| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 100, | |
| "global_step": 450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019555555555555556, | |
| "loss": 0.0098, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00019111111111111114, | |
| "loss": 0.0428, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0001866666666666667, | |
| "loss": 0.0188, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00018222222222222224, | |
| "loss": 0.0763, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.0098, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00017333333333333334, | |
| "loss": 0.0312, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00016888888888888889, | |
| "loss": 0.0579, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00016444444444444444, | |
| "loss": 0.024, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00016, | |
| "loss": 0.086, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.0009, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.8712962962962963, | |
| "eval_loss": 0.7067815661430359, | |
| "eval_runtime": 13.5776, | |
| "eval_samples_per_second": 79.543, | |
| "eval_steps_per_second": 9.943, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0001511111111111111, | |
| "loss": 0.0317, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00014666666666666666, | |
| "loss": 0.0388, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00014222222222222224, | |
| "loss": 0.0213, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0001377777777777778, | |
| "loss": 0.0258, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.0206, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00012888888888888892, | |
| "loss": 0.0481, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00012444444444444444, | |
| "loss": 0.0115, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00012, | |
| "loss": 0.015, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.00011555555555555555, | |
| "loss": 0.0517, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.0039, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 0.899074074074074, | |
| "eval_loss": 0.6390542387962341, | |
| "eval_runtime": 12.7742, | |
| "eval_samples_per_second": 84.545, | |
| "eval_steps_per_second": 10.568, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 0.0066, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.00010222222222222222, | |
| "loss": 0.002, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 9.777777777777778e-05, | |
| "loss": 0.001, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.0042, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.004, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 8.444444444444444e-05, | |
| "loss": 0.0018, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0019, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 7.555555555555556e-05, | |
| "loss": 0.0002, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 7.111111111111112e-05, | |
| "loss": 0.0002, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0014, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_accuracy": 0.9055555555555556, | |
| "eval_loss": 0.5535524487495422, | |
| "eval_runtime": 12.7035, | |
| "eval_samples_per_second": 85.016, | |
| "eval_steps_per_second": 10.627, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 6.222222222222222e-05, | |
| "loss": 0.0037, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 5.7777777777777776e-05, | |
| "loss": 0.0017, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.0033, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 0.0002, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0043, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0002, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.0002, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.0017, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.0002, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0018, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "eval_accuracy": 0.9083333333333333, | |
| "eval_loss": 0.5527949333190918, | |
| "eval_runtime": 12.8229, | |
| "eval_samples_per_second": 84.224, | |
| "eval_steps_per_second": 10.528, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.0002, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.002, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 0.0044, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 0.0032, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0002, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 450, | |
| "total_flos": 1.6739319895474176e+18, | |
| "train_loss": 0.015028678889438096, | |
| "train_runtime": 430.7019, | |
| "train_samples_per_second": 50.151, | |
| "train_steps_per_second": 1.045 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 450, | |
| "num_train_epochs": 5, | |
| "save_steps": 1000, | |
| "total_flos": 1.6739319895474176e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |