| { | |
| "best_metric": 0.989844278943805, | |
| "best_model_checkpoint": "salon_image_classifier_v1_convnext/checkpoint-1035", | |
| "epoch": 4.981949458483754, | |
| "eval_steps": 500, | |
| "global_step": 1035, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 6.277402877807617, | |
| "learning_rate": 2.403846153846154e-06, | |
| "loss": 1.4624, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 5.703274726867676, | |
| "learning_rate": 4.807692307692308e-06, | |
| "loss": 1.3971, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 7.412254810333252, | |
| "learning_rate": 7.211538461538461e-06, | |
| "loss": 1.3073, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 5.874199867248535, | |
| "learning_rate": 9.615384615384616e-06, | |
| "loss": 1.1877, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 4.8683342933654785, | |
| "learning_rate": 1.2019230769230771e-05, | |
| "loss": 1.0259, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 4.822791576385498, | |
| "learning_rate": 1.4423076923076923e-05, | |
| "loss": 0.8557, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 4.416042804718018, | |
| "learning_rate": 1.682692307692308e-05, | |
| "loss": 0.7254, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 3.968479633331299, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 0.5629, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 3.736450433731079, | |
| "learning_rate": 2.1634615384615387e-05, | |
| "loss": 0.4541, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 3.005037307739258, | |
| "learning_rate": 2.4038461538461542e-05, | |
| "loss": 0.3057, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 4.574312686920166, | |
| "learning_rate": 2.6442307692307694e-05, | |
| "loss": 0.2917, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 5.142052173614502, | |
| "learning_rate": 2.8846153846153845e-05, | |
| "loss": 0.2297, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 3.3775789737701416, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.2004, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 3.128685712814331, | |
| "learning_rate": 3.365384615384616e-05, | |
| "loss": 0.1851, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 5.491213798522949, | |
| "learning_rate": 3.605769230769231e-05, | |
| "loss": 0.1956, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 2.380528688430786, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.161, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 5.325184345245361, | |
| "learning_rate": 4.0865384615384615e-05, | |
| "loss": 0.1748, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 3.9706196784973145, | |
| "learning_rate": 4.326923076923077e-05, | |
| "loss": 0.1487, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 4.022750377655029, | |
| "learning_rate": 4.5673076923076925e-05, | |
| "loss": 0.1339, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 7.611146450042725, | |
| "learning_rate": 4.8076923076923084e-05, | |
| "loss": 0.1544, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 9.541460990905762, | |
| "learning_rate": 4.9946294307196566e-05, | |
| "loss": 0.1224, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 5.275872230529785, | |
| "learning_rate": 4.967776584317938e-05, | |
| "loss": 0.1511, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 2.8402061462402344, | |
| "learning_rate": 4.940923737916219e-05, | |
| "loss": 0.1081, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 6.35424280166626, | |
| "learning_rate": 4.9140708915145005e-05, | |
| "loss": 0.1288, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 6.0777435302734375, | |
| "learning_rate": 4.887218045112782e-05, | |
| "loss": 0.1277, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 3.709442377090454, | |
| "learning_rate": 4.860365198711064e-05, | |
| "loss": 0.136, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 6.1887593269348145, | |
| "learning_rate": 4.833512352309345e-05, | |
| "loss": 0.1758, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 4.514546871185303, | |
| "learning_rate": 4.806659505907626e-05, | |
| "loss": 0.156, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 5.032735824584961, | |
| "learning_rate": 4.7798066595059076e-05, | |
| "loss": 0.1593, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 9.502847671508789, | |
| "learning_rate": 4.7529538131041896e-05, | |
| "loss": 0.2014, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.6441538333892822, | |
| "learning_rate": 4.726100966702471e-05, | |
| "loss": 0.1331, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 2.8725085258483887, | |
| "learning_rate": 4.699248120300752e-05, | |
| "loss": 0.1261, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 2.6660351753234863, | |
| "learning_rate": 4.6723952738990334e-05, | |
| "loss": 0.1097, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 8.11793327331543, | |
| "learning_rate": 4.645542427497315e-05, | |
| "loss": 0.0933, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 3.4776053428649902, | |
| "learning_rate": 4.618689581095597e-05, | |
| "loss": 0.1543, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 3.1535816192626953, | |
| "learning_rate": 4.591836734693878e-05, | |
| "loss": 0.1245, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 4.248240947723389, | |
| "learning_rate": 4.564983888292159e-05, | |
| "loss": 0.1417, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 4.858673572540283, | |
| "learning_rate": 4.5381310418904406e-05, | |
| "loss": 0.1382, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.6954901218414307, | |
| "learning_rate": 4.511278195488722e-05, | |
| "loss": 0.1133, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.9366323947906494, | |
| "learning_rate": 4.484425349087004e-05, | |
| "loss": 0.1037, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 3.420692205429077, | |
| "learning_rate": 4.457572502685285e-05, | |
| "loss": 0.1269, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9834123222748815, | |
| "eval_loss": 0.050108980387449265, | |
| "eval_runtime": 6.0814, | |
| "eval_samples_per_second": 485.741, | |
| "eval_steps_per_second": 15.292, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 3.8513619899749756, | |
| "learning_rate": 4.4307196562835664e-05, | |
| "loss": 0.1179, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 3.5678696632385254, | |
| "learning_rate": 4.403866809881848e-05, | |
| "loss": 0.1067, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 3.045541286468506, | |
| "learning_rate": 4.3770139634801297e-05, | |
| "loss": 0.1014, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 2.9802796840667725, | |
| "learning_rate": 4.350161117078411e-05, | |
| "loss": 0.0839, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 2.974304437637329, | |
| "learning_rate": 4.323308270676692e-05, | |
| "loss": 0.096, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 3.204138994216919, | |
| "learning_rate": 4.2964554242749735e-05, | |
| "loss": 0.088, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 5.182232856750488, | |
| "learning_rate": 4.269602577873255e-05, | |
| "loss": 0.0926, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 2.830148458480835, | |
| "learning_rate": 4.242749731471536e-05, | |
| "loss": 0.0915, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 5.067986011505127, | |
| "learning_rate": 4.215896885069818e-05, | |
| "loss": 0.1051, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 3.488647937774658, | |
| "learning_rate": 4.1890440386680994e-05, | |
| "loss": 0.0948, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.4556047916412354, | |
| "learning_rate": 4.1621911922663806e-05, | |
| "loss": 0.0915, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 4.81022834777832, | |
| "learning_rate": 4.135338345864662e-05, | |
| "loss": 0.1068, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 4.263596057891846, | |
| "learning_rate": 4.108485499462943e-05, | |
| "loss": 0.106, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 2.286785840988159, | |
| "learning_rate": 4.0816326530612245e-05, | |
| "loss": 0.0784, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 3.0982165336608887, | |
| "learning_rate": 4.054779806659506e-05, | |
| "loss": 0.0774, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 3.3483657836914062, | |
| "learning_rate": 4.027926960257787e-05, | |
| "loss": 0.0732, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 3.933288335800171, | |
| "learning_rate": 4.0010741138560684e-05, | |
| "loss": 0.086, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 3.418642282485962, | |
| "learning_rate": 3.97422126745435e-05, | |
| "loss": 0.0803, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 3.975381374359131, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.1137, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 2.6510424613952637, | |
| "learning_rate": 3.920515574650913e-05, | |
| "loss": 0.1206, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 2.9062960147857666, | |
| "learning_rate": 3.893662728249194e-05, | |
| "loss": 0.1042, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 2.369260549545288, | |
| "learning_rate": 3.866809881847476e-05, | |
| "loss": 0.0856, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 5.03189754486084, | |
| "learning_rate": 3.8399570354457575e-05, | |
| "loss": 0.0831, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 4.001619338989258, | |
| "learning_rate": 3.813104189044039e-05, | |
| "loss": 0.0766, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 3.2500596046447754, | |
| "learning_rate": 3.78625134264232e-05, | |
| "loss": 0.0881, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 1.553835391998291, | |
| "learning_rate": 3.759398496240601e-05, | |
| "loss": 0.0714, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 4.914017200469971, | |
| "learning_rate": 3.732545649838883e-05, | |
| "loss": 0.0944, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 2.7451980113983154, | |
| "learning_rate": 3.7056928034371646e-05, | |
| "loss": 0.0733, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 4.39937162399292, | |
| "learning_rate": 3.678839957035446e-05, | |
| "loss": 0.1127, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 3.2537283897399902, | |
| "learning_rate": 3.651987110633727e-05, | |
| "loss": 0.1236, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 4.104055881500244, | |
| "learning_rate": 3.6251342642320084e-05, | |
| "loss": 0.114, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 2.0853748321533203, | |
| "learning_rate": 3.5982814178302904e-05, | |
| "loss": 0.0885, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 2.1333792209625244, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.0901, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 2.9660935401916504, | |
| "learning_rate": 3.544575725026853e-05, | |
| "loss": 0.0753, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 1.7010711431503296, | |
| "learning_rate": 3.517722878625134e-05, | |
| "loss": 0.094, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 3.7029120922088623, | |
| "learning_rate": 3.4908700322234156e-05, | |
| "loss": 0.1148, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 3.4257864952087402, | |
| "learning_rate": 3.4640171858216975e-05, | |
| "loss": 0.0879, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 6.967185020446777, | |
| "learning_rate": 3.437164339419979e-05, | |
| "loss": 0.1001, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 1.414320707321167, | |
| "learning_rate": 3.41031149301826e-05, | |
| "loss": 0.1218, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 1.8158583641052246, | |
| "learning_rate": 3.3834586466165414e-05, | |
| "loss": 0.0646, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "grad_norm": 2.169985055923462, | |
| "learning_rate": 3.3566058002148234e-05, | |
| "loss": 0.0776, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.488377571105957, | |
| "learning_rate": 3.3297529538131046e-05, | |
| "loss": 0.0572, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.987136086662153, | |
| "eval_loss": 0.03736753389239311, | |
| "eval_runtime": 6.0106, | |
| "eval_samples_per_second": 491.469, | |
| "eval_steps_per_second": 15.473, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 3.571716070175171, | |
| "learning_rate": 3.302900107411386e-05, | |
| "loss": 0.083, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 2.3017261028289795, | |
| "learning_rate": 3.276047261009667e-05, | |
| "loss": 0.0581, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 4.040441513061523, | |
| "learning_rate": 3.2491944146079485e-05, | |
| "loss": 0.0762, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 1.5620166063308716, | |
| "learning_rate": 3.2223415682062305e-05, | |
| "loss": 0.0778, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 2.1466052532196045, | |
| "learning_rate": 3.195488721804512e-05, | |
| "loss": 0.0876, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 1.7395237684249878, | |
| "learning_rate": 3.168635875402793e-05, | |
| "loss": 0.0996, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 5.349012851715088, | |
| "learning_rate": 3.1417830290010743e-05, | |
| "loss": 0.072, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 4.39583683013916, | |
| "learning_rate": 3.1149301825993556e-05, | |
| "loss": 0.0997, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 3.541804313659668, | |
| "learning_rate": 3.0880773361976376e-05, | |
| "loss": 0.0712, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 2.4857404232025146, | |
| "learning_rate": 3.061224489795919e-05, | |
| "loss": 0.0789, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 3.7146120071411133, | |
| "learning_rate": 3.0343716433942e-05, | |
| "loss": 0.088, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 2.6535651683807373, | |
| "learning_rate": 3.007518796992481e-05, | |
| "loss": 0.0797, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 1.3184159994125366, | |
| "learning_rate": 2.980665950590763e-05, | |
| "loss": 0.0846, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 2.7791616916656494, | |
| "learning_rate": 2.9538131041890444e-05, | |
| "loss": 0.074, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 4.284481048583984, | |
| "learning_rate": 2.9269602577873257e-05, | |
| "loss": 0.0943, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 3.457939386367798, | |
| "learning_rate": 2.900107411385607e-05, | |
| "loss": 0.117, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 1.2724689245224, | |
| "learning_rate": 2.8732545649838882e-05, | |
| "loss": 0.0803, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 2.6043756008148193, | |
| "learning_rate": 2.8464017185821702e-05, | |
| "loss": 0.0672, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 3.096151113510132, | |
| "learning_rate": 2.8195488721804515e-05, | |
| "loss": 0.0848, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 1.409250259399414, | |
| "learning_rate": 2.7926960257787328e-05, | |
| "loss": 0.0472, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.6517168283462524, | |
| "learning_rate": 2.765843179377014e-05, | |
| "loss": 0.0798, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 2.9068679809570312, | |
| "learning_rate": 2.7389903329752954e-05, | |
| "loss": 0.0658, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 3.5304489135742188, | |
| "learning_rate": 2.712137486573577e-05, | |
| "loss": 0.0512, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 4.021263122558594, | |
| "learning_rate": 2.6852846401718583e-05, | |
| "loss": 0.0889, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.863612413406372, | |
| "learning_rate": 2.6584317937701396e-05, | |
| "loss": 0.0671, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 2.903055191040039, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.071, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 4.729198455810547, | |
| "learning_rate": 2.6047261009667025e-05, | |
| "loss": 0.0753, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 4.479395866394043, | |
| "learning_rate": 2.577873254564984e-05, | |
| "loss": 0.088, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 1.303819179534912, | |
| "learning_rate": 2.5510204081632654e-05, | |
| "loss": 0.0606, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 1.7099323272705078, | |
| "learning_rate": 2.5241675617615467e-05, | |
| "loss": 0.0563, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 6.212317943572998, | |
| "learning_rate": 2.4973147153598283e-05, | |
| "loss": 0.0948, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "grad_norm": 2.4912729263305664, | |
| "learning_rate": 2.4704618689581096e-05, | |
| "loss": 0.0857, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 2.978109359741211, | |
| "learning_rate": 2.443609022556391e-05, | |
| "loss": 0.0976, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 1.6303725242614746, | |
| "learning_rate": 2.4167561761546725e-05, | |
| "loss": 0.0731, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 3.7856266498565674, | |
| "learning_rate": 2.3899033297529538e-05, | |
| "loss": 0.0833, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 3.6364758014678955, | |
| "learning_rate": 2.3630504833512354e-05, | |
| "loss": 0.0659, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 3.412216901779175, | |
| "learning_rate": 2.3361976369495167e-05, | |
| "loss": 0.1227, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 4.180442810058594, | |
| "learning_rate": 2.3093447905477984e-05, | |
| "loss": 0.0773, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 1.9002240896224976, | |
| "learning_rate": 2.2824919441460796e-05, | |
| "loss": 0.0656, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 3.6314845085144043, | |
| "learning_rate": 2.255639097744361e-05, | |
| "loss": 0.0768, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 2.4498918056488037, | |
| "learning_rate": 2.2287862513426426e-05, | |
| "loss": 0.0806, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9874746106973595, | |
| "eval_loss": 0.03359290212392807, | |
| "eval_runtime": 5.9679, | |
| "eval_samples_per_second": 494.983, | |
| "eval_steps_per_second": 15.583, | |
| "step": 623 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 3.5386340618133545, | |
| "learning_rate": 2.201933404940924e-05, | |
| "loss": 0.0704, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 2.9113972187042236, | |
| "learning_rate": 2.1750805585392055e-05, | |
| "loss": 0.0698, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 4.226195812225342, | |
| "learning_rate": 2.1482277121374868e-05, | |
| "loss": 0.0728, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 2.5996837615966797, | |
| "learning_rate": 2.121374865735768e-05, | |
| "loss": 0.0778, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 3.129270076751709, | |
| "learning_rate": 2.0945220193340497e-05, | |
| "loss": 0.0803, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 4.704379081726074, | |
| "learning_rate": 2.067669172932331e-05, | |
| "loss": 0.0605, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 3.2946126461029053, | |
| "learning_rate": 2.0408163265306123e-05, | |
| "loss": 0.0719, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "grad_norm": 3.0328078269958496, | |
| "learning_rate": 2.0139634801288935e-05, | |
| "loss": 0.064, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 3.618300199508667, | |
| "learning_rate": 1.987110633727175e-05, | |
| "loss": 0.0673, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 3.6683876514434814, | |
| "learning_rate": 1.9602577873254565e-05, | |
| "loss": 0.0571, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 1.6800568103790283, | |
| "learning_rate": 1.933404940923738e-05, | |
| "loss": 0.047, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 1.3418190479278564, | |
| "learning_rate": 1.9065520945220194e-05, | |
| "loss": 0.0633, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 2.5490853786468506, | |
| "learning_rate": 1.8796992481203007e-05, | |
| "loss": 0.0402, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 1.2604730129241943, | |
| "learning_rate": 1.8528464017185823e-05, | |
| "loss": 0.0605, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 4.190303325653076, | |
| "learning_rate": 1.8259935553168636e-05, | |
| "loss": 0.0886, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "grad_norm": 7.317651748657227, | |
| "learning_rate": 1.7991407089151452e-05, | |
| "loss": 0.0875, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 1.9360538721084595, | |
| "learning_rate": 1.7722878625134265e-05, | |
| "loss": 0.0557, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 4.210879802703857, | |
| "learning_rate": 1.7454350161117078e-05, | |
| "loss": 0.0735, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 5.373249053955078, | |
| "learning_rate": 1.7185821697099894e-05, | |
| "loss": 0.0975, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "grad_norm": 2.7579221725463867, | |
| "learning_rate": 1.6917293233082707e-05, | |
| "loss": 0.0705, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "grad_norm": 2.673476457595825, | |
| "learning_rate": 1.6648764769065523e-05, | |
| "loss": 0.0524, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 4.691239356994629, | |
| "learning_rate": 1.6380236305048336e-05, | |
| "loss": 0.0686, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 4.022672176361084, | |
| "learning_rate": 1.6111707841031152e-05, | |
| "loss": 0.0632, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "grad_norm": 2.69425630569458, | |
| "learning_rate": 1.5843179377013965e-05, | |
| "loss": 0.0636, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "grad_norm": 4.5648884773254395, | |
| "learning_rate": 1.5574650912996778e-05, | |
| "loss": 0.0627, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 3.1452832221984863, | |
| "learning_rate": 1.5306122448979594e-05, | |
| "loss": 0.0763, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "grad_norm": 5.030516147613525, | |
| "learning_rate": 1.5037593984962406e-05, | |
| "loss": 0.0581, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "grad_norm": 3.2779948711395264, | |
| "learning_rate": 1.4769065520945222e-05, | |
| "loss": 0.0549, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 1.7655272483825684, | |
| "learning_rate": 1.4500537056928035e-05, | |
| "loss": 0.0778, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 3.9161980152130127, | |
| "learning_rate": 1.4232008592910851e-05, | |
| "loss": 0.0773, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "grad_norm": 3.029256820678711, | |
| "learning_rate": 1.3963480128893664e-05, | |
| "loss": 0.056, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 4.829240322113037, | |
| "learning_rate": 1.3694951664876477e-05, | |
| "loss": 0.0722, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "grad_norm": 2.5825934410095215, | |
| "learning_rate": 1.3426423200859291e-05, | |
| "loss": 0.0372, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 1.706741452217102, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.0369, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 2.689628839492798, | |
| "learning_rate": 1.288936627282492e-05, | |
| "loss": 0.0676, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 3.5289785861968994, | |
| "learning_rate": 1.2620837808807733e-05, | |
| "loss": 0.0802, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 3.4799952507019043, | |
| "learning_rate": 1.2352309344790548e-05, | |
| "loss": 0.0589, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 1.8066169023513794, | |
| "learning_rate": 1.2083780880773363e-05, | |
| "loss": 0.0662, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 2.680537700653076, | |
| "learning_rate": 1.1815252416756177e-05, | |
| "loss": 0.0482, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 4.0623555183410645, | |
| "learning_rate": 1.1546723952738992e-05, | |
| "loss": 0.0786, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "grad_norm": 2.9296209812164307, | |
| "learning_rate": 1.1278195488721805e-05, | |
| "loss": 0.0406, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.4645802974700928, | |
| "learning_rate": 1.100966702470462e-05, | |
| "loss": 0.0589, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.988490182802979, | |
| "eval_loss": 0.03382693976163864, | |
| "eval_runtime": 6.0045, | |
| "eval_samples_per_second": 491.966, | |
| "eval_steps_per_second": 15.488, | |
| "step": 831 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 3.1075704097747803, | |
| "learning_rate": 1.0741138560687434e-05, | |
| "loss": 0.0602, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 0.9231489300727844, | |
| "learning_rate": 1.0472610096670248e-05, | |
| "loss": 0.0525, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 1.0160259008407593, | |
| "learning_rate": 1.0204081632653061e-05, | |
| "loss": 0.0409, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 1.9544317722320557, | |
| "learning_rate": 9.935553168635876e-06, | |
| "loss": 0.0504, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "grad_norm": 2.4267749786376953, | |
| "learning_rate": 9.66702470461869e-06, | |
| "loss": 0.059, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 2.689138889312744, | |
| "learning_rate": 9.398496240601503e-06, | |
| "loss": 0.0625, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 1.5462526082992554, | |
| "learning_rate": 9.129967776584318e-06, | |
| "loss": 0.0492, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "grad_norm": 4.044441223144531, | |
| "learning_rate": 8.861439312567132e-06, | |
| "loss": 0.05, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 5.436426162719727, | |
| "learning_rate": 8.592910848549947e-06, | |
| "loss": 0.0455, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 2.8710386753082275, | |
| "learning_rate": 8.324382384532762e-06, | |
| "loss": 0.0645, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "grad_norm": 3.6887547969818115, | |
| "learning_rate": 8.055853920515576e-06, | |
| "loss": 0.0623, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "grad_norm": 3.6889421939849854, | |
| "learning_rate": 7.787325456498389e-06, | |
| "loss": 0.0569, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "grad_norm": 3.734835624694824, | |
| "learning_rate": 7.518796992481203e-06, | |
| "loss": 0.0677, | |
| "step": 895 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "grad_norm": 2.4793076515197754, | |
| "learning_rate": 7.250268528464017e-06, | |
| "loss": 0.0445, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "grad_norm": 5.3167500495910645, | |
| "learning_rate": 6.981740064446832e-06, | |
| "loss": 0.0857, | |
| "step": 905 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "grad_norm": 2.9194352626800537, | |
| "learning_rate": 6.713211600429646e-06, | |
| "loss": 0.0704, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 2.78092885017395, | |
| "learning_rate": 6.44468313641246e-06, | |
| "loss": 0.062, | |
| "step": 915 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "grad_norm": 2.4215569496154785, | |
| "learning_rate": 6.176154672395274e-06, | |
| "loss": 0.063, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "grad_norm": 2.3797125816345215, | |
| "learning_rate": 5.907626208378089e-06, | |
| "loss": 0.052, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 0.7256618142127991, | |
| "learning_rate": 5.639097744360902e-06, | |
| "loss": 0.0468, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 1.283785104751587, | |
| "learning_rate": 5.370569280343717e-06, | |
| "loss": 0.0577, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 4.501251220703125, | |
| "learning_rate": 5.102040816326531e-06, | |
| "loss": 0.0595, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 2.7582945823669434, | |
| "learning_rate": 4.833512352309345e-06, | |
| "loss": 0.0693, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "grad_norm": 5.989513397216797, | |
| "learning_rate": 4.564983888292159e-06, | |
| "loss": 0.0746, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 3.373969554901123, | |
| "learning_rate": 4.2964554242749735e-06, | |
| "loss": 0.0529, | |
| "step": 955 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "grad_norm": 2.1218202114105225, | |
| "learning_rate": 4.027926960257788e-06, | |
| "loss": 0.0368, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 1.9887917041778564, | |
| "learning_rate": 3.7593984962406014e-06, | |
| "loss": 0.0354, | |
| "step": 965 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "grad_norm": 1.6485974788665771, | |
| "learning_rate": 3.490870032223416e-06, | |
| "loss": 0.0653, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "grad_norm": 2.1341357231140137, | |
| "learning_rate": 3.22234156820623e-06, | |
| "loss": 0.0421, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 5.573878765106201, | |
| "learning_rate": 2.9538131041890443e-06, | |
| "loss": 0.0741, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "grad_norm": 3.1527225971221924, | |
| "learning_rate": 2.6852846401718585e-06, | |
| "loss": 0.0327, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "grad_norm": 3.773310661315918, | |
| "learning_rate": 2.4167561761546726e-06, | |
| "loss": 0.0472, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "grad_norm": 3.0205063819885254, | |
| "learning_rate": 2.1482277121374868e-06, | |
| "loss": 0.0723, | |
| "step": 995 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "grad_norm": 3.098895788192749, | |
| "learning_rate": 1.8796992481203007e-06, | |
| "loss": 0.0743, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "grad_norm": 1.734661340713501, | |
| "learning_rate": 1.611170784103115e-06, | |
| "loss": 0.0337, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "grad_norm": 1.7586333751678467, | |
| "learning_rate": 1.3426423200859292e-06, | |
| "loss": 0.062, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "grad_norm": 2.720187187194824, | |
| "learning_rate": 1.0741138560687434e-06, | |
| "loss": 0.0791, | |
| "step": 1015 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "grad_norm": 1.8995425701141357, | |
| "learning_rate": 8.055853920515575e-07, | |
| "loss": 0.061, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "grad_norm": 4.119653224945068, | |
| "learning_rate": 5.370569280343717e-07, | |
| "loss": 0.0486, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 3.3452024459838867, | |
| "learning_rate": 2.6852846401718585e-07, | |
| "loss": 0.0572, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "grad_norm": 1.4279868602752686, | |
| "learning_rate": 0.0, | |
| "loss": 0.0326, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_accuracy": 0.989844278943805, | |
| "eval_loss": 0.029886629432439804, | |
| "eval_runtime": 5.9932, | |
| "eval_samples_per_second": 492.888, | |
| "eval_steps_per_second": 15.517, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "step": 1035, | |
| "total_flos": 3.2925088680822374e+18, | |
| "train_loss": 0.12655677452755434, | |
| "train_runtime": 660.3476, | |
| "train_samples_per_second": 201.303, | |
| "train_steps_per_second": 1.567 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 1035, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 3.2925088680822374e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |