{ "best_metric": 0.04956997185945511, "best_model_checkpoint": "vit_epochs5_batch32_lr5e-05_size224_tiles2_seed2_q1\\checkpoint-1876", "epoch": 5.0, "eval_steps": 500, "global_step": 2345, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.010660980810234541, "grad_norm": 0.004416467621922493, "learning_rate": 4.989339019189766e-05, "loss": 0.0006, "step": 5 }, { "epoch": 0.021321961620469083, "grad_norm": 0.014301459304988384, "learning_rate": 4.978678038379531e-05, "loss": 0.0004, "step": 10 }, { "epoch": 0.031982942430703626, "grad_norm": 0.0017900880193337798, "learning_rate": 4.9680170575692967e-05, "loss": 0.0002, "step": 15 }, { "epoch": 0.042643923240938165, "grad_norm": 0.0016476568998768926, "learning_rate": 4.957356076759062e-05, "loss": 0.0002, "step": 20 }, { "epoch": 0.053304904051172705, "grad_norm": 0.010723981074988842, "learning_rate": 4.9466950959488276e-05, "loss": 0.0003, "step": 25 }, { "epoch": 0.06396588486140725, "grad_norm": 0.012562624178826809, "learning_rate": 4.936034115138593e-05, "loss": 0.0003, "step": 30 }, { "epoch": 0.07462686567164178, "grad_norm": 0.0014855375047773123, "learning_rate": 4.9253731343283586e-05, "loss": 0.0002, "step": 35 }, { "epoch": 0.08528784648187633, "grad_norm": 0.7502368092536926, "learning_rate": 4.914712153518124e-05, "loss": 0.0007, "step": 40 }, { "epoch": 0.09594882729211088, "grad_norm": 0.0016878186725080013, "learning_rate": 4.904051172707889e-05, "loss": 0.0056, "step": 45 }, { "epoch": 0.10660980810234541, "grad_norm": 0.0038182407151907682, "learning_rate": 4.893390191897655e-05, "loss": 0.0002, "step": 50 }, { "epoch": 0.11727078891257996, "grad_norm": 0.002054449636489153, "learning_rate": 4.88272921108742e-05, "loss": 0.0072, "step": 55 }, { "epoch": 0.1279317697228145, "grad_norm": 0.07434427738189697, "learning_rate": 4.872068230277186e-05, "loss": 0.0406, "step": 60 }, { "epoch": 0.13859275053304904, "grad_norm": 13.703219413757324, "learning_rate": 4.861407249466951e-05, "loss": 0.0286, "step": 65 }, { "epoch": 0.14925373134328357, "grad_norm": 0.011812947690486908, "learning_rate": 4.850746268656717e-05, "loss": 0.0022, "step": 70 }, { "epoch": 0.15991471215351813, "grad_norm": 0.00310192140750587, "learning_rate": 4.840085287846482e-05, "loss": 0.0002, "step": 75 }, { "epoch": 0.17057569296375266, "grad_norm": 0.0011352329747751355, "learning_rate": 4.829424307036248e-05, "loss": 0.0012, "step": 80 }, { "epoch": 0.1812366737739872, "grad_norm": 14.11648178100586, "learning_rate": 4.8187633262260126e-05, "loss": 0.0195, "step": 85 }, { "epoch": 0.19189765458422176, "grad_norm": 5.982122898101807, "learning_rate": 4.808102345415779e-05, "loss": 0.003, "step": 90 }, { "epoch": 0.2025586353944563, "grad_norm": 0.01406597439199686, "learning_rate": 4.7974413646055436e-05, "loss": 0.0958, "step": 95 }, { "epoch": 0.21321961620469082, "grad_norm": 5.018512725830078, "learning_rate": 4.78678038379531e-05, "loss": 0.0917, "step": 100 }, { "epoch": 0.22388059701492538, "grad_norm": 0.0013859644532203674, "learning_rate": 4.7761194029850745e-05, "loss": 0.0613, "step": 105 }, { "epoch": 0.2345415778251599, "grad_norm": 0.11947944015264511, "learning_rate": 4.765458422174841e-05, "loss": 0.0377, "step": 110 }, { "epoch": 0.24520255863539445, "grad_norm": 0.003042879980057478, "learning_rate": 4.7547974413646055e-05, "loss": 0.0312, "step": 115 }, { "epoch": 0.255863539445629, "grad_norm": 0.004376358352601528, "learning_rate": 4.7441364605543716e-05, "loss": 0.0002, "step": 120 }, { "epoch": 0.26652452025586354, "grad_norm": 11.286640167236328, "learning_rate": 4.7334754797441364e-05, "loss": 0.0466, "step": 125 }, { "epoch": 0.2771855010660981, "grad_norm": 0.6653345823287964, "learning_rate": 4.7228144989339026e-05, "loss": 0.028, "step": 130 }, { "epoch": 0.2878464818763326, "grad_norm": 3.765774726867676, "learning_rate": 4.7121535181236674e-05, "loss": 0.0524, "step": 135 }, { "epoch": 0.29850746268656714, "grad_norm": 0.00767380278557539, "learning_rate": 4.7014925373134335e-05, "loss": 0.0289, "step": 140 }, { "epoch": 0.3091684434968017, "grad_norm": 0.004249317105859518, "learning_rate": 4.690831556503198e-05, "loss": 0.0308, "step": 145 }, { "epoch": 0.31982942430703626, "grad_norm": 0.008478735573589802, "learning_rate": 4.6801705756929645e-05, "loss": 0.0075, "step": 150 }, { "epoch": 0.3304904051172708, "grad_norm": 5.810318946838379, "learning_rate": 4.669509594882729e-05, "loss": 0.0108, "step": 155 }, { "epoch": 0.3411513859275053, "grad_norm": 0.0037327606696635485, "learning_rate": 4.658848614072495e-05, "loss": 0.0051, "step": 160 }, { "epoch": 0.35181236673773986, "grad_norm": 3.177417516708374, "learning_rate": 4.64818763326226e-05, "loss": 0.057, "step": 165 }, { "epoch": 0.3624733475479744, "grad_norm": 3.029132127761841, "learning_rate": 4.637526652452026e-05, "loss": 0.0054, "step": 170 }, { "epoch": 0.373134328358209, "grad_norm": 0.7287138104438782, "learning_rate": 4.626865671641791e-05, "loss": 0.0356, "step": 175 }, { "epoch": 0.3837953091684435, "grad_norm": 0.6818822026252747, "learning_rate": 4.6162046908315566e-05, "loss": 0.0019, "step": 180 }, { "epoch": 0.39445628997867804, "grad_norm": 0.10914435982704163, "learning_rate": 4.605543710021322e-05, "loss": 0.0018, "step": 185 }, { "epoch": 0.4051172707889126, "grad_norm": 0.007995429448783398, "learning_rate": 4.5948827292110876e-05, "loss": 0.0176, "step": 190 }, { "epoch": 0.4157782515991471, "grad_norm": 0.0048393430188298225, "learning_rate": 4.584221748400853e-05, "loss": 0.0009, "step": 195 }, { "epoch": 0.42643923240938164, "grad_norm": 0.011365322396159172, "learning_rate": 4.5735607675906185e-05, "loss": 0.0107, "step": 200 }, { "epoch": 0.43710021321961623, "grad_norm": 0.005557894706726074, "learning_rate": 4.562899786780384e-05, "loss": 0.0092, "step": 205 }, { "epoch": 0.44776119402985076, "grad_norm": 0.0014229494845494628, "learning_rate": 4.5522388059701495e-05, "loss": 0.0008, "step": 210 }, { "epoch": 0.4584221748400853, "grad_norm": 4.9571709632873535, "learning_rate": 4.541577825159915e-05, "loss": 0.0051, "step": 215 }, { "epoch": 0.4690831556503198, "grad_norm": 0.0018386637093499303, "learning_rate": 4.5309168443496804e-05, "loss": 0.0005, "step": 220 }, { "epoch": 0.47974413646055436, "grad_norm": 0.0016964643727988005, "learning_rate": 4.520255863539446e-05, "loss": 0.0012, "step": 225 }, { "epoch": 0.4904051172707889, "grad_norm": 2.2462761402130127, "learning_rate": 4.5095948827292114e-05, "loss": 0.0018, "step": 230 }, { "epoch": 0.5010660980810234, "grad_norm": 0.06354831159114838, "learning_rate": 4.498933901918977e-05, "loss": 0.0002, "step": 235 }, { "epoch": 0.511727078891258, "grad_norm": 1.238602638244629, "learning_rate": 4.488272921108742e-05, "loss": 0.0013, "step": 240 }, { "epoch": 0.5223880597014925, "grad_norm": 0.04044834151864052, "learning_rate": 4.477611940298508e-05, "loss": 0.0002, "step": 245 }, { "epoch": 0.5330490405117271, "grad_norm": 0.0027133615221828222, "learning_rate": 4.466950959488273e-05, "loss": 0.0092, "step": 250 }, { "epoch": 0.5437100213219617, "grad_norm": 0.007094858679920435, "learning_rate": 4.456289978678039e-05, "loss": 0.0008, "step": 255 }, { "epoch": 0.5543710021321961, "grad_norm": 0.011473302729427814, "learning_rate": 4.445628997867804e-05, "loss": 0.0003, "step": 260 }, { "epoch": 0.5650319829424307, "grad_norm": 0.0030412091873586178, "learning_rate": 4.43496801705757e-05, "loss": 0.0272, "step": 265 }, { "epoch": 0.5756929637526652, "grad_norm": 0.0024586496874690056, "learning_rate": 4.424307036247335e-05, "loss": 0.042, "step": 270 }, { "epoch": 0.5863539445628998, "grad_norm": 10.195365905761719, "learning_rate": 4.4136460554371006e-05, "loss": 0.0118, "step": 275 }, { "epoch": 0.5970149253731343, "grad_norm": 0.005107020493596792, "learning_rate": 4.402985074626866e-05, "loss": 0.0002, "step": 280 }, { "epoch": 0.6076759061833689, "grad_norm": 0.18854892253875732, "learning_rate": 4.3923240938166316e-05, "loss": 0.0273, "step": 285 }, { "epoch": 0.6183368869936035, "grad_norm": 3.8321614265441895, "learning_rate": 4.381663113006397e-05, "loss": 0.0507, "step": 290 }, { "epoch": 0.6289978678038379, "grad_norm": 0.012841924093663692, "learning_rate": 4.3710021321961625e-05, "loss": 0.0003, "step": 295 }, { "epoch": 0.6396588486140725, "grad_norm": 0.025611745193600655, "learning_rate": 4.360341151385928e-05, "loss": 0.0003, "step": 300 }, { "epoch": 0.650319829424307, "grad_norm": 0.22461196780204773, "learning_rate": 4.3496801705756935e-05, "loss": 0.038, "step": 305 }, { "epoch": 0.6609808102345416, "grad_norm": 0.005854998715221882, "learning_rate": 4.339019189765459e-05, "loss": 0.0013, "step": 310 }, { "epoch": 0.6716417910447762, "grad_norm": 0.13320687413215637, "learning_rate": 4.328358208955224e-05, "loss": 0.0006, "step": 315 }, { "epoch": 0.6823027718550106, "grad_norm": 0.0020415931940078735, "learning_rate": 4.31769722814499e-05, "loss": 0.0072, "step": 320 }, { "epoch": 0.6929637526652452, "grad_norm": 0.005883481353521347, "learning_rate": 4.307036247334755e-05, "loss": 0.0006, "step": 325 }, { "epoch": 0.7036247334754797, "grad_norm": 1.5399516820907593, "learning_rate": 4.29637526652452e-05, "loss": 0.055, "step": 330 }, { "epoch": 0.7142857142857143, "grad_norm": 0.011331605724990368, "learning_rate": 4.2857142857142856e-05, "loss": 0.0012, "step": 335 }, { "epoch": 0.7249466950959488, "grad_norm": 0.024298565462231636, "learning_rate": 4.275053304904051e-05, "loss": 0.0232, "step": 340 }, { "epoch": 0.7356076759061834, "grad_norm": 0.08052023500204086, "learning_rate": 4.2643923240938166e-05, "loss": 0.0192, "step": 345 }, { "epoch": 0.746268656716418, "grad_norm": 0.028125345706939697, "learning_rate": 4.253731343283582e-05, "loss": 0.0007, "step": 350 }, { "epoch": 0.7569296375266524, "grad_norm": 0.09456489235162735, "learning_rate": 4.2430703624733475e-05, "loss": 0.0029, "step": 355 }, { "epoch": 0.767590618336887, "grad_norm": 0.004745835904031992, "learning_rate": 4.232409381663113e-05, "loss": 0.0003, "step": 360 }, { "epoch": 0.7782515991471215, "grad_norm": 0.003011461813002825, "learning_rate": 4.2217484008528785e-05, "loss": 0.0003, "step": 365 }, { "epoch": 0.7889125799573561, "grad_norm": 0.6895479559898376, "learning_rate": 4.211087420042644e-05, "loss": 0.0302, "step": 370 }, { "epoch": 0.7995735607675906, "grad_norm": 0.011591597460210323, "learning_rate": 4.2004264392324094e-05, "loss": 0.0003, "step": 375 }, { "epoch": 0.8102345415778252, "grad_norm": 0.0027178144082427025, "learning_rate": 4.189765458422175e-05, "loss": 0.0003, "step": 380 }, { "epoch": 0.8208955223880597, "grad_norm": 0.026510847732424736, "learning_rate": 4.1791044776119404e-05, "loss": 0.0007, "step": 385 }, { "epoch": 0.8315565031982942, "grad_norm": 0.011108385398983955, "learning_rate": 4.168443496801706e-05, "loss": 0.001, "step": 390 }, { "epoch": 0.8422174840085288, "grad_norm": 0.013837439008057117, "learning_rate": 4.157782515991471e-05, "loss": 0.0649, "step": 395 }, { "epoch": 0.8528784648187633, "grad_norm": 0.03926901891827583, "learning_rate": 4.147121535181237e-05, "loss": 0.0011, "step": 400 }, { "epoch": 0.8635394456289979, "grad_norm": 0.0019680205732584, "learning_rate": 4.136460554371002e-05, "loss": 0.0048, "step": 405 }, { "epoch": 0.8742004264392325, "grad_norm": 0.001965946052223444, "learning_rate": 4.125799573560768e-05, "loss": 0.0207, "step": 410 }, { "epoch": 0.8848614072494669, "grad_norm": 5.410464763641357, "learning_rate": 4.115138592750533e-05, "loss": 0.0041, "step": 415 }, { "epoch": 0.8955223880597015, "grad_norm": 0.0015418330440297723, "learning_rate": 4.104477611940299e-05, "loss": 0.0002, "step": 420 }, { "epoch": 0.906183368869936, "grad_norm": 0.004000818356871605, "learning_rate": 4.093816631130064e-05, "loss": 0.0104, "step": 425 }, { "epoch": 0.9168443496801706, "grad_norm": 0.0018001300049945712, "learning_rate": 4.0831556503198296e-05, "loss": 0.0002, "step": 430 }, { "epoch": 0.9275053304904051, "grad_norm": 0.003767742309719324, "learning_rate": 4.072494669509595e-05, "loss": 0.0007, "step": 435 }, { "epoch": 0.9381663113006397, "grad_norm": 0.0016241801204159856, "learning_rate": 4.0618336886993606e-05, "loss": 0.0006, "step": 440 }, { "epoch": 0.9488272921108742, "grad_norm": 0.008778594434261322, "learning_rate": 4.051172707889126e-05, "loss": 0.0005, "step": 445 }, { "epoch": 0.9594882729211087, "grad_norm": 0.03206494823098183, "learning_rate": 4.0405117270788915e-05, "loss": 0.0002, "step": 450 }, { "epoch": 0.9701492537313433, "grad_norm": 0.01336848083883524, "learning_rate": 4.029850746268657e-05, "loss": 0.0003, "step": 455 }, { "epoch": 0.9808102345415778, "grad_norm": 0.003192806150764227, "learning_rate": 4.0191897654584225e-05, "loss": 0.0002, "step": 460 }, { "epoch": 0.9914712153518124, "grad_norm": 2.620249032974243, "learning_rate": 4.008528784648188e-05, "loss": 0.0052, "step": 465 }, { "epoch": 1.0, "eval_accuracy": 0.9866666666666667, "eval_loss": 0.05263223499059677, "eval_runtime": 18.697, "eval_samples_per_second": 200.567, "eval_steps_per_second": 6.311, "step": 469 }, { "epoch": 1.0021321961620469, "grad_norm": 0.0011655178386718035, "learning_rate": 3.997867803837953e-05, "loss": 0.0002, "step": 470 }, { "epoch": 1.0127931769722816, "grad_norm": 0.002307713497430086, "learning_rate": 3.987206823027719e-05, "loss": 0.0002, "step": 475 }, { "epoch": 1.023454157782516, "grad_norm": 0.001209950540214777, "learning_rate": 3.976545842217484e-05, "loss": 0.0029, "step": 480 }, { "epoch": 1.0341151385927505, "grad_norm": 0.0013398226583376527, "learning_rate": 3.96588486140725e-05, "loss": 0.0052, "step": 485 }, { "epoch": 1.044776119402985, "grad_norm": 0.0017092672642320395, "learning_rate": 3.9552238805970146e-05, "loss": 0.0108, "step": 490 }, { "epoch": 1.0554371002132197, "grad_norm": 0.08532125502824783, "learning_rate": 3.944562899786781e-05, "loss": 0.0003, "step": 495 }, { "epoch": 1.0660980810234542, "grad_norm": 0.023287221789360046, "learning_rate": 3.9339019189765456e-05, "loss": 0.0005, "step": 500 }, { "epoch": 1.0767590618336886, "grad_norm": 0.0011618522694334388, "learning_rate": 3.923240938166312e-05, "loss": 0.005, "step": 505 }, { "epoch": 1.0874200426439233, "grad_norm": 0.0068031721748411655, "learning_rate": 3.9125799573560765e-05, "loss": 0.0002, "step": 510 }, { "epoch": 1.0980810234541578, "grad_norm": 0.001230322872288525, "learning_rate": 3.901918976545843e-05, "loss": 0.0002, "step": 515 }, { "epoch": 1.1087420042643923, "grad_norm": 0.0025611193850636482, "learning_rate": 3.8912579957356075e-05, "loss": 0.022, "step": 520 }, { "epoch": 1.1194029850746268, "grad_norm": 0.4923318326473236, "learning_rate": 3.8805970149253736e-05, "loss": 0.0008, "step": 525 }, { "epoch": 1.1300639658848615, "grad_norm": 0.008376955054700375, "learning_rate": 3.8699360341151384e-05, "loss": 0.0001, "step": 530 }, { "epoch": 1.140724946695096, "grad_norm": 0.011232415214180946, "learning_rate": 3.8592750533049046e-05, "loss": 0.0002, "step": 535 }, { "epoch": 1.1513859275053304, "grad_norm": 0.0015362320700660348, "learning_rate": 3.8486140724946694e-05, "loss": 0.0001, "step": 540 }, { "epoch": 1.1620469083155651, "grad_norm": 0.055775176733732224, "learning_rate": 3.8379530916844355e-05, "loss": 0.0015, "step": 545 }, { "epoch": 1.1727078891257996, "grad_norm": 0.0011701452312991023, "learning_rate": 3.8272921108742e-05, "loss": 0.0001, "step": 550 }, { "epoch": 1.183368869936034, "grad_norm": 0.37267643213272095, "learning_rate": 3.8166311300639665e-05, "loss": 0.0006, "step": 555 }, { "epoch": 1.1940298507462686, "grad_norm": 0.001067770877853036, "learning_rate": 3.805970149253731e-05, "loss": 0.0001, "step": 560 }, { "epoch": 1.2046908315565032, "grad_norm": 0.0034073072019964457, "learning_rate": 3.7953091684434974e-05, "loss": 0.0002, "step": 565 }, { "epoch": 1.2153518123667377, "grad_norm": 0.0010213963687419891, "learning_rate": 3.784648187633262e-05, "loss": 0.0001, "step": 570 }, { "epoch": 1.2260127931769722, "grad_norm": 0.0018976283026859164, "learning_rate": 3.7739872068230284e-05, "loss": 0.033, "step": 575 }, { "epoch": 1.236673773987207, "grad_norm": 2.396184206008911, "learning_rate": 3.763326226012793e-05, "loss": 0.0017, "step": 580 }, { "epoch": 1.2473347547974414, "grad_norm": 0.0011223825858905911, "learning_rate": 3.752665245202559e-05, "loss": 0.0016, "step": 585 }, { "epoch": 1.2579957356076759, "grad_norm": 0.0015460484428331256, "learning_rate": 3.742004264392324e-05, "loss": 0.0001, "step": 590 }, { "epoch": 1.2686567164179103, "grad_norm": 0.01579880155622959, "learning_rate": 3.73134328358209e-05, "loss": 0.0228, "step": 595 }, { "epoch": 1.279317697228145, "grad_norm": 0.0013221129775047302, "learning_rate": 3.720682302771855e-05, "loss": 0.0002, "step": 600 }, { "epoch": 1.2899786780383795, "grad_norm": 0.006399068050086498, "learning_rate": 3.710021321961621e-05, "loss": 0.0001, "step": 605 }, { "epoch": 1.3006396588486142, "grad_norm": 0.0011100873816758394, "learning_rate": 3.699360341151386e-05, "loss": 0.0005, "step": 610 }, { "epoch": 1.3113006396588487, "grad_norm": 0.0012131993426010013, "learning_rate": 3.6886993603411515e-05, "loss": 0.0002, "step": 615 }, { "epoch": 1.3219616204690832, "grad_norm": 8.440406799316406, "learning_rate": 3.678038379530917e-05, "loss": 0.0089, "step": 620 }, { "epoch": 1.3326226012793176, "grad_norm": 0.001325496006757021, "learning_rate": 3.6673773987206824e-05, "loss": 0.0033, "step": 625 }, { "epoch": 1.3432835820895521, "grad_norm": 0.0011499589309096336, "learning_rate": 3.656716417910448e-05, "loss": 0.0002, "step": 630 }, { "epoch": 1.3539445628997868, "grad_norm": 0.0013248117174953222, "learning_rate": 3.6460554371002134e-05, "loss": 0.0006, "step": 635 }, { "epoch": 1.3646055437100213, "grad_norm": 0.0013559755170717835, "learning_rate": 3.635394456289979e-05, "loss": 0.0001, "step": 640 }, { "epoch": 1.375266524520256, "grad_norm": 0.002247630851343274, "learning_rate": 3.624733475479744e-05, "loss": 0.0001, "step": 645 }, { "epoch": 1.3859275053304905, "grad_norm": 0.30137932300567627, "learning_rate": 3.61407249466951e-05, "loss": 0.0005, "step": 650 }, { "epoch": 1.396588486140725, "grad_norm": 0.004751140717417002, "learning_rate": 3.603411513859275e-05, "loss": 0.0002, "step": 655 }, { "epoch": 1.4072494669509594, "grad_norm": 0.004453397821635008, "learning_rate": 3.592750533049041e-05, "loss": 0.0001, "step": 660 }, { "epoch": 1.417910447761194, "grad_norm": 0.1591702252626419, "learning_rate": 3.582089552238806e-05, "loss": 0.0003, "step": 665 }, { "epoch": 1.4285714285714286, "grad_norm": 0.11763138324022293, "learning_rate": 3.571428571428572e-05, "loss": 0.0002, "step": 670 }, { "epoch": 1.439232409381663, "grad_norm": 0.0011429776204749942, "learning_rate": 3.560767590618337e-05, "loss": 0.0001, "step": 675 }, { "epoch": 1.4498933901918978, "grad_norm": 0.0010695322416722775, "learning_rate": 3.5501066098081026e-05, "loss": 0.015, "step": 680 }, { "epoch": 1.4605543710021323, "grad_norm": 0.0012979171006008983, "learning_rate": 3.539445628997868e-05, "loss": 0.0001, "step": 685 }, { "epoch": 1.4712153518123667, "grad_norm": 0.5231314897537231, "learning_rate": 3.5287846481876336e-05, "loss": 0.0007, "step": 690 }, { "epoch": 1.4818763326226012, "grad_norm": 0.04535778611898422, "learning_rate": 3.518123667377399e-05, "loss": 0.0002, "step": 695 }, { "epoch": 1.4925373134328357, "grad_norm": 3.754040479660034, "learning_rate": 3.5074626865671645e-05, "loss": 0.0144, "step": 700 }, { "epoch": 1.5031982942430704, "grad_norm": 0.0009636252070777118, "learning_rate": 3.496801705756929e-05, "loss": 0.0003, "step": 705 }, { "epoch": 1.5138592750533049, "grad_norm": 0.0009455446852371097, "learning_rate": 3.4861407249466955e-05, "loss": 0.0001, "step": 710 }, { "epoch": 1.5245202558635396, "grad_norm": 0.0010165321873500943, "learning_rate": 3.47547974413646e-05, "loss": 0.0003, "step": 715 }, { "epoch": 1.535181236673774, "grad_norm": 0.0010783136822283268, "learning_rate": 3.4648187633262264e-05, "loss": 0.0002, "step": 720 }, { "epoch": 1.5458422174840085, "grad_norm": 0.017064280807971954, "learning_rate": 3.454157782515991e-05, "loss": 0.0003, "step": 725 }, { "epoch": 1.556503198294243, "grad_norm": 5.9620137214660645, "learning_rate": 3.4434968017057574e-05, "loss": 0.023, "step": 730 }, { "epoch": 1.5671641791044775, "grad_norm": 0.0008880856912583113, "learning_rate": 3.432835820895522e-05, "loss": 0.0002, "step": 735 }, { "epoch": 1.5778251599147122, "grad_norm": 0.0012469749199226499, "learning_rate": 3.422174840085288e-05, "loss": 0.0001, "step": 740 }, { "epoch": 1.5884861407249466, "grad_norm": 0.0010372394463047385, "learning_rate": 3.411513859275053e-05, "loss": 0.0001, "step": 745 }, { "epoch": 1.5991471215351813, "grad_norm": 0.0012505275662988424, "learning_rate": 3.400852878464819e-05, "loss": 0.0098, "step": 750 }, { "epoch": 1.6098081023454158, "grad_norm": 0.000909927417524159, "learning_rate": 3.390191897654584e-05, "loss": 0.0001, "step": 755 }, { "epoch": 1.6204690831556503, "grad_norm": 0.0009265166008844972, "learning_rate": 3.37953091684435e-05, "loss": 0.0001, "step": 760 }, { "epoch": 1.6311300639658848, "grad_norm": 0.0024810717441141605, "learning_rate": 3.368869936034115e-05, "loss": 0.0001, "step": 765 }, { "epoch": 1.6417910447761193, "grad_norm": 0.0009717804496176541, "learning_rate": 3.358208955223881e-05, "loss": 0.0235, "step": 770 }, { "epoch": 1.652452025586354, "grad_norm": 0.0009410977945663035, "learning_rate": 3.347547974413646e-05, "loss": 0.0002, "step": 775 }, { "epoch": 1.6631130063965884, "grad_norm": 0.0008978047408163548, "learning_rate": 3.336886993603412e-05, "loss": 0.0001, "step": 780 }, { "epoch": 1.6737739872068231, "grad_norm": 0.0012250030413269997, "learning_rate": 3.326226012793177e-05, "loss": 0.0012, "step": 785 }, { "epoch": 1.6844349680170576, "grad_norm": 0.0008970822673290968, "learning_rate": 3.3155650319829424e-05, "loss": 0.0001, "step": 790 }, { "epoch": 1.695095948827292, "grad_norm": 0.0009224070818163455, "learning_rate": 3.304904051172708e-05, "loss": 0.0001, "step": 795 }, { "epoch": 1.7057569296375266, "grad_norm": 0.021367406472563744, "learning_rate": 3.294243070362473e-05, "loss": 0.0004, "step": 800 }, { "epoch": 1.716417910447761, "grad_norm": 0.04242429509758949, "learning_rate": 3.283582089552239e-05, "loss": 0.0247, "step": 805 }, { "epoch": 1.7270788912579957, "grad_norm": 0.0014991400530561805, "learning_rate": 3.272921108742004e-05, "loss": 0.0106, "step": 810 }, { "epoch": 1.7377398720682304, "grad_norm": 0.06422313302755356, "learning_rate": 3.26226012793177e-05, "loss": 0.0007, "step": 815 }, { "epoch": 1.748400852878465, "grad_norm": 0.5072115659713745, "learning_rate": 3.251599147121535e-05, "loss": 0.0005, "step": 820 }, { "epoch": 1.7590618336886994, "grad_norm": 0.0009353103814646602, "learning_rate": 3.240938166311301e-05, "loss": 0.0001, "step": 825 }, { "epoch": 1.7697228144989339, "grad_norm": 0.0009094303823076189, "learning_rate": 3.230277185501066e-05, "loss": 0.0002, "step": 830 }, { "epoch": 1.7803837953091683, "grad_norm": 0.0011074618669226766, "learning_rate": 3.2196162046908317e-05, "loss": 0.0001, "step": 835 }, { "epoch": 1.7910447761194028, "grad_norm": 0.0009036573464982212, "learning_rate": 3.208955223880597e-05, "loss": 0.0001, "step": 840 }, { "epoch": 1.8017057569296375, "grad_norm": 0.002437415299937129, "learning_rate": 3.1982942430703626e-05, "loss": 0.0004, "step": 845 }, { "epoch": 1.8123667377398722, "grad_norm": 0.0009672676678746939, "learning_rate": 3.187633262260128e-05, "loss": 0.0001, "step": 850 }, { "epoch": 1.8230277185501067, "grad_norm": 0.000848081021104008, "learning_rate": 3.1769722814498935e-05, "loss": 0.0003, "step": 855 }, { "epoch": 1.8336886993603412, "grad_norm": 0.0010275078238919377, "learning_rate": 3.166311300639659e-05, "loss": 0.0001, "step": 860 }, { "epoch": 1.8443496801705757, "grad_norm": 0.0009115219581872225, "learning_rate": 3.1556503198294245e-05, "loss": 0.0001, "step": 865 }, { "epoch": 1.8550106609808101, "grad_norm": 0.0008501888951286674, "learning_rate": 3.14498933901919e-05, "loss": 0.0004, "step": 870 }, { "epoch": 1.8656716417910446, "grad_norm": 0.0009383116266690195, "learning_rate": 3.1343283582089554e-05, "loss": 0.0002, "step": 875 }, { "epoch": 1.8763326226012793, "grad_norm": 0.001120554399676621, "learning_rate": 3.123667377398721e-05, "loss": 0.0002, "step": 880 }, { "epoch": 1.886993603411514, "grad_norm": 0.0010722005972638726, "learning_rate": 3.1130063965884864e-05, "loss": 0.0001, "step": 885 }, { "epoch": 1.8976545842217485, "grad_norm": 0.0031061009503901005, "learning_rate": 3.102345415778252e-05, "loss": 0.0001, "step": 890 }, { "epoch": 1.908315565031983, "grad_norm": 0.0008591653313487768, "learning_rate": 3.0916844349680173e-05, "loss": 0.0001, "step": 895 }, { "epoch": 1.9189765458422174, "grad_norm": 0.0007861793274059892, "learning_rate": 3.081023454157783e-05, "loss": 0.0001, "step": 900 }, { "epoch": 1.929637526652452, "grad_norm": 0.0008735244628041983, "learning_rate": 3.070362473347548e-05, "loss": 0.0001, "step": 905 }, { "epoch": 1.9402985074626866, "grad_norm": 0.17977739870548248, "learning_rate": 3.059701492537314e-05, "loss": 0.0079, "step": 910 }, { "epoch": 1.950959488272921, "grad_norm": 0.0009820734849199653, "learning_rate": 3.0490405117270792e-05, "loss": 0.0001, "step": 915 }, { "epoch": 1.9616204690831558, "grad_norm": 0.1403326541185379, "learning_rate": 3.0383795309168444e-05, "loss": 0.0002, "step": 920 }, { "epoch": 1.9722814498933903, "grad_norm": 0.0009738589869812131, "learning_rate": 3.0277185501066102e-05, "loss": 0.0001, "step": 925 }, { "epoch": 1.9829424307036247, "grad_norm": 0.0007946694968268275, "learning_rate": 3.0170575692963753e-05, "loss": 0.0339, "step": 930 }, { "epoch": 1.9936034115138592, "grad_norm": 0.0010705440072342753, "learning_rate": 3.006396588486141e-05, "loss": 0.0002, "step": 935 }, { "epoch": 2.0, "eval_accuracy": 0.988, "eval_loss": 0.0637059360742569, "eval_runtime": 17.4485, "eval_samples_per_second": 214.918, "eval_steps_per_second": 6.763, "step": 938 }, { "epoch": 2.0042643923240937, "grad_norm": 0.0010233144275844097, "learning_rate": 2.9957356076759063e-05, "loss": 0.0001, "step": 940 }, { "epoch": 2.014925373134328, "grad_norm": 0.0008873817860148847, "learning_rate": 2.9850746268656714e-05, "loss": 0.0001, "step": 945 }, { "epoch": 2.025586353944563, "grad_norm": 0.15635471045970917, "learning_rate": 2.9744136460554372e-05, "loss": 0.0003, "step": 950 }, { "epoch": 2.0362473347547976, "grad_norm": 0.0008484420250169933, "learning_rate": 2.9637526652452023e-05, "loss": 0.0439, "step": 955 }, { "epoch": 2.046908315565032, "grad_norm": 0.0007956254994496703, "learning_rate": 2.953091684434968e-05, "loss": 0.0001, "step": 960 }, { "epoch": 2.0575692963752665, "grad_norm": 0.0015903839375823736, "learning_rate": 2.9424307036247333e-05, "loss": 0.0001, "step": 965 }, { "epoch": 2.068230277185501, "grad_norm": 0.008548264391720295, "learning_rate": 2.931769722814499e-05, "loss": 0.0001, "step": 970 }, { "epoch": 2.0788912579957355, "grad_norm": 0.02761167287826538, "learning_rate": 2.9211087420042642e-05, "loss": 0.0001, "step": 975 }, { "epoch": 2.08955223880597, "grad_norm": 0.0008219302981160581, "learning_rate": 2.91044776119403e-05, "loss": 0.0001, "step": 980 }, { "epoch": 2.100213219616205, "grad_norm": 0.026593640446662903, "learning_rate": 2.8997867803837952e-05, "loss": 0.0001, "step": 985 }, { "epoch": 2.1108742004264394, "grad_norm": 0.0008629567455500364, "learning_rate": 2.889125799573561e-05, "loss": 0.0322, "step": 990 }, { "epoch": 2.121535181236674, "grad_norm": 0.0008141687721945345, "learning_rate": 2.878464818763326e-05, "loss": 0.0001, "step": 995 }, { "epoch": 2.1321961620469083, "grad_norm": 0.002493669046089053, "learning_rate": 2.867803837953092e-05, "loss": 0.0003, "step": 1000 }, { "epoch": 2.142857142857143, "grad_norm": 0.0009892338421195745, "learning_rate": 2.857142857142857e-05, "loss": 0.0001, "step": 1005 }, { "epoch": 2.1535181236673773, "grad_norm": 7.557299613952637, "learning_rate": 2.846481876332623e-05, "loss": 0.0253, "step": 1010 }, { "epoch": 2.1641791044776117, "grad_norm": 0.000928084715269506, "learning_rate": 2.835820895522388e-05, "loss": 0.0001, "step": 1015 }, { "epoch": 2.1748400852878467, "grad_norm": 0.0019017774611711502, "learning_rate": 2.825159914712154e-05, "loss": 0.0001, "step": 1020 }, { "epoch": 2.185501066098081, "grad_norm": 0.002865538001060486, "learning_rate": 2.814498933901919e-05, "loss": 0.0002, "step": 1025 }, { "epoch": 2.1961620469083156, "grad_norm": 0.0010394201381132007, "learning_rate": 2.8038379530916848e-05, "loss": 0.0002, "step": 1030 }, { "epoch": 2.20682302771855, "grad_norm": 0.0007758212159387767, "learning_rate": 2.79317697228145e-05, "loss": 0.0001, "step": 1035 }, { "epoch": 2.2174840085287846, "grad_norm": 0.0008314052829518914, "learning_rate": 2.7825159914712157e-05, "loss": 0.0003, "step": 1040 }, { "epoch": 2.228144989339019, "grad_norm": 0.000982688507065177, "learning_rate": 2.771855010660981e-05, "loss": 0.0009, "step": 1045 }, { "epoch": 2.2388059701492535, "grad_norm": 0.12322326004505157, "learning_rate": 2.7611940298507467e-05, "loss": 0.0002, "step": 1050 }, { "epoch": 2.2494669509594885, "grad_norm": 0.5201085805892944, "learning_rate": 2.7505330490405118e-05, "loss": 0.0005, "step": 1055 }, { "epoch": 2.260127931769723, "grad_norm": 0.0016427789814770222, "learning_rate": 2.7398720682302776e-05, "loss": 0.0133, "step": 1060 }, { "epoch": 2.2707889125799574, "grad_norm": 0.0008199212606996298, "learning_rate": 2.7292110874200428e-05, "loss": 0.0001, "step": 1065 }, { "epoch": 2.281449893390192, "grad_norm": 0.0007976813358254731, "learning_rate": 2.7185501066098086e-05, "loss": 0.0001, "step": 1070 }, { "epoch": 2.2921108742004264, "grad_norm": 0.0008119718404486775, "learning_rate": 2.7078891257995737e-05, "loss": 0.0001, "step": 1075 }, { "epoch": 2.302771855010661, "grad_norm": 0.0007761920569464564, "learning_rate": 2.6972281449893395e-05, "loss": 0.0001, "step": 1080 }, { "epoch": 2.3134328358208958, "grad_norm": 0.0008879926754161716, "learning_rate": 2.6865671641791047e-05, "loss": 0.0002, "step": 1085 }, { "epoch": 2.3240938166311302, "grad_norm": 0.0008000338566489518, "learning_rate": 2.6759061833688705e-05, "loss": 0.0001, "step": 1090 }, { "epoch": 2.3347547974413647, "grad_norm": 0.0008171437657438219, "learning_rate": 2.6652452025586356e-05, "loss": 0.0003, "step": 1095 }, { "epoch": 2.345415778251599, "grad_norm": 0.0018164804205298424, "learning_rate": 2.6545842217484007e-05, "loss": 0.0002, "step": 1100 }, { "epoch": 2.3560767590618337, "grad_norm": 0.0008001861278899014, "learning_rate": 2.6439232409381666e-05, "loss": 0.0001, "step": 1105 }, { "epoch": 2.366737739872068, "grad_norm": 0.0018397425301373005, "learning_rate": 2.6332622601279317e-05, "loss": 0.0032, "step": 1110 }, { "epoch": 2.3773987206823026, "grad_norm": 0.0013512029545381665, "learning_rate": 2.6226012793176975e-05, "loss": 0.0001, "step": 1115 }, { "epoch": 2.388059701492537, "grad_norm": 0.004570353776216507, "learning_rate": 2.6119402985074626e-05, "loss": 0.0001, "step": 1120 }, { "epoch": 2.398720682302772, "grad_norm": 0.03125890716910362, "learning_rate": 2.6012793176972285e-05, "loss": 0.0001, "step": 1125 }, { "epoch": 2.4093816631130065, "grad_norm": 0.0008710200199857354, "learning_rate": 2.5906183368869936e-05, "loss": 0.0001, "step": 1130 }, { "epoch": 2.420042643923241, "grad_norm": 0.0007381685427390039, "learning_rate": 2.5799573560767594e-05, "loss": 0.043, "step": 1135 }, { "epoch": 2.4307036247334755, "grad_norm": 0.000781903974711895, "learning_rate": 2.5692963752665245e-05, "loss": 0.0001, "step": 1140 }, { "epoch": 2.44136460554371, "grad_norm": 0.0008003533584997058, "learning_rate": 2.5586353944562904e-05, "loss": 0.0001, "step": 1145 }, { "epoch": 2.4520255863539444, "grad_norm": 0.0024352369364351034, "learning_rate": 2.5479744136460555e-05, "loss": 0.0001, "step": 1150 }, { "epoch": 2.4626865671641793, "grad_norm": 0.0007893823203630745, "learning_rate": 2.537313432835821e-05, "loss": 0.0001, "step": 1155 }, { "epoch": 2.473347547974414, "grad_norm": 0.0008541466086171567, "learning_rate": 2.5266524520255864e-05, "loss": 0.0044, "step": 1160 }, { "epoch": 2.4840085287846483, "grad_norm": 0.0007681630668230355, "learning_rate": 2.515991471215352e-05, "loss": 0.0001, "step": 1165 }, { "epoch": 2.4946695095948828, "grad_norm": 0.0007641507545486093, "learning_rate": 2.5053304904051174e-05, "loss": 0.0001, "step": 1170 }, { "epoch": 2.5053304904051172, "grad_norm": 0.0016147271962836385, "learning_rate": 2.494669509594883e-05, "loss": 0.0001, "step": 1175 }, { "epoch": 2.5159914712153517, "grad_norm": 0.0031604429241269827, "learning_rate": 2.4840085287846483e-05, "loss": 0.0001, "step": 1180 }, { "epoch": 2.526652452025586, "grad_norm": 0.0007276069954968989, "learning_rate": 2.4733475479744138e-05, "loss": 0.0021, "step": 1185 }, { "epoch": 2.5373134328358207, "grad_norm": 0.01055662240833044, "learning_rate": 2.4626865671641793e-05, "loss": 0.0001, "step": 1190 }, { "epoch": 2.5479744136460556, "grad_norm": 0.0007680644630454481, "learning_rate": 2.4520255863539444e-05, "loss": 0.0001, "step": 1195 }, { "epoch": 2.55863539445629, "grad_norm": 0.0007043792866170406, "learning_rate": 2.44136460554371e-05, "loss": 0.0001, "step": 1200 }, { "epoch": 2.5692963752665245, "grad_norm": 0.0007545754197053611, "learning_rate": 2.4307036247334754e-05, "loss": 0.0001, "step": 1205 }, { "epoch": 2.579957356076759, "grad_norm": 0.0007419605972245336, "learning_rate": 2.420042643923241e-05, "loss": 0.0345, "step": 1210 }, { "epoch": 2.5906183368869935, "grad_norm": 0.0007108774152584374, "learning_rate": 2.4093816631130063e-05, "loss": 0.0001, "step": 1215 }, { "epoch": 2.6012793176972284, "grad_norm": 0.0007648534374311566, "learning_rate": 2.3987206823027718e-05, "loss": 0.0001, "step": 1220 }, { "epoch": 2.611940298507463, "grad_norm": 0.0007066840771585703, "learning_rate": 2.3880597014925373e-05, "loss": 0.0001, "step": 1225 }, { "epoch": 2.6226012793176974, "grad_norm": 0.0007613915367983282, "learning_rate": 2.3773987206823027e-05, "loss": 0.0001, "step": 1230 }, { "epoch": 2.633262260127932, "grad_norm": 0.0007352712564170361, "learning_rate": 2.3667377398720682e-05, "loss": 0.0001, "step": 1235 }, { "epoch": 2.6439232409381663, "grad_norm": 0.20012161135673523, "learning_rate": 2.3560767590618337e-05, "loss": 0.0056, "step": 1240 }, { "epoch": 2.654584221748401, "grad_norm": 0.0010511945001780987, "learning_rate": 2.345415778251599e-05, "loss": 0.0001, "step": 1245 }, { "epoch": 2.6652452025586353, "grad_norm": 0.001747147529385984, "learning_rate": 2.3347547974413646e-05, "loss": 0.0006, "step": 1250 }, { "epoch": 2.6759061833688698, "grad_norm": 0.0008538540569134057, "learning_rate": 2.32409381663113e-05, "loss": 0.0001, "step": 1255 }, { "epoch": 2.6865671641791042, "grad_norm": 0.000841671135276556, "learning_rate": 2.3134328358208956e-05, "loss": 0.0001, "step": 1260 }, { "epoch": 2.697228144989339, "grad_norm": 0.028351960703730583, "learning_rate": 2.302771855010661e-05, "loss": 0.0001, "step": 1265 }, { "epoch": 2.7078891257995736, "grad_norm": 0.0007217360544018447, "learning_rate": 2.2921108742004265e-05, "loss": 0.0001, "step": 1270 }, { "epoch": 2.718550106609808, "grad_norm": 0.0023344834335148335, "learning_rate": 2.281449893390192e-05, "loss": 0.0001, "step": 1275 }, { "epoch": 2.7292110874200426, "grad_norm": 0.05510537326335907, "learning_rate": 2.2707889125799575e-05, "loss": 0.0001, "step": 1280 }, { "epoch": 2.739872068230277, "grad_norm": 0.0008969557820819318, "learning_rate": 2.260127931769723e-05, "loss": 0.0001, "step": 1285 }, { "epoch": 2.750533049040512, "grad_norm": 0.0007515604957006872, "learning_rate": 2.2494669509594884e-05, "loss": 0.0001, "step": 1290 }, { "epoch": 2.7611940298507465, "grad_norm": 0.0014416479971259832, "learning_rate": 2.238805970149254e-05, "loss": 0.0001, "step": 1295 }, { "epoch": 2.771855010660981, "grad_norm": 0.0006909223739057779, "learning_rate": 2.2281449893390194e-05, "loss": 0.0001, "step": 1300 }, { "epoch": 2.7825159914712154, "grad_norm": 0.0010874925646930933, "learning_rate": 2.217484008528785e-05, "loss": 0.0001, "step": 1305 }, { "epoch": 2.79317697228145, "grad_norm": 0.0007274537929333746, "learning_rate": 2.2068230277185503e-05, "loss": 0.0001, "step": 1310 }, { "epoch": 2.8038379530916844, "grad_norm": 0.0007896247552707791, "learning_rate": 2.1961620469083158e-05, "loss": 0.0001, "step": 1315 }, { "epoch": 2.814498933901919, "grad_norm": 0.0006800142582505941, "learning_rate": 2.1855010660980813e-05, "loss": 0.0001, "step": 1320 }, { "epoch": 2.8251599147121533, "grad_norm": 0.0010969514260068536, "learning_rate": 2.1748400852878467e-05, "loss": 0.0007, "step": 1325 }, { "epoch": 2.835820895522388, "grad_norm": 0.001090347534045577, "learning_rate": 2.164179104477612e-05, "loss": 0.0001, "step": 1330 }, { "epoch": 2.8464818763326227, "grad_norm": 0.00072823854861781, "learning_rate": 2.1535181236673773e-05, "loss": 0.0001, "step": 1335 }, { "epoch": 2.857142857142857, "grad_norm": 0.0009573057759553194, "learning_rate": 2.1428571428571428e-05, "loss": 0.0001, "step": 1340 }, { "epoch": 2.8678038379530917, "grad_norm": 0.0007056166650727391, "learning_rate": 2.1321961620469083e-05, "loss": 0.0001, "step": 1345 }, { "epoch": 2.878464818763326, "grad_norm": 0.0007086369441822171, "learning_rate": 2.1215351812366738e-05, "loss": 0.0001, "step": 1350 }, { "epoch": 2.8891257995735606, "grad_norm": 0.0007490345742553473, "learning_rate": 2.1108742004264392e-05, "loss": 0.0003, "step": 1355 }, { "epoch": 2.8997867803837956, "grad_norm": 0.0014905696734786034, "learning_rate": 2.1002132196162047e-05, "loss": 0.0083, "step": 1360 }, { "epoch": 2.91044776119403, "grad_norm": 0.001380059402436018, "learning_rate": 2.0895522388059702e-05, "loss": 0.0001, "step": 1365 }, { "epoch": 2.9211087420042645, "grad_norm": 0.0008202097960747778, "learning_rate": 2.0788912579957357e-05, "loss": 0.0001, "step": 1370 }, { "epoch": 2.931769722814499, "grad_norm": 0.029673615470528603, "learning_rate": 2.068230277185501e-05, "loss": 0.0001, "step": 1375 }, { "epoch": 2.9424307036247335, "grad_norm": 0.0006965553038753569, "learning_rate": 2.0575692963752666e-05, "loss": 0.0001, "step": 1380 }, { "epoch": 2.953091684434968, "grad_norm": 0.003664041869342327, "learning_rate": 2.046908315565032e-05, "loss": 0.0001, "step": 1385 }, { "epoch": 2.9637526652452024, "grad_norm": 0.001040776027366519, "learning_rate": 2.0362473347547976e-05, "loss": 0.0001, "step": 1390 }, { "epoch": 2.974413646055437, "grad_norm": 0.0012956162681803107, "learning_rate": 2.025586353944563e-05, "loss": 0.0001, "step": 1395 }, { "epoch": 2.9850746268656714, "grad_norm": 0.0007179524400271475, "learning_rate": 2.0149253731343285e-05, "loss": 0.0001, "step": 1400 }, { "epoch": 2.9957356076759063, "grad_norm": 0.0007025228114798665, "learning_rate": 2.004264392324094e-05, "loss": 0.0281, "step": 1405 }, { "epoch": 3.0, "eval_accuracy": 0.9912, "eval_loss": 0.051877688616514206, "eval_runtime": 17.6181, "eval_samples_per_second": 212.85, "eval_steps_per_second": 6.698, "step": 1407 }, { "epoch": 3.0063965884861408, "grad_norm": 0.0008881425019353628, "learning_rate": 1.9936034115138594e-05, "loss": 0.0001, "step": 1410 }, { "epoch": 3.0170575692963753, "grad_norm": 0.0006929202936589718, "learning_rate": 1.982942430703625e-05, "loss": 0.0001, "step": 1415 }, { "epoch": 3.0277185501066097, "grad_norm": 0.0014414478791877627, "learning_rate": 1.9722814498933904e-05, "loss": 0.0001, "step": 1420 }, { "epoch": 3.038379530916844, "grad_norm": 0.0006831750506535172, "learning_rate": 1.961620469083156e-05, "loss": 0.0001, "step": 1425 }, { "epoch": 3.0490405117270787, "grad_norm": 0.0011796592734754086, "learning_rate": 1.9509594882729213e-05, "loss": 0.0001, "step": 1430 }, { "epoch": 3.0597014925373136, "grad_norm": 0.0013477298198267817, "learning_rate": 1.9402985074626868e-05, "loss": 0.0046, "step": 1435 }, { "epoch": 3.070362473347548, "grad_norm": 0.0031290818005800247, "learning_rate": 1.9296375266524523e-05, "loss": 0.0001, "step": 1440 }, { "epoch": 3.0810234541577826, "grad_norm": 0.0006676053162664175, "learning_rate": 1.9189765458422178e-05, "loss": 0.0001, "step": 1445 }, { "epoch": 3.091684434968017, "grad_norm": 0.000660720921587199, "learning_rate": 1.9083155650319832e-05, "loss": 0.0001, "step": 1450 }, { "epoch": 3.1023454157782515, "grad_norm": 0.0006679501966573298, "learning_rate": 1.8976545842217487e-05, "loss": 0.0001, "step": 1455 }, { "epoch": 3.113006396588486, "grad_norm": 0.001204867148771882, "learning_rate": 1.8869936034115142e-05, "loss": 0.0001, "step": 1460 }, { "epoch": 3.1236673773987205, "grad_norm": 0.0007060498464852571, "learning_rate": 1.8763326226012797e-05, "loss": 0.0001, "step": 1465 }, { "epoch": 3.1343283582089554, "grad_norm": 0.0007371944957412779, "learning_rate": 1.865671641791045e-05, "loss": 0.0001, "step": 1470 }, { "epoch": 3.14498933901919, "grad_norm": 0.0008442083490081131, "learning_rate": 1.8550106609808106e-05, "loss": 0.0001, "step": 1475 }, { "epoch": 3.1556503198294243, "grad_norm": 0.002831502119079232, "learning_rate": 1.8443496801705757e-05, "loss": 0.0028, "step": 1480 }, { "epoch": 3.166311300639659, "grad_norm": 0.0008311690762639046, "learning_rate": 1.8336886993603412e-05, "loss": 0.0001, "step": 1485 }, { "epoch": 3.1769722814498933, "grad_norm": 0.0006683879182673991, "learning_rate": 1.8230277185501067e-05, "loss": 0.0001, "step": 1490 }, { "epoch": 3.1876332622601278, "grad_norm": 0.0036641780752688646, "learning_rate": 1.812366737739872e-05, "loss": 0.0001, "step": 1495 }, { "epoch": 3.1982942430703627, "grad_norm": 0.0006455176044255495, "learning_rate": 1.8017057569296376e-05, "loss": 0.0001, "step": 1500 }, { "epoch": 3.208955223880597, "grad_norm": 0.0007483024382963777, "learning_rate": 1.791044776119403e-05, "loss": 0.0001, "step": 1505 }, { "epoch": 3.2196162046908317, "grad_norm": 0.0015509811928495765, "learning_rate": 1.7803837953091686e-05, "loss": 0.0001, "step": 1510 }, { "epoch": 3.230277185501066, "grad_norm": 0.0006412238581106067, "learning_rate": 1.769722814498934e-05, "loss": 0.0001, "step": 1515 }, { "epoch": 3.2409381663113006, "grad_norm": 0.0008321650675497949, "learning_rate": 1.7590618336886995e-05, "loss": 0.0001, "step": 1520 }, { "epoch": 3.251599147121535, "grad_norm": 0.0006530891987495124, "learning_rate": 1.7484008528784647e-05, "loss": 0.0001, "step": 1525 }, { "epoch": 3.2622601279317696, "grad_norm": 0.0006848535849712789, "learning_rate": 1.73773987206823e-05, "loss": 0.0001, "step": 1530 }, { "epoch": 3.272921108742004, "grad_norm": 0.0006426739855669439, "learning_rate": 1.7270788912579956e-05, "loss": 0.0001, "step": 1535 }, { "epoch": 3.283582089552239, "grad_norm": 0.0006685128319077194, "learning_rate": 1.716417910447761e-05, "loss": 0.0001, "step": 1540 }, { "epoch": 3.2942430703624734, "grad_norm": 0.0007420861511491239, "learning_rate": 1.7057569296375266e-05, "loss": 0.0001, "step": 1545 }, { "epoch": 3.304904051172708, "grad_norm": 0.002152465283870697, "learning_rate": 1.695095948827292e-05, "loss": 0.0001, "step": 1550 }, { "epoch": 3.3155650319829424, "grad_norm": 0.05875202640891075, "learning_rate": 1.6844349680170575e-05, "loss": 0.0001, "step": 1555 }, { "epoch": 3.326226012793177, "grad_norm": 0.0006523785996250808, "learning_rate": 1.673773987206823e-05, "loss": 0.0001, "step": 1560 }, { "epoch": 3.3368869936034113, "grad_norm": 0.001634942484088242, "learning_rate": 1.6631130063965885e-05, "loss": 0.0001, "step": 1565 }, { "epoch": 3.3475479744136463, "grad_norm": 0.0006406964967027307, "learning_rate": 1.652452025586354e-05, "loss": 0.0001, "step": 1570 }, { "epoch": 3.3582089552238807, "grad_norm": 0.0006093855481594801, "learning_rate": 1.6417910447761194e-05, "loss": 0.0001, "step": 1575 }, { "epoch": 3.368869936034115, "grad_norm": 0.0006211912841536105, "learning_rate": 1.631130063965885e-05, "loss": 0.0001, "step": 1580 }, { "epoch": 3.3795309168443497, "grad_norm": 0.0010416092118248343, "learning_rate": 1.6204690831556504e-05, "loss": 0.0001, "step": 1585 }, { "epoch": 3.390191897654584, "grad_norm": 0.0006299918168224394, "learning_rate": 1.6098081023454158e-05, "loss": 0.0001, "step": 1590 }, { "epoch": 3.4008528784648187, "grad_norm": 0.0015466258628293872, "learning_rate": 1.5991471215351813e-05, "loss": 0.0001, "step": 1595 }, { "epoch": 3.411513859275053, "grad_norm": 0.0010751493973657489, "learning_rate": 1.5884861407249468e-05, "loss": 0.0001, "step": 1600 }, { "epoch": 3.4221748400852876, "grad_norm": 0.0006764642894268036, "learning_rate": 1.5778251599147122e-05, "loss": 0.0001, "step": 1605 }, { "epoch": 3.4328358208955225, "grad_norm": 0.020116742700338364, "learning_rate": 1.5671641791044777e-05, "loss": 0.0001, "step": 1610 }, { "epoch": 3.443496801705757, "grad_norm": 0.0006016920087859035, "learning_rate": 1.5565031982942432e-05, "loss": 0.0001, "step": 1615 }, { "epoch": 3.4541577825159915, "grad_norm": 0.10637471824884415, "learning_rate": 1.5458422174840087e-05, "loss": 0.0002, "step": 1620 }, { "epoch": 3.464818763326226, "grad_norm": 0.0006321455002762377, "learning_rate": 1.535181236673774e-05, "loss": 0.0001, "step": 1625 }, { "epoch": 3.4754797441364604, "grad_norm": 0.000656594056636095, "learning_rate": 1.5245202558635396e-05, "loss": 0.0001, "step": 1630 }, { "epoch": 3.486140724946695, "grad_norm": 0.0006587958778254688, "learning_rate": 1.5138592750533051e-05, "loss": 0.0001, "step": 1635 }, { "epoch": 3.49680170575693, "grad_norm": 0.0009308066801168025, "learning_rate": 1.5031982942430706e-05, "loss": 0.0001, "step": 1640 }, { "epoch": 3.5074626865671643, "grad_norm": 0.0006502924370579422, "learning_rate": 1.4925373134328357e-05, "loss": 0.0001, "step": 1645 }, { "epoch": 3.518123667377399, "grad_norm": 0.0009144488140009344, "learning_rate": 1.4818763326226012e-05, "loss": 0.0001, "step": 1650 }, { "epoch": 3.5287846481876333, "grad_norm": 0.0006393847288563848, "learning_rate": 1.4712153518123666e-05, "loss": 0.0001, "step": 1655 }, { "epoch": 3.5394456289978677, "grad_norm": 0.0006517107249237597, "learning_rate": 1.4605543710021321e-05, "loss": 0.0001, "step": 1660 }, { "epoch": 3.550106609808102, "grad_norm": 0.0006136732990853488, "learning_rate": 1.4498933901918976e-05, "loss": 0.0003, "step": 1665 }, { "epoch": 3.5607675906183367, "grad_norm": 0.000644017243757844, "learning_rate": 1.439232409381663e-05, "loss": 0.0001, "step": 1670 }, { "epoch": 3.571428571428571, "grad_norm": 0.010297813452780247, "learning_rate": 1.4285714285714285e-05, "loss": 0.0001, "step": 1675 }, { "epoch": 3.582089552238806, "grad_norm": 0.0020787366665899754, "learning_rate": 1.417910447761194e-05, "loss": 0.0001, "step": 1680 }, { "epoch": 3.5927505330490406, "grad_norm": 0.0006698886281810701, "learning_rate": 1.4072494669509595e-05, "loss": 0.0031, "step": 1685 }, { "epoch": 3.603411513859275, "grad_norm": 0.0007361291209235787, "learning_rate": 1.396588486140725e-05, "loss": 0.0001, "step": 1690 }, { "epoch": 3.6140724946695095, "grad_norm": 0.0006179155898280442, "learning_rate": 1.3859275053304904e-05, "loss": 0.0001, "step": 1695 }, { "epoch": 3.624733475479744, "grad_norm": 0.0006052028038538992, "learning_rate": 1.3752665245202559e-05, "loss": 0.0001, "step": 1700 }, { "epoch": 3.635394456289979, "grad_norm": 0.0007120109512470663, "learning_rate": 1.3646055437100214e-05, "loss": 0.0001, "step": 1705 }, { "epoch": 3.6460554371002134, "grad_norm": 0.0005918610841035843, "learning_rate": 1.3539445628997869e-05, "loss": 0.0001, "step": 1710 }, { "epoch": 3.656716417910448, "grad_norm": 0.0010003502247855067, "learning_rate": 1.3432835820895523e-05, "loss": 0.0001, "step": 1715 }, { "epoch": 3.6673773987206824, "grad_norm": 0.000583444896619767, "learning_rate": 1.3326226012793178e-05, "loss": 0.0001, "step": 1720 }, { "epoch": 3.678038379530917, "grad_norm": 0.0006132448907010257, "learning_rate": 1.3219616204690833e-05, "loss": 0.0001, "step": 1725 }, { "epoch": 3.6886993603411513, "grad_norm": 0.0006164976512081921, "learning_rate": 1.3113006396588488e-05, "loss": 0.0001, "step": 1730 }, { "epoch": 3.699360341151386, "grad_norm": 0.001158764585852623, "learning_rate": 1.3006396588486142e-05, "loss": 0.0066, "step": 1735 }, { "epoch": 3.7100213219616203, "grad_norm": 0.0006081092869862914, "learning_rate": 1.2899786780383797e-05, "loss": 0.0001, "step": 1740 }, { "epoch": 3.7206823027718547, "grad_norm": 0.0006186039536260068, "learning_rate": 1.2793176972281452e-05, "loss": 0.0001, "step": 1745 }, { "epoch": 3.7313432835820897, "grad_norm": 0.0006305409478954971, "learning_rate": 1.2686567164179105e-05, "loss": 0.0001, "step": 1750 }, { "epoch": 3.742004264392324, "grad_norm": 0.0006021612789481878, "learning_rate": 1.257995735607676e-05, "loss": 0.0001, "step": 1755 }, { "epoch": 3.7526652452025586, "grad_norm": 0.0006244622054509819, "learning_rate": 1.2473347547974414e-05, "loss": 0.0001, "step": 1760 }, { "epoch": 3.763326226012793, "grad_norm": 0.004119617398828268, "learning_rate": 1.2366737739872069e-05, "loss": 0.0001, "step": 1765 }, { "epoch": 3.7739872068230276, "grad_norm": 0.0006019301945343614, "learning_rate": 1.2260127931769722e-05, "loss": 0.0001, "step": 1770 }, { "epoch": 3.7846481876332625, "grad_norm": 0.0006072636460885406, "learning_rate": 1.2153518123667377e-05, "loss": 0.0001, "step": 1775 }, { "epoch": 3.795309168443497, "grad_norm": 0.000580860476475209, "learning_rate": 1.2046908315565032e-05, "loss": 0.0001, "step": 1780 }, { "epoch": 3.8059701492537314, "grad_norm": 0.0006087400251999497, "learning_rate": 1.1940298507462686e-05, "loss": 0.0001, "step": 1785 }, { "epoch": 3.816631130063966, "grad_norm": 0.0005997332045808434, "learning_rate": 1.1833688699360341e-05, "loss": 0.0001, "step": 1790 }, { "epoch": 3.8272921108742004, "grad_norm": 0.0005745955277234316, "learning_rate": 1.1727078891257996e-05, "loss": 0.0001, "step": 1795 }, { "epoch": 3.837953091684435, "grad_norm": 0.0006252889288589358, "learning_rate": 1.162046908315565e-05, "loss": 0.0001, "step": 1800 }, { "epoch": 3.8486140724946694, "grad_norm": 0.0005680587491951883, "learning_rate": 1.1513859275053305e-05, "loss": 0.0032, "step": 1805 }, { "epoch": 3.859275053304904, "grad_norm": 0.0013071306748315692, "learning_rate": 1.140724946695096e-05, "loss": 0.0001, "step": 1810 }, { "epoch": 3.8699360341151388, "grad_norm": 0.0008826420526020229, "learning_rate": 1.1300639658848615e-05, "loss": 0.0001, "step": 1815 }, { "epoch": 3.8805970149253732, "grad_norm": 0.0017494745552539825, "learning_rate": 1.119402985074627e-05, "loss": 0.0001, "step": 1820 }, { "epoch": 3.8912579957356077, "grad_norm": 0.0005758851184509695, "learning_rate": 1.1087420042643924e-05, "loss": 0.0001, "step": 1825 }, { "epoch": 3.901918976545842, "grad_norm": 0.0006469243671745062, "learning_rate": 1.0980810234541579e-05, "loss": 0.0001, "step": 1830 }, { "epoch": 3.9125799573560767, "grad_norm": 0.0006032307865098119, "learning_rate": 1.0874200426439234e-05, "loss": 0.0001, "step": 1835 }, { "epoch": 3.923240938166311, "grad_norm": 0.0005521726561710238, "learning_rate": 1.0767590618336887e-05, "loss": 0.0001, "step": 1840 }, { "epoch": 3.933901918976546, "grad_norm": 0.001913422136567533, "learning_rate": 1.0660980810234541e-05, "loss": 0.0001, "step": 1845 }, { "epoch": 3.9445628997867805, "grad_norm": 0.0005568794440478086, "learning_rate": 1.0554371002132196e-05, "loss": 0.0001, "step": 1850 }, { "epoch": 3.955223880597015, "grad_norm": 0.0005806397530250251, "learning_rate": 1.0447761194029851e-05, "loss": 0.0001, "step": 1855 }, { "epoch": 3.9658848614072495, "grad_norm": 0.0005738319596275687, "learning_rate": 1.0341151385927506e-05, "loss": 0.0001, "step": 1860 }, { "epoch": 3.976545842217484, "grad_norm": 0.0006032651290297508, "learning_rate": 1.023454157782516e-05, "loss": 0.0001, "step": 1865 }, { "epoch": 3.9872068230277184, "grad_norm": 0.0005541262216866016, "learning_rate": 1.0127931769722815e-05, "loss": 0.0001, "step": 1870 }, { "epoch": 3.997867803837953, "grad_norm": 0.0005812663584947586, "learning_rate": 1.002132196162047e-05, "loss": 0.0001, "step": 1875 }, { "epoch": 4.0, "eval_accuracy": 0.9909333333333333, "eval_loss": 0.04956997185945511, "eval_runtime": 18.2062, "eval_samples_per_second": 205.973, "eval_steps_per_second": 6.481, "step": 1876 }, { "epoch": 4.008528784648187, "grad_norm": 0.000589254719670862, "learning_rate": 9.914712153518125e-06, "loss": 0.0001, "step": 1880 }, { "epoch": 4.019189765458422, "grad_norm": 0.0005932562635280192, "learning_rate": 9.80810234541578e-06, "loss": 0.0001, "step": 1885 }, { "epoch": 4.029850746268656, "grad_norm": 0.0007711496436968446, "learning_rate": 9.701492537313434e-06, "loss": 0.0001, "step": 1890 }, { "epoch": 4.040511727078891, "grad_norm": 0.0011907704174518585, "learning_rate": 9.594882729211089e-06, "loss": 0.0001, "step": 1895 }, { "epoch": 4.051172707889126, "grad_norm": 0.001914159394800663, "learning_rate": 9.488272921108744e-06, "loss": 0.0001, "step": 1900 }, { "epoch": 4.061833688699361, "grad_norm": 0.0005822319653816521, "learning_rate": 9.381663113006398e-06, "loss": 0.0001, "step": 1905 }, { "epoch": 4.072494669509595, "grad_norm": 0.0005916253430768847, "learning_rate": 9.275053304904053e-06, "loss": 0.0001, "step": 1910 }, { "epoch": 4.08315565031983, "grad_norm": 0.0009646411053836346, "learning_rate": 9.168443496801706e-06, "loss": 0.0001, "step": 1915 }, { "epoch": 4.093816631130064, "grad_norm": 0.0005698201130144298, "learning_rate": 9.06183368869936e-06, "loss": 0.0001, "step": 1920 }, { "epoch": 4.104477611940299, "grad_norm": 0.0005831252201460302, "learning_rate": 8.955223880597016e-06, "loss": 0.0001, "step": 1925 }, { "epoch": 4.115138592750533, "grad_norm": 0.0005589622305706143, "learning_rate": 8.84861407249467e-06, "loss": 0.0001, "step": 1930 }, { "epoch": 4.1257995735607675, "grad_norm": 0.0006069365772418678, "learning_rate": 8.742004264392323e-06, "loss": 0.0001, "step": 1935 }, { "epoch": 4.136460554371002, "grad_norm": 0.0005879868404008448, "learning_rate": 8.635394456289978e-06, "loss": 0.0001, "step": 1940 }, { "epoch": 4.1471215351812365, "grad_norm": 0.0006471486412920058, "learning_rate": 8.528784648187633e-06, "loss": 0.0001, "step": 1945 }, { "epoch": 4.157782515991471, "grad_norm": 0.000561364518944174, "learning_rate": 8.422174840085288e-06, "loss": 0.0001, "step": 1950 }, { "epoch": 4.1684434968017055, "grad_norm": 0.0006335545913316309, "learning_rate": 8.315565031982942e-06, "loss": 0.0009, "step": 1955 }, { "epoch": 4.17910447761194, "grad_norm": 0.0005844077677465975, "learning_rate": 8.208955223880597e-06, "loss": 0.0001, "step": 1960 }, { "epoch": 4.189765458422174, "grad_norm": 0.0005894974456168711, "learning_rate": 8.102345415778252e-06, "loss": 0.0001, "step": 1965 }, { "epoch": 4.20042643923241, "grad_norm": 0.0005456387880258262, "learning_rate": 7.995735607675907e-06, "loss": 0.0001, "step": 1970 }, { "epoch": 4.211087420042644, "grad_norm": 0.0007201472762972116, "learning_rate": 7.889125799573561e-06, "loss": 0.0001, "step": 1975 }, { "epoch": 4.221748400852879, "grad_norm": 0.0005597337149083614, "learning_rate": 7.782515991471216e-06, "loss": 0.0001, "step": 1980 }, { "epoch": 4.232409381663113, "grad_norm": 0.1250966489315033, "learning_rate": 7.67590618336887e-06, "loss": 0.0037, "step": 1985 }, { "epoch": 4.243070362473348, "grad_norm": 0.0005643576150760055, "learning_rate": 7.5692963752665255e-06, "loss": 0.0001, "step": 1990 }, { "epoch": 4.253731343283582, "grad_norm": 0.0005553606315515935, "learning_rate": 7.4626865671641785e-06, "loss": 0.0001, "step": 1995 }, { "epoch": 4.264392324093817, "grad_norm": 0.0007400668691843748, "learning_rate": 7.356076759061833e-06, "loss": 0.0001, "step": 2000 }, { "epoch": 4.275053304904051, "grad_norm": 0.0005373511230573058, "learning_rate": 7.249466950959488e-06, "loss": 0.0001, "step": 2005 }, { "epoch": 4.285714285714286, "grad_norm": 0.0005943583673797548, "learning_rate": 7.142857142857143e-06, "loss": 0.0001, "step": 2010 }, { "epoch": 4.29637526652452, "grad_norm": 0.0005727015668526292, "learning_rate": 7.0362473347547975e-06, "loss": 0.0001, "step": 2015 }, { "epoch": 4.3070362473347545, "grad_norm": 0.0005763427470810711, "learning_rate": 6.929637526652452e-06, "loss": 0.0001, "step": 2020 }, { "epoch": 4.317697228144989, "grad_norm": 0.08408217132091522, "learning_rate": 6.823027718550107e-06, "loss": 0.0023, "step": 2025 }, { "epoch": 4.3283582089552235, "grad_norm": 0.0005716944579035044, "learning_rate": 6.716417910447762e-06, "loss": 0.0001, "step": 2030 }, { "epoch": 4.339019189765459, "grad_norm": 0.0020967533346265554, "learning_rate": 6.609808102345416e-06, "loss": 0.0001, "step": 2035 }, { "epoch": 4.349680170575693, "grad_norm": 0.0007090475410223007, "learning_rate": 6.503198294243071e-06, "loss": 0.0001, "step": 2040 }, { "epoch": 4.360341151385928, "grad_norm": 0.0008146331529133022, "learning_rate": 6.396588486140726e-06, "loss": 0.0001, "step": 2045 }, { "epoch": 4.371002132196162, "grad_norm": 0.0005934647633694112, "learning_rate": 6.28997867803838e-06, "loss": 0.0001, "step": 2050 }, { "epoch": 4.381663113006397, "grad_norm": 0.000562927860300988, "learning_rate": 6.1833688699360345e-06, "loss": 0.0001, "step": 2055 }, { "epoch": 4.392324093816631, "grad_norm": 0.0020479762461036444, "learning_rate": 6.076759061833688e-06, "loss": 0.0001, "step": 2060 }, { "epoch": 4.402985074626866, "grad_norm": 0.0006749996682628989, "learning_rate": 5.970149253731343e-06, "loss": 0.0001, "step": 2065 }, { "epoch": 4.4136460554371, "grad_norm": 0.0005582980229519308, "learning_rate": 5.863539445628998e-06, "loss": 0.0001, "step": 2070 }, { "epoch": 4.424307036247335, "grad_norm": 0.0018319780938327312, "learning_rate": 5.756929637526653e-06, "loss": 0.0001, "step": 2075 }, { "epoch": 4.434968017057569, "grad_norm": 0.0005455813370645046, "learning_rate": 5.650319829424307e-06, "loss": 0.0001, "step": 2080 }, { "epoch": 4.445628997867804, "grad_norm": 0.0005394156323745847, "learning_rate": 5.543710021321962e-06, "loss": 0.0001, "step": 2085 }, { "epoch": 4.456289978678038, "grad_norm": 0.0008809953578747809, "learning_rate": 5.437100213219617e-06, "loss": 0.0001, "step": 2090 }, { "epoch": 4.466950959488273, "grad_norm": 0.0005550557980313897, "learning_rate": 5.330490405117271e-06, "loss": 0.0001, "step": 2095 }, { "epoch": 4.477611940298507, "grad_norm": 0.000527980737388134, "learning_rate": 5.2238805970149255e-06, "loss": 0.0001, "step": 2100 }, { "epoch": 4.4882729211087415, "grad_norm": 0.0005987246404401958, "learning_rate": 5.11727078891258e-06, "loss": 0.0001, "step": 2105 }, { "epoch": 4.498933901918977, "grad_norm": 0.0005716937012039125, "learning_rate": 5.010660980810235e-06, "loss": 0.0001, "step": 2110 }, { "epoch": 4.509594882729211, "grad_norm": 0.000552869460079819, "learning_rate": 4.90405117270789e-06, "loss": 0.0001, "step": 2115 }, { "epoch": 4.520255863539446, "grad_norm": 0.0005779630737379193, "learning_rate": 4.797441364605544e-06, "loss": 0.0001, "step": 2120 }, { "epoch": 4.53091684434968, "grad_norm": 0.0006005573086440563, "learning_rate": 4.690831556503199e-06, "loss": 0.0001, "step": 2125 }, { "epoch": 4.541577825159915, "grad_norm": 0.0005399554502218962, "learning_rate": 4.584221748400853e-06, "loss": 0.0001, "step": 2130 }, { "epoch": 4.552238805970149, "grad_norm": 0.0005871957982890308, "learning_rate": 4.477611940298508e-06, "loss": 0.0001, "step": 2135 }, { "epoch": 4.562899786780384, "grad_norm": 0.0005549567868001759, "learning_rate": 4.371002132196162e-06, "loss": 0.0001, "step": 2140 }, { "epoch": 4.573560767590618, "grad_norm": 0.0005498258979059756, "learning_rate": 4.264392324093816e-06, "loss": 0.0001, "step": 2145 }, { "epoch": 4.584221748400853, "grad_norm": 0.0021440256386995316, "learning_rate": 4.157782515991471e-06, "loss": 0.0001, "step": 2150 }, { "epoch": 4.594882729211087, "grad_norm": 0.0005465818685479462, "learning_rate": 4.051172707889126e-06, "loss": 0.0001, "step": 2155 }, { "epoch": 4.605543710021322, "grad_norm": 0.0005637862486764789, "learning_rate": 3.944562899786781e-06, "loss": 0.0001, "step": 2160 }, { "epoch": 4.616204690831556, "grad_norm": 0.0005367129342630506, "learning_rate": 3.837953091684435e-06, "loss": 0.0001, "step": 2165 }, { "epoch": 4.6268656716417915, "grad_norm": 0.000578960869461298, "learning_rate": 3.7313432835820893e-06, "loss": 0.0001, "step": 2170 }, { "epoch": 4.637526652452026, "grad_norm": 0.0008364486857317388, "learning_rate": 3.624733475479744e-06, "loss": 0.0001, "step": 2175 }, { "epoch": 4.6481876332622605, "grad_norm": 0.0005582637386396527, "learning_rate": 3.5181236673773987e-06, "loss": 0.0001, "step": 2180 }, { "epoch": 4.658848614072495, "grad_norm": 0.0022017708979547024, "learning_rate": 3.4115138592750535e-06, "loss": 0.0001, "step": 2185 }, { "epoch": 4.669509594882729, "grad_norm": 0.0005324066733010113, "learning_rate": 3.304904051172708e-06, "loss": 0.0001, "step": 2190 }, { "epoch": 4.680170575692964, "grad_norm": 0.0005428901058621705, "learning_rate": 3.198294243070363e-06, "loss": 0.0001, "step": 2195 }, { "epoch": 4.690831556503198, "grad_norm": 0.0005897351074963808, "learning_rate": 3.0916844349680173e-06, "loss": 0.0001, "step": 2200 }, { "epoch": 4.701492537313433, "grad_norm": 0.0005692556733265519, "learning_rate": 2.9850746268656716e-06, "loss": 0.0001, "step": 2205 }, { "epoch": 4.712153518123667, "grad_norm": 0.0005513677024282515, "learning_rate": 2.8784648187633263e-06, "loss": 0.0001, "step": 2210 }, { "epoch": 4.722814498933902, "grad_norm": 0.33751627802848816, "learning_rate": 2.771855010660981e-06, "loss": 0.0058, "step": 2215 }, { "epoch": 4.733475479744136, "grad_norm": 0.0008201355230994523, "learning_rate": 2.6652452025586354e-06, "loss": 0.0001, "step": 2220 }, { "epoch": 4.744136460554371, "grad_norm": 0.0005522746359929442, "learning_rate": 2.55863539445629e-06, "loss": 0.0001, "step": 2225 }, { "epoch": 4.754797441364605, "grad_norm": 0.0005784935783594847, "learning_rate": 2.452025586353945e-06, "loss": 0.0001, "step": 2230 }, { "epoch": 4.76545842217484, "grad_norm": 0.0005910536856390536, "learning_rate": 2.3454157782515996e-06, "loss": 0.0001, "step": 2235 }, { "epoch": 4.776119402985074, "grad_norm": 0.0005423520342446864, "learning_rate": 2.238805970149254e-06, "loss": 0.0001, "step": 2240 }, { "epoch": 4.786780383795309, "grad_norm": 0.0005804357933811843, "learning_rate": 2.132196162046908e-06, "loss": 0.0001, "step": 2245 }, { "epoch": 4.797441364605544, "grad_norm": 0.0005387410637922585, "learning_rate": 2.025586353944563e-06, "loss": 0.0001, "step": 2250 }, { "epoch": 4.8081023454157785, "grad_norm": 0.0005506675806827843, "learning_rate": 1.9189765458422177e-06, "loss": 0.0001, "step": 2255 }, { "epoch": 4.818763326226013, "grad_norm": 0.000627724512014538, "learning_rate": 1.812366737739872e-06, "loss": 0.0001, "step": 2260 }, { "epoch": 4.8294243070362475, "grad_norm": 0.0008332194411195815, "learning_rate": 1.7057569296375267e-06, "loss": 0.0001, "step": 2265 }, { "epoch": 4.840085287846482, "grad_norm": 0.0006300428067333996, "learning_rate": 1.5991471215351815e-06, "loss": 0.0001, "step": 2270 }, { "epoch": 4.850746268656716, "grad_norm": 0.0006895289407111704, "learning_rate": 1.4925373134328358e-06, "loss": 0.0001, "step": 2275 }, { "epoch": 4.861407249466951, "grad_norm": 0.0005459040985442698, "learning_rate": 1.3859275053304905e-06, "loss": 0.0001, "step": 2280 }, { "epoch": 4.872068230277185, "grad_norm": 0.0005209689261391759, "learning_rate": 1.279317697228145e-06, "loss": 0.0001, "step": 2285 }, { "epoch": 4.88272921108742, "grad_norm": 0.0010145273990929127, "learning_rate": 1.1727078891257998e-06, "loss": 0.0001, "step": 2290 }, { "epoch": 4.893390191897654, "grad_norm": 0.0006381816347129643, "learning_rate": 1.066098081023454e-06, "loss": 0.0022, "step": 2295 }, { "epoch": 4.904051172707889, "grad_norm": 0.0012976997531950474, "learning_rate": 9.594882729211088e-07, "loss": 0.0001, "step": 2300 }, { "epoch": 4.914712153518123, "grad_norm": 0.0005350754363462329, "learning_rate": 8.528784648187634e-07, "loss": 0.0001, "step": 2305 }, { "epoch": 4.925373134328359, "grad_norm": 0.0005413396283984184, "learning_rate": 7.462686567164179e-07, "loss": 0.0001, "step": 2310 }, { "epoch": 4.936034115138593, "grad_norm": 0.0007284008897840977, "learning_rate": 6.396588486140725e-07, "loss": 0.0001, "step": 2315 }, { "epoch": 4.946695095948828, "grad_norm": 0.0005542092840187252, "learning_rate": 5.33049040511727e-07, "loss": 0.0001, "step": 2320 }, { "epoch": 4.957356076759062, "grad_norm": 0.0005262020858936012, "learning_rate": 4.264392324093817e-07, "loss": 0.0001, "step": 2325 }, { "epoch": 4.968017057569297, "grad_norm": 0.0010335709666833282, "learning_rate": 3.1982942430703626e-07, "loss": 0.0001, "step": 2330 }, { "epoch": 4.978678038379531, "grad_norm": 0.0017745111836120486, "learning_rate": 2.1321961620469084e-07, "loss": 0.0001, "step": 2335 }, { "epoch": 4.9893390191897655, "grad_norm": 0.0005233931588009, "learning_rate": 1.0660980810234542e-07, "loss": 0.0001, "step": 2340 }, { "epoch": 5.0, "grad_norm": 0.0006056681158952415, "learning_rate": 0.0, "loss": 0.0001, "step": 2345 }, { "epoch": 5.0, "eval_accuracy": 0.9906666666666667, "eval_loss": 0.050289787352085114, "eval_runtime": 18.6045, "eval_samples_per_second": 201.564, "eval_steps_per_second": 6.343, "step": 2345 }, { "epoch": 5.0, "step": 2345, "total_flos": 5.8118992210944e+18, "train_loss": 0.003966709892754492, "train_runtime": 820.258, "train_samples_per_second": 91.435, "train_steps_per_second": 2.859 } ], "logging_steps": 5, "max_steps": 2345, "num_input_tokens_seen": 0, "num_train_epochs": 5, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 5.8118992210944e+18, "train_batch_size": 32, "trial_name": null, "trial_params": null }