.9571, | |
| "eval_gen_len": 18.8752, | |
| "eval_loss": 1.475421667098999, | |
| "eval_runtime": 306.5307, | |
| "eval_samples_per_second": 249.535, | |
| "eval_steps_per_second": 1.951, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 23.69, | |
| "learning_rate": 0.00033777777777777777, | |
| "loss": 0.891, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 24.45, | |
| "learning_rate": 0.00032, | |
| "loss": 0.8772, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 25.21, | |
| "learning_rate": 0.0003022222222222222, | |
| "loss": 0.8642, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 25.98, | |
| "learning_rate": 0.0002844444444444445, | |
| "loss": 0.8509, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 26.74, | |
| "learning_rate": 0.0002666666666666667, | |
| "loss": 0.833, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 27.51, | |
| "learning_rate": 0.0002488888888888889, | |
| "loss": 0.8229, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 28.27, | |
| "learning_rate": 0.0002311111111111111, | |
| "loss": 0.8119, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 29.04, | |
| "learning_rate": 0.00021333333333333333, | |
| "loss": 0.8025, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 29.8, | |
| "learning_rate": 0.00019555555555555556, | |
| "loss": 0.7901, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.7807, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "eval_bleu": 3.9767, | |
| "eval_gen_len": 18.8761, | |
| "eval_loss": 1.5373053550720215, | |
| "eval_runtime": 305.2497, | |
| "eval_samples_per_second": 250.582, | |
| "eval_steps_per_second": 1.959, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 31.33, | |
| "learning_rate": 0.00016, | |
| "loss": 0.773, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 32.09, | |
| "learning_rate": 0.00014222222222222224, | |
| "loss": 0.7664, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 32.86, | |
| "learning_rate": 0.00012444444444444444, | |
| "loss": 0.7577, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 33.62, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 0.7505, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 34.38, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.7456, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 35.15, | |
| "learning_rate": 7.111111111111112e-05, | |
| "loss": 0.7414, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 35.91, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.7363, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 36.68, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.7324, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 37.44, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.7301, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 38.2, | |
| "learning_rate": 0.0, | |
| "loss": 0.7288, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 38.2, | |
| "eval_bleu": 3.9838, | |
| "eval_gen_len": 18.8778, | |
| "eval_loss": 1.5656607151031494, | |
| "eval_runtime": 304.6836, | |
| "eval_samples_per_second": 251.047, | |
| "eval_steps_per_second": 1.963, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 38.2, | |
| "step": 5000, | |
| "total_flos": 6.138296014525694e+18, | |
| "train_loss": 1.3865033660888673, | |
| "train_runtime": 14019.5814, | |
| "train_samples_per_second": 1460.814, | |
| "train_steps_per_second": 0.357 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 39, | |
| "total_flos": 6.138296014525694e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |