| { | |
| "best_metric": 0.9383116883116883, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-1218", | |
| "epoch": 29.655172413793103, | |
| "eval_steps": 500, | |
| "global_step": 1290, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.875968992248062e-06, | |
| "loss": 2.033, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 7.751937984496124e-06, | |
| "loss": 1.819, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.1627906976744187e-05, | |
| "loss": 1.3652, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.5503875968992248e-05, | |
| "loss": 0.9843, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.6948051948051948, | |
| "eval_loss": 0.850001335144043, | |
| "eval_runtime": 6.318, | |
| "eval_samples_per_second": 97.499, | |
| "eval_steps_per_second": 3.166, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.937984496124031e-05, | |
| "loss": 0.816, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.3255813953488374e-05, | |
| "loss": 0.7631, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.7131782945736434e-05, | |
| "loss": 0.6499, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.1007751937984497e-05, | |
| "loss": 0.5335, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7824675324675324, | |
| "eval_loss": 0.5584108233451843, | |
| "eval_runtime": 6.0937, | |
| "eval_samples_per_second": 101.088, | |
| "eval_steps_per_second": 3.282, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 3.488372093023256e-05, | |
| "loss": 0.5594, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 3.875968992248062e-05, | |
| "loss": 0.4565, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.263565891472868e-05, | |
| "loss": 0.4759, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 4.651162790697675e-05, | |
| "loss": 0.4542, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.995693367786391e-05, | |
| "loss": 0.4263, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.8116883116883117, | |
| "eval_loss": 0.4790755808353424, | |
| "eval_runtime": 6.348, | |
| "eval_samples_per_second": 97.038, | |
| "eval_steps_per_second": 3.151, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 4.952627045650302e-05, | |
| "loss": 0.3769, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.9095607235142123e-05, | |
| "loss": 0.3852, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 4.866494401378123e-05, | |
| "loss": 0.3409, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.823428079242033e-05, | |
| "loss": 0.3308, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8344155844155844, | |
| "eval_loss": 0.4268946349620819, | |
| "eval_runtime": 6.5538, | |
| "eval_samples_per_second": 93.991, | |
| "eval_steps_per_second": 3.052, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 4.780361757105943e-05, | |
| "loss": 0.3093, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 4.737295434969854e-05, | |
| "loss": 0.3429, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 4.694229112833764e-05, | |
| "loss": 0.2697, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 4.651162790697675e-05, | |
| "loss": 0.2882, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.35674524307250977, | |
| "eval_runtime": 6.3226, | |
| "eval_samples_per_second": 97.428, | |
| "eval_steps_per_second": 3.163, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 4.6080964685615853e-05, | |
| "loss": 0.2729, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 4.565030146425496e-05, | |
| "loss": 0.2543, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 4.521963824289406e-05, | |
| "loss": 0.2526, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 4.478897502153316e-05, | |
| "loss": 0.255, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 4.435831180017227e-05, | |
| "loss": 0.2517, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8701298701298701, | |
| "eval_loss": 0.33171918988227844, | |
| "eval_runtime": 5.9688, | |
| "eval_samples_per_second": 103.204, | |
| "eval_steps_per_second": 3.351, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 4.392764857881137e-05, | |
| "loss": 0.2317, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 4.349698535745048e-05, | |
| "loss": 0.1944, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 4.306632213608958e-05, | |
| "loss": 0.248, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 4.263565891472868e-05, | |
| "loss": 0.1908, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.8814935064935064, | |
| "eval_loss": 0.3081734776496887, | |
| "eval_runtime": 6.4591, | |
| "eval_samples_per_second": 95.369, | |
| "eval_steps_per_second": 3.096, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 4.220499569336779e-05, | |
| "loss": 0.1726, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 4.177433247200689e-05, | |
| "loss": 0.1663, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 4.1343669250646e-05, | |
| "loss": 0.1602, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 4.09130060292851e-05, | |
| "loss": 0.187, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8798701298701299, | |
| "eval_loss": 0.3230064809322357, | |
| "eval_runtime": 6.0202, | |
| "eval_samples_per_second": 102.322, | |
| "eval_steps_per_second": 3.322, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.048234280792421e-05, | |
| "loss": 0.1676, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 4.005167958656331e-05, | |
| "loss": 0.1681, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 3.962101636520241e-05, | |
| "loss": 0.1818, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 3.919035314384152e-05, | |
| "loss": 0.1299, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 3.875968992248062e-05, | |
| "loss": 0.1434, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.900974025974026, | |
| "eval_loss": 0.3322867453098297, | |
| "eval_runtime": 6.3218, | |
| "eval_samples_per_second": 97.44, | |
| "eval_steps_per_second": 3.164, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 3.832902670111973e-05, | |
| "loss": 0.1432, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 3.789836347975883e-05, | |
| "loss": 0.1313, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 3.746770025839794e-05, | |
| "loss": 0.1546, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.1277, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9074675324675324, | |
| "eval_loss": 0.2488991618156433, | |
| "eval_runtime": 6.4317, | |
| "eval_samples_per_second": 95.776, | |
| "eval_steps_per_second": 3.11, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 3.660637381567614e-05, | |
| "loss": 0.1349, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 3.617571059431525e-05, | |
| "loss": 0.1074, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 10.57, | |
| "learning_rate": 3.5745047372954346e-05, | |
| "loss": 0.1218, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "learning_rate": 3.531438415159346e-05, | |
| "loss": 0.156, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.887987012987013, | |
| "eval_loss": 0.3245611786842346, | |
| "eval_runtime": 6.5235, | |
| "eval_samples_per_second": 94.429, | |
| "eval_steps_per_second": 3.066, | |
| "step": 478 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 3.488372093023256e-05, | |
| "loss": 0.1113, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 11.26, | |
| "learning_rate": 3.445305770887167e-05, | |
| "loss": 0.0888, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 3.4022394487510767e-05, | |
| "loss": 0.1083, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 3.359173126614987e-05, | |
| "loss": 0.1126, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 11.95, | |
| "learning_rate": 3.316106804478898e-05, | |
| "loss": 0.0781, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.900974025974026, | |
| "eval_loss": 0.31213870644569397, | |
| "eval_runtime": 5.9923, | |
| "eval_samples_per_second": 102.799, | |
| "eval_steps_per_second": 3.338, | |
| "step": 522 | |
| }, | |
| { | |
| "epoch": 12.18, | |
| "learning_rate": 3.273040482342808e-05, | |
| "loss": 0.0906, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 3.229974160206719e-05, | |
| "loss": 0.0864, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 3.186907838070629e-05, | |
| "loss": 0.0854, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 12.87, | |
| "learning_rate": 3.143841515934539e-05, | |
| "loss": 0.1001, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.9058441558441559, | |
| "eval_loss": 0.27083244919776917, | |
| "eval_runtime": 6.2499, | |
| "eval_samples_per_second": 98.561, | |
| "eval_steps_per_second": 3.2, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 3.1007751937984497e-05, | |
| "loss": 0.0923, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 3.05770887166236e-05, | |
| "loss": 0.0771, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 3.0146425495262704e-05, | |
| "loss": 0.0873, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 2.971576227390181e-05, | |
| "loss": 0.0892, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.913961038961039, | |
| "eval_loss": 0.2581653296947479, | |
| "eval_runtime": 6.4729, | |
| "eval_samples_per_second": 95.166, | |
| "eval_steps_per_second": 3.09, | |
| "step": 609 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 2.9285099052540914e-05, | |
| "loss": 0.0802, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 2.8854435831180023e-05, | |
| "loss": 0.073, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 2.842377260981912e-05, | |
| "loss": 0.0584, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 14.71, | |
| "learning_rate": 2.7993109388458226e-05, | |
| "loss": 0.0552, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 14.94, | |
| "learning_rate": 2.7562446167097332e-05, | |
| "loss": 0.0644, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.922077922077922, | |
| "eval_loss": 0.24860848486423492, | |
| "eval_runtime": 6.4617, | |
| "eval_samples_per_second": 95.332, | |
| "eval_steps_per_second": 3.095, | |
| "step": 652 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 2.7131782945736434e-05, | |
| "loss": 0.0906, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 15.4, | |
| "learning_rate": 2.670111972437554e-05, | |
| "loss": 0.0507, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 15.63, | |
| "learning_rate": 2.6270456503014644e-05, | |
| "loss": 0.0597, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 2.5839793281653746e-05, | |
| "loss": 0.0689, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9237012987012987, | |
| "eval_loss": 0.246454119682312, | |
| "eval_runtime": 6.0878, | |
| "eval_samples_per_second": 101.186, | |
| "eval_steps_per_second": 3.285, | |
| "step": 696 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "learning_rate": 2.540913006029285e-05, | |
| "loss": 0.0696, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 16.32, | |
| "learning_rate": 2.4978466838931956e-05, | |
| "loss": 0.0563, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 2.4547803617571062e-05, | |
| "loss": 0.058, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 16.78, | |
| "learning_rate": 2.4117140396210164e-05, | |
| "loss": 0.0547, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.9334415584415584, | |
| "eval_loss": 0.24022094905376434, | |
| "eval_runtime": 6.0529, | |
| "eval_samples_per_second": 101.769, | |
| "eval_steps_per_second": 3.304, | |
| "step": 739 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 2.368647717484927e-05, | |
| "loss": 0.068, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.3255813953488374e-05, | |
| "loss": 0.0556, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 17.47, | |
| "learning_rate": 2.282515073212748e-05, | |
| "loss": 0.0541, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 2.239448751076658e-05, | |
| "loss": 0.0444, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "learning_rate": 2.1963824289405686e-05, | |
| "loss": 0.0597, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9237012987012987, | |
| "eval_loss": 0.2533910572528839, | |
| "eval_runtime": 6.1928, | |
| "eval_samples_per_second": 99.47, | |
| "eval_steps_per_second": 3.23, | |
| "step": 783 | |
| }, | |
| { | |
| "epoch": 18.16, | |
| "learning_rate": 2.153316106804479e-05, | |
| "loss": 0.0461, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 2.1102497846683894e-05, | |
| "loss": 0.044, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 2.0671834625323e-05, | |
| "loss": 0.0474, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 18.85, | |
| "learning_rate": 2.0241171403962104e-05, | |
| "loss": 0.0512, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.24004700779914856, | |
| "eval_runtime": 6.4994, | |
| "eval_samples_per_second": 94.777, | |
| "eval_steps_per_second": 3.077, | |
| "step": 826 | |
| }, | |
| { | |
| "epoch": 19.08, | |
| "learning_rate": 1.9810508182601206e-05, | |
| "loss": 0.056, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 1.937984496124031e-05, | |
| "loss": 0.0527, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 19.54, | |
| "learning_rate": 1.8949181739879416e-05, | |
| "loss": 0.0282, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 19.77, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0486, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.8087855297157624e-05, | |
| "loss": 0.041, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9285714285714286, | |
| "eval_loss": 0.23969942331314087, | |
| "eval_runtime": 6.4042, | |
| "eval_samples_per_second": 96.187, | |
| "eval_steps_per_second": 3.123, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 20.23, | |
| "learning_rate": 1.765719207579673e-05, | |
| "loss": 0.0347, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 20.46, | |
| "learning_rate": 1.7226528854435834e-05, | |
| "loss": 0.0523, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 20.69, | |
| "learning_rate": 1.6795865633074936e-05, | |
| "loss": 0.0334, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 20.92, | |
| "learning_rate": 1.636520241171404e-05, | |
| "loss": 0.0376, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.926948051948052, | |
| "eval_loss": 0.26630115509033203, | |
| "eval_runtime": 6.6399, | |
| "eval_samples_per_second": 92.772, | |
| "eval_steps_per_second": 3.012, | |
| "step": 913 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "learning_rate": 1.5934539190353146e-05, | |
| "loss": 0.0334, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 21.38, | |
| "learning_rate": 1.5503875968992248e-05, | |
| "loss": 0.042, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 21.61, | |
| "learning_rate": 1.5073212747631352e-05, | |
| "loss": 0.0444, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 21.84, | |
| "learning_rate": 1.4642549526270457e-05, | |
| "loss": 0.0412, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.922077922077922, | |
| "eval_loss": 0.3025703430175781, | |
| "eval_runtime": 6.5146, | |
| "eval_samples_per_second": 94.556, | |
| "eval_steps_per_second": 3.07, | |
| "step": 957 | |
| }, | |
| { | |
| "epoch": 22.07, | |
| "learning_rate": 1.421188630490956e-05, | |
| "loss": 0.0321, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 22.3, | |
| "learning_rate": 1.3781223083548666e-05, | |
| "loss": 0.0383, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 22.53, | |
| "learning_rate": 1.335055986218777e-05, | |
| "loss": 0.027, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 22.76, | |
| "learning_rate": 1.2919896640826873e-05, | |
| "loss": 0.0446, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 22.99, | |
| "learning_rate": 1.2489233419465978e-05, | |
| "loss": 0.0423, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 22.99, | |
| "eval_accuracy": 0.9301948051948052, | |
| "eval_loss": 0.26784056425094604, | |
| "eval_runtime": 6.0495, | |
| "eval_samples_per_second": 101.827, | |
| "eval_steps_per_second": 3.306, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 23.22, | |
| "learning_rate": 1.2058570198105082e-05, | |
| "loss": 0.0327, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 23.45, | |
| "learning_rate": 1.1627906976744187e-05, | |
| "loss": 0.0463, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "learning_rate": 1.119724375538329e-05, | |
| "loss": 0.0314, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 23.91, | |
| "learning_rate": 1.0766580534022396e-05, | |
| "loss": 0.0266, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.25100043416023254, | |
| "eval_runtime": 6.0247, | |
| "eval_samples_per_second": 102.246, | |
| "eval_steps_per_second": 3.32, | |
| "step": 1044 | |
| }, | |
| { | |
| "epoch": 24.14, | |
| "learning_rate": 1.03359173126615e-05, | |
| "loss": 0.0257, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 24.37, | |
| "learning_rate": 9.905254091300603e-06, | |
| "loss": 0.0292, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 24.6, | |
| "learning_rate": 9.474590869939708e-06, | |
| "loss": 0.0329, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 24.83, | |
| "learning_rate": 9.043927648578812e-06, | |
| "loss": 0.0313, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.9334415584415584, | |
| "eval_loss": 0.2541840672492981, | |
| "eval_runtime": 6.4751, | |
| "eval_samples_per_second": 95.134, | |
| "eval_steps_per_second": 3.089, | |
| "step": 1087 | |
| }, | |
| { | |
| "epoch": 25.06, | |
| "learning_rate": 8.613264427217917e-06, | |
| "loss": 0.0178, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 25.29, | |
| "learning_rate": 8.18260120585702e-06, | |
| "loss": 0.0189, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 25.52, | |
| "learning_rate": 7.751937984496124e-06, | |
| "loss": 0.0399, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 25.75, | |
| "learning_rate": 7.3212747631352285e-06, | |
| "loss": 0.0269, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 25.98, | |
| "learning_rate": 6.890611541774333e-06, | |
| "loss": 0.0207, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9334415584415584, | |
| "eval_loss": 0.2742658257484436, | |
| "eval_runtime": 6.484, | |
| "eval_samples_per_second": 95.003, | |
| "eval_steps_per_second": 3.084, | |
| "step": 1131 | |
| }, | |
| { | |
| "epoch": 26.21, | |
| "learning_rate": 6.4599483204134365e-06, | |
| "loss": 0.0412, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 26.44, | |
| "learning_rate": 6.029285099052541e-06, | |
| "loss": 0.0271, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 5.598621877691645e-06, | |
| "loss": 0.0268, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 26.9, | |
| "learning_rate": 5.16795865633075e-06, | |
| "loss": 0.0292, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 26.99, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2614338994026184, | |
| "eval_runtime": 6.1763, | |
| "eval_samples_per_second": 99.736, | |
| "eval_steps_per_second": 3.238, | |
| "step": 1174 | |
| }, | |
| { | |
| "epoch": 27.13, | |
| "learning_rate": 4.737295434969854e-06, | |
| "loss": 0.0248, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "learning_rate": 4.3066322136089585e-06, | |
| "loss": 0.0329, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 27.59, | |
| "learning_rate": 3.875968992248062e-06, | |
| "loss": 0.0307, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 27.82, | |
| "learning_rate": 3.4453057708871665e-06, | |
| "loss": 0.0242, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9383116883116883, | |
| "eval_loss": 0.24687811732292175, | |
| "eval_runtime": 5.9381, | |
| "eval_samples_per_second": 103.737, | |
| "eval_steps_per_second": 3.368, | |
| "step": 1218 | |
| }, | |
| { | |
| "epoch": 28.05, | |
| "learning_rate": 3.0146425495262704e-06, | |
| "loss": 0.0295, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 28.28, | |
| "learning_rate": 2.583979328165375e-06, | |
| "loss": 0.0161, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 28.51, | |
| "learning_rate": 2.1533161068044793e-06, | |
| "loss": 0.034, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 28.74, | |
| "learning_rate": 1.7226528854435832e-06, | |
| "loss": 0.0247, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 28.97, | |
| "learning_rate": 1.2919896640826874e-06, | |
| "loss": 0.0201, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_accuracy": 0.9366883116883117, | |
| "eval_loss": 0.2533850371837616, | |
| "eval_runtime": 6.3893, | |
| "eval_samples_per_second": 96.411, | |
| "eval_steps_per_second": 3.13, | |
| "step": 1261 | |
| }, | |
| { | |
| "epoch": 29.2, | |
| "learning_rate": 8.613264427217916e-07, | |
| "loss": 0.0262, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 29.43, | |
| "learning_rate": 4.306632213608958e-07, | |
| "loss": 0.0226, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 29.66, | |
| "learning_rate": 0.0, | |
| "loss": 0.0354, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 29.66, | |
| "eval_accuracy": 0.9366883116883117, | |
| "eval_loss": 0.2525167465209961, | |
| "eval_runtime": 6.455, | |
| "eval_samples_per_second": 95.43, | |
| "eval_steps_per_second": 3.098, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 29.66, | |
| "step": 1290, | |
| "total_flos": 4.0838875031628657e+18, | |
| "train_loss": 0.17642173106356185, | |
| "train_runtime": 3123.2223, | |
| "train_samples_per_second": 53.205, | |
| "train_steps_per_second": 0.413 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1290, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 4.0838875031628657e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |