| { | |
| "best_metric": 0.8666666666666667, | |
| "best_model_checkpoint": "vit-base-patch16-224-dmae-va-U5-100-iN\\checkpoint-85", | |
| "epoch": 90.3225806451613, | |
| "eval_steps": 500, | |
| "global_step": 700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.3812448978424072, | |
| "eval_runtime": 0.9939, | |
| "eval_samples_per_second": 60.369, | |
| "eval_steps_per_second": 2.012, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 1.3848, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.3605996370315552, | |
| "eval_runtime": 1.0777, | |
| "eval_samples_per_second": 55.675, | |
| "eval_steps_per_second": 1.856, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.3686, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 1.3074811697006226, | |
| "eval_runtime": 1.0069, | |
| "eval_samples_per_second": 59.587, | |
| "eval_steps_per_second": 1.986, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 1.2965, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.2370479106903076, | |
| "eval_runtime": 1.023, | |
| "eval_samples_per_second": 58.649, | |
| "eval_steps_per_second": 1.955, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 1.1167830228805542, | |
| "eval_runtime": 1.0537, | |
| "eval_samples_per_second": 56.944, | |
| "eval_steps_per_second": 1.898, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.9624060150375936e-05, | |
| "loss": 1.1753, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 1.0309817790985107, | |
| "eval_runtime": 0.9955, | |
| "eval_samples_per_second": 60.27, | |
| "eval_steps_per_second": 2.009, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 4.887218045112782e-05, | |
| "loss": 1.0294, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.9315768480300903, | |
| "eval_runtime": 1.0112, | |
| "eval_samples_per_second": 59.337, | |
| "eval_steps_per_second": 1.978, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 0.902, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6833333333333333, | |
| "eval_loss": 0.8727719187736511, | |
| "eval_runtime": 1.0259, | |
| "eval_samples_per_second": 58.484, | |
| "eval_steps_per_second": 1.949, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.7666666666666667, | |
| "eval_loss": 0.8128864765167236, | |
| "eval_runtime": 1.0791, | |
| "eval_samples_per_second": 55.601, | |
| "eval_steps_per_second": 1.853, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.736842105263158e-05, | |
| "loss": 0.7812, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.7006206512451172, | |
| "eval_runtime": 1.0876, | |
| "eval_samples_per_second": 55.167, | |
| "eval_steps_per_second": 1.839, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 4.6616541353383456e-05, | |
| "loss": 0.6419, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.8666666666666667, | |
| "eval_loss": 0.6381439566612244, | |
| "eval_runtime": 1.0437, | |
| "eval_samples_per_second": 57.486, | |
| "eval_steps_per_second": 1.916, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 4.586466165413534e-05, | |
| "loss": 0.5109, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.6327125430107117, | |
| "eval_runtime": 1.0397, | |
| "eval_samples_per_second": 57.707, | |
| "eval_steps_per_second": 1.924, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 4.511278195488722e-05, | |
| "loss": 0.3838, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.8666666666666667, | |
| "eval_loss": 0.5441552996635437, | |
| "eval_runtime": 1.0509, | |
| "eval_samples_per_second": 57.092, | |
| "eval_steps_per_second": 1.903, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6755495667457581, | |
| "eval_runtime": 1.041, | |
| "eval_samples_per_second": 57.635, | |
| "eval_steps_per_second": 1.921, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 0.285, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.7755728960037231, | |
| "eval_runtime": 1.0199, | |
| "eval_samples_per_second": 58.828, | |
| "eval_steps_per_second": 1.961, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 4.3609022556390975e-05, | |
| "loss": 0.2672, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.8106668591499329, | |
| "eval_runtime": 1.0429, | |
| "eval_samples_per_second": 57.531, | |
| "eval_steps_per_second": 1.918, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 0.2466, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.5219495892524719, | |
| "eval_runtime": 1.0179, | |
| "eval_samples_per_second": 58.943, | |
| "eval_steps_per_second": 1.965, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.7040973901748657, | |
| "eval_runtime": 1.0682, | |
| "eval_samples_per_second": 56.169, | |
| "eval_steps_per_second": 1.872, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 0.2312, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7878958582878113, | |
| "eval_runtime": 0.9838, | |
| "eval_samples_per_second": 60.989, | |
| "eval_steps_per_second": 2.033, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 4.135338345864662e-05, | |
| "loss": 0.1933, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.709021806716919, | |
| "eval_runtime": 1.0518, | |
| "eval_samples_per_second": 57.046, | |
| "eval_steps_per_second": 1.902, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 0.1692, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.5394979119300842, | |
| "eval_runtime": 1.0183, | |
| "eval_samples_per_second": 58.921, | |
| "eval_steps_per_second": 1.964, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 3.9849624060150376e-05, | |
| "loss": 0.1578, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.6418856382369995, | |
| "eval_runtime": 1.0177, | |
| "eval_samples_per_second": 58.959, | |
| "eval_steps_per_second": 1.965, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.5736449956893921, | |
| "eval_runtime": 1.0279, | |
| "eval_samples_per_second": 58.369, | |
| "eval_steps_per_second": 1.946, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 3.909774436090226e-05, | |
| "loss": 0.1321, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.747129499912262, | |
| "eval_runtime": 1.021, | |
| "eval_samples_per_second": 58.768, | |
| "eval_steps_per_second": 1.959, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 3.834586466165413e-05, | |
| "loss": 0.1114, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.7666666666666667, | |
| "eval_loss": 0.6447189450263977, | |
| "eval_runtime": 1.0464, | |
| "eval_samples_per_second": 57.341, | |
| "eval_steps_per_second": 1.911, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 3.759398496240601e-05, | |
| "loss": 0.1385, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.61579829454422, | |
| "eval_runtime": 0.9804, | |
| "eval_samples_per_second": 61.197, | |
| "eval_steps_per_second": 2.04, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.6467480063438416, | |
| "eval_runtime": 1.0425, | |
| "eval_samples_per_second": 57.556, | |
| "eval_steps_per_second": 1.919, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 0.1136, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.6180290579795837, | |
| "eval_runtime": 1.0892, | |
| "eval_samples_per_second": 55.084, | |
| "eval_steps_per_second": 1.836, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 3.6090225563909776e-05, | |
| "loss": 0.0997, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.8578386902809143, | |
| "eval_runtime": 0.9851, | |
| "eval_samples_per_second": 60.908, | |
| "eval_steps_per_second": 2.03, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 3.533834586466165e-05, | |
| "loss": 0.1064, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.6778053641319275, | |
| "eval_runtime": 0.9651, | |
| "eval_samples_per_second": 62.17, | |
| "eval_steps_per_second": 2.072, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 3.458646616541353e-05, | |
| "loss": 0.0775, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8123681545257568, | |
| "eval_runtime": 0.9886, | |
| "eval_samples_per_second": 60.693, | |
| "eval_steps_per_second": 2.023, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.7782633900642395, | |
| "eval_runtime": 1.0238, | |
| "eval_samples_per_second": 58.605, | |
| "eval_steps_per_second": 1.953, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 3.3834586466165414e-05, | |
| "loss": 0.0921, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.8320418000221252, | |
| "eval_runtime": 1.0125, | |
| "eval_samples_per_second": 59.258, | |
| "eval_steps_per_second": 1.975, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 0.0919, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.8309503197669983, | |
| "eval_runtime": 0.9631, | |
| "eval_samples_per_second": 62.3, | |
| "eval_steps_per_second": 2.077, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 3.233082706766917e-05, | |
| "loss": 0.0888, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.6575559377670288, | |
| "eval_runtime": 0.9897, | |
| "eval_samples_per_second": 60.621, | |
| "eval_steps_per_second": 2.021, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.7043873071670532, | |
| "eval_runtime": 0.9919, | |
| "eval_samples_per_second": 60.489, | |
| "eval_steps_per_second": 2.016, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 3.157894736842105e-05, | |
| "loss": 0.0693, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.9, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.7607947587966919, | |
| "eval_runtime": 0.9426, | |
| "eval_samples_per_second": 63.653, | |
| "eval_steps_per_second": 2.122, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 37.42, | |
| "learning_rate": 3.082706766917293e-05, | |
| "loss": 0.061, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.7802255153656006, | |
| "eval_runtime": 0.9812, | |
| "eval_samples_per_second": 61.149, | |
| "eval_steps_per_second": 2.038, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "learning_rate": 3.007518796992481e-05, | |
| "loss": 0.0699, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 38.97, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.7761934995651245, | |
| "eval_runtime": 0.9881, | |
| "eval_samples_per_second": 60.722, | |
| "eval_steps_per_second": 2.024, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 0.0652, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.7579043507575989, | |
| "eval_runtime": 1.0374, | |
| "eval_samples_per_second": 57.837, | |
| "eval_steps_per_second": 1.928, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 40.9, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.998461902141571, | |
| "eval_runtime": 0.9702, | |
| "eval_samples_per_second": 61.842, | |
| "eval_steps_per_second": 2.061, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 41.29, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.0562, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 41.94, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.8027188777923584, | |
| "eval_runtime": 1.0453, | |
| "eval_samples_per_second": 57.397, | |
| "eval_steps_per_second": 1.913, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 42.58, | |
| "learning_rate": 2.7819548872180452e-05, | |
| "loss": 0.0534, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 42.97, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9704654812812805, | |
| "eval_runtime": 1.0187, | |
| "eval_samples_per_second": 58.899, | |
| "eval_steps_per_second": 1.963, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 43.87, | |
| "learning_rate": 2.706766917293233e-05, | |
| "loss": 0.0519, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.7301105260848999, | |
| "eval_runtime": 0.998, | |
| "eval_samples_per_second": 60.123, | |
| "eval_steps_per_second": 2.004, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 44.9, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8432899713516235, | |
| "eval_runtime": 1.0546, | |
| "eval_samples_per_second": 56.895, | |
| "eval_steps_per_second": 1.897, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.0529, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 45.94, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8533557653427124, | |
| "eval_runtime": 1.0201, | |
| "eval_samples_per_second": 58.815, | |
| "eval_steps_per_second": 1.961, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.0772, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 46.97, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8561839461326599, | |
| "eval_runtime": 1.0725, | |
| "eval_samples_per_second": 55.942, | |
| "eval_steps_per_second": 1.865, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 47.74, | |
| "learning_rate": 2.4812030075187968e-05, | |
| "loss": 0.0644, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8419451117515564, | |
| "eval_runtime": 1.0154, | |
| "eval_samples_per_second": 59.091, | |
| "eval_steps_per_second": 1.97, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 48.9, | |
| "eval_accuracy": 0.7666666666666667, | |
| "eval_loss": 1.125100016593933, | |
| "eval_runtime": 1.3342, | |
| "eval_samples_per_second": 44.969, | |
| "eval_steps_per_second": 1.499, | |
| "step": 379 | |
| }, | |
| { | |
| "epoch": 49.03, | |
| "learning_rate": 2.406015037593985e-05, | |
| "loss": 0.0467, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 49.94, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.7536706924438477, | |
| "eval_runtime": 1.6625, | |
| "eval_samples_per_second": 36.09, | |
| "eval_steps_per_second": 1.203, | |
| "step": 387 | |
| }, | |
| { | |
| "epoch": 50.32, | |
| "learning_rate": 2.3308270676691728e-05, | |
| "loss": 0.0576, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 50.97, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.7517222166061401, | |
| "eval_runtime": 1.0574, | |
| "eval_samples_per_second": 56.741, | |
| "eval_steps_per_second": 1.891, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "learning_rate": 2.255639097744361e-05, | |
| "loss": 0.0344, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8343304395675659, | |
| "eval_runtime": 1.0509, | |
| "eval_samples_per_second": 57.096, | |
| "eval_steps_per_second": 1.903, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 52.9, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.0663, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 52.9, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.7636159062385559, | |
| "eval_runtime": 0.952, | |
| "eval_samples_per_second": 63.026, | |
| "eval_steps_per_second": 2.101, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 53.94, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.8253324031829834, | |
| "eval_runtime": 0.9847, | |
| "eval_samples_per_second": 60.932, | |
| "eval_steps_per_second": 2.031, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 54.19, | |
| "learning_rate": 2.105263157894737e-05, | |
| "loss": 0.0353, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 54.97, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.934796154499054, | |
| "eval_runtime": 1.0691, | |
| "eval_samples_per_second": 56.12, | |
| "eval_steps_per_second": 1.871, | |
| "step": 426 | |
| }, | |
| { | |
| "epoch": 55.48, | |
| "learning_rate": 2.0300751879699247e-05, | |
| "loss": 0.0524, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.8216551542282104, | |
| "eval_runtime": 0.9842, | |
| "eval_samples_per_second": 60.966, | |
| "eval_steps_per_second": 2.032, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 56.77, | |
| "learning_rate": 1.954887218045113e-05, | |
| "loss": 0.0479, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 56.9, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.758594810962677, | |
| "eval_runtime": 1.0267, | |
| "eval_samples_per_second": 58.439, | |
| "eval_steps_per_second": 1.948, | |
| "step": 441 | |
| }, | |
| { | |
| "epoch": 57.94, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8146753311157227, | |
| "eval_runtime": 0.965, | |
| "eval_samples_per_second": 62.179, | |
| "eval_steps_per_second": 2.073, | |
| "step": 449 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "learning_rate": 1.8796992481203007e-05, | |
| "loss": 0.0595, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 58.97, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9999889135360718, | |
| "eval_runtime": 1.0193, | |
| "eval_samples_per_second": 58.862, | |
| "eval_steps_per_second": 1.962, | |
| "step": 457 | |
| }, | |
| { | |
| "epoch": 59.35, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.0475, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9291054010391235, | |
| "eval_runtime": 0.9846, | |
| "eval_samples_per_second": 60.941, | |
| "eval_steps_per_second": 2.031, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 60.65, | |
| "learning_rate": 1.7293233082706766e-05, | |
| "loss": 0.049, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 60.9, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9588444828987122, | |
| "eval_runtime": 1.0072, | |
| "eval_samples_per_second": 59.57, | |
| "eval_steps_per_second": 1.986, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 61.94, | |
| "learning_rate": 1.6541353383458648e-05, | |
| "loss": 0.0398, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 61.94, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9500763416290283, | |
| "eval_runtime": 1.1967, | |
| "eval_samples_per_second": 50.137, | |
| "eval_steps_per_second": 1.671, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 62.97, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9498623013496399, | |
| "eval_runtime": 1.0197, | |
| "eval_samples_per_second": 58.839, | |
| "eval_steps_per_second": 1.961, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 63.23, | |
| "learning_rate": 1.5789473684210526e-05, | |
| "loss": 0.0496, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9279070496559143, | |
| "eval_runtime": 1.006, | |
| "eval_samples_per_second": 59.643, | |
| "eval_steps_per_second": 1.988, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 64.52, | |
| "learning_rate": 1.5037593984962406e-05, | |
| "loss": 0.0354, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 64.9, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.9677076935768127, | |
| "eval_runtime": 0.9989, | |
| "eval_samples_per_second": 60.065, | |
| "eval_steps_per_second": 2.002, | |
| "step": 503 | |
| }, | |
| { | |
| "epoch": 65.81, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.0325, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 65.94, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.8371397852897644, | |
| "eval_runtime": 0.9656, | |
| "eval_samples_per_second": 62.134, | |
| "eval_steps_per_second": 2.071, | |
| "step": 511 | |
| }, | |
| { | |
| "epoch": 66.97, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9682708978652954, | |
| "eval_runtime": 0.956, | |
| "eval_samples_per_second": 62.76, | |
| "eval_steps_per_second": 2.092, | |
| "step": 519 | |
| }, | |
| { | |
| "epoch": 67.1, | |
| "learning_rate": 1.3533834586466165e-05, | |
| "loss": 0.0335, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 1.0455142259597778, | |
| "eval_runtime": 0.9995, | |
| "eval_samples_per_second": 60.033, | |
| "eval_steps_per_second": 2.001, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 68.39, | |
| "learning_rate": 1.2781954887218045e-05, | |
| "loss": 0.0375, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 68.9, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.9026990532875061, | |
| "eval_runtime": 1.016, | |
| "eval_samples_per_second": 59.058, | |
| "eval_steps_per_second": 1.969, | |
| "step": 534 | |
| }, | |
| { | |
| "epoch": 69.68, | |
| "learning_rate": 1.2030075187969925e-05, | |
| "loss": 0.0424, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 69.94, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.8043019771575928, | |
| "eval_runtime": 0.9715, | |
| "eval_samples_per_second": 61.76, | |
| "eval_steps_per_second": 2.059, | |
| "step": 542 | |
| }, | |
| { | |
| "epoch": 70.97, | |
| "learning_rate": 1.1278195488721805e-05, | |
| "loss": 0.0383, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 70.97, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9034760594367981, | |
| "eval_runtime": 0.9985, | |
| "eval_samples_per_second": 60.088, | |
| "eval_steps_per_second": 2.003, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9360395073890686, | |
| "eval_runtime": 1.0923, | |
| "eval_samples_per_second": 54.931, | |
| "eval_steps_per_second": 1.831, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 72.26, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.0295, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 72.9, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9840652346611023, | |
| "eval_runtime": 1.1084, | |
| "eval_samples_per_second": 54.132, | |
| "eval_steps_per_second": 1.804, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 73.55, | |
| "learning_rate": 9.774436090225564e-06, | |
| "loss": 0.0307, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 73.94, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9300164580345154, | |
| "eval_runtime": 1.152, | |
| "eval_samples_per_second": 52.082, | |
| "eval_steps_per_second": 1.736, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 74.84, | |
| "learning_rate": 9.022556390977444e-06, | |
| "loss": 0.0376, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 74.97, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9629533290863037, | |
| "eval_runtime": 1.3504, | |
| "eval_samples_per_second": 44.432, | |
| "eval_steps_per_second": 1.481, | |
| "step": 581 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.977742612361908, | |
| "eval_runtime": 1.044, | |
| "eval_samples_per_second": 57.469, | |
| "eval_steps_per_second": 1.916, | |
| "step": 589 | |
| }, | |
| { | |
| "epoch": 76.13, | |
| "learning_rate": 8.270676691729324e-06, | |
| "loss": 0.0259, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 76.9, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9322593808174133, | |
| "eval_runtime": 1.0858, | |
| "eval_samples_per_second": 55.256, | |
| "eval_steps_per_second": 1.842, | |
| "step": 596 | |
| }, | |
| { | |
| "epoch": 77.42, | |
| "learning_rate": 7.518796992481203e-06, | |
| "loss": 0.0345, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 77.94, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9075172543525696, | |
| "eval_runtime": 1.1281, | |
| "eval_samples_per_second": 53.185, | |
| "eval_steps_per_second": 1.773, | |
| "step": 604 | |
| }, | |
| { | |
| "epoch": 78.71, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.0346, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 78.97, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8950561881065369, | |
| "eval_runtime": 0.9932, | |
| "eval_samples_per_second": 60.412, | |
| "eval_steps_per_second": 2.014, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 6.015037593984962e-06, | |
| "loss": 0.0319, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9676039814949036, | |
| "eval_runtime": 0.9844, | |
| "eval_samples_per_second": 60.95, | |
| "eval_steps_per_second": 2.032, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 80.9, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.988412618637085, | |
| "eval_runtime": 0.9973, | |
| "eval_samples_per_second": 60.161, | |
| "eval_steps_per_second": 2.005, | |
| "step": 627 | |
| }, | |
| { | |
| "epoch": 81.29, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 0.0226, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 81.94, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9850680232048035, | |
| "eval_runtime": 1.0518, | |
| "eval_samples_per_second": 57.046, | |
| "eval_steps_per_second": 1.902, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 82.58, | |
| "learning_rate": 4.511278195488722e-06, | |
| "loss": 0.033, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 82.97, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9710198640823364, | |
| "eval_runtime": 1.0106, | |
| "eval_samples_per_second": 59.371, | |
| "eval_steps_per_second": 1.979, | |
| "step": 643 | |
| }, | |
| { | |
| "epoch": 83.87, | |
| "learning_rate": 3.7593984962406014e-06, | |
| "loss": 0.0262, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.985089123249054, | |
| "eval_runtime": 1.0015, | |
| "eval_samples_per_second": 59.912, | |
| "eval_steps_per_second": 1.997, | |
| "step": 651 | |
| }, | |
| { | |
| "epoch": 84.9, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9868298172950745, | |
| "eval_runtime": 1.0224, | |
| "eval_samples_per_second": 58.683, | |
| "eval_steps_per_second": 1.956, | |
| "step": 658 | |
| }, | |
| { | |
| "epoch": 85.16, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.0345, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 85.94, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9702271819114685, | |
| "eval_runtime": 1.0333, | |
| "eval_samples_per_second": 58.065, | |
| "eval_steps_per_second": 1.936, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 86.45, | |
| "learning_rate": 2.255639097744361e-06, | |
| "loss": 0.0299, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 86.97, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9888620376586914, | |
| "eval_runtime": 0.9649, | |
| "eval_samples_per_second": 62.184, | |
| "eval_steps_per_second": 2.073, | |
| "step": 674 | |
| }, | |
| { | |
| "epoch": 87.74, | |
| "learning_rate": 1.5037593984962406e-06, | |
| "loss": 0.0347, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 1.000300645828247, | |
| "eval_runtime": 1.0185, | |
| "eval_samples_per_second": 58.91, | |
| "eval_steps_per_second": 1.964, | |
| "step": 682 | |
| }, | |
| { | |
| "epoch": 88.9, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9913238286972046, | |
| "eval_runtime": 1.0979, | |
| "eval_samples_per_second": 54.647, | |
| "eval_steps_per_second": 1.822, | |
| "step": 689 | |
| }, | |
| { | |
| "epoch": 89.03, | |
| "learning_rate": 7.518796992481203e-07, | |
| "loss": 0.0288, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 89.94, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9858866333961487, | |
| "eval_runtime": 1.063, | |
| "eval_samples_per_second": 56.445, | |
| "eval_steps_per_second": 1.882, | |
| "step": 697 | |
| }, | |
| { | |
| "epoch": 90.32, | |
| "learning_rate": 0.0, | |
| "loss": 0.0198, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 90.32, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.9857978820800781, | |
| "eval_runtime": 1.01, | |
| "eval_samples_per_second": 59.406, | |
| "eval_steps_per_second": 1.98, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 90.32, | |
| "step": 700, | |
| "total_flos": 6.817867456880148e+18, | |
| "train_loss": 0.19939448043704033, | |
| "train_runtime": 1530.1698, | |
| "train_samples_per_second": 63.653, | |
| "train_steps_per_second": 0.457 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 6.817867456880148e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |