| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.1103594409128905, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00016, | |
| "loss": 8.4604, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00032, | |
| "loss": 4.2921, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00048, | |
| "loss": 3.3774, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00064, | |
| "loss": 2.9308, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0008, | |
| "loss": 2.641, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0007822222222222222, | |
| "loss": 2.4213, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0007644444444444445, | |
| "loss": 2.2593, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0007466666666666667, | |
| "loss": 2.1346, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0007288888888888889, | |
| "loss": 2.0382, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0007111111111111111, | |
| "loss": 1.955, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0006933333333333333, | |
| "loss": 1.8814, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0006755555555555555, | |
| "loss": 1.8179, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0006577777777777777, | |
| "loss": 1.7623, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00064, | |
| "loss": 1.7135, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0006222222222222223, | |
| "loss": 1.6689, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0006044444444444445, | |
| "loss": 1.6295, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0005866666666666667, | |
| "loss": 1.5857, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.000568888888888889, | |
| "loss": 1.5541, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0005511111111111112, | |
| "loss": 1.5221, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0005333333333333334, | |
| "loss": 1.4947, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0005155555555555557, | |
| "loss": 1.4683, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0004977777777777778, | |
| "loss": 1.4415, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00048, | |
| "loss": 1.4183, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.0004622222222222222, | |
| "loss": 1.3975, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00044444444444444447, | |
| "loss": 1.3756, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_bleu": 2.3992, | |
| "eval_gen_len": 18.9403, | |
| "eval_loss": 1.535530686378479, | |
| "eval_runtime": 7626.9892, | |
| "eval_samples_per_second": 248.341, | |
| "eval_steps_per_second": 1.94, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00042666666666666667, | |
| "loss": 1.358, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00040888888888888887, | |
| "loss": 1.339, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.0003911111111111111, | |
| "loss": 1.3213, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.0003733333333333334, | |
| "loss": 1.3068, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00035555555555555557, | |
| "loss": 1.2927, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.00033777777777777777, | |
| "loss": 1.279, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00032, | |
| "loss": 1.2641, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0003022222222222222, | |
| "loss": 1.2455, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.0002844444444444445, | |
| "loss": 1.2329, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0002666666666666667, | |
| "loss": 1.2239, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0002488888888888889, | |
| "loss": 1.2123, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.0002311111111111111, | |
| "loss": 1.2036, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.00021333333333333333, | |
| "loss": 1.1937, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.00019555555555555556, | |
| "loss": 1.1851, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 1.1772, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.00016, | |
| "loss": 1.1708, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.00014222222222222224, | |
| "loss": 1.1634, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.00012444444444444444, | |
| "loss": 1.158, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 1.1541, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 1.1494, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 7.111111111111112e-05, | |
| "loss": 1.1435, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 1.1406, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 1.1375, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 1.1315, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0, | |
| "loss": 1.1295, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "eval_bleu": 2.543, | |
| "eval_gen_len": 18.9389, | |
| "eval_loss": 1.5099098682403564, | |
| "eval_runtime": 7653.3773, | |
| "eval_samples_per_second": 247.485, | |
| "eval_steps_per_second": 1.934, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "step": 10000, | |
| "total_flos": 1.2278131895086088e+19, | |
| "train_loss": 1.7310921798706054, | |
| "train_runtime": 38945.4898, | |
| "train_samples_per_second": 1051.726, | |
| "train_steps_per_second": 0.257 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.2278131895086088e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |