| { | |
| "best_metric": 0.8787128712871287, | |
| "best_model_checkpoint": "Electrcical-IMAGE-finetuned-eurosat/checkpoint-199", | |
| "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": 6.11484432220459, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 1.726, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 4.320251941680908, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.5532, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.6163366336633663, | |
| "eval_loss": 1.170404314994812, | |
| "eval_runtime": 5.3474, | |
| "eval_samples_per_second": 75.551, | |
| "eval_steps_per_second": 2.431, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 4.363418102264404, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 1.2686, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 6.138254165649414, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.0232, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 7.400917053222656, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.8115, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7673267326732673, | |
| "eval_loss": 0.6826810836791992, | |
| "eval_runtime": 3.1989, | |
| "eval_samples_per_second": 126.295, | |
| "eval_steps_per_second": 4.064, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 7.412561893463135, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 0.6904, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 8.475811004638672, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.5908, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 32.85942077636719, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.5513, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.8415841584158416, | |
| "eval_loss": 0.4524537920951843, | |
| "eval_runtime": 3.2851, | |
| "eval_samples_per_second": 122.978, | |
| "eval_steps_per_second": 3.957, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 5.583408355712891, | |
| "learning_rate": 4.662698412698413e-05, | |
| "loss": 0.5107, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 7.957437038421631, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.4767, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "grad_norm": 6.342191219329834, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.455, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8539603960396039, | |
| "eval_loss": 0.40117621421813965, | |
| "eval_runtime": 3.1407, | |
| "eval_samples_per_second": 128.633, | |
| "eval_steps_per_second": 4.139, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 5.71556282043457, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.4405, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "grad_norm": 6.5584306716918945, | |
| "learning_rate": 4.265873015873016e-05, | |
| "loss": 0.4179, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "grad_norm": 6.672788143157959, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.3901, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_accuracy": 0.8613861386138614, | |
| "eval_loss": 0.38241395354270935, | |
| "eval_runtime": 3.2535, | |
| "eval_samples_per_second": 124.173, | |
| "eval_steps_per_second": 3.996, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "grad_norm": 8.260834693908691, | |
| "learning_rate": 4.067460317460318e-05, | |
| "loss": 0.3765, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "grad_norm": 9.854930877685547, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.3419, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 8.068675994873047, | |
| "learning_rate": 3.8690476190476195e-05, | |
| "loss": 0.4042, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8638613861386139, | |
| "eval_loss": 0.37970858812332153, | |
| "eval_runtime": 3.1263, | |
| "eval_samples_per_second": 129.226, | |
| "eval_steps_per_second": 4.158, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "grad_norm": 5.019181728363037, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.3335, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "grad_norm": 5.983124256134033, | |
| "learning_rate": 3.6706349206349205e-05, | |
| "loss": 0.3591, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.8787128712871287, | |
| "eval_loss": 0.3504551351070404, | |
| "eval_runtime": 3.3207, | |
| "eval_samples_per_second": 121.661, | |
| "eval_steps_per_second": 3.915, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 5.728125095367432, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.3308, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 6.73638916015625, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.3228, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "grad_norm": 9.998391151428223, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.2989, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8613861386138614, | |
| "eval_loss": 0.355110764503479, | |
| "eval_runtime": 3.1811, | |
| "eval_samples_per_second": 127.002, | |
| "eval_steps_per_second": 4.087, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 6.521678924560547, | |
| "learning_rate": 3.273809523809524e-05, | |
| "loss": 0.2857, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "grad_norm": 6.126306056976318, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.2748, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "grad_norm": 7.768061637878418, | |
| "learning_rate": 3.075396825396826e-05, | |
| "loss": 0.3029, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_accuracy": 0.8663366336633663, | |
| "eval_loss": 0.36247390508651733, | |
| "eval_runtime": 3.1822, | |
| "eval_samples_per_second": 126.958, | |
| "eval_steps_per_second": 4.085, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "grad_norm": 6.541659355163574, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.3138, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "grad_norm": 4.587341785430908, | |
| "learning_rate": 2.876984126984127e-05, | |
| "loss": 0.2556, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "grad_norm": 3.2555856704711914, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2606, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.849009900990099, | |
| "eval_loss": 0.3615105450153351, | |
| "eval_runtime": 3.1469, | |
| "eval_samples_per_second": 128.38, | |
| "eval_steps_per_second": 4.131, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "grad_norm": 5.79010534286499, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.2589, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "grad_norm": 7.236560821533203, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.2615, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "grad_norm": 6.618343353271484, | |
| "learning_rate": 2.4801587301587305e-05, | |
| "loss": 0.2413, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.8787128712871287, | |
| "eval_loss": 0.34345176815986633, | |
| "eval_runtime": 3.2537, | |
| "eval_samples_per_second": 124.168, | |
| "eval_steps_per_second": 3.996, | |
| "step": 313 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "grad_norm": 7.067264556884766, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.2329, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "grad_norm": 4.156674861907959, | |
| "learning_rate": 2.281746031746032e-05, | |
| "loss": 0.2286, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "grad_norm": 6.207752227783203, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.2051, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8663366336633663, | |
| "eval_loss": 0.3371087908744812, | |
| "eval_runtime": 3.105, | |
| "eval_samples_per_second": 130.112, | |
| "eval_steps_per_second": 4.187, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "grad_norm": 4.795997142791748, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.2052, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "grad_norm": 5.227004528045654, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.1962, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "grad_norm": 6.650285720825195, | |
| "learning_rate": 1.884920634920635e-05, | |
| "loss": 0.2477, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "eval_accuracy": 0.8638613861386139, | |
| "eval_loss": 0.34505516290664673, | |
| "eval_runtime": 3.309, | |
| "eval_samples_per_second": 122.091, | |
| "eval_steps_per_second": 3.929, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "grad_norm": 4.851602554321289, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.2151, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "grad_norm": 6.44926118850708, | |
| "learning_rate": 1.6865079365079367e-05, | |
| "loss": 0.2271, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8737623762376238, | |
| "eval_loss": 0.3363986909389496, | |
| "eval_runtime": 3.1022, | |
| "eval_samples_per_second": 130.23, | |
| "eval_steps_per_second": 4.191, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "grad_norm": 4.615256309509277, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.2164, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "grad_norm": 4.279232025146484, | |
| "learning_rate": 1.4880952380952381e-05, | |
| "loss": 0.1937, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "grad_norm": 5.792173385620117, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2112, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.8638613861386139, | |
| "eval_loss": 0.35587263107299805, | |
| "eval_runtime": 3.6554, | |
| "eval_samples_per_second": 110.522, | |
| "eval_steps_per_second": 3.556, | |
| "step": 427 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "grad_norm": 4.082373142242432, | |
| "learning_rate": 1.2896825396825398e-05, | |
| "loss": 0.1815, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "grad_norm": 6.473873138427734, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.1884, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "grad_norm": 3.9221067428588867, | |
| "learning_rate": 1.0912698412698414e-05, | |
| "loss": 0.1902, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8737623762376238, | |
| "eval_loss": 0.36296331882476807, | |
| "eval_runtime": 3.2193, | |
| "eval_samples_per_second": 125.494, | |
| "eval_steps_per_second": 4.038, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "grad_norm": 3.29888916015625, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.1834, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "grad_norm": 5.7562432289123535, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.1797, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "grad_norm": 4.14888858795166, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.1739, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_accuracy": 0.8712871287128713, | |
| "eval_loss": 0.36299142241477966, | |
| "eval_runtime": 3.2007, | |
| "eval_samples_per_second": 126.223, | |
| "eval_steps_per_second": 4.062, | |
| "step": 484 | |
| }, | |
| { | |
| "epoch": 17.19, | |
| "grad_norm": 5.445935249328613, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1723, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "grad_norm": 10.137904167175293, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.1721, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "grad_norm": 6.753917217254639, | |
| "learning_rate": 4.96031746031746e-06, | |
| "loss": 0.195, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8663366336633663, | |
| "eval_loss": 0.3625456690788269, | |
| "eval_runtime": 3.1415, | |
| "eval_samples_per_second": 128.599, | |
| "eval_steps_per_second": 4.138, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 18.25, | |
| "grad_norm": 5.913514614105225, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.138, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "grad_norm": 5.631173610687256, | |
| "learning_rate": 2.9761904761904763e-06, | |
| "loss": 0.1784, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "grad_norm": 3.3359873294830322, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.1621, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.8762376237623762, | |
| "eval_loss": 0.35711079835891724, | |
| "eval_runtime": 3.3137, | |
| "eval_samples_per_second": 121.916, | |
| "eval_steps_per_second": 3.923, | |
| "step": 541 | |
| }, | |
| { | |
| "epoch": 19.3, | |
| "grad_norm": 4.744650363922119, | |
| "learning_rate": 9.92063492063492e-07, | |
| "loss": 0.1726, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "grad_norm": 5.039443492889404, | |
| "learning_rate": 0.0, | |
| "loss": 0.154, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "eval_accuracy": 0.8737623762376238, | |
| "eval_loss": 0.3554615080356598, | |
| "eval_runtime": 3.3898, | |
| "eval_samples_per_second": 119.18, | |
| "eval_steps_per_second": 3.835, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "step": 560, | |
| "total_flos": 1.7752639958663823e+18, | |
| "train_loss": 0.3740995813693319, | |
| "train_runtime": 1232.3243, | |
| "train_samples_per_second": 58.978, | |
| "train_steps_per_second": 0.454 | |
| } | |
| ], | |
| "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 | |
| } | |