{ "best_metric": 0.022642984986305237, "best_model_checkpoint": "vit_epochs5_batch32_lr5e-05_size224_tiles3_seed2_q3_DA/checkpoint-2345", "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": 1.253076434135437, "learning_rate": 4.989339019189766e-05, "loss": 0.6371, "step": 5 }, { "epoch": 0.021321961620469083, "grad_norm": 1.2961275577545166, "learning_rate": 4.978678038379531e-05, "loss": 0.5657, "step": 10 }, { "epoch": 0.031982942430703626, "grad_norm": 1.244076132774353, "learning_rate": 4.9680170575692967e-05, "loss": 0.4588, "step": 15 }, { "epoch": 0.042643923240938165, "grad_norm": 2.153555154800415, "learning_rate": 4.957356076759062e-05, "loss": 0.3802, "step": 20 }, { "epoch": 0.053304904051172705, "grad_norm": 1.3869515657424927, "learning_rate": 4.9466950959488276e-05, "loss": 0.3136, "step": 25 }, { "epoch": 0.06396588486140725, "grad_norm": 1.4363356828689575, "learning_rate": 4.936034115138593e-05, "loss": 0.2547, "step": 30 }, { "epoch": 0.07462686567164178, "grad_norm": 1.7793512344360352, "learning_rate": 4.9253731343283586e-05, "loss": 0.3012, "step": 35 }, { "epoch": 0.08528784648187633, "grad_norm": 0.7354267239570618, "learning_rate": 4.914712153518124e-05, "loss": 0.2187, "step": 40 }, { "epoch": 0.09594882729211088, "grad_norm": 4.01702880859375, "learning_rate": 4.904051172707889e-05, "loss": 0.2756, "step": 45 }, { "epoch": 0.10660980810234541, "grad_norm": 1.72369384765625, "learning_rate": 4.893390191897655e-05, "loss": 0.1623, "step": 50 }, { "epoch": 0.11727078891257996, "grad_norm": 2.391291379928589, "learning_rate": 4.88272921108742e-05, "loss": 0.1671, "step": 55 }, { "epoch": 0.1279317697228145, "grad_norm": 2.076777458190918, "learning_rate": 4.872068230277186e-05, "loss": 0.19, "step": 60 }, { "epoch": 0.13859275053304904, "grad_norm": 1.2287567853927612, "learning_rate": 4.861407249466951e-05, "loss": 0.1482, "step": 65 }, { "epoch": 0.14925373134328357, "grad_norm": 2.1280577182769775, "learning_rate": 4.850746268656717e-05, "loss": 0.1428, "step": 70 }, { "epoch": 0.15991471215351813, "grad_norm": 3.1935970783233643, "learning_rate": 4.840085287846482e-05, "loss": 0.163, "step": 75 }, { "epoch": 0.17057569296375266, "grad_norm": 2.945774555206299, "learning_rate": 4.829424307036248e-05, "loss": 0.1641, "step": 80 }, { "epoch": 0.1812366737739872, "grad_norm": 0.569043755531311, "learning_rate": 4.8187633262260126e-05, "loss": 0.0801, "step": 85 }, { "epoch": 0.19189765458422176, "grad_norm": 9.947427749633789, "learning_rate": 4.808102345415779e-05, "loss": 0.1377, "step": 90 }, { "epoch": 0.2025586353944563, "grad_norm": 2.9618964195251465, "learning_rate": 4.7974413646055436e-05, "loss": 0.1672, "step": 95 }, { "epoch": 0.21321961620469082, "grad_norm": 4.840129375457764, "learning_rate": 4.78678038379531e-05, "loss": 0.1918, "step": 100 }, { "epoch": 0.22388059701492538, "grad_norm": 4.555376052856445, "learning_rate": 4.7761194029850745e-05, "loss": 0.1764, "step": 105 }, { "epoch": 0.2345415778251599, "grad_norm": 2.924715280532837, "learning_rate": 4.765458422174841e-05, "loss": 0.1183, "step": 110 }, { "epoch": 0.24520255863539445, "grad_norm": 1.4472227096557617, "learning_rate": 4.7547974413646055e-05, "loss": 0.1399, "step": 115 }, { "epoch": 0.255863539445629, "grad_norm": 4.603981971740723, "learning_rate": 4.7441364605543716e-05, "loss": 0.1188, "step": 120 }, { "epoch": 0.26652452025586354, "grad_norm": 2.4801743030548096, "learning_rate": 4.7334754797441364e-05, "loss": 0.1915, "step": 125 }, { "epoch": 0.2771855010660981, "grad_norm": 5.715509414672852, "learning_rate": 4.7228144989339026e-05, "loss": 0.1665, "step": 130 }, { "epoch": 0.2878464818763326, "grad_norm": 1.5166106224060059, "learning_rate": 4.7121535181236674e-05, "loss": 0.1342, "step": 135 }, { "epoch": 0.29850746268656714, "grad_norm": 4.280544757843018, "learning_rate": 4.7014925373134335e-05, "loss": 0.1233, "step": 140 }, { "epoch": 0.3091684434968017, "grad_norm": 2.712048053741455, "learning_rate": 4.690831556503198e-05, "loss": 0.1222, "step": 145 }, { "epoch": 0.31982942430703626, "grad_norm": 3.7164363861083984, "learning_rate": 4.6801705756929645e-05, "loss": 0.0921, "step": 150 }, { "epoch": 0.3304904051172708, "grad_norm": 5.086050510406494, "learning_rate": 4.669509594882729e-05, "loss": 0.1714, "step": 155 }, { "epoch": 0.3411513859275053, "grad_norm": 5.839847564697266, "learning_rate": 4.658848614072495e-05, "loss": 0.1846, "step": 160 }, { "epoch": 0.35181236673773986, "grad_norm": 9.39326286315918, "learning_rate": 4.64818763326226e-05, "loss": 0.201, "step": 165 }, { "epoch": 0.3624733475479744, "grad_norm": 5.713239669799805, "learning_rate": 4.637526652452026e-05, "loss": 0.1344, "step": 170 }, { "epoch": 0.373134328358209, "grad_norm": 4.050966262817383, "learning_rate": 4.626865671641791e-05, "loss": 0.1261, "step": 175 }, { "epoch": 0.3837953091684435, "grad_norm": 4.296970844268799, "learning_rate": 4.6162046908315566e-05, "loss": 0.125, "step": 180 }, { "epoch": 0.39445628997867804, "grad_norm": 3.1301257610321045, "learning_rate": 4.605543710021322e-05, "loss": 0.1064, "step": 185 }, { "epoch": 0.4051172707889126, "grad_norm": 1.3620059490203857, "learning_rate": 4.5948827292110876e-05, "loss": 0.1282, "step": 190 }, { "epoch": 0.4157782515991471, "grad_norm": 1.2779539823532104, "learning_rate": 4.584221748400853e-05, "loss": 0.148, "step": 195 }, { "epoch": 0.42643923240938164, "grad_norm": 4.439069747924805, "learning_rate": 4.5735607675906185e-05, "loss": 0.0742, "step": 200 }, { "epoch": 0.43710021321961623, "grad_norm": 2.933945655822754, "learning_rate": 4.562899786780384e-05, "loss": 0.1823, "step": 205 }, { "epoch": 0.44776119402985076, "grad_norm": 3.5814318656921387, "learning_rate": 4.5522388059701495e-05, "loss": 0.0947, "step": 210 }, { "epoch": 0.4584221748400853, "grad_norm": 2.5738556385040283, "learning_rate": 4.541577825159915e-05, "loss": 0.1157, "step": 215 }, { "epoch": 0.4690831556503198, "grad_norm": 1.4425334930419922, "learning_rate": 4.5309168443496804e-05, "loss": 0.1326, "step": 220 }, { "epoch": 0.47974413646055436, "grad_norm": 5.78731107711792, "learning_rate": 4.520255863539446e-05, "loss": 0.128, "step": 225 }, { "epoch": 0.4904051172707889, "grad_norm": 13.519885063171387, "learning_rate": 4.5095948827292114e-05, "loss": 0.2299, "step": 230 }, { "epoch": 0.5010660980810234, "grad_norm": 1.8650281429290771, "learning_rate": 4.498933901918977e-05, "loss": 0.1771, "step": 235 }, { "epoch": 0.511727078891258, "grad_norm": 1.5189687013626099, "learning_rate": 4.488272921108742e-05, "loss": 0.0958, "step": 240 }, { "epoch": 0.5223880597014925, "grad_norm": 2.778465509414673, "learning_rate": 4.477611940298508e-05, "loss": 0.1259, "step": 245 }, { "epoch": 0.5330490405117271, "grad_norm": 4.404140949249268, "learning_rate": 4.466950959488273e-05, "loss": 0.1011, "step": 250 }, { "epoch": 0.5437100213219617, "grad_norm": 4.144996166229248, "learning_rate": 4.456289978678039e-05, "loss": 0.1591, "step": 255 }, { "epoch": 0.5543710021321961, "grad_norm": 0.8145550489425659, "learning_rate": 4.445628997867804e-05, "loss": 0.1067, "step": 260 }, { "epoch": 0.5650319829424307, "grad_norm": 4.139211654663086, "learning_rate": 4.43496801705757e-05, "loss": 0.1423, "step": 265 }, { "epoch": 0.5756929637526652, "grad_norm": 4.686325550079346, "learning_rate": 4.424307036247335e-05, "loss": 0.1749, "step": 270 }, { "epoch": 0.5863539445628998, "grad_norm": 0.23579491674900055, "learning_rate": 4.4136460554371006e-05, "loss": 0.1033, "step": 275 }, { "epoch": 0.5970149253731343, "grad_norm": 1.7055755853652954, "learning_rate": 4.402985074626866e-05, "loss": 0.0949, "step": 280 }, { "epoch": 0.6076759061833689, "grad_norm": 0.27129459381103516, "learning_rate": 4.3923240938166316e-05, "loss": 0.0924, "step": 285 }, { "epoch": 0.6183368869936035, "grad_norm": 3.5735697746276855, "learning_rate": 4.381663113006397e-05, "loss": 0.1483, "step": 290 }, { "epoch": 0.6289978678038379, "grad_norm": 2.463090181350708, "learning_rate": 4.3710021321961625e-05, "loss": 0.1386, "step": 295 }, { "epoch": 0.6396588486140725, "grad_norm": 0.7783936262130737, "learning_rate": 4.360341151385928e-05, "loss": 0.1247, "step": 300 }, { "epoch": 0.650319829424307, "grad_norm": 1.5393179655075073, "learning_rate": 4.3496801705756935e-05, "loss": 0.1255, "step": 305 }, { "epoch": 0.6609808102345416, "grad_norm": 3.372725486755371, "learning_rate": 4.339019189765459e-05, "loss": 0.0836, "step": 310 }, { "epoch": 0.6716417910447762, "grad_norm": 3.007035493850708, "learning_rate": 4.328358208955224e-05, "loss": 0.0814, "step": 315 }, { "epoch": 0.6823027718550106, "grad_norm": 3.2889404296875, "learning_rate": 4.31769722814499e-05, "loss": 0.0856, "step": 320 }, { "epoch": 0.6929637526652452, "grad_norm": 0.41842836141586304, "learning_rate": 4.307036247334755e-05, "loss": 0.0685, "step": 325 }, { "epoch": 0.7036247334754797, "grad_norm": 4.353408336639404, "learning_rate": 4.29637526652452e-05, "loss": 0.0743, "step": 330 }, { "epoch": 0.7142857142857143, "grad_norm": 6.490444660186768, "learning_rate": 4.2857142857142856e-05, "loss": 0.1079, "step": 335 }, { "epoch": 0.7249466950959488, "grad_norm": 1.1309465169906616, "learning_rate": 4.275053304904051e-05, "loss": 0.0949, "step": 340 }, { "epoch": 0.7356076759061834, "grad_norm": 4.720521450042725, "learning_rate": 4.2643923240938166e-05, "loss": 0.1684, "step": 345 }, { "epoch": 0.746268656716418, "grad_norm": 0.9798983931541443, "learning_rate": 4.253731343283582e-05, "loss": 0.1594, "step": 350 }, { "epoch": 0.7569296375266524, "grad_norm": 2.3568029403686523, "learning_rate": 4.2430703624733475e-05, "loss": 0.1418, "step": 355 }, { "epoch": 0.767590618336887, "grad_norm": 2.3879332542419434, "learning_rate": 4.232409381663113e-05, "loss": 0.1388, "step": 360 }, { "epoch": 0.7782515991471215, "grad_norm": 2.574840545654297, "learning_rate": 4.2217484008528785e-05, "loss": 0.0623, "step": 365 }, { "epoch": 0.7889125799573561, "grad_norm": 2.5726804733276367, "learning_rate": 4.211087420042644e-05, "loss": 0.1364, "step": 370 }, { "epoch": 0.7995735607675906, "grad_norm": 2.8437912464141846, "learning_rate": 4.2004264392324094e-05, "loss": 0.1069, "step": 375 }, { "epoch": 0.8102345415778252, "grad_norm": 2.232976198196411, "learning_rate": 4.189765458422175e-05, "loss": 0.0834, "step": 380 }, { "epoch": 0.8208955223880597, "grad_norm": 1.5938422679901123, "learning_rate": 4.1791044776119404e-05, "loss": 0.1152, "step": 385 }, { "epoch": 0.8315565031982942, "grad_norm": 4.246496677398682, "learning_rate": 4.168443496801706e-05, "loss": 0.1332, "step": 390 }, { "epoch": 0.8422174840085288, "grad_norm": 4.515913963317871, "learning_rate": 4.157782515991471e-05, "loss": 0.1044, "step": 395 }, { "epoch": 0.8528784648187633, "grad_norm": 0.9609928131103516, "learning_rate": 4.147121535181237e-05, "loss": 0.0396, "step": 400 }, { "epoch": 0.8635394456289979, "grad_norm": 4.7857465744018555, "learning_rate": 4.136460554371002e-05, "loss": 0.1239, "step": 405 }, { "epoch": 0.8742004264392325, "grad_norm": 7.453455448150635, "learning_rate": 4.125799573560768e-05, "loss": 0.1026, "step": 410 }, { "epoch": 0.8848614072494669, "grad_norm": 0.1801329404115677, "learning_rate": 4.115138592750533e-05, "loss": 0.1547, "step": 415 }, { "epoch": 0.8955223880597015, "grad_norm": 2.3846054077148438, "learning_rate": 4.104477611940299e-05, "loss": 0.1513, "step": 420 }, { "epoch": 0.906183368869936, "grad_norm": 3.5320534706115723, "learning_rate": 4.093816631130064e-05, "loss": 0.0633, "step": 425 }, { "epoch": 0.9168443496801706, "grad_norm": 3.5563056468963623, "learning_rate": 4.0831556503198296e-05, "loss": 0.1433, "step": 430 }, { "epoch": 0.9275053304904051, "grad_norm": 4.565072059631348, "learning_rate": 4.072494669509595e-05, "loss": 0.13, "step": 435 }, { "epoch": 0.9381663113006397, "grad_norm": 2.07597279548645, "learning_rate": 4.0618336886993606e-05, "loss": 0.1345, "step": 440 }, { "epoch": 0.9488272921108742, "grad_norm": 2.309739589691162, "learning_rate": 4.051172707889126e-05, "loss": 0.0783, "step": 445 }, { "epoch": 0.9594882729211087, "grad_norm": 3.955974817276001, "learning_rate": 4.0405117270788915e-05, "loss": 0.1065, "step": 450 }, { "epoch": 0.9701492537313433, "grad_norm": 3.2670273780822754, "learning_rate": 4.029850746268657e-05, "loss": 0.1497, "step": 455 }, { "epoch": 0.9808102345415778, "grad_norm": 3.6428298950195312, "learning_rate": 4.0191897654584225e-05, "loss": 0.1598, "step": 460 }, { "epoch": 0.9914712153518124, "grad_norm": 3.081238269805908, "learning_rate": 4.008528784648188e-05, "loss": 0.0924, "step": 465 }, { "epoch": 1.0, "eval_accuracy": 0.9805333333333334, "eval_loss": 0.06031749024987221, "eval_runtime": 29.4395, "eval_samples_per_second": 127.38, "eval_steps_per_second": 4.008, "step": 469 }, { "epoch": 1.0021321961620469, "grad_norm": 1.2290693521499634, "learning_rate": 3.997867803837953e-05, "loss": 0.1032, "step": 470 }, { "epoch": 1.0127931769722816, "grad_norm": 3.4809911251068115, "learning_rate": 3.987206823027719e-05, "loss": 0.097, "step": 475 }, { "epoch": 1.023454157782516, "grad_norm": 0.7603719830513, "learning_rate": 3.976545842217484e-05, "loss": 0.1017, "step": 480 }, { "epoch": 1.0341151385927505, "grad_norm": 2.5602762699127197, "learning_rate": 3.96588486140725e-05, "loss": 0.1063, "step": 485 }, { "epoch": 1.044776119402985, "grad_norm": 3.698781728744507, "learning_rate": 3.9552238805970146e-05, "loss": 0.2007, "step": 490 }, { "epoch": 1.0554371002132197, "grad_norm": 4.1277008056640625, "learning_rate": 3.944562899786781e-05, "loss": 0.1423, "step": 495 }, { "epoch": 1.0660980810234542, "grad_norm": 2.0080020427703857, "learning_rate": 3.9339019189765456e-05, "loss": 0.1029, "step": 500 }, { "epoch": 1.0767590618336886, "grad_norm": 3.3581795692443848, "learning_rate": 3.923240938166312e-05, "loss": 0.0757, "step": 505 }, { "epoch": 1.0874200426439233, "grad_norm": 1.021774411201477, "learning_rate": 3.9125799573560765e-05, "loss": 0.0635, "step": 510 }, { "epoch": 1.0980810234541578, "grad_norm": 1.3954639434814453, "learning_rate": 3.901918976545843e-05, "loss": 0.1092, "step": 515 }, { "epoch": 1.1087420042643923, "grad_norm": 2.2165262699127197, "learning_rate": 3.8912579957356075e-05, "loss": 0.1182, "step": 520 }, { "epoch": 1.1194029850746268, "grad_norm": 0.5253220200538635, "learning_rate": 3.8805970149253736e-05, "loss": 0.0756, "step": 525 }, { "epoch": 1.1300639658848615, "grad_norm": 1.7623412609100342, "learning_rate": 3.8699360341151384e-05, "loss": 0.0948, "step": 530 }, { "epoch": 1.140724946695096, "grad_norm": 3.1822142601013184, "learning_rate": 3.8592750533049046e-05, "loss": 0.145, "step": 535 }, { "epoch": 1.1513859275053304, "grad_norm": 0.556550145149231, "learning_rate": 3.8486140724946694e-05, "loss": 0.1264, "step": 540 }, { "epoch": 1.1620469083155651, "grad_norm": 4.393623352050781, "learning_rate": 3.8379530916844355e-05, "loss": 0.1199, "step": 545 }, { "epoch": 1.1727078891257996, "grad_norm": 3.06972336769104, "learning_rate": 3.8272921108742e-05, "loss": 0.0881, "step": 550 }, { "epoch": 1.183368869936034, "grad_norm": 2.290722608566284, "learning_rate": 3.8166311300639665e-05, "loss": 0.0578, "step": 555 }, { "epoch": 1.1940298507462686, "grad_norm": 1.2358784675598145, "learning_rate": 3.805970149253731e-05, "loss": 0.0494, "step": 560 }, { "epoch": 1.2046908315565032, "grad_norm": 1.2099443674087524, "learning_rate": 3.7953091684434974e-05, "loss": 0.0542, "step": 565 }, { "epoch": 1.2153518123667377, "grad_norm": 3.388763427734375, "learning_rate": 3.784648187633262e-05, "loss": 0.0781, "step": 570 }, { "epoch": 1.2260127931769722, "grad_norm": 0.21282166242599487, "learning_rate": 3.7739872068230284e-05, "loss": 0.1315, "step": 575 }, { "epoch": 1.236673773987207, "grad_norm": 2.9546573162078857, "learning_rate": 3.763326226012793e-05, "loss": 0.0955, "step": 580 }, { "epoch": 1.2473347547974414, "grad_norm": 0.39389169216156006, "learning_rate": 3.752665245202559e-05, "loss": 0.0616, "step": 585 }, { "epoch": 1.2579957356076759, "grad_norm": 3.0497117042541504, "learning_rate": 3.742004264392324e-05, "loss": 0.0808, "step": 590 }, { "epoch": 1.2686567164179103, "grad_norm": 2.4725160598754883, "learning_rate": 3.73134328358209e-05, "loss": 0.1912, "step": 595 }, { "epoch": 1.279317697228145, "grad_norm": 1.4559518098831177, "learning_rate": 3.720682302771855e-05, "loss": 0.075, "step": 600 }, { "epoch": 1.2899786780383795, "grad_norm": 2.7683145999908447, "learning_rate": 3.710021321961621e-05, "loss": 0.0874, "step": 605 }, { "epoch": 1.3006396588486142, "grad_norm": 4.364225387573242, "learning_rate": 3.699360341151386e-05, "loss": 0.1016, "step": 610 }, { "epoch": 1.3113006396588487, "grad_norm": 5.513325214385986, "learning_rate": 3.6886993603411515e-05, "loss": 0.1085, "step": 615 }, { "epoch": 1.3219616204690832, "grad_norm": 1.3611972332000732, "learning_rate": 3.678038379530917e-05, "loss": 0.0632, "step": 620 }, { "epoch": 1.3326226012793176, "grad_norm": 3.379392147064209, "learning_rate": 3.6673773987206824e-05, "loss": 0.0774, "step": 625 }, { "epoch": 1.3432835820895521, "grad_norm": 0.5443459749221802, "learning_rate": 3.656716417910448e-05, "loss": 0.0912, "step": 630 }, { "epoch": 1.3539445628997868, "grad_norm": 0.21950693428516388, "learning_rate": 3.6460554371002134e-05, "loss": 0.0865, "step": 635 }, { "epoch": 1.3646055437100213, "grad_norm": 1.2746403217315674, "learning_rate": 3.635394456289979e-05, "loss": 0.0491, "step": 640 }, { "epoch": 1.375266524520256, "grad_norm": 1.9499658346176147, "learning_rate": 3.624733475479744e-05, "loss": 0.0634, "step": 645 }, { "epoch": 1.3859275053304905, "grad_norm": 0.28151124715805054, "learning_rate": 3.61407249466951e-05, "loss": 0.1227, "step": 650 }, { "epoch": 1.396588486140725, "grad_norm": 5.18129825592041, "learning_rate": 3.603411513859275e-05, "loss": 0.0583, "step": 655 }, { "epoch": 1.4072494669509594, "grad_norm": 4.646321773529053, "learning_rate": 3.592750533049041e-05, "loss": 0.0827, "step": 660 }, { "epoch": 1.417910447761194, "grad_norm": 0.6517966389656067, "learning_rate": 3.582089552238806e-05, "loss": 0.0656, "step": 665 }, { "epoch": 1.4285714285714286, "grad_norm": 0.6320937871932983, "learning_rate": 3.571428571428572e-05, "loss": 0.0899, "step": 670 }, { "epoch": 1.439232409381663, "grad_norm": 3.7697184085845947, "learning_rate": 3.560767590618337e-05, "loss": 0.0973, "step": 675 }, { "epoch": 1.4498933901918978, "grad_norm": 7.1296610832214355, "learning_rate": 3.5501066098081026e-05, "loss": 0.1663, "step": 680 }, { "epoch": 1.4605543710021323, "grad_norm": 2.1503608226776123, "learning_rate": 3.539445628997868e-05, "loss": 0.0993, "step": 685 }, { "epoch": 1.4712153518123667, "grad_norm": 4.715859889984131, "learning_rate": 3.5287846481876336e-05, "loss": 0.0881, "step": 690 }, { "epoch": 1.4818763326226012, "grad_norm": 0.4445961117744446, "learning_rate": 3.518123667377399e-05, "loss": 0.0216, "step": 695 }, { "epoch": 1.4925373134328357, "grad_norm": 1.1854264736175537, "learning_rate": 3.5074626865671645e-05, "loss": 0.0988, "step": 700 }, { "epoch": 1.5031982942430704, "grad_norm": 3.872591495513916, "learning_rate": 3.496801705756929e-05, "loss": 0.1236, "step": 705 }, { "epoch": 1.5138592750533049, "grad_norm": 0.996483564376831, "learning_rate": 3.4861407249466955e-05, "loss": 0.0725, "step": 710 }, { "epoch": 1.5245202558635396, "grad_norm": 2.6979429721832275, "learning_rate": 3.47547974413646e-05, "loss": 0.1071, "step": 715 }, { "epoch": 1.535181236673774, "grad_norm": 2.521070957183838, "learning_rate": 3.4648187633262264e-05, "loss": 0.0906, "step": 720 }, { "epoch": 1.5458422174840085, "grad_norm": 3.037982702255249, "learning_rate": 3.454157782515991e-05, "loss": 0.0524, "step": 725 }, { "epoch": 1.556503198294243, "grad_norm": 1.728571891784668, "learning_rate": 3.4434968017057574e-05, "loss": 0.0741, "step": 730 }, { "epoch": 1.5671641791044775, "grad_norm": 1.142603874206543, "learning_rate": 3.432835820895522e-05, "loss": 0.0374, "step": 735 }, { "epoch": 1.5778251599147122, "grad_norm": 4.268226146697998, "learning_rate": 3.422174840085288e-05, "loss": 0.1548, "step": 740 }, { "epoch": 1.5884861407249466, "grad_norm": 0.666825532913208, "learning_rate": 3.411513859275053e-05, "loss": 0.0659, "step": 745 }, { "epoch": 1.5991471215351813, "grad_norm": 4.599668502807617, "learning_rate": 3.400852878464819e-05, "loss": 0.0637, "step": 750 }, { "epoch": 1.6098081023454158, "grad_norm": 1.2907124757766724, "learning_rate": 3.390191897654584e-05, "loss": 0.1011, "step": 755 }, { "epoch": 1.6204690831556503, "grad_norm": 4.451695442199707, "learning_rate": 3.37953091684435e-05, "loss": 0.1136, "step": 760 }, { "epoch": 1.6311300639658848, "grad_norm": 2.0966134071350098, "learning_rate": 3.368869936034115e-05, "loss": 0.0783, "step": 765 }, { "epoch": 1.6417910447761193, "grad_norm": 0.4260331988334656, "learning_rate": 3.358208955223881e-05, "loss": 0.0651, "step": 770 }, { "epoch": 1.652452025586354, "grad_norm": 3.2982163429260254, "learning_rate": 3.347547974413646e-05, "loss": 0.0829, "step": 775 }, { "epoch": 1.6631130063965884, "grad_norm": 2.151705265045166, "learning_rate": 3.336886993603412e-05, "loss": 0.1155, "step": 780 }, { "epoch": 1.6737739872068231, "grad_norm": 3.8690342903137207, "learning_rate": 3.326226012793177e-05, "loss": 0.0539, "step": 785 }, { "epoch": 1.6844349680170576, "grad_norm": 1.0087538957595825, "learning_rate": 3.3155650319829424e-05, "loss": 0.0725, "step": 790 }, { "epoch": 1.695095948827292, "grad_norm": 0.7453776597976685, "learning_rate": 3.304904051172708e-05, "loss": 0.0837, "step": 795 }, { "epoch": 1.7057569296375266, "grad_norm": 1.1071205139160156, "learning_rate": 3.294243070362473e-05, "loss": 0.0608, "step": 800 }, { "epoch": 1.716417910447761, "grad_norm": 0.3818202614784241, "learning_rate": 3.283582089552239e-05, "loss": 0.081, "step": 805 }, { "epoch": 1.7270788912579957, "grad_norm": 3.340217351913452, "learning_rate": 3.272921108742004e-05, "loss": 0.0452, "step": 810 }, { "epoch": 1.7377398720682304, "grad_norm": 7.753445148468018, "learning_rate": 3.26226012793177e-05, "loss": 0.1454, "step": 815 }, { "epoch": 1.748400852878465, "grad_norm": 0.72555011510849, "learning_rate": 3.251599147121535e-05, "loss": 0.0821, "step": 820 }, { "epoch": 1.7590618336886994, "grad_norm": 3.5226473808288574, "learning_rate": 3.240938166311301e-05, "loss": 0.1097, "step": 825 }, { "epoch": 1.7697228144989339, "grad_norm": 6.742045879364014, "learning_rate": 3.230277185501066e-05, "loss": 0.1397, "step": 830 }, { "epoch": 1.7803837953091683, "grad_norm": 4.19145393371582, "learning_rate": 3.2196162046908317e-05, "loss": 0.1146, "step": 835 }, { "epoch": 1.7910447761194028, "grad_norm": 3.483776807785034, "learning_rate": 3.208955223880597e-05, "loss": 0.1214, "step": 840 }, { "epoch": 1.8017057569296375, "grad_norm": 1.7062664031982422, "learning_rate": 3.1982942430703626e-05, "loss": 0.1022, "step": 845 }, { "epoch": 1.8123667377398722, "grad_norm": 0.6988951563835144, "learning_rate": 3.187633262260128e-05, "loss": 0.0568, "step": 850 }, { "epoch": 1.8230277185501067, "grad_norm": 3.394421339035034, "learning_rate": 3.1769722814498935e-05, "loss": 0.0678, "step": 855 }, { "epoch": 1.8336886993603412, "grad_norm": 2.9316060543060303, "learning_rate": 3.166311300639659e-05, "loss": 0.1152, "step": 860 }, { "epoch": 1.8443496801705757, "grad_norm": 2.0208592414855957, "learning_rate": 3.1556503198294245e-05, "loss": 0.088, "step": 865 }, { "epoch": 1.8550106609808101, "grad_norm": 3.43327260017395, "learning_rate": 3.14498933901919e-05, "loss": 0.1156, "step": 870 }, { "epoch": 1.8656716417910446, "grad_norm": 1.629196286201477, "learning_rate": 3.1343283582089554e-05, "loss": 0.0228, "step": 875 }, { "epoch": 1.8763326226012793, "grad_norm": 1.165114164352417, "learning_rate": 3.123667377398721e-05, "loss": 0.1128, "step": 880 }, { "epoch": 1.886993603411514, "grad_norm": 0.8657171726226807, "learning_rate": 3.1130063965884864e-05, "loss": 0.0759, "step": 885 }, { "epoch": 1.8976545842217485, "grad_norm": 1.936814546585083, "learning_rate": 3.102345415778252e-05, "loss": 0.0665, "step": 890 }, { "epoch": 1.908315565031983, "grad_norm": 0.4630189836025238, "learning_rate": 3.0916844349680173e-05, "loss": 0.0558, "step": 895 }, { "epoch": 1.9189765458422174, "grad_norm": 2.1839075088500977, "learning_rate": 3.081023454157783e-05, "loss": 0.0362, "step": 900 }, { "epoch": 1.929637526652452, "grad_norm": 2.0083508491516113, "learning_rate": 3.070362473347548e-05, "loss": 0.086, "step": 905 }, { "epoch": 1.9402985074626866, "grad_norm": 0.6220686435699463, "learning_rate": 3.059701492537314e-05, "loss": 0.0968, "step": 910 }, { "epoch": 1.950959488272921, "grad_norm": 3.643712043762207, "learning_rate": 3.0490405117270792e-05, "loss": 0.0933, "step": 915 }, { "epoch": 1.9616204690831558, "grad_norm": 0.08588960021734238, "learning_rate": 3.0383795309168444e-05, "loss": 0.0574, "step": 920 }, { "epoch": 1.9722814498933903, "grad_norm": 0.6295548677444458, "learning_rate": 3.0277185501066102e-05, "loss": 0.0852, "step": 925 }, { "epoch": 1.9829424307036247, "grad_norm": 0.248437762260437, "learning_rate": 3.0170575692963753e-05, "loss": 0.1085, "step": 930 }, { "epoch": 1.9936034115138592, "grad_norm": 1.4025636911392212, "learning_rate": 3.006396588486141e-05, "loss": 0.0766, "step": 935 }, { "epoch": 2.0, "eval_accuracy": 0.9882666666666666, "eval_loss": 0.03895151615142822, "eval_runtime": 29.2543, "eval_samples_per_second": 128.186, "eval_steps_per_second": 4.034, "step": 938 }, { "epoch": 2.0042643923240937, "grad_norm": 2.564427614212036, "learning_rate": 2.9957356076759063e-05, "loss": 0.1046, "step": 940 }, { "epoch": 2.014925373134328, "grad_norm": 1.40167236328125, "learning_rate": 2.9850746268656714e-05, "loss": 0.0522, "step": 945 }, { "epoch": 2.025586353944563, "grad_norm": 0.8157774209976196, "learning_rate": 2.9744136460554372e-05, "loss": 0.0219, "step": 950 }, { "epoch": 2.0362473347547976, "grad_norm": 2.9061124324798584, "learning_rate": 2.9637526652452023e-05, "loss": 0.0811, "step": 955 }, { "epoch": 2.046908315565032, "grad_norm": 0.7512766718864441, "learning_rate": 2.953091684434968e-05, "loss": 0.0349, "step": 960 }, { "epoch": 2.0575692963752665, "grad_norm": 5.608509063720703, "learning_rate": 2.9424307036247333e-05, "loss": 0.108, "step": 965 }, { "epoch": 2.068230277185501, "grad_norm": 1.4548602104187012, "learning_rate": 2.931769722814499e-05, "loss": 0.0949, "step": 970 }, { "epoch": 2.0788912579957355, "grad_norm": 3.231023073196411, "learning_rate": 2.9211087420042642e-05, "loss": 0.0782, "step": 975 }, { "epoch": 2.08955223880597, "grad_norm": 6.984096050262451, "learning_rate": 2.91044776119403e-05, "loss": 0.1128, "step": 980 }, { "epoch": 2.100213219616205, "grad_norm": 0.13570423424243927, "learning_rate": 2.8997867803837952e-05, "loss": 0.0262, "step": 985 }, { "epoch": 2.1108742004264394, "grad_norm": 1.98729407787323, "learning_rate": 2.889125799573561e-05, "loss": 0.0495, "step": 990 }, { "epoch": 2.121535181236674, "grad_norm": 1.8757045269012451, "learning_rate": 2.878464818763326e-05, "loss": 0.0687, "step": 995 }, { "epoch": 2.1321961620469083, "grad_norm": 2.904733896255493, "learning_rate": 2.867803837953092e-05, "loss": 0.0364, "step": 1000 }, { "epoch": 2.142857142857143, "grad_norm": 2.2336878776550293, "learning_rate": 2.857142857142857e-05, "loss": 0.0409, "step": 1005 }, { "epoch": 2.1535181236673773, "grad_norm": 4.610574245452881, "learning_rate": 2.846481876332623e-05, "loss": 0.049, "step": 1010 }, { "epoch": 2.1641791044776117, "grad_norm": 2.506977081298828, "learning_rate": 2.835820895522388e-05, "loss": 0.2106, "step": 1015 }, { "epoch": 2.1748400852878467, "grad_norm": 1.8474242687225342, "learning_rate": 2.825159914712154e-05, "loss": 0.0611, "step": 1020 }, { "epoch": 2.185501066098081, "grad_norm": 2.107987642288208, "learning_rate": 2.814498933901919e-05, "loss": 0.1191, "step": 1025 }, { "epoch": 2.1961620469083156, "grad_norm": 1.3394441604614258, "learning_rate": 2.8038379530916848e-05, "loss": 0.0812, "step": 1030 }, { "epoch": 2.20682302771855, "grad_norm": 2.769834518432617, "learning_rate": 2.79317697228145e-05, "loss": 0.0587, "step": 1035 }, { "epoch": 2.2174840085287846, "grad_norm": 0.04353656619787216, "learning_rate": 2.7825159914712157e-05, "loss": 0.0083, "step": 1040 }, { "epoch": 2.228144989339019, "grad_norm": 0.6627817749977112, "learning_rate": 2.771855010660981e-05, "loss": 0.0731, "step": 1045 }, { "epoch": 2.2388059701492535, "grad_norm": 0.1595916748046875, "learning_rate": 2.7611940298507467e-05, "loss": 0.0593, "step": 1050 }, { "epoch": 2.2494669509594885, "grad_norm": 2.4652700424194336, "learning_rate": 2.7505330490405118e-05, "loss": 0.0197, "step": 1055 }, { "epoch": 2.260127931769723, "grad_norm": 1.91483736038208, "learning_rate": 2.7398720682302776e-05, "loss": 0.0984, "step": 1060 }, { "epoch": 2.2707889125799574, "grad_norm": 2.1739375591278076, "learning_rate": 2.7292110874200428e-05, "loss": 0.1015, "step": 1065 }, { "epoch": 2.281449893390192, "grad_norm": 6.4230828285217285, "learning_rate": 2.7185501066098086e-05, "loss": 0.0454, "step": 1070 }, { "epoch": 2.2921108742004264, "grad_norm": 2.000547409057617, "learning_rate": 2.7078891257995737e-05, "loss": 0.0722, "step": 1075 }, { "epoch": 2.302771855010661, "grad_norm": 5.884206295013428, "learning_rate": 2.6972281449893395e-05, "loss": 0.0764, "step": 1080 }, { "epoch": 2.3134328358208958, "grad_norm": 0.7102048993110657, "learning_rate": 2.6865671641791047e-05, "loss": 0.0724, "step": 1085 }, { "epoch": 2.3240938166311302, "grad_norm": 1.670265793800354, "learning_rate": 2.6759061833688705e-05, "loss": 0.0967, "step": 1090 }, { "epoch": 2.3347547974413647, "grad_norm": 0.5737790465354919, "learning_rate": 2.6652452025586356e-05, "loss": 0.0509, "step": 1095 }, { "epoch": 2.345415778251599, "grad_norm": 0.8868186473846436, "learning_rate": 2.6545842217484007e-05, "loss": 0.0614, "step": 1100 }, { "epoch": 2.3560767590618337, "grad_norm": 7.29151725769043, "learning_rate": 2.6439232409381666e-05, "loss": 0.0438, "step": 1105 }, { "epoch": 2.366737739872068, "grad_norm": 1.0474796295166016, "learning_rate": 2.6332622601279317e-05, "loss": 0.0528, "step": 1110 }, { "epoch": 2.3773987206823026, "grad_norm": 4.946176052093506, "learning_rate": 2.6226012793176975e-05, "loss": 0.1153, "step": 1115 }, { "epoch": 2.388059701492537, "grad_norm": 0.3259846866130829, "learning_rate": 2.6119402985074626e-05, "loss": 0.0526, "step": 1120 }, { "epoch": 2.398720682302772, "grad_norm": 1.7642853260040283, "learning_rate": 2.6012793176972285e-05, "loss": 0.0792, "step": 1125 }, { "epoch": 2.4093816631130065, "grad_norm": 3.1603195667266846, "learning_rate": 2.5906183368869936e-05, "loss": 0.0339, "step": 1130 }, { "epoch": 2.420042643923241, "grad_norm": 0.149946391582489, "learning_rate": 2.5799573560767594e-05, "loss": 0.0388, "step": 1135 }, { "epoch": 2.4307036247334755, "grad_norm": 0.10843349248170853, "learning_rate": 2.5692963752665245e-05, "loss": 0.0581, "step": 1140 }, { "epoch": 2.44136460554371, "grad_norm": 3.565418004989624, "learning_rate": 2.5586353944562904e-05, "loss": 0.0773, "step": 1145 }, { "epoch": 2.4520255863539444, "grad_norm": 3.206373453140259, "learning_rate": 2.5479744136460555e-05, "loss": 0.0788, "step": 1150 }, { "epoch": 2.4626865671641793, "grad_norm": 0.7184895873069763, "learning_rate": 2.537313432835821e-05, "loss": 0.096, "step": 1155 }, { "epoch": 2.473347547974414, "grad_norm": 3.5069594383239746, "learning_rate": 2.5266524520255864e-05, "loss": 0.1027, "step": 1160 }, { "epoch": 2.4840085287846483, "grad_norm": 0.37467339634895325, "learning_rate": 2.515991471215352e-05, "loss": 0.0573, "step": 1165 }, { "epoch": 2.4946695095948828, "grad_norm": 0.6130836606025696, "learning_rate": 2.5053304904051174e-05, "loss": 0.0651, "step": 1170 }, { "epoch": 2.5053304904051172, "grad_norm": 1.5621353387832642, "learning_rate": 2.494669509594883e-05, "loss": 0.0668, "step": 1175 }, { "epoch": 2.5159914712153517, "grad_norm": 5.647741317749023, "learning_rate": 2.4840085287846483e-05, "loss": 0.0966, "step": 1180 }, { "epoch": 2.526652452025586, "grad_norm": 2.5682458877563477, "learning_rate": 2.4733475479744138e-05, "loss": 0.0478, "step": 1185 }, { "epoch": 2.5373134328358207, "grad_norm": 0.39340585470199585, "learning_rate": 2.4626865671641793e-05, "loss": 0.0561, "step": 1190 }, { "epoch": 2.5479744136460556, "grad_norm": 0.33411240577697754, "learning_rate": 2.4520255863539444e-05, "loss": 0.0418, "step": 1195 }, { "epoch": 2.55863539445629, "grad_norm": 0.18760384619235992, "learning_rate": 2.44136460554371e-05, "loss": 0.0458, "step": 1200 }, { "epoch": 2.5692963752665245, "grad_norm": 2.615713119506836, "learning_rate": 2.4307036247334754e-05, "loss": 0.061, "step": 1205 }, { "epoch": 2.579957356076759, "grad_norm": 3.9026904106140137, "learning_rate": 2.420042643923241e-05, "loss": 0.0418, "step": 1210 }, { "epoch": 2.5906183368869935, "grad_norm": 8.405099868774414, "learning_rate": 2.4093816631130063e-05, "loss": 0.0729, "step": 1215 }, { "epoch": 2.6012793176972284, "grad_norm": 1.3211092948913574, "learning_rate": 2.3987206823027718e-05, "loss": 0.0508, "step": 1220 }, { "epoch": 2.611940298507463, "grad_norm": 1.9056097269058228, "learning_rate": 2.3880597014925373e-05, "loss": 0.1017, "step": 1225 }, { "epoch": 2.6226012793176974, "grad_norm": 4.098701477050781, "learning_rate": 2.3773987206823027e-05, "loss": 0.0628, "step": 1230 }, { "epoch": 2.633262260127932, "grad_norm": 0.12911789119243622, "learning_rate": 2.3667377398720682e-05, "loss": 0.0553, "step": 1235 }, { "epoch": 2.6439232409381663, "grad_norm": 4.947664737701416, "learning_rate": 2.3560767590618337e-05, "loss": 0.0381, "step": 1240 }, { "epoch": 2.654584221748401, "grad_norm": 2.0366995334625244, "learning_rate": 2.345415778251599e-05, "loss": 0.0666, "step": 1245 }, { "epoch": 2.6652452025586353, "grad_norm": 0.7470578551292419, "learning_rate": 2.3347547974413646e-05, "loss": 0.037, "step": 1250 }, { "epoch": 2.6759061833688698, "grad_norm": 2.562520742416382, "learning_rate": 2.32409381663113e-05, "loss": 0.0476, "step": 1255 }, { "epoch": 2.6865671641791042, "grad_norm": 0.5824430584907532, "learning_rate": 2.3134328358208956e-05, "loss": 0.0416, "step": 1260 }, { "epoch": 2.697228144989339, "grad_norm": 5.459311485290527, "learning_rate": 2.302771855010661e-05, "loss": 0.1057, "step": 1265 }, { "epoch": 2.7078891257995736, "grad_norm": 4.892146587371826, "learning_rate": 2.2921108742004265e-05, "loss": 0.0431, "step": 1270 }, { "epoch": 2.718550106609808, "grad_norm": 5.7114033699035645, "learning_rate": 2.281449893390192e-05, "loss": 0.0683, "step": 1275 }, { "epoch": 2.7292110874200426, "grad_norm": 0.9516615271568298, "learning_rate": 2.2707889125799575e-05, "loss": 0.0579, "step": 1280 }, { "epoch": 2.739872068230277, "grad_norm": 3.072978973388672, "learning_rate": 2.260127931769723e-05, "loss": 0.0931, "step": 1285 }, { "epoch": 2.750533049040512, "grad_norm": 1.261651873588562, "learning_rate": 2.2494669509594884e-05, "loss": 0.0902, "step": 1290 }, { "epoch": 2.7611940298507465, "grad_norm": 6.886789798736572, "learning_rate": 2.238805970149254e-05, "loss": 0.0863, "step": 1295 }, { "epoch": 2.771855010660981, "grad_norm": 5.224747180938721, "learning_rate": 2.2281449893390194e-05, "loss": 0.0762, "step": 1300 }, { "epoch": 2.7825159914712154, "grad_norm": 0.15022771060466766, "learning_rate": 2.217484008528785e-05, "loss": 0.0379, "step": 1305 }, { "epoch": 2.79317697228145, "grad_norm": 2.850597620010376, "learning_rate": 2.2068230277185503e-05, "loss": 0.0501, "step": 1310 }, { "epoch": 2.8038379530916844, "grad_norm": 0.5554096102714539, "learning_rate": 2.1961620469083158e-05, "loss": 0.0419, "step": 1315 }, { "epoch": 2.814498933901919, "grad_norm": 0.3497494161128998, "learning_rate": 2.1855010660980813e-05, "loss": 0.0701, "step": 1320 }, { "epoch": 2.8251599147121533, "grad_norm": 0.4103732705116272, "learning_rate": 2.1748400852878467e-05, "loss": 0.0375, "step": 1325 }, { "epoch": 2.835820895522388, "grad_norm": 4.490753650665283, "learning_rate": 2.164179104477612e-05, "loss": 0.0379, "step": 1330 }, { "epoch": 2.8464818763326227, "grad_norm": 0.15869414806365967, "learning_rate": 2.1535181236673773e-05, "loss": 0.0483, "step": 1335 }, { "epoch": 2.857142857142857, "grad_norm": 1.1513383388519287, "learning_rate": 2.1428571428571428e-05, "loss": 0.062, "step": 1340 }, { "epoch": 2.8678038379530917, "grad_norm": 1.6376324892044067, "learning_rate": 2.1321961620469083e-05, "loss": 0.0311, "step": 1345 }, { "epoch": 2.878464818763326, "grad_norm": 1.549334168434143, "learning_rate": 2.1215351812366738e-05, "loss": 0.0285, "step": 1350 }, { "epoch": 2.8891257995735606, "grad_norm": 0.6793725490570068, "learning_rate": 2.1108742004264392e-05, "loss": 0.0303, "step": 1355 }, { "epoch": 2.8997867803837956, "grad_norm": 6.082405090332031, "learning_rate": 2.1002132196162047e-05, "loss": 0.0591, "step": 1360 }, { "epoch": 2.91044776119403, "grad_norm": 1.481665849685669, "learning_rate": 2.0895522388059702e-05, "loss": 0.0476, "step": 1365 }, { "epoch": 2.9211087420042645, "grad_norm": 4.322070121765137, "learning_rate": 2.0788912579957357e-05, "loss": 0.0277, "step": 1370 }, { "epoch": 2.931769722814499, "grad_norm": 2.1505560874938965, "learning_rate": 2.068230277185501e-05, "loss": 0.0964, "step": 1375 }, { "epoch": 2.9424307036247335, "grad_norm": 5.245396614074707, "learning_rate": 2.0575692963752666e-05, "loss": 0.0395, "step": 1380 }, { "epoch": 2.953091684434968, "grad_norm": 0.7271926999092102, "learning_rate": 2.046908315565032e-05, "loss": 0.0643, "step": 1385 }, { "epoch": 2.9637526652452024, "grad_norm": 0.16647839546203613, "learning_rate": 2.0362473347547976e-05, "loss": 0.0678, "step": 1390 }, { "epoch": 2.974413646055437, "grad_norm": 6.918933868408203, "learning_rate": 2.025586353944563e-05, "loss": 0.0349, "step": 1395 }, { "epoch": 2.9850746268656714, "grad_norm": 4.1197919845581055, "learning_rate": 2.0149253731343285e-05, "loss": 0.0777, "step": 1400 }, { "epoch": 2.9957356076759063, "grad_norm": 4.218685626983643, "learning_rate": 2.004264392324094e-05, "loss": 0.0501, "step": 1405 }, { "epoch": 3.0, "eval_accuracy": 0.9893333333333333, "eval_loss": 0.03320843353867531, "eval_runtime": 29.2407, "eval_samples_per_second": 128.246, "eval_steps_per_second": 4.035, "step": 1407 }, { "epoch": 3.0063965884861408, "grad_norm": 0.03237692266702652, "learning_rate": 1.9936034115138594e-05, "loss": 0.0642, "step": 1410 }, { "epoch": 3.0170575692963753, "grad_norm": 0.7085376381874084, "learning_rate": 1.982942430703625e-05, "loss": 0.0409, "step": 1415 }, { "epoch": 3.0277185501066097, "grad_norm": 3.4055681228637695, "learning_rate": 1.9722814498933904e-05, "loss": 0.059, "step": 1420 }, { "epoch": 3.038379530916844, "grad_norm": 2.316131114959717, "learning_rate": 1.961620469083156e-05, "loss": 0.1122, "step": 1425 }, { "epoch": 3.0490405117270787, "grad_norm": 0.7073843479156494, "learning_rate": 1.9509594882729213e-05, "loss": 0.0474, "step": 1430 }, { "epoch": 3.0597014925373136, "grad_norm": 0.6925232410430908, "learning_rate": 1.9402985074626868e-05, "loss": 0.0724, "step": 1435 }, { "epoch": 3.070362473347548, "grad_norm": 1.898357629776001, "learning_rate": 1.9296375266524523e-05, "loss": 0.0362, "step": 1440 }, { "epoch": 3.0810234541577826, "grad_norm": 0.5176880955696106, "learning_rate": 1.9189765458422178e-05, "loss": 0.0532, "step": 1445 }, { "epoch": 3.091684434968017, "grad_norm": 0.27506181597709656, "learning_rate": 1.9083155650319832e-05, "loss": 0.0682, "step": 1450 }, { "epoch": 3.1023454157782515, "grad_norm": 2.064847469329834, "learning_rate": 1.8976545842217487e-05, "loss": 0.0832, "step": 1455 }, { "epoch": 3.113006396588486, "grad_norm": 1.1283111572265625, "learning_rate": 1.8869936034115142e-05, "loss": 0.0704, "step": 1460 }, { "epoch": 3.1236673773987205, "grad_norm": 2.303053855895996, "learning_rate": 1.8763326226012797e-05, "loss": 0.0774, "step": 1465 }, { "epoch": 3.1343283582089554, "grad_norm": 3.597050428390503, "learning_rate": 1.865671641791045e-05, "loss": 0.0998, "step": 1470 }, { "epoch": 3.14498933901919, "grad_norm": 0.05906692147254944, "learning_rate": 1.8550106609808106e-05, "loss": 0.0188, "step": 1475 }, { "epoch": 3.1556503198294243, "grad_norm": 0.2777367830276489, "learning_rate": 1.8443496801705757e-05, "loss": 0.0529, "step": 1480 }, { "epoch": 3.166311300639659, "grad_norm": 3.8748035430908203, "learning_rate": 1.8336886993603412e-05, "loss": 0.0446, "step": 1485 }, { "epoch": 3.1769722814498933, "grad_norm": 1.977518081665039, "learning_rate": 1.8230277185501067e-05, "loss": 0.0732, "step": 1490 }, { "epoch": 3.1876332622601278, "grad_norm": 7.160504341125488, "learning_rate": 1.812366737739872e-05, "loss": 0.0549, "step": 1495 }, { "epoch": 3.1982942430703627, "grad_norm": 0.3385598361492157, "learning_rate": 1.8017057569296376e-05, "loss": 0.0116, "step": 1500 }, { "epoch": 3.208955223880597, "grad_norm": 0.21955998241901398, "learning_rate": 1.791044776119403e-05, "loss": 0.0491, "step": 1505 }, { "epoch": 3.2196162046908317, "grad_norm": 2.725681781768799, "learning_rate": 1.7803837953091686e-05, "loss": 0.0313, "step": 1510 }, { "epoch": 3.230277185501066, "grad_norm": 0.04153380170464516, "learning_rate": 1.769722814498934e-05, "loss": 0.0123, "step": 1515 }, { "epoch": 3.2409381663113006, "grad_norm": 0.49530500173568726, "learning_rate": 1.7590618336886995e-05, "loss": 0.0525, "step": 1520 }, { "epoch": 3.251599147121535, "grad_norm": 5.86508846282959, "learning_rate": 1.7484008528784647e-05, "loss": 0.0583, "step": 1525 }, { "epoch": 3.2622601279317696, "grad_norm": 11.29727840423584, "learning_rate": 1.73773987206823e-05, "loss": 0.1008, "step": 1530 }, { "epoch": 3.272921108742004, "grad_norm": 2.0493924617767334, "learning_rate": 1.7270788912579956e-05, "loss": 0.0163, "step": 1535 }, { "epoch": 3.283582089552239, "grad_norm": 0.6533018350601196, "learning_rate": 1.716417910447761e-05, "loss": 0.0796, "step": 1540 }, { "epoch": 3.2942430703624734, "grad_norm": 0.9406389594078064, "learning_rate": 1.7057569296375266e-05, "loss": 0.0193, "step": 1545 }, { "epoch": 3.304904051172708, "grad_norm": 0.7454879283905029, "learning_rate": 1.695095948827292e-05, "loss": 0.0908, "step": 1550 }, { "epoch": 3.3155650319829424, "grad_norm": 0.1846035122871399, "learning_rate": 1.6844349680170575e-05, "loss": 0.1085, "step": 1555 }, { "epoch": 3.326226012793177, "grad_norm": 3.9499289989471436, "learning_rate": 1.673773987206823e-05, "loss": 0.046, "step": 1560 }, { "epoch": 3.3368869936034113, "grad_norm": 0.14025430381298065, "learning_rate": 1.6631130063965885e-05, "loss": 0.0327, "step": 1565 }, { "epoch": 3.3475479744136463, "grad_norm": 1.1740471124649048, "learning_rate": 1.652452025586354e-05, "loss": 0.0562, "step": 1570 }, { "epoch": 3.3582089552238807, "grad_norm": 3.9793527126312256, "learning_rate": 1.6417910447761194e-05, "loss": 0.0441, "step": 1575 }, { "epoch": 3.368869936034115, "grad_norm": 4.597386360168457, "learning_rate": 1.631130063965885e-05, "loss": 0.0723, "step": 1580 }, { "epoch": 3.3795309168443497, "grad_norm": 0.02715531922876835, "learning_rate": 1.6204690831556504e-05, "loss": 0.0128, "step": 1585 }, { "epoch": 3.390191897654584, "grad_norm": 7.36865234375, "learning_rate": 1.6098081023454158e-05, "loss": 0.0926, "step": 1590 }, { "epoch": 3.4008528784648187, "grad_norm": 4.011542797088623, "learning_rate": 1.5991471215351813e-05, "loss": 0.1018, "step": 1595 }, { "epoch": 3.411513859275053, "grad_norm": 0.9654463529586792, "learning_rate": 1.5884861407249468e-05, "loss": 0.0163, "step": 1600 }, { "epoch": 3.4221748400852876, "grad_norm": 4.028111457824707, "learning_rate": 1.5778251599147122e-05, "loss": 0.0944, "step": 1605 }, { "epoch": 3.4328358208955225, "grad_norm": 0.027582503855228424, "learning_rate": 1.5671641791044777e-05, "loss": 0.1142, "step": 1610 }, { "epoch": 3.443496801705757, "grad_norm": 0.1788293868303299, "learning_rate": 1.5565031982942432e-05, "loss": 0.0204, "step": 1615 }, { "epoch": 3.4541577825159915, "grad_norm": 6.049529075622559, "learning_rate": 1.5458422174840087e-05, "loss": 0.0842, "step": 1620 }, { "epoch": 3.464818763326226, "grad_norm": 1.3860670328140259, "learning_rate": 1.535181236673774e-05, "loss": 0.0488, "step": 1625 }, { "epoch": 3.4754797441364604, "grad_norm": 2.400519609451294, "learning_rate": 1.5245202558635396e-05, "loss": 0.1417, "step": 1630 }, { "epoch": 3.486140724946695, "grad_norm": 0.5289180278778076, "learning_rate": 1.5138592750533051e-05, "loss": 0.0299, "step": 1635 }, { "epoch": 3.49680170575693, "grad_norm": 1.9841607809066772, "learning_rate": 1.5031982942430706e-05, "loss": 0.1442, "step": 1640 }, { "epoch": 3.5074626865671643, "grad_norm": 0.6429154872894287, "learning_rate": 1.4925373134328357e-05, "loss": 0.0771, "step": 1645 }, { "epoch": 3.518123667377399, "grad_norm": 0.40449410676956177, "learning_rate": 1.4818763326226012e-05, "loss": 0.016, "step": 1650 }, { "epoch": 3.5287846481876333, "grad_norm": 3.1155402660369873, "learning_rate": 1.4712153518123666e-05, "loss": 0.0727, "step": 1655 }, { "epoch": 3.5394456289978677, "grad_norm": 1.3135961294174194, "learning_rate": 1.4605543710021321e-05, "loss": 0.048, "step": 1660 }, { "epoch": 3.550106609808102, "grad_norm": 0.13344089686870575, "learning_rate": 1.4498933901918976e-05, "loss": 0.0299, "step": 1665 }, { "epoch": 3.5607675906183367, "grad_norm": 4.653741359710693, "learning_rate": 1.439232409381663e-05, "loss": 0.082, "step": 1670 }, { "epoch": 3.571428571428571, "grad_norm": 0.11814280599355698, "learning_rate": 1.4285714285714285e-05, "loss": 0.0128, "step": 1675 }, { "epoch": 3.582089552238806, "grad_norm": 7.6886467933654785, "learning_rate": 1.417910447761194e-05, "loss": 0.0961, "step": 1680 }, { "epoch": 3.5927505330490406, "grad_norm": 5.170369625091553, "learning_rate": 1.4072494669509595e-05, "loss": 0.0766, "step": 1685 }, { "epoch": 3.603411513859275, "grad_norm": 5.314568996429443, "learning_rate": 1.396588486140725e-05, "loss": 0.067, "step": 1690 }, { "epoch": 3.6140724946695095, "grad_norm": 0.07409282773733139, "learning_rate": 1.3859275053304904e-05, "loss": 0.041, "step": 1695 }, { "epoch": 3.624733475479744, "grad_norm": 6.101559638977051, "learning_rate": 1.3752665245202559e-05, "loss": 0.0513, "step": 1700 }, { "epoch": 3.635394456289979, "grad_norm": 2.0448861122131348, "learning_rate": 1.3646055437100214e-05, "loss": 0.0272, "step": 1705 }, { "epoch": 3.6460554371002134, "grad_norm": 3.6384999752044678, "learning_rate": 1.3539445628997869e-05, "loss": 0.0287, "step": 1710 }, { "epoch": 3.656716417910448, "grad_norm": 0.03651139512658119, "learning_rate": 1.3432835820895523e-05, "loss": 0.0484, "step": 1715 }, { "epoch": 3.6673773987206824, "grad_norm": 4.424672603607178, "learning_rate": 1.3326226012793178e-05, "loss": 0.0254, "step": 1720 }, { "epoch": 3.678038379530917, "grad_norm": 0.863547146320343, "learning_rate": 1.3219616204690833e-05, "loss": 0.0379, "step": 1725 }, { "epoch": 3.6886993603411513, "grad_norm": 0.2523600161075592, "learning_rate": 1.3113006396588488e-05, "loss": 0.0789, "step": 1730 }, { "epoch": 3.699360341151386, "grad_norm": 0.797463059425354, "learning_rate": 1.3006396588486142e-05, "loss": 0.0573, "step": 1735 }, { "epoch": 3.7100213219616203, "grad_norm": 2.2122867107391357, "learning_rate": 1.2899786780383797e-05, "loss": 0.0529, "step": 1740 }, { "epoch": 3.7206823027718547, "grad_norm": 0.34208062291145325, "learning_rate": 1.2793176972281452e-05, "loss": 0.0377, "step": 1745 }, { "epoch": 3.7313432835820897, "grad_norm": 0.27025628089904785, "learning_rate": 1.2686567164179105e-05, "loss": 0.1455, "step": 1750 }, { "epoch": 3.742004264392324, "grad_norm": 4.990874290466309, "learning_rate": 1.257995735607676e-05, "loss": 0.0728, "step": 1755 }, { "epoch": 3.7526652452025586, "grad_norm": 2.51625919342041, "learning_rate": 1.2473347547974414e-05, "loss": 0.0161, "step": 1760 }, { "epoch": 3.763326226012793, "grad_norm": 1.0285941362380981, "learning_rate": 1.2366737739872069e-05, "loss": 0.025, "step": 1765 }, { "epoch": 3.7739872068230276, "grad_norm": 3.464445114135742, "learning_rate": 1.2260127931769722e-05, "loss": 0.0906, "step": 1770 }, { "epoch": 3.7846481876332625, "grad_norm": 1.4850032329559326, "learning_rate": 1.2153518123667377e-05, "loss": 0.1252, "step": 1775 }, { "epoch": 3.795309168443497, "grad_norm": 0.8000995516777039, "learning_rate": 1.2046908315565032e-05, "loss": 0.0658, "step": 1780 }, { "epoch": 3.8059701492537314, "grad_norm": 6.374350547790527, "learning_rate": 1.1940298507462686e-05, "loss": 0.0388, "step": 1785 }, { "epoch": 3.816631130063966, "grad_norm": 0.35630708932876587, "learning_rate": 1.1833688699360341e-05, "loss": 0.0844, "step": 1790 }, { "epoch": 3.8272921108742004, "grad_norm": 4.863065242767334, "learning_rate": 1.1727078891257996e-05, "loss": 0.0664, "step": 1795 }, { "epoch": 3.837953091684435, "grad_norm": 6.536571502685547, "learning_rate": 1.162046908315565e-05, "loss": 0.044, "step": 1800 }, { "epoch": 3.8486140724946694, "grad_norm": 1.5584124326705933, "learning_rate": 1.1513859275053305e-05, "loss": 0.03, "step": 1805 }, { "epoch": 3.859275053304904, "grad_norm": 6.224547863006592, "learning_rate": 1.140724946695096e-05, "loss": 0.045, "step": 1810 }, { "epoch": 3.8699360341151388, "grad_norm": 8.088601112365723, "learning_rate": 1.1300639658848615e-05, "loss": 0.09, "step": 1815 }, { "epoch": 3.8805970149253732, "grad_norm": 4.969430923461914, "learning_rate": 1.119402985074627e-05, "loss": 0.081, "step": 1820 }, { "epoch": 3.8912579957356077, "grad_norm": 7.029250621795654, "learning_rate": 1.1087420042643924e-05, "loss": 0.016, "step": 1825 }, { "epoch": 3.901918976545842, "grad_norm": 1.4717775583267212, "learning_rate": 1.0980810234541579e-05, "loss": 0.0081, "step": 1830 }, { "epoch": 3.9125799573560767, "grad_norm": 1.592339277267456, "learning_rate": 1.0874200426439234e-05, "loss": 0.1263, "step": 1835 }, { "epoch": 3.923240938166311, "grad_norm": 0.2614741325378418, "learning_rate": 1.0767590618336887e-05, "loss": 0.0914, "step": 1840 }, { "epoch": 3.933901918976546, "grad_norm": 3.768350124359131, "learning_rate": 1.0660980810234541e-05, "loss": 0.1056, "step": 1845 }, { "epoch": 3.9445628997867805, "grad_norm": 2.808711290359497, "learning_rate": 1.0554371002132196e-05, "loss": 0.1152, "step": 1850 }, { "epoch": 3.955223880597015, "grad_norm": 4.131954193115234, "learning_rate": 1.0447761194029851e-05, "loss": 0.0246, "step": 1855 }, { "epoch": 3.9658848614072495, "grad_norm": 6.506793022155762, "learning_rate": 1.0341151385927506e-05, "loss": 0.0447, "step": 1860 }, { "epoch": 3.976545842217484, "grad_norm": 1.3220537900924683, "learning_rate": 1.023454157782516e-05, "loss": 0.0484, "step": 1865 }, { "epoch": 3.9872068230277184, "grad_norm": 1.1393160820007324, "learning_rate": 1.0127931769722815e-05, "loss": 0.0273, "step": 1870 }, { "epoch": 3.997867803837953, "grad_norm": 1.9221222400665283, "learning_rate": 1.002132196162047e-05, "loss": 0.0421, "step": 1875 }, { "epoch": 4.0, "eval_accuracy": 0.9925333333333334, "eval_loss": 0.024030033499002457, "eval_runtime": 29.4459, "eval_samples_per_second": 127.352, "eval_steps_per_second": 4.007, "step": 1876 }, { "epoch": 4.008528784648187, "grad_norm": 0.16687144339084625, "learning_rate": 9.914712153518125e-06, "loss": 0.0577, "step": 1880 }, { "epoch": 4.019189765458422, "grad_norm": 2.212986946105957, "learning_rate": 9.80810234541578e-06, "loss": 0.0496, "step": 1885 }, { "epoch": 4.029850746268656, "grad_norm": 4.696678161621094, "learning_rate": 9.701492537313434e-06, "loss": 0.0555, "step": 1890 }, { "epoch": 4.040511727078891, "grad_norm": 3.450464963912964, "learning_rate": 9.594882729211089e-06, "loss": 0.0534, "step": 1895 }, { "epoch": 4.051172707889126, "grad_norm": 6.538852691650391, "learning_rate": 9.488272921108744e-06, "loss": 0.0808, "step": 1900 }, { "epoch": 4.061833688699361, "grad_norm": 2.80407977104187, "learning_rate": 9.381663113006398e-06, "loss": 0.0544, "step": 1905 }, { "epoch": 4.072494669509595, "grad_norm": 0.021833008155226707, "learning_rate": 9.275053304904053e-06, "loss": 0.0161, "step": 1910 }, { "epoch": 4.08315565031983, "grad_norm": 4.794767379760742, "learning_rate": 9.168443496801706e-06, "loss": 0.0525, "step": 1915 }, { "epoch": 4.093816631130064, "grad_norm": 5.79359769821167, "learning_rate": 9.06183368869936e-06, "loss": 0.0318, "step": 1920 }, { "epoch": 4.104477611940299, "grad_norm": 1.9805699586868286, "learning_rate": 8.955223880597016e-06, "loss": 0.0931, "step": 1925 }, { "epoch": 4.115138592750533, "grad_norm": 0.21113300323486328, "learning_rate": 8.84861407249467e-06, "loss": 0.0425, "step": 1930 }, { "epoch": 4.1257995735607675, "grad_norm": 1.3944600820541382, "learning_rate": 8.742004264392323e-06, "loss": 0.0346, "step": 1935 }, { "epoch": 4.136460554371002, "grad_norm": 6.401991844177246, "learning_rate": 8.635394456289978e-06, "loss": 0.0243, "step": 1940 }, { "epoch": 4.1471215351812365, "grad_norm": 0.6892251968383789, "learning_rate": 8.528784648187633e-06, "loss": 0.0598, "step": 1945 }, { "epoch": 4.157782515991471, "grad_norm": 0.46206191182136536, "learning_rate": 8.422174840085288e-06, "loss": 0.0701, "step": 1950 }, { "epoch": 4.1684434968017055, "grad_norm": 2.1541647911071777, "learning_rate": 8.315565031982942e-06, "loss": 0.0198, "step": 1955 }, { "epoch": 4.17910447761194, "grad_norm": 1.302574872970581, "learning_rate": 8.208955223880597e-06, "loss": 0.0259, "step": 1960 }, { "epoch": 4.189765458422174, "grad_norm": 3.6631381511688232, "learning_rate": 8.102345415778252e-06, "loss": 0.0213, "step": 1965 }, { "epoch": 4.20042643923241, "grad_norm": 0.8104808926582336, "learning_rate": 7.995735607675907e-06, "loss": 0.0426, "step": 1970 }, { "epoch": 4.211087420042644, "grad_norm": 0.38910865783691406, "learning_rate": 7.889125799573561e-06, "loss": 0.0169, "step": 1975 }, { "epoch": 4.221748400852879, "grad_norm": 1.2753266096115112, "learning_rate": 7.782515991471216e-06, "loss": 0.0596, "step": 1980 }, { "epoch": 4.232409381663113, "grad_norm": 0.4680202603340149, "learning_rate": 7.67590618336887e-06, "loss": 0.0178, "step": 1985 }, { "epoch": 4.243070362473348, "grad_norm": 3.6173112392425537, "learning_rate": 7.5692963752665255e-06, "loss": 0.0415, "step": 1990 }, { "epoch": 4.253731343283582, "grad_norm": 0.45237183570861816, "learning_rate": 7.4626865671641785e-06, "loss": 0.0402, "step": 1995 }, { "epoch": 4.264392324093817, "grad_norm": 2.6849756240844727, "learning_rate": 7.356076759061833e-06, "loss": 0.0223, "step": 2000 }, { "epoch": 4.275053304904051, "grad_norm": 1.1324074268341064, "learning_rate": 7.249466950959488e-06, "loss": 0.0621, "step": 2005 }, { "epoch": 4.285714285714286, "grad_norm": 0.6863396763801575, "learning_rate": 7.142857142857143e-06, "loss": 0.0684, "step": 2010 }, { "epoch": 4.29637526652452, "grad_norm": 4.6197285652160645, "learning_rate": 7.0362473347547975e-06, "loss": 0.0416, "step": 2015 }, { "epoch": 4.3070362473347545, "grad_norm": 7.268411159515381, "learning_rate": 6.929637526652452e-06, "loss": 0.0934, "step": 2020 }, { "epoch": 4.317697228144989, "grad_norm": 6.680026054382324, "learning_rate": 6.823027718550107e-06, "loss": 0.0636, "step": 2025 }, { "epoch": 4.3283582089552235, "grad_norm": 6.048337936401367, "learning_rate": 6.716417910447762e-06, "loss": 0.0733, "step": 2030 }, { "epoch": 4.339019189765459, "grad_norm": 0.7440837621688843, "learning_rate": 6.609808102345416e-06, "loss": 0.0223, "step": 2035 }, { "epoch": 4.349680170575693, "grad_norm": 3.257054328918457, "learning_rate": 6.503198294243071e-06, "loss": 0.0377, "step": 2040 }, { "epoch": 4.360341151385928, "grad_norm": 1.4893893003463745, "learning_rate": 6.396588486140726e-06, "loss": 0.0328, "step": 2045 }, { "epoch": 4.371002132196162, "grad_norm": 0.021775823086500168, "learning_rate": 6.28997867803838e-06, "loss": 0.0614, "step": 2050 }, { "epoch": 4.381663113006397, "grad_norm": 3.6540675163269043, "learning_rate": 6.1833688699360345e-06, "loss": 0.0117, "step": 2055 }, { "epoch": 4.392324093816631, "grad_norm": 2.4917829036712646, "learning_rate": 6.076759061833688e-06, "loss": 0.0313, "step": 2060 }, { "epoch": 4.402985074626866, "grad_norm": 0.16013827919960022, "learning_rate": 5.970149253731343e-06, "loss": 0.0202, "step": 2065 }, { "epoch": 4.4136460554371, "grad_norm": 21.597875595092773, "learning_rate": 5.863539445628998e-06, "loss": 0.0809, "step": 2070 }, { "epoch": 4.424307036247335, "grad_norm": 9.23674201965332, "learning_rate": 5.756929637526653e-06, "loss": 0.0521, "step": 2075 }, { "epoch": 4.434968017057569, "grad_norm": 0.6427888870239258, "learning_rate": 5.650319829424307e-06, "loss": 0.0681, "step": 2080 }, { "epoch": 4.445628997867804, "grad_norm": 0.2844518721103668, "learning_rate": 5.543710021321962e-06, "loss": 0.0648, "step": 2085 }, { "epoch": 4.456289978678038, "grad_norm": 0.03346753865480423, "learning_rate": 5.437100213219617e-06, "loss": 0.0198, "step": 2090 }, { "epoch": 4.466950959488273, "grad_norm": 0.8559691905975342, "learning_rate": 5.330490405117271e-06, "loss": 0.0465, "step": 2095 }, { "epoch": 4.477611940298507, "grad_norm": 1.7826225757598877, "learning_rate": 5.2238805970149255e-06, "loss": 0.0227, "step": 2100 }, { "epoch": 4.4882729211087415, "grad_norm": 3.5491533279418945, "learning_rate": 5.11727078891258e-06, "loss": 0.0179, "step": 2105 }, { "epoch": 4.498933901918977, "grad_norm": 4.218719482421875, "learning_rate": 5.010660980810235e-06, "loss": 0.0666, "step": 2110 }, { "epoch": 4.509594882729211, "grad_norm": 3.1942098140716553, "learning_rate": 4.90405117270789e-06, "loss": 0.0633, "step": 2115 }, { "epoch": 4.520255863539446, "grad_norm": 0.4261585772037506, "learning_rate": 4.797441364605544e-06, "loss": 0.0327, "step": 2120 }, { "epoch": 4.53091684434968, "grad_norm": 1.4427200555801392, "learning_rate": 4.690831556503199e-06, "loss": 0.0159, "step": 2125 }, { "epoch": 4.541577825159915, "grad_norm": 0.7061243653297424, "learning_rate": 4.584221748400853e-06, "loss": 0.0179, "step": 2130 }, { "epoch": 4.552238805970149, "grad_norm": 4.287113666534424, "learning_rate": 4.477611940298508e-06, "loss": 0.0722, "step": 2135 }, { "epoch": 4.562899786780384, "grad_norm": 3.5793492794036865, "learning_rate": 4.371002132196162e-06, "loss": 0.0549, "step": 2140 }, { "epoch": 4.573560767590618, "grad_norm": 2.7709929943084717, "learning_rate": 4.264392324093816e-06, "loss": 0.0282, "step": 2145 }, { "epoch": 4.584221748400853, "grad_norm": 3.4923176765441895, "learning_rate": 4.157782515991471e-06, "loss": 0.0288, "step": 2150 }, { "epoch": 4.594882729211087, "grad_norm": 0.33699101209640503, "learning_rate": 4.051172707889126e-06, "loss": 0.0076, "step": 2155 }, { "epoch": 4.605543710021322, "grad_norm": 0.03753482550382614, "learning_rate": 3.944562899786781e-06, "loss": 0.0354, "step": 2160 }, { "epoch": 4.616204690831556, "grad_norm": 0.46259766817092896, "learning_rate": 3.837953091684435e-06, "loss": 0.0416, "step": 2165 }, { "epoch": 4.6268656716417915, "grad_norm": 2.0245108604431152, "learning_rate": 3.7313432835820893e-06, "loss": 0.0501, "step": 2170 }, { "epoch": 4.637526652452026, "grad_norm": 0.6268620491027832, "learning_rate": 3.624733475479744e-06, "loss": 0.0257, "step": 2175 }, { "epoch": 4.6481876332622605, "grad_norm": 0.07455793023109436, "learning_rate": 3.5181236673773987e-06, "loss": 0.1272, "step": 2180 }, { "epoch": 4.658848614072495, "grad_norm": 5.205850124359131, "learning_rate": 3.4115138592750535e-06, "loss": 0.0806, "step": 2185 }, { "epoch": 4.669509594882729, "grad_norm": 0.3119026720523834, "learning_rate": 3.304904051172708e-06, "loss": 0.0223, "step": 2190 }, { "epoch": 4.680170575692964, "grad_norm": 2.940166711807251, "learning_rate": 3.198294243070363e-06, "loss": 0.0836, "step": 2195 }, { "epoch": 4.690831556503198, "grad_norm": 0.5660606622695923, "learning_rate": 3.0916844349680173e-06, "loss": 0.0899, "step": 2200 }, { "epoch": 4.701492537313433, "grad_norm": 3.191072702407837, "learning_rate": 2.9850746268656716e-06, "loss": 0.0749, "step": 2205 }, { "epoch": 4.712153518123667, "grad_norm": 0.02729104645550251, "learning_rate": 2.8784648187633263e-06, "loss": 0.0166, "step": 2210 }, { "epoch": 4.722814498933902, "grad_norm": 0.022408660501241684, "learning_rate": 2.771855010660981e-06, "loss": 0.0097, "step": 2215 }, { "epoch": 4.733475479744136, "grad_norm": 0.1295965313911438, "learning_rate": 2.6652452025586354e-06, "loss": 0.0269, "step": 2220 }, { "epoch": 4.744136460554371, "grad_norm": 1.248518466949463, "learning_rate": 2.55863539445629e-06, "loss": 0.0438, "step": 2225 }, { "epoch": 4.754797441364605, "grad_norm": 1.6471575498580933, "learning_rate": 2.452025586353945e-06, "loss": 0.0197, "step": 2230 }, { "epoch": 4.76545842217484, "grad_norm": 6.227173328399658, "learning_rate": 2.3454157782515996e-06, "loss": 0.074, "step": 2235 }, { "epoch": 4.776119402985074, "grad_norm": 0.06783439964056015, "learning_rate": 2.238805970149254e-06, "loss": 0.0247, "step": 2240 }, { "epoch": 4.786780383795309, "grad_norm": 4.055930137634277, "learning_rate": 2.132196162046908e-06, "loss": 0.0761, "step": 2245 }, { "epoch": 4.797441364605544, "grad_norm": 0.8484672904014587, "learning_rate": 2.025586353944563e-06, "loss": 0.0438, "step": 2250 }, { "epoch": 4.8081023454157785, "grad_norm": 6.8790106773376465, "learning_rate": 1.9189765458422177e-06, "loss": 0.0481, "step": 2255 }, { "epoch": 4.818763326226013, "grad_norm": 0.7811554074287415, "learning_rate": 1.812366737739872e-06, "loss": 0.0718, "step": 2260 }, { "epoch": 4.8294243070362475, "grad_norm": 0.04351401701569557, "learning_rate": 1.7057569296375267e-06, "loss": 0.0131, "step": 2265 }, { "epoch": 4.840085287846482, "grad_norm": 5.299257278442383, "learning_rate": 1.5991471215351815e-06, "loss": 0.0319, "step": 2270 }, { "epoch": 4.850746268656716, "grad_norm": 3.5330796241760254, "learning_rate": 1.4925373134328358e-06, "loss": 0.0702, "step": 2275 }, { "epoch": 4.861407249466951, "grad_norm": 4.28450345993042, "learning_rate": 1.3859275053304905e-06, "loss": 0.0484, "step": 2280 }, { "epoch": 4.872068230277185, "grad_norm": 0.263338178396225, "learning_rate": 1.279317697228145e-06, "loss": 0.0118, "step": 2285 }, { "epoch": 4.88272921108742, "grad_norm": 0.022164933383464813, "learning_rate": 1.1727078891257998e-06, "loss": 0.0086, "step": 2290 }, { "epoch": 4.893390191897654, "grad_norm": 4.788683891296387, "learning_rate": 1.066098081023454e-06, "loss": 0.0372, "step": 2295 }, { "epoch": 4.904051172707889, "grad_norm": 1.400497555732727, "learning_rate": 9.594882729211088e-07, "loss": 0.0436, "step": 2300 }, { "epoch": 4.914712153518123, "grad_norm": 0.06390980631113052, "learning_rate": 8.528784648187634e-07, "loss": 0.0653, "step": 2305 }, { "epoch": 4.925373134328359, "grad_norm": 2.249465227127075, "learning_rate": 7.462686567164179e-07, "loss": 0.0602, "step": 2310 }, { "epoch": 4.936034115138593, "grad_norm": 3.8415865898132324, "learning_rate": 6.396588486140725e-07, "loss": 0.0123, "step": 2315 }, { "epoch": 4.946695095948828, "grad_norm": 3.278261423110962, "learning_rate": 5.33049040511727e-07, "loss": 0.0477, "step": 2320 }, { "epoch": 4.957356076759062, "grad_norm": 3.061466693878174, "learning_rate": 4.264392324093817e-07, "loss": 0.0856, "step": 2325 }, { "epoch": 4.968017057569297, "grad_norm": 2.182039260864258, "learning_rate": 3.1982942430703626e-07, "loss": 0.0476, "step": 2330 }, { "epoch": 4.978678038379531, "grad_norm": 0.6122257709503174, "learning_rate": 2.1321961620469084e-07, "loss": 0.0151, "step": 2335 }, { "epoch": 4.9893390191897655, "grad_norm": 0.5850390791893005, "learning_rate": 1.0660980810234542e-07, "loss": 0.009, "step": 2340 }, { "epoch": 5.0, "grad_norm": 3.1123313903808594, "learning_rate": 0.0, "loss": 0.0266, "step": 2345 }, { "epoch": 5.0, "eval_accuracy": 0.9930666666666667, "eval_loss": 0.022642984986305237, "eval_runtime": 29.5306, "eval_samples_per_second": 126.987, "eval_steps_per_second": 3.996, "step": 2345 }, { "epoch": 5.0, "step": 2345, "total_flos": 5.8118992210944e+18, "train_loss": 0.08223572542100573, "train_runtime": 1446.703, "train_samples_per_second": 51.842, "train_steps_per_second": 1.621 } ], "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 }