| { | |
| "best_metric": 0.7278010101558682, | |
| "best_model_checkpoint": ".\\output\\checkpoint-450", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6e-06, | |
| "loss": 1.8646, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.2666666666666667e-05, | |
| "loss": 1.7999, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9333333333333333e-05, | |
| "loss": 1.6122, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.5558, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.9703703703703707e-05, | |
| "loss": 1.4972, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.8962962962962965e-05, | |
| "loss": 1.482, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.8222222222222223e-05, | |
| "loss": 1.4642, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.7481481481481482e-05, | |
| "loss": 1.4455, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.6740740740740743e-05, | |
| "loss": 1.4228, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.4093, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 2.525925925925926e-05, | |
| "loss": 1.3687, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.451851851851852e-05, | |
| "loss": 1.4491, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.377777777777778e-05, | |
| "loss": 1.4152, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 2.303703703703704e-05, | |
| "loss": 1.4672, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.2296296296296297e-05, | |
| "loss": 1.3758, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7276595744680852, | |
| "eval_loss": 1.2825729846954346, | |
| "eval_runtime": 14.8174, | |
| "eval_samples_per_second": 33.474, | |
| "eval_steps_per_second": 4.184, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 2.155555555555556e-05, | |
| "loss": 1.372, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 2.0814814814814817e-05, | |
| "loss": 1.4247, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.0074074074074075e-05, | |
| "loss": 1.3985, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.9333333333333333e-05, | |
| "loss": 1.3752, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.8592592592592595e-05, | |
| "loss": 1.4129, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.7851851851851853e-05, | |
| "loss": 1.3847, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.7111111111111112e-05, | |
| "loss": 1.3855, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.6370370370370374e-05, | |
| "loss": 1.3916, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.5629629629629632e-05, | |
| "loss": 1.4184, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 1.4888888888888888e-05, | |
| "loss": 1.3827, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.4148148148148148e-05, | |
| "loss": 1.3742, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.3407407407407408e-05, | |
| "loss": 1.4016, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.2666666666666667e-05, | |
| "loss": 1.3888, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.1925925925925927e-05, | |
| "loss": 1.3748, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.1185185185185185e-05, | |
| "loss": 1.3763, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.727228590694538, | |
| "eval_loss": 1.2746729850769043, | |
| "eval_runtime": 14.4425, | |
| "eval_samples_per_second": 34.343, | |
| "eval_steps_per_second": 4.293, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.0444444444444445e-05, | |
| "loss": 1.3781, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 9.703703703703703e-06, | |
| "loss": 1.4019, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.962962962962963e-06, | |
| "loss": 1.3862, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 8.222222222222222e-06, | |
| "loss": 1.3784, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 7.481481481481482e-06, | |
| "loss": 1.3407, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.740740740740741e-06, | |
| "loss": 1.3496, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 6e-06, | |
| "loss": 1.3681, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 5.25925925925926e-06, | |
| "loss": 1.3453, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.518518518518519e-06, | |
| "loss": 1.3294, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3.7777777777777777e-06, | |
| "loss": 1.3361, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 3.0370370370370372e-06, | |
| "loss": 1.3615, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.2962962962962964e-06, | |
| "loss": 1.3104, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 1.5555555555555556e-06, | |
| "loss": 1.3395, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 8.148148148148149e-07, | |
| "loss": 1.3499, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.407407407407407e-08, | |
| "loss": 1.3558, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7278010101558682, | |
| "eval_loss": 1.260701298713684, | |
| "eval_runtime": 14.198, | |
| "eval_samples_per_second": 34.935, | |
| "eval_steps_per_second": 4.367, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 450, | |
| "total_flos": 3789443078682624.0, | |
| "train_loss": 1.4182749218410915, | |
| "train_runtime": 1171.1026, | |
| "train_samples_per_second": 12.291, | |
| "train_steps_per_second": 0.384 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 450, | |
| "num_train_epochs": 3, | |
| "save_steps": 100.0, | |
| "total_flos": 3789443078682624.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |