| { | |
| "best_metric": 0.1326947957277298, | |
| "best_model_checkpoint": "./vit-base-noise-classifier-merge\\checkpoint-500", | |
| "epoch": 10.0, | |
| "eval_steps": 100, | |
| "global_step": 1010, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.900990099009901e-06, | |
| "loss": 0.6809, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.801980198019802e-06, | |
| "loss": 0.6748, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.702970297029704e-06, | |
| "loss": 0.6559, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.603960396039604e-06, | |
| "loss": 0.6097, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.504950495049505e-06, | |
| "loss": 0.6009, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 9.405940594059405e-06, | |
| "loss": 0.5522, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 9.306930693069308e-06, | |
| "loss": 0.5324, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 9.20792079207921e-06, | |
| "loss": 0.4652, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 9.10891089108911e-06, | |
| "loss": 0.4078, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 9.009900990099011e-06, | |
| "loss": 0.3836, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.9213483146067416, | |
| "eval_loss": 0.35515740513801575, | |
| "eval_runtime": 7.2882, | |
| "eval_samples_per_second": 24.423, | |
| "eval_steps_per_second": 3.156, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.910891089108911e-06, | |
| "loss": 0.3241, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 8.811881188118812e-06, | |
| "loss": 0.3047, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 8.712871287128714e-06, | |
| "loss": 0.2842, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 8.613861386138615e-06, | |
| "loss": 0.2481, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 8.514851485148515e-06, | |
| "loss": 0.2672, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 8.415841584158416e-06, | |
| "loss": 0.2617, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.316831683168318e-06, | |
| "loss": 0.1862, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 8.217821782178218e-06, | |
| "loss": 0.1962, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 8.11881188118812e-06, | |
| "loss": 0.1716, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 8.019801980198021e-06, | |
| "loss": 0.1923, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_accuracy": 0.9325842696629213, | |
| "eval_loss": 0.19820977747440338, | |
| "eval_runtime": 7.3054, | |
| "eval_samples_per_second": 24.365, | |
| "eval_steps_per_second": 3.148, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 7.920792079207921e-06, | |
| "loss": 0.201, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 7.821782178217822e-06, | |
| "loss": 0.1257, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 7.722772277227724e-06, | |
| "loss": 0.1639, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 7.6237623762376246e-06, | |
| "loss": 0.094, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 7.524752475247525e-06, | |
| "loss": 0.1489, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 7.425742574257426e-06, | |
| "loss": 0.1405, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 7.326732673267327e-06, | |
| "loss": 0.1139, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 7.227722772277228e-06, | |
| "loss": 0.1405, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 7.128712871287129e-06, | |
| "loss": 0.1273, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 7.02970297029703e-06, | |
| "loss": 0.0953, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.9606741573033708, | |
| "eval_loss": 0.13618314266204834, | |
| "eval_runtime": 7.2985, | |
| "eval_samples_per_second": 24.389, | |
| "eval_steps_per_second": 3.151, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 6.930693069306931e-06, | |
| "loss": 0.0708, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 6.831683168316833e-06, | |
| "loss": 0.1164, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 6.732673267326733e-06, | |
| "loss": 0.1114, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 6.633663366336635e-06, | |
| "loss": 0.1032, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 6.534653465346535e-06, | |
| "loss": 0.0625, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 6.4356435643564364e-06, | |
| "loss": 0.0401, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 6.336633663366337e-06, | |
| "loss": 0.0829, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 6.237623762376238e-06, | |
| "loss": 0.1053, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 6.138613861386139e-06, | |
| "loss": 0.0642, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 6.03960396039604e-06, | |
| "loss": 0.0622, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "eval_accuracy": 0.9662921348314607, | |
| "eval_loss": 0.14083810150623322, | |
| "eval_runtime": 8.2574, | |
| "eval_samples_per_second": 21.556, | |
| "eval_steps_per_second": 2.785, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 5.940594059405941e-06, | |
| "loss": 0.0481, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 5.841584158415842e-06, | |
| "loss": 0.0648, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 5.7425742574257425e-06, | |
| "loss": 0.049, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 5.643564356435644e-06, | |
| "loss": 0.0459, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 5.544554455445545e-06, | |
| "loss": 0.0986, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 5.4455445544554465e-06, | |
| "loss": 0.0899, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 5.346534653465347e-06, | |
| "loss": 0.0484, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 5.247524752475248e-06, | |
| "loss": 0.0699, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 5.148514851485149e-06, | |
| "loss": 0.0248, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 5.04950495049505e-06, | |
| "loss": 0.0392, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_accuracy": 0.9719101123595506, | |
| "eval_loss": 0.1326947957277298, | |
| "eval_runtime": 8.3115, | |
| "eval_samples_per_second": 21.416, | |
| "eval_steps_per_second": 2.767, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 4.950495049504951e-06, | |
| "loss": 0.0226, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 4.851485148514852e-06, | |
| "loss": 0.064, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 4.7524752475247525e-06, | |
| "loss": 0.0675, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 4.653465346534654e-06, | |
| "loss": 0.0452, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 4.554455445544555e-06, | |
| "loss": 0.0365, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 4.455445544554456e-06, | |
| "loss": 0.0394, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 4.356435643564357e-06, | |
| "loss": 0.0272, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 4.2574257425742575e-06, | |
| "loss": 0.0672, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 4.158415841584159e-06, | |
| "loss": 0.0404, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 4.05940594059406e-06, | |
| "loss": 0.0197, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.9719101123595506, | |
| "eval_loss": 0.13345670700073242, | |
| "eval_runtime": 8.0014, | |
| "eval_samples_per_second": 22.246, | |
| "eval_steps_per_second": 2.875, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 3.960396039603961e-06, | |
| "loss": 0.028, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 3.861386138613862e-06, | |
| "loss": 0.0381, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 3.7623762376237625e-06, | |
| "loss": 0.0403, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 3.6633663366336635e-06, | |
| "loss": 0.0183, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 3.5643564356435644e-06, | |
| "loss": 0.0324, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 3.4653465346534653e-06, | |
| "loss": 0.0411, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 3.3663366336633666e-06, | |
| "loss": 0.0247, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 3.2673267326732676e-06, | |
| "loss": 0.0381, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 3.1683168316831685e-06, | |
| "loss": 0.017, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 3.0693069306930694e-06, | |
| "loss": 0.0278, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "eval_accuracy": 0.9606741573033708, | |
| "eval_loss": 0.13798531889915466, | |
| "eval_runtime": 8.0678, | |
| "eval_samples_per_second": 22.063, | |
| "eval_steps_per_second": 2.851, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 2.9702970297029703e-06, | |
| "loss": 0.0349, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 2.8712871287128712e-06, | |
| "loss": 0.0224, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 2.7722772277227726e-06, | |
| "loss": 0.0565, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 2.6732673267326735e-06, | |
| "loss": 0.0394, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 2.5742574257425744e-06, | |
| "loss": 0.0265, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 2.4752475247524753e-06, | |
| "loss": 0.0157, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 2.3762376237623762e-06, | |
| "loss": 0.0164, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 2.2772277227722776e-06, | |
| "loss": 0.0153, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 2.1782178217821785e-06, | |
| "loss": 0.0153, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 2.0792079207920794e-06, | |
| "loss": 0.0153, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "eval_accuracy": 0.9662921348314607, | |
| "eval_loss": 0.13522128760814667, | |
| "eval_runtime": 7.8876, | |
| "eval_samples_per_second": 22.567, | |
| "eval_steps_per_second": 2.916, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 1.9801980198019803e-06, | |
| "loss": 0.0305, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 1.8811881188118813e-06, | |
| "loss": 0.0351, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 1.7821782178217822e-06, | |
| "loss": 0.0148, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 1.6831683168316833e-06, | |
| "loss": 0.0381, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 1.5841584158415842e-06, | |
| "loss": 0.0151, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 1.4851485148514852e-06, | |
| "loss": 0.0145, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 1.3861386138613863e-06, | |
| "loss": 0.0266, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 1.2871287128712872e-06, | |
| "loss": 0.0146, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 1.1881188118811881e-06, | |
| "loss": 0.0147, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "learning_rate": 1.0891089108910893e-06, | |
| "loss": 0.0284, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "eval_accuracy": 0.9662921348314607, | |
| "eval_loss": 0.13962529599666595, | |
| "eval_runtime": 7.6158, | |
| "eval_samples_per_second": 23.372, | |
| "eval_steps_per_second": 3.02, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 9.900990099009902e-07, | |
| "loss": 0.0221, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 8.910891089108911e-07, | |
| "loss": 0.014, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 7.920792079207921e-07, | |
| "loss": 0.0266, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 6.930693069306931e-07, | |
| "loss": 0.0138, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 5.940594059405941e-07, | |
| "loss": 0.033, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 4.950495049504951e-07, | |
| "loss": 0.0371, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 3.9603960396039606e-07, | |
| "loss": 0.0148, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 2.9702970297029703e-07, | |
| "loss": 0.017, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 1.9801980198019803e-07, | |
| "loss": 0.0138, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 9.900990099009901e-08, | |
| "loss": 0.0252, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "eval_accuracy": 0.9662921348314607, | |
| "eval_loss": 0.1401771754026413, | |
| "eval_runtime": 7.8442, | |
| "eval_samples_per_second": 22.692, | |
| "eval_steps_per_second": 2.932, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.014, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1010, | |
| "total_flos": 1.250720712379515e+18, | |
| "train_loss": 0.12084352607184118, | |
| "train_runtime": 814.8954, | |
| "train_samples_per_second": 19.806, | |
| "train_steps_per_second": 1.239 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1010, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 1.250720712379515e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |