| { | |
| "best_metric": 0.8960396039603961, | |
| "best_model_checkpoint": "Electrcical-IMAGE-finetuned-eurosat/checkpoint-313", | |
| "epoch": 19.649122807017545, | |
| "eval_steps": 500, | |
| "global_step": 560, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 10.917478561401367, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 1.8574, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 7.904279708862305, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.6143, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.5346534653465347, | |
| "eval_loss": 1.2882142066955566, | |
| "eval_runtime": 3.1902, | |
| "eval_samples_per_second": 126.638, | |
| "eval_steps_per_second": 4.075, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 12.139655113220215, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 1.3248, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 6.75371789932251, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.1002, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 16.99814224243164, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.8597, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7648514851485149, | |
| "eval_loss": 0.7302265763282776, | |
| "eval_runtime": 3.1986, | |
| "eval_samples_per_second": 126.304, | |
| "eval_steps_per_second": 4.064, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 6.605724811553955, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 0.7264, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 9.7291841506958, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.6225, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 7.020862579345703, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.5858, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.8465346534653465, | |
| "eval_loss": 0.48486289381980896, | |
| "eval_runtime": 3.1164, | |
| "eval_samples_per_second": 129.637, | |
| "eval_steps_per_second": 4.171, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 5.680540561676025, | |
| "learning_rate": 4.662698412698413e-05, | |
| "loss": 0.5214, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 7.502087116241455, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.4695, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "grad_norm": 7.262702465057373, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.4332, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8613861386138614, | |
| "eval_loss": 0.4274178743362427, | |
| "eval_runtime": 3.1661, | |
| "eval_samples_per_second": 127.603, | |
| "eval_steps_per_second": 4.106, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 6.890863418579102, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.4297, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "grad_norm": 6.25393533706665, | |
| "learning_rate": 4.265873015873016e-05, | |
| "loss": 0.4149, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "grad_norm": 7.106054782867432, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.4054, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_accuracy": 0.8787128712871287, | |
| "eval_loss": 0.3687044382095337, | |
| "eval_runtime": 3.1309, | |
| "eval_samples_per_second": 129.037, | |
| "eval_steps_per_second": 4.152, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "grad_norm": 6.924484729766846, | |
| "learning_rate": 4.067460317460318e-05, | |
| "loss": 0.3727, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "grad_norm": 5.540406703948975, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.3299, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 6.109258651733398, | |
| "learning_rate": 3.8690476190476195e-05, | |
| "loss": 0.3826, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8613861386138614, | |
| "eval_loss": 0.37880900502204895, | |
| "eval_runtime": 3.1895, | |
| "eval_samples_per_second": 126.667, | |
| "eval_steps_per_second": 4.076, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "grad_norm": 4.945355415344238, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.3249, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "grad_norm": 4.232621669769287, | |
| "learning_rate": 3.6706349206349205e-05, | |
| "loss": 0.3561, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.8935643564356436, | |
| "eval_loss": 0.36995938420295715, | |
| "eval_runtime": 3.1439, | |
| "eval_samples_per_second": 128.503, | |
| "eval_steps_per_second": 4.135, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 5.742563247680664, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.3097, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 6.4120402336120605, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.3167, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "grad_norm": 6.971658229827881, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.2838, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8811881188118812, | |
| "eval_loss": 0.3550195097923279, | |
| "eval_runtime": 3.0809, | |
| "eval_samples_per_second": 131.129, | |
| "eval_steps_per_second": 4.22, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 6.908318042755127, | |
| "learning_rate": 3.273809523809524e-05, | |
| "loss": 0.2828, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "grad_norm": 5.122198104858398, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.267, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "grad_norm": 4.9148077964782715, | |
| "learning_rate": 3.075396825396826e-05, | |
| "loss": 0.2897, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_accuracy": 0.8886138613861386, | |
| "eval_loss": 0.3698354661464691, | |
| "eval_runtime": 3.2226, | |
| "eval_samples_per_second": 125.366, | |
| "eval_steps_per_second": 4.034, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "grad_norm": 6.60095739364624, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.3077, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "grad_norm": 5.082594394683838, | |
| "learning_rate": 2.876984126984127e-05, | |
| "loss": 0.245, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "grad_norm": 3.717313766479492, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2519, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8836633663366337, | |
| "eval_loss": 0.34588712453842163, | |
| "eval_runtime": 3.2343, | |
| "eval_samples_per_second": 124.913, | |
| "eval_steps_per_second": 4.019, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "grad_norm": 4.068792819976807, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.2456, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "grad_norm": 4.915719509124756, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.2599, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "grad_norm": 4.958366870880127, | |
| "learning_rate": 2.4801587301587305e-05, | |
| "loss": 0.2194, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.8960396039603961, | |
| "eval_loss": 0.3583451807498932, | |
| "eval_runtime": 3.1613, | |
| "eval_samples_per_second": 127.797, | |
| "eval_steps_per_second": 4.112, | |
| "step": 313 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "grad_norm": 8.63839054107666, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.2206, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "grad_norm": 6.567727565765381, | |
| "learning_rate": 2.281746031746032e-05, | |
| "loss": 0.22, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "grad_norm": 5.235540866851807, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.1955, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8886138613861386, | |
| "eval_loss": 0.3442213535308838, | |
| "eval_runtime": 3.1157, | |
| "eval_samples_per_second": 129.666, | |
| "eval_steps_per_second": 4.172, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "grad_norm": 5.771058559417725, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.1896, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "grad_norm": 4.845823287963867, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.1914, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "grad_norm": 6.198949813842773, | |
| "learning_rate": 1.884920634920635e-05, | |
| "loss": 0.2443, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "eval_accuracy": 0.8787128712871287, | |
| "eval_loss": 0.38009530305862427, | |
| "eval_runtime": 3.1836, | |
| "eval_samples_per_second": 126.902, | |
| "eval_steps_per_second": 4.083, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "grad_norm": 5.5330023765563965, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.1938, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "grad_norm": 6.819355487823486, | |
| "learning_rate": 1.6865079365079367e-05, | |
| "loss": 0.207, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8861386138613861, | |
| "eval_loss": 0.34993720054626465, | |
| "eval_runtime": 3.1367, | |
| "eval_samples_per_second": 128.798, | |
| "eval_steps_per_second": 4.144, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "grad_norm": 3.808520555496216, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.2176, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "grad_norm": 4.6104631423950195, | |
| "learning_rate": 1.4880952380952381e-05, | |
| "loss": 0.1725, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "grad_norm": 7.437196731567383, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2078, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.8836633663366337, | |
| "eval_loss": 0.3701125383377075, | |
| "eval_runtime": 3.2627, | |
| "eval_samples_per_second": 123.824, | |
| "eval_steps_per_second": 3.984, | |
| "step": 427 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "grad_norm": 4.747229099273682, | |
| "learning_rate": 1.2896825396825398e-05, | |
| "loss": 0.1775, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "grad_norm": 5.044639587402344, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.1731, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "grad_norm": 5.209843635559082, | |
| "learning_rate": 1.0912698412698414e-05, | |
| "loss": 0.1873, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8861386138613861, | |
| "eval_loss": 0.37732651829719543, | |
| "eval_runtime": 3.1869, | |
| "eval_samples_per_second": 126.767, | |
| "eval_steps_per_second": 4.079, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "grad_norm": 3.205880880355835, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.1728, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "grad_norm": 5.715033054351807, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.1746, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "grad_norm": 3.9077258110046387, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.1697, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_accuracy": 0.8861386138613861, | |
| "eval_loss": 0.37531015276908875, | |
| "eval_runtime": 3.1372, | |
| "eval_samples_per_second": 128.778, | |
| "eval_steps_per_second": 4.144, | |
| "step": 484 | |
| }, | |
| { | |
| "epoch": 17.19, | |
| "grad_norm": 6.2585344314575195, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1774, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "grad_norm": 4.743368625640869, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.1602, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "grad_norm": 5.987907409667969, | |
| "learning_rate": 4.96031746031746e-06, | |
| "loss": 0.1812, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8910891089108911, | |
| "eval_loss": 0.3747256398200989, | |
| "eval_runtime": 3.3841, | |
| "eval_samples_per_second": 119.38, | |
| "eval_steps_per_second": 3.841, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 18.25, | |
| "grad_norm": 4.54545259475708, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.1375, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "grad_norm": 5.2062249183654785, | |
| "learning_rate": 2.9761904761904763e-06, | |
| "loss": 0.1713, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "grad_norm": 5.484841823577881, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.151, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.8861386138613861, | |
| "eval_loss": 0.3735988438129425, | |
| "eval_runtime": 3.2943, | |
| "eval_samples_per_second": 122.636, | |
| "eval_steps_per_second": 3.946, | |
| "step": 541 | |
| }, | |
| { | |
| "epoch": 19.3, | |
| "grad_norm": 5.083439350128174, | |
| "learning_rate": 9.92063492063492e-07, | |
| "loss": 0.163, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "grad_norm": 6.915902137756348, | |
| "learning_rate": 0.0, | |
| "loss": 0.1567, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "eval_accuracy": 0.8861386138613861, | |
| "eval_loss": 0.3726496696472168, | |
| "eval_runtime": 3.3611, | |
| "eval_samples_per_second": 120.199, | |
| "eval_steps_per_second": 3.868, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "step": 560, | |
| "total_flos": 1.7752639958663823e+18, | |
| "train_loss": 0.37542668517146793, | |
| "train_runtime": 934.7322, | |
| "train_samples_per_second": 77.755, | |
| "train_steps_per_second": 0.599 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 560, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.7752639958663823e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |