| { | |
| "best_metric": 0.06115679070353508, | |
| "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-1700", | |
| "epoch": 3.0, | |
| "eval_steps": 100, | |
| "global_step": 1788, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.9720357941834455e-05, | |
| "loss": 0.001, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.944071588366891e-05, | |
| "loss": 0.0021, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.9161073825503354e-05, | |
| "loss": 0.0007, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.888143176733781e-05, | |
| "loss": 0.0596, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.8601789709172266e-05, | |
| "loss": 0.0007, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.832214765100672e-05, | |
| "loss": 0.0021, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.8042505592841165e-05, | |
| "loss": 0.0276, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.776286353467562e-05, | |
| "loss": 0.0008, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.748322147651007e-05, | |
| "loss": 0.0235, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.720357941834452e-05, | |
| "loss": 0.0007, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.9748110831234257, | |
| "eval_loss": 0.12113822996616364, | |
| "eval_runtime": 39.5676, | |
| "eval_samples_per_second": 60.201, | |
| "eval_steps_per_second": 7.531, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.692393736017897e-05, | |
| "loss": 0.0362, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.664429530201342e-05, | |
| "loss": 0.0006, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.6364653243847875e-05, | |
| "loss": 0.0446, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.608501118568233e-05, | |
| "loss": 0.0006, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.580536912751678e-05, | |
| "loss": 0.0795, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.552572706935123e-05, | |
| "loss": 0.002, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.5246085011185686e-05, | |
| "loss": 0.0177, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.496644295302014e-05, | |
| "loss": 0.0005, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.4686800894854585e-05, | |
| "loss": 0.0005, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.440715883668904e-05, | |
| "loss": 0.0005, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.9785894206549118, | |
| "eval_loss": 0.10272762924432755, | |
| "eval_runtime": 40.3638, | |
| "eval_samples_per_second": 59.013, | |
| "eval_steps_per_second": 7.383, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.412751677852349e-05, | |
| "loss": 0.0005, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.384787472035794e-05, | |
| "loss": 0.0005, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.3568232662192396e-05, | |
| "loss": 0.0012, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.328859060402685e-05, | |
| "loss": 0.0005, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.30089485458613e-05, | |
| "loss": 0.0004, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.2729306487695754e-05, | |
| "loss": 0.0006, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.244966442953021e-05, | |
| "loss": 0.0005, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.217002237136465e-05, | |
| "loss": 0.0107, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.1890380313199106e-05, | |
| "loss": 0.0015, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.161073825503356e-05, | |
| "loss": 0.0195, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.9836272040302267, | |
| "eval_loss": 0.08691323548555374, | |
| "eval_runtime": 40.569, | |
| "eval_samples_per_second": 58.715, | |
| "eval_steps_per_second": 7.346, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.133109619686801e-05, | |
| "loss": 0.0005, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.1051454138702464e-05, | |
| "loss": 0.0006, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.077181208053692e-05, | |
| "loss": 0.0924, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.052013422818792e-05, | |
| "loss": 0.0389, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.0240492170022376e-05, | |
| "loss": 0.0007, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3.996085011185682e-05, | |
| "loss": 0.0438, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.9681208053691275e-05, | |
| "loss": 0.0005, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.940156599552573e-05, | |
| "loss": 0.0005, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 3.912192393736018e-05, | |
| "loss": 0.0027, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.884228187919463e-05, | |
| "loss": 0.0025, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.9844668345927792, | |
| "eval_loss": 0.0822625532746315, | |
| "eval_runtime": 42.2419, | |
| "eval_samples_per_second": 56.389, | |
| "eval_steps_per_second": 7.055, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.8562639821029086e-05, | |
| "loss": 0.0005, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.828299776286354e-05, | |
| "loss": 0.0004, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.800335570469799e-05, | |
| "loss": 0.0004, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3.772371364653244e-05, | |
| "loss": 0.0005, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.744407158836689e-05, | |
| "loss": 0.0005, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.716442953020134e-05, | |
| "loss": 0.0004, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.6884787472035796e-05, | |
| "loss": 0.0004, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 3.660514541387025e-05, | |
| "loss": 0.0731, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.63255033557047e-05, | |
| "loss": 0.0004, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.6045861297539154e-05, | |
| "loss": 0.0154, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.9827875734676742, | |
| "eval_loss": 0.0887603908777237, | |
| "eval_runtime": 38.897, | |
| "eval_samples_per_second": 61.239, | |
| "eval_steps_per_second": 7.661, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 3.576621923937361e-05, | |
| "loss": 0.0513, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3.548657718120805e-05, | |
| "loss": 0.0025, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 3.5206935123042505e-05, | |
| "loss": 0.0004, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 3.492729306487696e-05, | |
| "loss": 0.0004, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3.464765100671141e-05, | |
| "loss": 0.0106, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.4368008948545864e-05, | |
| "loss": 0.0004, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 3.4088366890380317e-05, | |
| "loss": 0.0005, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3.380872483221477e-05, | |
| "loss": 0.0005, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.352908277404922e-05, | |
| "loss": 0.0005, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 3.3249440715883675e-05, | |
| "loss": 0.0004, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.9853064651553316, | |
| "eval_loss": 0.07808367908000946, | |
| "eval_runtime": 39.8136, | |
| "eval_samples_per_second": 59.829, | |
| "eval_steps_per_second": 7.485, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 3.296979865771812e-05, | |
| "loss": 0.0298, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.2690156599552574e-05, | |
| "loss": 0.0004, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 3.2410514541387026e-05, | |
| "loss": 0.008, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.213087248322148e-05, | |
| "loss": 0.0229, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 3.1851230425055925e-05, | |
| "loss": 0.0044, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 3.157158836689038e-05, | |
| "loss": 0.0003, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 3.129194630872484e-05, | |
| "loss": 0.026, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3.101230425055929e-05, | |
| "loss": 0.0004, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 3.0732662192393736e-05, | |
| "loss": 0.0008, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 3.045302013422819e-05, | |
| "loss": 0.0004, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_accuracy": 0.9832073887489504, | |
| "eval_loss": 0.09313122928142548, | |
| "eval_runtime": 38.9358, | |
| "eval_samples_per_second": 61.178, | |
| "eval_steps_per_second": 7.654, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.0173378076062642e-05, | |
| "loss": 0.0004, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 2.9893736017897095e-05, | |
| "loss": 0.0004, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 2.9614093959731544e-05, | |
| "loss": 0.0003, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 2.9334451901565997e-05, | |
| "loss": 0.0004, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.905480984340045e-05, | |
| "loss": 0.0283, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.8775167785234902e-05, | |
| "loss": 0.0004, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 2.8495525727069348e-05, | |
| "loss": 0.0004, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.8215883668903804e-05, | |
| "loss": 0.0016, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.7936241610738257e-05, | |
| "loss": 0.0003, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.765659955257271e-05, | |
| "loss": 0.0004, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_accuracy": 0.9811083123425692, | |
| "eval_loss": 0.09949999302625656, | |
| "eval_runtime": 40.1971, | |
| "eval_samples_per_second": 59.258, | |
| "eval_steps_per_second": 7.413, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.7376957494407163e-05, | |
| "loss": 0.0004, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.709731543624161e-05, | |
| "loss": 0.0004, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.6817673378076065e-05, | |
| "loss": 0.0174, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.6538031319910518e-05, | |
| "loss": 0.0075, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.625838926174497e-05, | |
| "loss": 0.0003, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 2.5978747203579416e-05, | |
| "loss": 0.0004, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 2.569910514541387e-05, | |
| "loss": 0.0003, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 2.5419463087248325e-05, | |
| "loss": 0.0004, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 2.5139821029082778e-05, | |
| "loss": 0.0004, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.4860178970917228e-05, | |
| "loss": 0.0004, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_accuracy": 0.9832073887489504, | |
| "eval_loss": 0.09245476126670837, | |
| "eval_runtime": 39.9023, | |
| "eval_samples_per_second": 59.696, | |
| "eval_steps_per_second": 7.468, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 2.4580536912751677e-05, | |
| "loss": 0.009, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 2.4300894854586133e-05, | |
| "loss": 0.0004, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 2.4021252796420583e-05, | |
| "loss": 0.0003, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 2.3741610738255035e-05, | |
| "loss": 0.0004, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 2.3461968680089485e-05, | |
| "loss": 0.0003, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.3182326621923937e-05, | |
| "loss": 0.0137, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 2.290268456375839e-05, | |
| "loss": 0.0003, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.2623042505592843e-05, | |
| "loss": 0.0003, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.2343400447427292e-05, | |
| "loss": 0.0003, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 2.2063758389261745e-05, | |
| "loss": 0.0003, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.9836272040302267, | |
| "eval_loss": 0.08568126708269119, | |
| "eval_runtime": 40.7121, | |
| "eval_samples_per_second": 58.508, | |
| "eval_steps_per_second": 7.32, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 2.1784116331096198e-05, | |
| "loss": 0.0003, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.150447427293065e-05, | |
| "loss": 0.0037, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 2.1224832214765103e-05, | |
| "loss": 0.0003, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.0945190156599553e-05, | |
| "loss": 0.0003, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 2.0665548098434006e-05, | |
| "loss": 0.0004, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.038590604026846e-05, | |
| "loss": 0.0008, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.010626398210291e-05, | |
| "loss": 0.0008, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.982662192393736e-05, | |
| "loss": 0.0003, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.9546979865771813e-05, | |
| "loss": 0.0003, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.9267337807606263e-05, | |
| "loss": 0.0364, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.9844668345927792, | |
| "eval_loss": 0.07881016284227371, | |
| "eval_runtime": 40.0549, | |
| "eval_samples_per_second": 59.468, | |
| "eval_steps_per_second": 7.44, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.898769574944072e-05, | |
| "loss": 0.0022, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.870805369127517e-05, | |
| "loss": 0.0233, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.842841163310962e-05, | |
| "loss": 0.0003, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 1.814876957494407e-05, | |
| "loss": 0.0004, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.7869127516778523e-05, | |
| "loss": 0.0014, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.7589485458612976e-05, | |
| "loss": 0.0003, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.730984340044743e-05, | |
| "loss": 0.0003, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.7030201342281878e-05, | |
| "loss": 0.0004, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.675055928411633e-05, | |
| "loss": 0.0041, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.6470917225950784e-05, | |
| "loss": 0.0003, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_accuracy": 0.984047019311503, | |
| "eval_loss": 0.07748164236545563, | |
| "eval_runtime": 39.6939, | |
| "eval_samples_per_second": 60.009, | |
| "eval_steps_per_second": 7.507, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.6191275167785237e-05, | |
| "loss": 0.0003, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.5911633109619686e-05, | |
| "loss": 0.0003, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.563199105145414e-05, | |
| "loss": 0.0058, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.535234899328859e-05, | |
| "loss": 0.0003, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.5072706935123043e-05, | |
| "loss": 0.0005, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1.4793064876957497e-05, | |
| "loss": 0.0003, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.4513422818791946e-05, | |
| "loss": 0.0003, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.4233780760626401e-05, | |
| "loss": 0.0003, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.395413870246085e-05, | |
| "loss": 0.0003, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.3674496644295303e-05, | |
| "loss": 0.0003, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_accuracy": 0.9857262804366079, | |
| "eval_loss": 0.07177738100290298, | |
| "eval_runtime": 40.7042, | |
| "eval_samples_per_second": 58.52, | |
| "eval_steps_per_second": 7.321, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.3394854586129754e-05, | |
| "loss": 0.0014, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.3115212527964207e-05, | |
| "loss": 0.0003, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1.2835570469798658e-05, | |
| "loss": 0.0003, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.255592841163311e-05, | |
| "loss": 0.0003, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.2276286353467562e-05, | |
| "loss": 0.0003, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.1996644295302013e-05, | |
| "loss": 0.0003, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.1717002237136467e-05, | |
| "loss": 0.0003, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1.1437360178970918e-05, | |
| "loss": 0.0003, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 1.115771812080537e-05, | |
| "loss": 0.0003, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.0878076062639822e-05, | |
| "loss": 0.0003, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_accuracy": 0.9848866498740554, | |
| "eval_loss": 0.08044111728668213, | |
| "eval_runtime": 40.6091, | |
| "eval_samples_per_second": 58.657, | |
| "eval_steps_per_second": 7.338, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 1.0598434004474273e-05, | |
| "loss": 0.0003, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.0318791946308726e-05, | |
| "loss": 0.0003, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.0039149888143177e-05, | |
| "loss": 0.0003, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 9.75950782997763e-06, | |
| "loss": 0.001, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 9.479865771812081e-06, | |
| "loss": 0.0003, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 9.200223713646534e-06, | |
| "loss": 0.0003, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 8.920581655480985e-06, | |
| "loss": 0.0003, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 8.640939597315436e-06, | |
| "loss": 0.0003, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 8.361297539149889e-06, | |
| "loss": 0.0003, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 8.08165548098434e-06, | |
| "loss": 0.0003, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_accuracy": 0.9836272040302267, | |
| "eval_loss": 0.07507482171058655, | |
| "eval_runtime": 40.8067, | |
| "eval_samples_per_second": 58.373, | |
| "eval_steps_per_second": 7.303, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 7.802013422818793e-06, | |
| "loss": 0.0003, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 7.522371364653244e-06, | |
| "loss": 0.0028, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 7.242729306487696e-06, | |
| "loss": 0.0003, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 6.963087248322148e-06, | |
| "loss": 0.0003, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 6.6834451901565995e-06, | |
| "loss": 0.0003, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 6.4038031319910514e-06, | |
| "loss": 0.0002, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 6.124161073825504e-06, | |
| "loss": 0.0078, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 5.844519015659955e-06, | |
| "loss": 0.0003, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 5.564876957494407e-06, | |
| "loss": 0.0003, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 5.285234899328859e-06, | |
| "loss": 0.0003, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_accuracy": 0.9869857262804366, | |
| "eval_loss": 0.06587431579828262, | |
| "eval_runtime": 41.2817, | |
| "eval_samples_per_second": 57.701, | |
| "eval_steps_per_second": 7.219, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 5.005592841163311e-06, | |
| "loss": 0.0003, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 4.725950782997763e-06, | |
| "loss": 0.0002, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 4.446308724832215e-06, | |
| "loss": 0.0003, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0003, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.887024608501119e-06, | |
| "loss": 0.0003, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 3.607382550335571e-06, | |
| "loss": 0.0002, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.3277404921700227e-06, | |
| "loss": 0.0003, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 3.048098434004474e-06, | |
| "loss": 0.0007, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 2.7684563758389265e-06, | |
| "loss": 0.0003, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.4888143176733784e-06, | |
| "loss": 0.0002, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.9874055415617129, | |
| "eval_loss": 0.06115679070353508, | |
| "eval_runtime": 41.2926, | |
| "eval_samples_per_second": 57.686, | |
| "eval_steps_per_second": 7.217, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 2.20917225950783e-06, | |
| "loss": 0.0003, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.929530201342282e-06, | |
| "loss": 0.0003, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.6498881431767338e-06, | |
| "loss": 0.0003, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.3702460850111857e-06, | |
| "loss": 0.0002, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.0906040268456377e-06, | |
| "loss": 0.0003, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 8.109619686800895e-07, | |
| "loss": 0.0002, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 5.313199105145415e-07, | |
| "loss": 0.0002, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 2.516778523489933e-07, | |
| "loss": 0.0002, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 1788, | |
| "total_flos": 2.2152891152361554e+18, | |
| "train_loss": 0.005421906567993314, | |
| "train_runtime": 1587.8678, | |
| "train_samples_per_second": 18.001, | |
| "train_steps_per_second": 1.126 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1788, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.2152891152361554e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |