{ "capture": "/cap/s3/c2/cap", "n_sequences": 1472, "steps": 1472, "block": 6, "lr_peak": 5e-05, "warmup_steps": 73, "pos_per_seq": 16, "a_conf": 0.0, "trainable_params": 517318215, "n_tensors": 72, "loss_first": 0.9736005757004023, "loss_last": 0.04045040856999549, "loss_tail_mean": 0.4446075509756595, "ce_last": 0.10928969697118873, "tv_last": 0.03280160060967319, "conf_last": 0.1864870432536918, "diag_draft_eq_target_argmax_pct": 64.7758, "diag_draft_eq_ground_truth_pct": 64.0476, "diag_target_top1_p": 0.8416373158027655, "diag_draft_top1_p": 0.756630592423065, "history": [ { "step": 0, "T": 541, "positions": 16, "loss": 0.9736005757004023, "ce": 4.103468403220177, "tv": 0.6258375099860132, "conf": 0.8456984991207719, "grad_norm": 8.604708671569824, "lr": 6.849315068493151e-07 }, { "step": 1, "T": 645, "positions": 16, "loss": 1.1650708355009556, "ce": 5.098860539495945, "tv": 0.7279830891638994, "conf": 0.5731722395867109, "grad_norm": 8.94845199584961, "lr": 1.3698630136986302e-06 }, { "step": 2, "T": 527, "positions": 16, "loss": 0.9520305097103119, "ce": 4.645798405632377, "tv": 0.5416118334978819, "conf": 0.681897304020822, "grad_norm": 9.359341621398926, "lr": 2.0547945205479454e-06 }, { "step": 3, "T": 848, "positions": 16, "loss": 0.89791115000844, "ce": 4.158722909167409, "tv": 0.5355987418442965, "conf": 0.5099201295524836, "grad_norm": 8.483851432800293, "lr": 2.7397260273972604e-06 }, { "step": 4, "T": 675, "positions": 16, "loss": 0.8645978941058274, "ce": 4.881395278847776, "tv": 0.41828708062530495, "conf": 0.8835052555659786, "grad_norm": 2.9840455055236816, "lr": 3.4246575342465754e-06 }, { "step": 5, "T": 536, "positions": 16, "loss": 0.5382030799519271, "ce": 3.1305344496504404, "tv": 0.2501662809518166, "conf": 0.36709686007816344, "grad_norm": 6.440013885498047, "lr": 4.109589041095891e-06 }, { "step": 6, "T": 619, "positions": 16, "loss": 0.35148079212012817, "ce": 2.1148639842949706, "tv": 0.15554929609788815, "conf": 0.45762898738030344, "grad_norm": 3.5477817058563232, "lr": 4.7945205479452054e-06 }, { "step": 7, "T": 632, "positions": 16, "loss": 1.1161452643573284, "ce": 4.358837820589542, "tv": 0.7558461055159569, "conf": 0.4706646641716361, "grad_norm": 5.947121620178223, "lr": 5.479452054794521e-06 }, { "step": 8, "T": 529, "positions": 16, "loss": 0.7332931626588106, "ce": 2.601118030026555, "tv": 0.5257570762187243, "conf": 0.6974833495914936, "grad_norm": 8.625568389892578, "lr": 6.164383561643836e-06 }, { "step": 9, "T": 530, "positions": 16, "loss": 0.6387204985367134, "ce": 2.2961771072587, "tv": 0.45455866283737123, "conf": 0.649827066808939, "grad_norm": 5.531002998352051, "lr": 6.849315068493151e-06 }, { "step": 10, "T": 660, "positions": 16, "loss": 1.1681065410375595, "ce": 4.91364523768425, "tv": 0.7519355714321136, "conf": 0.4650520272552967, "grad_norm": 5.741738319396973, "lr": 7.534246575342466e-06 }, { "step": 11, "T": 1056, "positions": 16, "loss": 1.0378636537643615, "ce": 5.150473581044935, "tv": 0.5809069797833217, "conf": 0.6447352857794613, "grad_norm": 5.425670623779297, "lr": 8.219178082191782e-06 }, { "step": 12, "T": 537, "positions": 16, "loss": 0.4966072801107657, "ce": 2.1168663696735166, "tv": 0.316578480276803, "conf": 0.7192296587163582, "grad_norm": 4.631267070770264, "lr": 8.904109589041097e-06 }, { "step": 13, "T": 539, "positions": 16, "loss": 0.9045861531049013, "ce": 3.646146487444639, "tv": 0.5999683141708374, "conf": 0.7634636852890253, "grad_norm": 7.865741729736328, "lr": 9.589041095890411e-06 }, { "step": 14, "T": 627, "positions": 16, "loss": 0.42294090341601986, "ce": 2.1238138299668208, "tv": 0.23395503535721218, "conf": 0.4767517487052828, "grad_norm": 4.917176723480225, "lr": 1.0273972602739726e-05 }, { "step": 15, "T": 526, "positions": 16, "loss": 1.05423004925251, "ce": 4.276482105255127, "tv": 0.6962020471692085, "conf": 0.6619575582444668, "grad_norm": 5.997800827026367, "lr": 1.0958904109589042e-05 }, { "step": 16, "T": 1697, "positions": 16, "loss": 1.2991811372339725, "ce": 6.2386767119169235, "tv": 0.7503483351320028, "conf": 0.46465608291327953, "grad_norm": 5.154723644256592, "lr": 1.1643835616438357e-05 }, { "step": 17, "T": 608, "positions": 16, "loss": 0.8404683331027627, "ce": 3.552533818408847, "tv": 0.539127723313868, "conf": 0.7686185147613287, "grad_norm": 9.464700698852539, "lr": 1.2328767123287673e-05 }, { "step": 18, "T": 541, "positions": 16, "loss": 0.41044411104667233, "ce": 1.6482284497533328, "tv": 0.27291252876420913, "conf": 0.5125109604850877, "grad_norm": 11.461861610412598, "lr": 1.3013698630136986e-05 }, { "step": 19, "T": 528, "positions": 16, "loss": 1.0526292761787772, "ce": 4.490327004343271, "tv": 0.6706628380343318, "conf": 0.46241509541869164, "grad_norm": 4.324149131774902, "lr": 1.3698630136986302e-05 }, { "step": 20, "T": 531, "positions": 16, "loss": 0.721891619352391, "ce": 3.0972620104348607, "tv": 0.457961580628762, "conf": 0.5193047804641537, "grad_norm": 11.135403633117676, "lr": 1.4383561643835619e-05 }, { "step": 21, "T": 531, "positions": 16, "loss": 0.7330177621915936, "ce": 2.5237028654664755, "tv": 0.5340527622029185, "conf": 0.730816868133843, "grad_norm": 6.102502822875977, "lr": 1.5068493150684933e-05 }, { "step": 22, "T": 536, "positions": 16, "loss": 1.0728369634598494, "ce": 5.312575673684478, "tv": 0.6017549000680447, "conf": 0.8842977192252874, "grad_norm": 5.835565567016602, "lr": 1.5753424657534245e-05 }, { "step": 23, "T": 571, "positions": 16, "loss": 1.1208831295371056, "ce": 4.831208880990744, "tv": 0.7086247503757477, "conf": 0.6022511348128319, "grad_norm": 9.505934715270996, "lr": 1.6438356164383563e-05 }, { "step": 24, "T": 524, "positions": 16, "loss": 0.44190010367310606, "ce": 2.2941019360077917, "tv": 0.2360998950098292, "conf": 0.3515120398369618, "grad_norm": 4.392905235290527, "lr": 1.7123287671232875e-05 }, { "step": 25, "T": 675, "positions": 16, "loss": 1.1811357401311398, "ce": 5.269399601966143, "tv": 0.7268841825425625, "conf": 0.5382811008021235, "grad_norm": 9.132583618164062, "lr": 1.7808219178082194e-05 }, { "step": 26, "T": 615, "positions": 16, "loss": 0.5458293867413886, "ce": 2.292967173270881, "tv": 0.35170297615695745, "conf": 0.5635244222357869, "grad_norm": 7.673511028289795, "lr": 1.8493150684931506e-05 }, { "step": 27, "T": 677, "positions": 16, "loss": 0.8064987286925316, "ce": 3.2254896238446236, "tv": 0.5377219906076789, "conf": 0.5978517644107342, "grad_norm": 4.773327827453613, "lr": 1.9178082191780822e-05 }, { "step": 28, "T": 537, "positions": 16, "loss": 1.1795028038322926, "ce": 6.237723298370838, "tv": 0.6174782924354076, "conf": 0.7148046996444464, "grad_norm": 11.255009651184082, "lr": 1.9863013698630137e-05 }, { "step": 29, "T": 652, "positions": 16, "loss": 1.1090794876217842, "ce": 4.558595605194569, "tv": 0.7257999181747437, "conf": 0.48547845892608166, "grad_norm": 4.095385551452637, "lr": 2.0547945205479453e-05 }, { "step": 30, "T": 553, "positions": 16, "loss": 0.4575979275396094, "ce": 1.537441626744112, "tv": 0.3376153095450718, "conf": 0.47790140751749277, "grad_norm": 5.747526168823242, "lr": 2.1232876712328768e-05 }, { "step": 31, "T": 528, "positions": 16, "loss": 0.5157193157356232, "ce": 1.6821052324958146, "tv": 0.38612089795060456, "conf": 0.5346699170768261, "grad_norm": 8.445945739746094, "lr": 2.1917808219178083e-05 }, { "step": 32, "T": 591, "positions": 16, "loss": 0.5149173795798561, "ce": 2.347422612569062, "tv": 0.31130568886874244, "conf": 0.9052272131666541, "grad_norm": 6.018770694732666, "lr": 2.2602739726027396e-05 }, { "step": 33, "T": 528, "positions": 16, "loss": 0.36452705565898214, "ce": 1.1609663358976832, "tv": 0.27603380313667003, "conf": 0.4734267109306529, "grad_norm": 10.466574668884277, "lr": 2.3287671232876714e-05 }, { "step": 34, "T": 539, "positions": 16, "loss": 0.5044501752126962, "ce": 2.4292961187893525, "tv": 0.29057839908637106, "conf": 0.7155023869127035, "grad_norm": 12.22764778137207, "lr": 2.3972602739726026e-05 }, { "step": 35, "T": 608, "positions": 16, "loss": 0.611599181778729, "ce": 2.929032551124692, "tv": 0.35410656686872244, "conf": 0.720551423728466, "grad_norm": 4.813377857208252, "lr": 2.4657534246575345e-05 }, { "step": 36, "T": 538, "positions": 16, "loss": 0.6802999004721642, "ce": 2.3916324069723487, "tv": 0.4901518439874053, "conf": 0.5164088467136025, "grad_norm": 6.301151752471924, "lr": 2.5342465753424657e-05 }, { "step": 37, "T": 542, "positions": 16, "loss": 0.7616388835012913, "ce": 2.680804308503866, "tv": 0.548398282378912, "conf": 0.6598878521472216, "grad_norm": 8.335856437683105, "lr": 2.6027397260273973e-05 }, { "step": 38, "T": 672, "positions": 16, "loss": 1.2638854831457138, "ce": 5.777718611061573, "tv": 0.7623484786599874, "conf": 0.5095349512994289, "grad_norm": 2.5607364177703857, "lr": 2.6712328767123288e-05 }, { "step": 39, "T": 672, "positions": 16, "loss": 0.5954403111245483, "ce": 2.8853687621158315, "tv": 0.3410038223664742, "conf": 0.32999890425708145, "grad_norm": 5.925373554229736, "lr": 2.7397260273972603e-05 }, { "step": 40, "T": 823, "positions": 16, "loss": 1.2870489433407784, "ce": 6.934098389931023, "tv": 0.6595990066416562, "conf": 0.6900983592495322, "grad_norm": 6.9769768714904785, "lr": 2.808219178082192e-05 }, { "step": 41, "T": 542, "positions": 16, "loss": 1.0206173956394196, "ce": 5.3323445320129395, "tv": 0.5415366012603045, "conf": 0.9186799172312021, "grad_norm": 6.250906944274902, "lr": 2.8767123287671238e-05 }, { "step": 42, "T": 528, "positions": 16, "loss": 1.5909864231944084, "ce": 9.253238156437874, "tv": 0.7396251112222672, "conf": 0.7891044542193413, "grad_norm": 8.384148597717285, "lr": 2.9452054794520546e-05 }, { "step": 43, "T": 892, "positions": 16, "loss": 0.7954454341634118, "ce": 4.333897519225502, "tv": 0.40228409893097705, "conf": 0.5584721607156098, "grad_norm": 3.718093156814575, "lr": 3.0136986301369865e-05 }, { "step": 44, "T": 641, "positions": 16, "loss": 1.0862663444131613, "ce": 4.435256592929363, "tv": 0.7141563277691603, "conf": 0.5997447744011879, "grad_norm": 3.2327535152435303, "lr": 3.0821917808219184e-05 }, { "step": 45, "T": 527, "positions": 16, "loss": 0.7208257708698511, "ce": 3.2530478723347187, "tv": 0.439467771910131, "conf": 0.7478622701019049, "grad_norm": 4.916807651519775, "lr": 3.150684931506849e-05 }, { "step": 46, "T": 527, "positions": 16, "loss": 0.6698723346926272, "ce": 3.4630847894586623, "tv": 0.3595154113136232, "conf": 0.6565905921161175, "grad_norm": 6.818506240844727, "lr": 3.219178082191781e-05 }, { "step": 47, "T": 989, "positions": 16, "loss": 1.1373991803338868, "ce": 6.32270706347299, "tv": 0.5612538662098814, "conf": 0.6534444847638952, "grad_norm": 5.661182403564453, "lr": 3.287671232876713e-05 }, { "step": 48, "T": 611, "positions": 16, "loss": 1.042689255438745, "ce": 5.0050909970887005, "tv": 0.6024223933927715, "conf": 0.9223770573735237, "grad_norm": 8.12209701538086, "lr": 3.356164383561644e-05 }, { "step": 49, "T": 558, "positions": 16, "loss": 0.9114466314204037, "ce": 4.393782772589475, "tv": 0.5245203897356987, "conf": 0.9174098782241344, "grad_norm": 6.599114418029785, "lr": 3.424657534246575e-05 }, { "step": 50, "T": 542, "positions": 16, "loss": 0.5098556165758055, "ce": 3.1934209344981355, "tv": 0.21168165267590666, "conf": 0.3450841908052098, "grad_norm": 3.5403695106506348, "lr": 3.493150684931507e-05 }, { "step": 51, "T": 527, "positions": 16, "loss": 0.7612294307909906, "ce": 3.6909247850999236, "tv": 0.43570773769170046, "conf": 0.8773499568924308, "grad_norm": 3.6913700103759766, "lr": 3.561643835616439e-05 }, { "step": 52, "T": 540, "positions": 16, "loss": 0.3106450108316494, "ce": 1.4984196341247298, "tv": 0.17867006176675204, "conf": 0.5217722924426198, "grad_norm": 2.20211124420166, "lr": 3.63013698630137e-05 }, { "step": 53, "T": 673, "positions": 16, "loss": 1.0540857259184122, "ce": 4.2965593710541725, "tv": 0.6938109006732702, "conf": 0.5910787209868431, "grad_norm": 4.751731872558594, "lr": 3.698630136986301e-05 }, { "step": 54, "T": 613, "positions": 16, "loss": 0.804573565692408, "ce": 3.2755268573528156, "tv": 0.5300231956061907, "conf": 0.552842409349978, "grad_norm": 6.639126777648926, "lr": 3.767123287671233e-05 }, { "step": 55, "T": 608, "positions": 16, "loss": 0.8658500586170703, "ce": 3.7947696569608524, "tv": 0.5404145730426535, "conf": 0.7506331726908684, "grad_norm": 4.5846781730651855, "lr": 3.8356164383561644e-05 }, { "step": 56, "T": 535, "positions": 16, "loss": 0.5113611225970089, "ce": 1.6677639605477452, "tv": 0.38287193118594587, "conf": 0.5979495793581009, "grad_norm": 5.728870391845703, "lr": 3.904109589041096e-05 }, { "step": 57, "T": 531, "positions": 16, "loss": 0.887166696600616, "ce": 4.5100856041535735, "tv": 0.48462012968957424, "conf": 1.011107537895441, "grad_norm": 7.6554341316223145, "lr": 3.9726027397260274e-05 }, { "step": 58, "T": 668, "positions": 16, "loss": 0.5150611526332796, "ce": 2.756675901953713, "tv": 0.2659928584616864, "conf": 0.7082216902635992, "grad_norm": 2.8747403621673584, "lr": 4.0410958904109586e-05 }, { "step": 59, "T": 677, "positions": 16, "loss": 1.0606720075011253, "ce": 4.792873360216618, "tv": 0.6459829695522785, "conf": 0.7012704089283943, "grad_norm": 6.4728217124938965, "lr": 4.1095890410958905e-05 }, { "step": 60, "T": 531, "positions": 16, "loss": 0.8193874857388437, "ce": 3.102549040224403, "tv": 0.56570287887007, "conf": 0.5570799084380269, "grad_norm": 6.500418663024902, "lr": 4.1780821917808224e-05 }, { "step": 61, "T": 672, "positions": 16, "loss": 0.975808551069349, "ce": 4.003396535525098, "tv": 0.6394098949385807, "conf": 0.5587444500997663, "grad_norm": 8.127198219299316, "lr": 4.2465753424657536e-05 }, { "step": 62, "T": 582, "positions": 16, "loss": 1.0475762952119112, "ce": 4.89735871180892, "tv": 0.6198227126151323, "conf": 0.9431830029934645, "grad_norm": 16.364200592041016, "lr": 4.315068493150685e-05 }, { "step": 63, "T": 583, "positions": 16, "loss": 0.2416343896275066, "ce": 0.9151962835694576, "tv": 0.16679417441991973, "conf": 0.3102233672980219, "grad_norm": 8.334650039672852, "lr": 4.383561643835617e-05 }, { "step": 64, "T": 1047, "positions": 16, "loss": 1.0859813226707047, "ce": 5.5045019283061265, "tv": 0.5950345802120864, "conf": 0.8583614113740623, "grad_norm": 5.589264392852783, "lr": 4.4520547945205486e-05 }, { "step": 65, "T": 673, "positions": 16, "loss": 1.1101486440747976, "ce": 4.748782008886337, "tv": 0.7058560512959957, "conf": 0.5144200082868338, "grad_norm": 5.149134635925293, "lr": 4.520547945205479e-05 }, { "step": 66, "T": 600, "positions": 16, "loss": 0.7255914655979723, "ce": 3.0293923378922045, "tv": 0.4696135986596346, "conf": 0.6948263142257929, "grad_norm": 4.901910781860352, "lr": 4.589041095890411e-05 }, { "step": 67, "T": 672, "positions": 16, "loss": 0.9602568261325359, "ce": 4.1847834549844265, "tv": 0.6019760984927416, "conf": 0.49623459577560425, "grad_norm": 5.8659772872924805, "lr": 4.657534246575343e-05 }, { "step": 68, "T": 622, "positions": 16, "loss": 0.18133089888942777, "ce": 0.675665860522713, "tv": 0.12640478441971936, "conf": 0.38468922208994627, "grad_norm": 4.972193717956543, "lr": 4.726027397260275e-05 }, { "step": 69, "T": 531, "positions": 16, "loss": 0.98713094368577, "ce": 4.0177303701639175, "tv": 0.6503977049142122, "conf": 0.5234889667481184, "grad_norm": 5.142071723937988, "lr": 4.794520547945205e-05 }, { "step": 70, "T": 531, "positions": 16, "loss": 0.6818272743184934, "ce": 3.0811797676378774, "tv": 0.41523252677870914, "conf": 0.4341576335718855, "grad_norm": 4.662594318389893, "lr": 4.863013698630137e-05 }, { "step": 71, "T": 527, "positions": 16, "loss": 1.1003346145153046, "ce": 4.540608383715153, "tv": 0.7180819772183895, "conf": 0.5400676662102342, "grad_norm": 5.159008026123047, "lr": 4.931506849315069e-05 }, { "step": 72, "T": 542, "positions": 16, "loss": 0.7596705802716315, "ce": 2.5284353881143034, "tv": 0.5631411839276552, "conf": 0.6837142072618008, "grad_norm": 6.983900547027588, "lr": 5e-05 }, { "step": 73, "T": 527, "positions": 16, "loss": 0.9679593592882156, "ce": 3.6915847063064575, "tv": 0.6653343327343464, "conf": 0.5749659016728401, "grad_norm": 5.573715686798096, "lr": 5e-05 }, { "step": 74, "T": 629, "positions": 16, "loss": 1.1491634715348482, "ce": 5.667654439806938, "tv": 0.6471088863909245, "conf": 0.8114075604826212, "grad_norm": 3.4387104511260986, "lr": 4.999993696610457e-05 }, { "step": 75, "T": 646, "positions": 16, "loss": 0.6946093402802944, "ce": 2.660929575562477, "tv": 0.47612930927425623, "conf": 0.6131476238369942, "grad_norm": 4.575263977050781, "lr": 4.999974786473614e-05 }, { "step": 76, "T": 527, "positions": 16, "loss": 0.9616811908781528, "ce": 3.7024849727749825, "tv": 0.6571474578231573, "conf": 0.731815043836832, "grad_norm": 5.8502655029296875, "lr": 4.9999432696848295e-05 }, { "step": 77, "T": 581, "positions": 16, "loss": 0.9906067084521055, "ce": 4.296550206840038, "tv": 0.6232796963304281, "conf": 0.8213168438524008, "grad_norm": 6.929669380187988, "lr": 4.999899146403034e-05 }, { "step": 78, "T": 528, "positions": 16, "loss": 0.5550302340998314, "ce": 2.291715249826666, "tv": 0.36206524493172765, "conf": 0.5563018526881933, "grad_norm": 5.292882442474365, "lr": 4.9998424168507275e-05 }, { "step": 79, "T": 534, "positions": 16, "loss": 1.3229399472475052, "ce": 6.6808876395225525, "tv": 0.727612417191267, "conf": 0.9418812915682793, "grad_norm": 5.9611101150512695, "lr": 4.999773081313981e-05 }, { "step": 80, "T": 532, "positions": 16, "loss": 1.0079914517700672, "ce": 3.851860199123621, "tv": 0.6920060347765684, "conf": 0.5771194845438004, "grad_norm": 4.468182563781738, "lr": 4.999691140142434e-05 }, { "step": 81, "T": 533, "positions": 16, "loss": 0.7353272285690764, "ce": 3.5360884054898634, "tv": 0.4241315590697923, "conf": 0.45705095888115466, "grad_norm": 7.321213722229004, "lr": 4.9995965937492927e-05 }, { "step": 82, "T": 527, "positions": 16, "loss": 0.9444464119151235, "ce": 3.887215979397297, "tv": 0.6174720199778676, "conf": 0.7083096988499165, "grad_norm": 4.389327526092529, "lr": 4.999489442611327e-05 }, { "step": 83, "T": 580, "positions": 16, "loss": 0.8394649736583233, "ce": 3.7533947490155697, "tv": 0.5156950084492564, "conf": 0.8827461786568165, "grad_norm": 9.841110229492188, "lr": 4.999369687268868e-05 }, { "step": 84, "T": 1391, "positions": 16, "loss": 0.8649645298719406, "ce": 3.805579751729965, "tv": 0.5382295232266188, "conf": 0.562032300978899, "grad_norm": 3.3267524242401123, "lr": 4.999237328325809e-05 }, { "step": 85, "T": 527, "positions": 16, "loss": 1.055350847542286, "ce": 4.493004322052002, "tv": 0.6733893547207117, "conf": 0.7408777885138988, "grad_norm": 2.569875478744507, "lr": 4.9990923664495964e-05 }, { "step": 86, "T": 538, "positions": 16, "loss": 0.33130711983540095, "ce": 1.202784063440049, "tv": 0.23447634669719264, "conf": 0.3666240442544222, "grad_norm": 9.553722381591797, "lr": 4.9989348023712325e-05 }, { "step": 87, "T": 648, "positions": 16, "loss": 0.6128523870720528, "ce": 2.659736013854854, "tv": 0.3854208893026225, "conf": 0.5982153499498963, "grad_norm": 10.126078605651855, "lr": 4.998764636885267e-05 }, { "step": 88, "T": 672, "positions": 16, "loss": 1.0000308901071548, "ce": 4.512725509703159, "tv": 0.6097314767539501, "conf": 0.6715234741568565, "grad_norm": 8.460880279541016, "lr": 4.998581870849795e-05 }, { "step": 89, "T": 599, "positions": 16, "loss": 0.8547478774562478, "ce": 3.1347516495734453, "tv": 0.6014141393825412, "conf": 0.6468070987612009, "grad_norm": 5.213296890258789, "lr": 4.998386505186453e-05 }, { "step": 90, "T": 655, "positions": 16, "loss": 0.8453810093924403, "ce": 3.6837168522179127, "tv": 0.5300103696063161, "conf": 0.647123109549284, "grad_norm": 4.437480926513672, "lr": 4.998178540880415e-05 }, { "step": 91, "T": 646, "positions": 16, "loss": 0.7474642340093851, "ce": 2.8123402893543243, "tv": 0.5180335603654385, "conf": 0.582741517573595, "grad_norm": 6.288307189941406, "lr": 4.997957978980383e-05 }, { "step": 92, "T": 606, "positions": 16, "loss": 0.5977941194432788, "ce": 2.7466535659623332, "tv": 0.35903195368882734, "conf": 0.49369800789281726, "grad_norm": 7.905003547668457, "lr": 4.9977248205985895e-05 }, { "step": 93, "T": 526, "positions": 16, "loss": 0.4244630387984216, "ce": 1.3928750164341182, "tv": 0.31686171260662377, "conf": 0.6314868796616793, "grad_norm": 9.038926124572754, "lr": 4.997479066910782e-05 }, { "step": 94, "T": 541, "positions": 16, "loss": 0.5828181151300669, "ce": 2.191572837997228, "tv": 0.4040675750002265, "conf": 0.48741695703938603, "grad_norm": 4.19784688949585, "lr": 4.997220719156228e-05 }, { "step": 95, "T": 546, "positions": 16, "loss": 0.8136079423129559, "ce": 2.9558106623589993, "tv": 0.5755854351446033, "conf": 0.7507130932062864, "grad_norm": 4.102855205535889, "lr": 4.9969497786376994e-05 }, { "step": 96, "T": 532, "positions": 16, "loss": 0.40655394255009014, "ce": 1.4620972606244322, "tv": 0.2892713578839903, "conf": 0.32965609640814364, "grad_norm": 3.367610454559326, "lr": 4.996666246721472e-05 }, { "step": 97, "T": 527, "positions": 16, "loss": 1.3961007446050644, "ce": 6.92672361433506, "tv": 0.7815871350467205, "conf": 0.8843356044963002, "grad_norm": 4.941042423248291, "lr": 4.996370124837314e-05 }, { "step": 98, "T": 678, "positions": 16, "loss": 0.5305983126163483, "ce": 1.9386745300143957, "tv": 0.3741454090923071, "conf": 0.6786253117024899, "grad_norm": 10.755980491638184, "lr": 4.996061414478485e-05 }, { "step": 99, "T": 542, "positions": 16, "loss": 0.7165676094591618, "ce": 2.4740011347457767, "tv": 0.5212972308509052, "conf": 0.6190966693684459, "grad_norm": 12.47455883026123, "lr": 4.995740117201719e-05 }, { "step": 100, "T": 536, "positions": 16, "loss": 0.1779547521728091, "ce": 0.7510770877706818, "tv": 0.11427449386974331, "conf": 0.23715003003599122, "grad_norm": 2.639315605163574, "lr": 4.995406234627229e-05 }, { "step": 101, "T": 540, "positions": 16, "loss": 0.9561688005924225, "ce": 4.0164792984724045, "tv": 0.6161343157291412, "conf": 0.7206463082693517, "grad_norm": 3.1795308589935303, "lr": 4.995059768438686e-05 }, { "step": 102, "T": 628, "positions": 16, "loss": 0.44717550711357035, "ce": 2.4496102900011465, "tv": 0.22468275495339185, "conf": 0.8051649332046509, "grad_norm": 3.359431743621826, "lr": 4.994700720383221e-05 }, { "step": 103, "T": 599, "positions": 16, "loss": 0.6131446482613683, "ce": 2.2958394326269627, "tv": 0.426178568508476, "conf": 0.7921089846640825, "grad_norm": 4.553620338439941, "lr": 4.994329092271408e-05 }, { "step": 104, "T": 608, "positions": 16, "loss": 0.39930934621952474, "ce": 1.477703224751167, "tv": 0.2794878118438646, "conf": 0.6414226554334164, "grad_norm": 3.831770420074463, "lr": 4.993944885977262e-05 }, { "step": 105, "T": 618, "positions": 16, "loss": 0.7264483943581581, "ce": 2.9238723665475845, "tv": 0.4822901701554656, "conf": 0.7814396061003208, "grad_norm": 10.932870864868164, "lr": 4.993548103438224e-05 }, { "step": 106, "T": 673, "positions": 16, "loss": 0.8998259734362364, "ce": 3.521142115816474, "tv": 0.6085686376318336, "conf": 0.674904522486031, "grad_norm": 6.5808796882629395, "lr": 4.993138746655154e-05 }, { "step": 107, "T": 607, "positions": 16, "loss": 0.4852791824378073, "ce": 2.015262472559698, "tv": 0.31528102583251894, "conf": 0.6814459739252925, "grad_norm": 4.917659759521484, "lr": 4.992716817692319e-05 }, { "step": 108, "T": 580, "positions": 16, "loss": 0.5710910041816533, "ce": 2.459710008231923, "tv": 0.36124445172026753, "conf": 0.6014880817383528, "grad_norm": 3.8673222064971924, "lr": 4.992282318677387e-05 }, { "step": 109, "T": 565, "positions": 16, "loss": 0.40860475646331906, "ce": 1.7999325467972085, "tv": 0.2540127763641067, "conf": 0.6855113916099072, "grad_norm": 2.535212755203247, "lr": 4.9918352518014087e-05 }, { "step": 110, "T": 533, "positions": 16, "loss": 0.8162246540014166, "ce": 3.752543456386775, "tv": 0.48996704968158156, "conf": 0.8880772069096565, "grad_norm": 10.622612953186035, "lr": 4.991375619318817e-05 }, { "step": 111, "T": 610, "positions": 16, "loss": 0.6188186130666509, "ce": 2.4200376964217867, "tv": 0.41868314696148445, "conf": 0.7798149972222745, "grad_norm": 4.280676364898682, "lr": 4.990903423547403e-05 }, { "step": 112, "T": 1026, "positions": 16, "loss": 0.5520138075589784, "ce": 2.069802141839318, "tv": 0.3833706699369941, "conf": 0.41271557146683335, "grad_norm": 3.950188636779785, "lr": 4.990418666868315e-05 }, { "step": 113, "T": 613, "positions": 16, "loss": 0.6738875517621636, "ce": 2.738589160144329, "tv": 0.44447627663612366, "conf": 0.543517455458641, "grad_norm": 6.654091835021973, "lr": 4.9899213517260416e-05 }, { "step": 114, "T": 675, "positions": 16, "loss": 0.7382939420640469, "ce": 3.0632521333172917, "tv": 0.47996527422219515, "conf": 0.5841060299426317, "grad_norm": 2.8831610679626465, "lr": 4.989411480628397e-05 }, { "step": 115, "T": 1001, "positions": 16, "loss": 0.7193340724334121, "ce": 2.813184428960085, "tv": 0.48668402852490544, "conf": 0.54766546189785, "grad_norm": 8.226829528808594, "lr": 4.988889056146518e-05 }, { "step": 116, "T": 553, "positions": 16, "loss": 0.770853322930634, "ce": 3.4919467754662037, "tv": 0.4685096060857177, "conf": 0.7356504518538713, "grad_norm": 8.378266334533691, "lr": 4.988354080914839e-05 }, { "step": 117, "T": 676, "positions": 16, "loss": 0.3115275579948502, "ce": 1.306610601160628, "tv": 0.20096277942502638, "conf": 0.511673049069941, "grad_norm": 2.73419451713562, "lr": 4.9878065576310847e-05 }, { "step": 118, "T": 530, "positions": 16, "loss": 0.43897787714377046, "ce": 1.60636720608818, "tv": 0.3092679511901224, "conf": 0.5412371819838881, "grad_norm": 9.424474716186523, "lr": 4.9872464890562576e-05 }, { "step": 119, "T": 531, "positions": 16, "loss": 0.23951164912432432, "ce": 0.867732453654753, "tv": 0.16970932978438213, "conf": 0.26749471318908036, "grad_norm": 2.456723690032959, "lr": 4.986673878014624e-05 }, { "step": 120, "T": 560, "positions": 16, "loss": 0.7256422992795706, "ce": 2.82433607801795, "tv": 0.49245412088930607, "conf": 0.6407336723059416, "grad_norm": 9.157952308654785, "lr": 4.9860887273936927e-05 }, { "step": 121, "T": 1390, "positions": 16, "loss": 1.1684034732170403, "ce": 5.998920660465956, "tv": 0.631679336540401, "conf": 0.48561627976596355, "grad_norm": 4.623064994812012, "lr": 4.9854910401442134e-05 }, { "step": 122, "T": 536, "positions": 16, "loss": 0.5553716327995062, "ce": 2.009574375813827, "tv": 0.39379355893470347, "conf": 0.6202099770307541, "grad_norm": 5.5472588539123535, "lr": 4.9848808192801475e-05 }, { "step": 123, "T": 622, "positions": 16, "loss": 0.8324969038367271, "ce": 3.1557724326848984, "tv": 0.5743551980704069, "conf": 0.5404198374599218, "grad_norm": 2.518662214279175, "lr": 4.9842580678786645e-05 }, { "step": 124, "T": 557, "positions": 16, "loss": 0.6394535857252777, "ce": 2.2482780879363418, "tv": 0.460695315618068, "conf": 0.6267956960946321, "grad_norm": 6.160676956176758, "lr": 4.9836227890801196e-05 }, { "step": 125, "T": 563, "positions": 16, "loss": 0.6365206693299115, "ce": 2.3326965533196926, "tv": 0.4480566857382655, "conf": 0.5058357762172818, "grad_norm": 4.171810150146484, "lr": 4.982974986088041e-05 }, { "step": 126, "T": 653, "positions": 16, "loss": 0.5048361198278144, "ce": 1.6973021656740457, "tv": 0.37233991210814565, "conf": 0.48943023197352886, "grad_norm": 3.7019031047821045, "lr": 4.982314662169111e-05 }, { "step": 127, "T": 554, "positions": 16, "loss": 0.807654766831547, "ce": 3.522915313951671, "tv": 0.505959156434983, "conf": 0.5608986131846905, "grad_norm": 4.488377094268799, "lr": 4.9816418206531544e-05 }, { "step": 128, "T": 672, "positions": 16, "loss": 0.8825835100142285, "ce": 4.122654420614708, "tv": 0.5225756336003542, "conf": 0.5583643862046301, "grad_norm": 6.783528804779053, "lr": 4.980956464933116e-05 }, { "step": 129, "T": 609, "positions": 16, "loss": 0.4190823187818751, "ce": 1.7215479905717075, "tv": 0.2743638986139558, "conf": 0.4792203784454614, "grad_norm": 6.944522857666016, "lr": 4.980258598465049e-05 }, { "step": 130, "T": 571, "positions": 16, "loss": 0.5600557110738009, "ce": 1.9051922966027632, "tv": 0.41059610579395667, "conf": 0.6446389192715287, "grad_norm": 5.529710292816162, "lr": 4.979548224768089e-05 }, { "step": 131, "T": 561, "positions": 16, "loss": 0.4542123503051698, "ce": 1.5964835563208908, "tv": 0.32729332079179585, "conf": 0.5424399459734559, "grad_norm": 2.9320297241210938, "lr": 4.978825347424448e-05 }, { "step": 132, "T": 531, "positions": 16, "loss": 0.8601277191191912, "ce": 3.3114866837859154, "tv": 0.58775451220572, "conf": 0.6499015553854406, "grad_norm": 5.531436920166016, "lr": 4.9780899700793895e-05 }, { "step": 133, "T": 591, "positions": 16, "loss": 0.5743461516321986, "ce": 2.5825754420638987, "tv": 0.3512095661535568, "conf": 0.616059081570711, "grad_norm": 7.29761266708374, "lr": 4.9773420964412064e-05 }, { "step": 134, "T": 560, "positions": 16, "loss": 0.7176119876094162, "ce": 2.696677979780361, "tv": 0.49771577632054687, "conf": 0.5440781922079623, "grad_norm": 4.769730567932129, "lr": 4.9765817302812114e-05 }, { "step": 135, "T": 626, "positions": 16, "loss": 0.4353287696649204, "ce": 1.7262094916004571, "tv": 0.29189757839776576, "conf": 0.7670788811519742, "grad_norm": 5.66165018081665, "lr": 4.975808875433711e-05 }, { "step": 136, "T": 1018, "positions": 16, "loss": 0.879697899101302, "ce": 4.08145990211051, "tv": 0.5239466006169096, "conf": 0.6613445968832821, "grad_norm": 5.4206461906433105, "lr": 4.9750235357959904e-05 }, { "step": 137, "T": 527, "positions": 16, "loss": 0.9211372639983892, "ce": 3.955695489421487, "tv": 0.5839641392230988, "conf": 0.8519279826432467, "grad_norm": 5.36994743347168, "lr": 4.97422571532829e-05 }, { "step": 138, "T": 673, "positions": 16, "loss": 1.2460977286100388, "ce": 6.132188871502876, "tv": 0.7031987402588129, "conf": 0.8160538282245398, "grad_norm": 9.899136543273926, "lr": 4.973415418053789e-05 }, { "step": 139, "T": 936, "positions": 16, "loss": 0.7625119650037959, "ce": 3.437615579692647, "tv": 0.4652782464399934, "conf": 0.5348074547946453, "grad_norm": 3.9103147983551025, "lr": 4.972592648058583e-05 }, { "step": 140, "T": 593, "positions": 16, "loss": 0.6696217940188944, "ce": 2.5758343054912984, "tv": 0.45782040897756815, "conf": 0.7043949700891972, "grad_norm": 16.105531692504883, "lr": 4.971757409491663e-05 }, { "step": 141, "T": 531, "positions": 16, "loss": 0.6973414579406381, "ce": 2.4127133898437023, "tv": 0.5067445756867528, "conf": 0.763328418135643, "grad_norm": 17.915372848510742, "lr": 4.970909706564896e-05 }, { "step": 142, "T": 540, "positions": 16, "loss": 0.3885051539691631, "ce": 1.2643025661163847, "tv": 0.2911943393482943, "conf": 0.44331610202789307, "grad_norm": 5.75253963470459, "lr": 4.9700495435530046e-05 }, { "step": 143, "T": 543, "positions": 16, "loss": 0.30927530031476635, "ce": 1.0405078171461355, "tv": 0.2280272436037194, "conf": 0.39234012039378285, "grad_norm": 3.014336585998535, "lr": 4.969176924793543e-05 }, { "step": 144, "T": 573, "positions": 16, "loss": 0.20255710769197321, "ce": 0.9626604611803486, "tv": 0.11810117481945781, "conf": 0.4442789340391755, "grad_norm": 3.204442262649536, "lr": 4.9682918546868744e-05 }, { "step": 145, "T": 535, "positions": 16, "loss": 0.4657424025826913, "ce": 1.7388229874741228, "tv": 0.32428899705337244, "conf": 0.5778441443108022, "grad_norm": 10.297890663146973, "lr": 4.967394337696153e-05 }, { "step": 146, "T": 1001, "positions": 16, "loss": 0.9527868032455444, "ce": 4.751979129388928, "tv": 0.5306543093174696, "conf": 0.6025696247816086, "grad_norm": 5.715545177459717, "lr": 4.966484378347299e-05 }, { "step": 147, "T": 620, "positions": 16, "loss": 0.5183991498779505, "ce": 1.7576492892112583, "tv": 0.38070469396188855, "conf": 0.9055821746587753, "grad_norm": 13.128976821899414, "lr": 4.9655619812289734e-05 }, { "step": 148, "T": 630, "positions": 16, "loss": 0.620286294259131, "ce": 2.0603743225801736, "tv": 0.46027651196345687, "conf": 0.8110915273427963, "grad_norm": 13.14883041381836, "lr": 4.96462715099256e-05 }, { "step": 149, "T": 1178, "positions": 16, "loss": 1.1246098168194294, "ce": 5.599772669374943, "tv": 0.627369505353272, "conf": 0.529413846321404, "grad_norm": 5.806455135345459, "lr": 4.963679892352137e-05 }, { "step": 150, "T": 614, "positions": 16, "loss": 0.5117230471805669, "ce": 1.7957317292457446, "tv": 0.3690554201602936, "conf": 0.5637907711789012, "grad_norm": 3.885570764541626, "lr": 4.962720210084458e-05 }, { "step": 151, "T": 676, "positions": 16, "loss": 0.4270708438998554, "ce": 1.936763665638864, "tv": 0.25932720335549675, "conf": 0.5354925808496773, "grad_norm": 4.198619365692139, "lr": 4.961748109028923e-05 }, { "step": 152, "T": 677, "positions": 16, "loss": 0.7923881663009524, "ce": 3.48859622515738, "tv": 0.4928095121867955, "conf": 0.5439499011263251, "grad_norm": 4.054403781890869, "lr": 4.960763594087558e-05 }, { "step": 153, "T": 538, "positions": 16, "loss": 0.3542362407897599, "ce": 1.2536039247643203, "tv": 0.25430650237831287, "conf": 0.4434480487834662, "grad_norm": 4.514420986175537, "lr": 4.9597666702249865e-05 }, { "step": 154, "T": 544, "positions": 16, "loss": 0.3197513585619163, "ce": 0.9541299335687654, "tv": 0.24926485966716427, "conf": 0.4047634315211326, "grad_norm": 3.132352590560913, "lr": 4.95875734246841e-05 }, { "step": 155, "T": 539, "positions": 16, "loss": 0.5142405307124136, "ce": 2.022075859742472, "tv": 0.3467032707412727, "conf": 0.6012525111436844, "grad_norm": 3.798156261444092, "lr": 4.957735615907576e-05 }, { "step": 156, "T": 541, "positions": 16, "loss": 0.5547849375288934, "ce": 1.7812642180360854, "tv": 0.4185094656422734, "conf": 1.3169352608383633, "grad_norm": 6.109062194824219, "lr": 4.9567014956947574e-05 }, { "step": 157, "T": 902, "positions": 16, "loss": 0.849695804528892, "ce": 3.6030971333384514, "tv": 0.5437623169273138, "conf": 0.7021031430922449, "grad_norm": 4.959951400756836, "lr": 4.955654987044724e-05 }, { "step": 158, "T": 648, "positions": 16, "loss": 0.2670937523362227, "ce": 0.8727437408815604, "tv": 0.19979931059060618, "conf": 0.8844176409766078, "grad_norm": 3.0961573123931885, "lr": 4.954596095234718e-05 }, { "step": 159, "T": 672, "positions": 16, "loss": 0.9220202453434467, "ce": 3.757065288722515, "tv": 0.6070152558386326, "conf": 1.0179175464436412, "grad_norm": 3.365452766418457, "lr": 4.953524825604424e-05 }, { "step": 160, "T": 1629, "positions": 16, "loss": 0.9457015844673151, "ce": 4.632099893475242, "tv": 0.5361017625618842, "conf": 0.6403387757600285, "grad_norm": 5.288125038146973, "lr": 4.952441183555947e-05 }, { "step": 161, "T": 594, "positions": 16, "loss": 0.41567400460189674, "ce": 1.748657552874647, "tv": 0.2675647367723286, "conf": 0.5424844780936837, "grad_norm": 2.2490363121032715, "lr": 4.9513451745537806e-05 }, { "step": 162, "T": 672, "positions": 16, "loss": 0.9261262491345406, "ce": 3.579004503786564, "tv": 0.6313620209693909, "conf": 1.1655094483867288, "grad_norm": 4.490217685699463, "lr": 4.950236804124783e-05 }, { "step": 163, "T": 538, "positions": 16, "loss": 0.5380471151438542, "ce": 2.05612687417306, "tv": 0.3693715959379915, "conf": 0.8658165340311825, "grad_norm": 7.259298324584961, "lr": 4.9491160778581445e-05 }, { "step": 164, "T": 563, "positions": 16, "loss": 0.7157008368521929, "ce": 3.0847285203635693, "tv": 0.45247553940862417, "conf": 0.6654177405871451, "grad_norm": 2.5869622230529785, "lr": 4.9479830014053675e-05 }, { "step": 165, "T": 555, "positions": 16, "loss": 0.4005256164018647, "ce": 1.5524009785003727, "tv": 0.27253947939607315, "conf": 0.4675913915343699, "grad_norm": 5.5756731033325195, "lr": 4.9468375804802284e-05 }, { "step": 166, "T": 539, "positions": 16, "loss": 0.38006353577657137, "ce": 1.42682820760092, "tv": 0.2637563442622195, "conf": 0.7188072251155972, "grad_norm": 5.17134952545166, "lr": 4.945679820858754e-05 }, { "step": 167, "T": 528, "positions": 16, "loss": 0.8677110932767391, "ce": 3.2070248872041702, "tv": 0.60778733715415, "conf": 0.8210888681933284, "grad_norm": 4.959220886230469, "lr": 4.9445097283791935e-05 }, { "step": 168, "T": 568, "positions": 16, "loss": 0.30836798914606334, "ce": 1.1163936661264415, "tv": 0.2185873578673636, "conf": 0.2792238781694323, "grad_norm": 3.995931386947632, "lr": 4.943327308941985e-05 }, { "step": 169, "T": 638, "positions": 16, "loss": 0.2731282886943518, "ce": 1.1304899897354517, "tv": 0.17786587854971003, "conf": 0.5875042038969696, "grad_norm": 2.0426676273345947, "lr": 4.9421325685097286e-05 }, { "step": 170, "T": 541, "positions": 16, "loss": 0.19973757652951463, "ce": 0.7194592098758221, "tv": 0.14199072478731978, "conf": 0.5358689513523132, "grad_norm": 4.501649379730225, "lr": 4.940925513107157e-05 }, { "step": 171, "T": 546, "positions": 16, "loss": 0.8030872885137796, "ce": 3.005022931843996, "tv": 0.5584277883172035, "conf": 0.583853586576879, "grad_norm": 4.223526477813721, "lr": 4.939706148821101e-05 }, { "step": 172, "T": 960, "positions": 16, "loss": 0.8561102030798793, "ce": 3.4976915009319782, "tv": 0.5626011635176837, "conf": 0.6557402512989938, "grad_norm": 3.0001978874206543, "lr": 4.9384744818004636e-05 }, { "step": 173, "T": 645, "positions": 16, "loss": 0.8881691060960293, "ce": 3.3208636455237865, "tv": 0.6178697217255831, "conf": 0.6321421284228563, "grad_norm": 5.3572468757629395, "lr": 4.9372305182561874e-05 }, { "step": 174, "T": 595, "positions": 16, "loss": 0.5867251814197516, "ce": 2.2267478105641203, "tv": 0.4045004375220742, "conf": 0.7271821266040206, "grad_norm": 3.444694995880127, "lr": 4.93597426446122e-05 }, { "step": 175, "T": 535, "positions": 16, "loss": 0.3870968942064792, "ce": 1.3096382826042827, "tv": 0.2845923042623326, "conf": 0.6148591660894454, "grad_norm": 6.12309455871582, "lr": 4.9347057267504887e-05 }, { "step": 176, "T": 672, "positions": 16, "loss": 0.9028911180794239, "ce": 3.528240904211998, "tv": 0.6111855953931808, "conf": 0.518225590698421, "grad_norm": 3.42195725440979, "lr": 4.933424911520862e-05 }, { "step": 177, "T": 526, "positions": 16, "loss": 1.0731891617178917, "ce": 4.518917813897133, "tv": 0.6903304513543844, "conf": 1.1295823846012354, "grad_norm": 5.327836990356445, "lr": 4.9321318252311234e-05 }, { "step": 178, "T": 588, "positions": 16, "loss": 0.46218502920964966, "ce": 2.143347706140048, "tv": 0.2753891834363458, "conf": 0.4266157513484359, "grad_norm": 2.40580415725708, "lr": 4.9308264744019326e-05 }, { "step": 179, "T": 634, "positions": 16, "loss": 0.9131493046879768, "ce": 3.7722063213586807, "tv": 0.5954763405025005, "conf": 0.6785690104588866, "grad_norm": 2.4440908432006836, "lr": 4.9295088656157985e-05 }, { "step": 180, "T": 608, "positions": 16, "loss": 0.3297344225429697, "ce": 1.2268770689661324, "tv": 0.2300519069613074, "conf": 0.5756149378139526, "grad_norm": 2.909464120864868, "lr": 4.928179005517041e-05 }, { "step": 181, "T": 675, "positions": 16, "loss": 0.8544380203820765, "ce": 3.2415549363940954, "tv": 0.5892028324306011, "conf": 0.7333353944122791, "grad_norm": 2.6862359046936035, "lr": 4.9268369008117624e-05 }, { "step": 182, "T": 667, "positions": 16, "loss": 1.2422667983919382, "ce": 6.432938024401665, "tv": 0.6655255639925599, "conf": 0.7906825020909309, "grad_norm": 9.904930114746094, "lr": 4.9254825582678094e-05 }, { "step": 183, "T": 576, "positions": 16, "loss": 0.4682273796061054, "ce": 1.9310258771511144, "tv": 0.30569421133986907, "conf": 0.40604432206600904, "grad_norm": 5.232869625091553, "lr": 4.9241159847147405e-05 }, { "step": 184, "T": 534, "positions": 16, "loss": 0.620829995909844, "ce": 2.7781575197998336, "tv": 0.3811269259049368, "conf": 0.8214865000918508, "grad_norm": 2.9092321395874023, "lr": 4.922737187043792e-05 }, { "step": 185, "T": 543, "positions": 16, "loss": 0.5346728786826134, "ce": 2.1363548648077995, "tv": 0.3567082219524309, "conf": 0.5002086153253913, "grad_norm": 8.482879638671875, "lr": 4.9213461722078436e-05 }, { "step": 186, "T": 524, "positions": 16, "loss": 0.6945020128041506, "ce": 2.3698102603666484, "tv": 0.5083566424436867, "conf": 0.7419015765190125, "grad_norm": 5.495217323303223, "lr": 4.9199429472213815e-05 }, { "step": 187, "T": 674, "positions": 16, "loss": 0.6228135228157043, "ce": 2.4678096882998943, "tv": 0.4178139651194215, "conf": 0.769628370180726, "grad_norm": 10.338973045349121, "lr": 4.918527519160465e-05 }, { "step": 188, "T": 542, "positions": 16, "loss": 0.4521659487982106, "ce": 1.6122919326844567, "tv": 0.32326305719834636, "conf": 0.48529351269826293, "grad_norm": 3.4833197593688965, "lr": 4.917099895162689e-05 }, { "step": 189, "T": 672, "positions": 16, "loss": 0.6636529021780007, "ce": 2.7878064226242714, "tv": 0.42763584526255727, "conf": 0.7316254982724786, "grad_norm": 6.033681869506836, "lr": 4.91566008242715e-05 }, { "step": 190, "T": 528, "positions": 16, "loss": 0.5987086672903388, "ce": 2.358189764112467, "tv": 0.4032107818275108, "conf": 0.9204472559504211, "grad_norm": 4.024230003356934, "lr": 4.914208088214409e-05 }, { "step": 191, "T": 542, "positions": 16, "loss": 0.36275114411546383, "ce": 1.4814141984815024, "tv": 0.23845525105207344, "conf": 0.46882691280916333, "grad_norm": 2.406580924987793, "lr": 4.9127439198464536e-05 }, { "step": 192, "T": 584, "positions": 16, "loss": 0.7040338236838579, "ce": 3.016614153981209, "tv": 0.44708046689629555, "conf": 0.7854097168892622, "grad_norm": 17.188480377197266, "lr": 4.9112675847066625e-05 }, { "step": 193, "T": 650, "positions": 16, "loss": 0.7880095610562421, "ce": 3.029250056046294, "tv": 0.5389828476654657, "conf": 0.7232895200140774, "grad_norm": 5.132096767425537, "lr": 4.9097790902397686e-05 }, { "step": 194, "T": 1710, "positions": 16, "loss": 0.9491116143763065, "ce": 4.925688486197032, "tv": 0.5072697302093729, "conf": 0.6065160473808646, "grad_norm": 3.968583345413208, "lr": 4.908278443951819e-05 }, { "step": 195, "T": 672, "positions": 16, "loss": 0.8106053851079196, "ce": 3.736941726063378, "tv": 0.4854569174349308, "conf": 0.8271798440255225, "grad_norm": 3.6774349212646484, "lr": 4.906765653410142e-05 }, { "step": 196, "T": 534, "positions": 16, "loss": 0.6457019338267855, "ce": 2.5421123356500175, "tv": 0.4349896736384835, "conf": 0.5913891277741641, "grad_norm": 5.1713547706604, "lr": 4.905240726243303e-05 }, { "step": 197, "T": 541, "positions": 16, "loss": 0.6091865030466579, "ce": 3.0193404526798986, "tv": 0.34139162756036967, "conf": 0.7761037821182981, "grad_norm": 4.041895389556885, "lr": 4.90370367014107e-05 }, { "step": 198, "T": 915, "positions": 16, "loss": 0.738386111377622, "ce": 3.38732694691862, "tv": 0.44405934948736103, "conf": 0.7526766043156385, "grad_norm": 5.415581703186035, "lr": 4.902154492854374e-05 }, { "step": 199, "T": 1176, "positions": 16, "loss": 0.6856984924379503, "ce": 3.0154054001163786, "tv": 0.4268421656670398, "conf": 0.49664791536633857, "grad_norm": 4.9440460205078125, "lr": 4.9005932021952694e-05 }, { "step": 200, "T": 621, "positions": 16, "loss": 0.8575038574635983, "ce": 3.0242701172828674, "tv": 0.6167520601302385, "conf": 0.8725285287946463, "grad_norm": 5.0139055252075195, "lr": 4.8990198060368945e-05 }, { "step": 201, "T": 672, "positions": 16, "loss": 0.9075799742713571, "ce": 3.8272582702338696, "tv": 0.5831712912768126, "conf": 0.6096938122063875, "grad_norm": 3.8100051879882812, "lr": 4.897434312313431e-05 }, { "step": 202, "T": 600, "positions": 16, "loss": 0.3270591215696186, "ce": 0.8400600946042687, "tv": 0.2700590128079057, "conf": 0.6404489292763174, "grad_norm": 14.932677268981934, "lr": 4.89583672902007e-05 }, { "step": 203, "T": 537, "positions": 16, "loss": 0.5193713265107363, "ce": 1.7751387679818436, "tv": 0.37984161036729347, "conf": 0.6635610745288432, "grad_norm": 5.058139801025391, "lr": 4.8942270642129604e-05 }, { "step": 204, "T": 540, "positions": 16, "loss": 1.0289915800094604, "ce": 4.625749662518501, "tv": 0.6293518058955669, "conf": 0.9259479437023401, "grad_norm": 4.539547920227051, "lr": 4.8926053260091784e-05 }, { "step": 205, "T": 538, "positions": 16, "loss": 0.29075850799199543, "ce": 0.8417394038006023, "tv": 0.22953841036360245, "conf": 0.399115240201354, "grad_norm": 3.9666004180908203, "lr": 4.8909715225866836e-05 }, { "step": 206, "T": 543, "positions": 16, "loss": 0.18403792126991902, "ce": 0.6531443527246665, "tv": 0.13191499070671853, "conf": 0.4321207073517144, "grad_norm": 2.277495861053467, "lr": 4.8893256621842734e-05 }, { "step": 207, "T": 532, "positions": 16, "loss": 0.6111885253340006, "ce": 2.4454175964929163, "tv": 0.4073852961882949, "conf": 0.8442473169416189, "grad_norm": 5.323704719543457, "lr": 4.887667753101548e-05 }, { "step": 208, "T": 672, "positions": 16, "loss": 0.6555363319857861, "ce": 2.367552647148841, "tv": 0.46531229966058163, "conf": 0.6808699764078483, "grad_norm": 3.424941301345825, "lr": 4.8859978036988644e-05 }, { "step": 209, "T": 621, "positions": 16, "loss": 0.42417272739112377, "ce": 1.8111486225388944, "tv": 0.2700643017888069, "conf": 0.7944882865995169, "grad_norm": 3.473167896270752, "lr": 4.884315822397297e-05 }, { "step": 210, "T": 678, "positions": 16, "loss": 0.5284697510069236, "ce": 2.1273951583571034, "tv": 0.35081137777888216, "conf": 1.1765866712667048, "grad_norm": 2.800339698791504, "lr": 4.882621817678592e-05 }, { "step": 211, "T": 658, "positions": 16, "loss": 0.4260348821990192, "ce": 1.5167076399666257, "tv": 0.30484903091564775, "conf": 0.5643689353018999, "grad_norm": 3.786862850189209, "lr": 4.880915798085126e-05 }, { "step": 212, "T": 954, "positions": 16, "loss": 0.9733953895047307, "ce": 4.996972429973539, "tv": 0.5263312729075551, "conf": 0.6231417274102569, "grad_norm": 5.5054240226745605, "lr": 4.879197772219865e-05 }, { "step": 213, "T": 534, "positions": 16, "loss": 0.5205120448954403, "ce": 1.742140272865072, "tv": 0.3847755941096693, "conf": 0.8813933893106878, "grad_norm": 12.941115379333496, "lr": 4.8774677487463175e-05 }, { "step": 214, "T": 538, "positions": 16, "loss": 0.4597923948895186, "ce": 1.549083185615018, "tv": 0.33876008726656437, "conf": 0.724271603859961, "grad_norm": 7.223724365234375, "lr": 4.875725736388492e-05 }, { "step": 215, "T": 674, "positions": 16, "loss": 0.7394791403785348, "ce": 2.654589332640171, "tv": 0.5266890991479158, "conf": 0.8134747799485922, "grad_norm": 2.7882022857666016, "lr": 4.873971743930855e-05 }, { "step": 216, "T": 623, "positions": 16, "loss": 0.7578958850353956, "ce": 2.903876915574074, "tv": 0.5194535544142127, "conf": 0.7555536134168506, "grad_norm": 3.33088755607605, "lr": 4.8722057802182866e-05 }, { "step": 217, "T": 577, "positions": 16, "loss": 0.5503803697647527, "ce": 2.2138354426715523, "tv": 0.36555204971227795, "conf": 0.7140346707310528, "grad_norm": 3.264549493789673, "lr": 4.87042785415603e-05 }, { "step": 218, "T": 597, "positions": 16, "loss": 0.553226772841299, "ce": 2.2447066790336976, "tv": 0.3652845713950228, "conf": 0.9820270808413625, "grad_norm": 4.769550800323486, "lr": 4.8686379747096556e-05 }, { "step": 219, "T": 569, "positions": 16, "loss": 0.699766356498003, "ce": 2.679560426622629, "tv": 0.47978921979665756, "conf": 0.752060383092612, "grad_norm": 4.5847954750061035, "lr": 4.8668361509050074e-05 }, { "step": 220, "T": 527, "positions": 16, "loss": 0.30477824094123207, "ce": 1.1492728473531315, "tv": 0.21094549655390438, "conf": 0.65462157712318, "grad_norm": 3.044372797012329, "lr": 4.8650223918281645e-05 }, { "step": 221, "T": 532, "positions": 16, "loss": 0.5576589373777097, "ce": 2.289372188875859, "tv": 0.36524637170805363, "conf": 0.8935150634497404, "grad_norm": 5.99857759475708, "lr": 4.863196706625391e-05 }, { "step": 222, "T": 532, "positions": 16, "loss": 1.2904005274176598, "ce": 6.500642411410809, "tv": 0.7114847749471664, "conf": 0.9618087708950043, "grad_norm": 6.276761531829834, "lr": 4.86135910450309e-05 }, { "step": 223, "T": 673, "positions": 16, "loss": 0.8032342698425055, "ce": 3.0601073019206524, "tv": 0.5524706076830626, "conf": 0.8719844203442335, "grad_norm": 2.999279022216797, "lr": 4.85950959472776e-05 }, { "step": 224, "T": 584, "positions": 16, "loss": 0.41419953892182093, "ce": 1.475727343800827, "tv": 0.29625201516319066, "conf": 0.43963599659036845, "grad_norm": 2.0949628353118896, "lr": 4.857648186625945e-05 }, { "step": 225, "T": 538, "positions": 16, "loss": 0.727023262064904, "ce": 2.76356313819997, "tv": 0.5007410563994199, "conf": 0.8446452766656876, "grad_norm": 4.316161632537842, "lr": 4.8557748895841895e-05 }, { "step": 226, "T": 548, "positions": 16, "loss": 0.5138174092571717, "ce": 2.015526243048953, "tv": 0.34696088630880695, "conf": 0.7047801297158003, "grad_norm": 2.6646578311920166, "lr": 4.85388971304899e-05 }, { "step": 227, "T": 541, "positions": 16, "loss": 0.30479827253657277, "ce": 1.1807735951915674, "tv": 0.20746768917888403, "conf": 0.6235933422576636, "grad_norm": 4.833371162414551, "lr": 4.851992666526749e-05 }, { "step": 228, "T": 607, "positions": 16, "loss": 0.24361242982922704, "ce": 0.8676571791638708, "tv": 0.17427411760218092, "conf": 0.5977660060161725, "grad_norm": 3.596076250076294, "lr": 4.850083759583723e-05 }, { "step": 229, "T": 530, "positions": 16, "loss": 0.32619766255811555, "ce": 1.2484718527157384, "tv": 0.22372275874658953, "conf": 0.5383565180236474, "grad_norm": 3.716749429702759, "lr": 4.8481630018459825e-05 }, { "step": 230, "T": 561, "positions": 16, "loss": 0.25727977602218743, "ce": 0.9916000039193023, "tv": 0.1756886466537253, "conf": 0.49528741650283337, "grad_norm": 3.3313772678375244, "lr": 4.8462304029993514e-05 }, { "step": 231, "T": 976, "positions": 16, "loss": 0.8530048756040287, "ce": 3.8771197344615302, "tv": 0.5169920816369995, "conf": 0.6806846959807444, "grad_norm": 4.460178375244141, "lr": 4.844285972789371e-05 }, { "step": 232, "T": 557, "positions": 16, "loss": 0.7924851709976792, "ce": 3.1509059390518814, "tv": 0.5304384133778512, "conf": 0.758378338534385, "grad_norm": 5.754398345947266, "lr": 4.842329721021241e-05 }, { "step": 233, "T": 648, "positions": 16, "loss": 0.46519285466638394, "ce": 1.5725400870433077, "tv": 0.34215429791947827, "conf": 0.9886178923770785, "grad_norm": 4.374191761016846, "lr": 4.840361657559775e-05 }, { "step": 234, "T": 542, "positions": 16, "loss": 0.36130176699771255, "ce": 1.5794713851264532, "tv": 0.2259495899634203, "conf": 0.2859910516999662, "grad_norm": 1.7441186904907227, "lr": 4.8383817923293494e-05 }, { "step": 235, "T": 644, "positions": 16, "loss": 0.5217998220141453, "ce": 1.938408546109713, "tv": 0.3643988543699379, "conf": 0.5502534558763728, "grad_norm": 4.394278526306152, "lr": 4.836390135313854e-05 }, { "step": 236, "T": 534, "positions": 16, "loss": 0.6893211046699435, "ce": 2.714894323144108, "tv": 0.4642574144527316, "conf": 0.862470526360994, "grad_norm": 4.721604824066162, "lr": 4.834386696556641e-05 }, { "step": 237, "T": 1054, "positions": 16, "loss": 0.7603295826338581, "ce": 3.0897518182864587, "tv": 0.5015048737841425, "conf": 0.8102282374165952, "grad_norm": 3.87031626701355, "lr": 4.8323714861604744e-05 }, { "step": 238, "T": 675, "positions": 16, "loss": 0.46519837182131596, "ce": 1.9063416565422813, "tv": 0.30507134487561416, "conf": 0.5097696541342884, "grad_norm": 2.8271565437316895, "lr": 4.830344514287478e-05 }, { "step": 239, "T": 619, "positions": 16, "loss": 0.7596620063268347, "ce": 3.1381562002788996, "tv": 0.4953848923469195, "conf": 0.780728137557162, "grad_norm": 2.5028443336486816, "lr": 4.8283057911590865e-05 }, { "step": 240, "T": 552, "positions": 16, "loss": 0.9671581590082496, "ce": 4.934422773891129, "tv": 0.526350989472121, "conf": 0.7697336059063673, "grad_norm": 6.472486972808838, "lr": 4.826255327055995e-05 }, { "step": 241, "T": 542, "positions": 16, "loss": 0.7537002895624028, "ce": 2.7210659637821664, "tv": 0.5351041275489479, "conf": 0.8218558291555382, "grad_norm": 3.154975175857544, "lr": 4.8241931323181e-05 }, { "step": 242, "T": 927, "positions": 16, "loss": 0.6136954792846154, "ce": 2.5723282281019237, "tv": 0.3960696123403977, "conf": 0.9689203286543489, "grad_norm": 5.068091869354248, "lr": 4.822119217344456e-05 }, { "step": 243, "T": 581, "positions": 16, "loss": 0.2128361796385434, "ce": 0.8320933299983153, "tv": 0.14402984418120468, "conf": 0.6005016115959734, "grad_norm": 4.179928302764893, "lr": 4.8200335925932185e-05 }, { "step": 244, "T": 541, "positions": 16, "loss": 0.6655402323231101, "ce": 2.589991091750562, "tv": 0.45171235781162977, "conf": 0.7873521959409118, "grad_norm": 3.6495115756988525, "lr": 4.817936268581592e-05 }, { "step": 245, "T": 531, "positions": 16, "loss": 0.9495353456586599, "ce": 3.865395943634212, "tv": 0.6255508419126272, "conf": 0.6677528340369463, "grad_norm": 6.4959187507629395, "lr": 4.815827255885775e-05 }, { "step": 246, "T": 577, "positions": 16, "loss": 0.4478509758482687, "ce": 1.6029949855292216, "tv": 0.3195016296231188, "conf": 1.1287927515804768, "grad_norm": 4.426764965057373, "lr": 4.813706565140912e-05 }, { "step": 247, "T": 827, "positions": 16, "loss": 0.8027468074578792, "ce": 3.4135404595872387, "tv": 0.5126586416736245, "conf": 1.0545414322987199, "grad_norm": 5.456354141235352, "lr": 4.811574207041034e-05 }, { "step": 248, "T": 532, "positions": 16, "loss": 0.25644731742795557, "ce": 0.6833735329564661, "tv": 0.20901106973178685, "conf": 0.9223735183477402, "grad_norm": 6.176387786865234, "lr": 4.809430192339008e-05 }, { "step": 249, "T": 532, "positions": 16, "loss": 0.43229467132186983, "ce": 1.3897967837765464, "tv": 0.32590555823844625, "conf": 0.9018165294583014, "grad_norm": 3.917154550552368, "lr": 4.8072745318464826e-05 }, { "step": 250, "T": 587, "positions": 16, "loss": 0.5003142568748444, "ce": 1.6963942715228768, "tv": 0.3674164938274771, "conf": 0.985528773162514, "grad_norm": 3.910722494125366, "lr": 4.8051072364338315e-05 }, { "step": 251, "T": 620, "positions": 16, "loss": 0.27531135759818426, "ce": 0.944131439536477, "tv": 0.20099801600008504, "conf": 0.6110421558842063, "grad_norm": 2.529996871948242, "lr": 4.802928317030101e-05 }, { "step": 252, "T": 535, "positions": 16, "loss": 0.8463917160406709, "ce": 3.456091072410345, "tv": 0.5564251355826855, "conf": 0.6529518631286919, "grad_norm": 4.422735214233398, "lr": 4.800737784622952e-05 }, { "step": 253, "T": 532, "positions": 16, "loss": 0.6640487438999116, "ce": 2.8599524591118097, "tv": 0.4200594574213028, "conf": 0.61015918222256, "grad_norm": 4.974478721618652, "lr": 4.79853565025861e-05 }, { "step": 254, "T": 553, "positions": 16, "loss": 0.3286423019017093, "ce": 1.2926198533314164, "tv": 0.2215336920999107, "conf": 0.6494392039894592, "grad_norm": 4.692689418792725, "lr": 4.7963219250418014e-05 }, { "step": 255, "T": 567, "positions": 16, "loss": 0.20054019896269892, "ce": 0.601546962636121, "tv": 0.15598388854050427, "conf": 0.5187542482744902, "grad_norm": 2.9232702255249023, "lr": 4.794096620135705e-05 }, { "step": 256, "T": 676, "positions": 16, "loss": 0.46593455696711317, "ce": 1.8266764808140579, "tv": 0.31474102503125323, "conf": 0.7425714737837552, "grad_norm": 2.1864724159240723, "lr": 4.791859746761892e-05 }, { "step": 257, "T": 571, "positions": 16, "loss": 0.8241903606103733, "ce": 3.3766940071072895, "tv": 0.5405788498464972, "conf": 0.49664421880152076, "grad_norm": 4.8265838623046875, "lr": 4.7896113162002694e-05 }, { "step": 258, "T": 675, "positions": 16, "loss": 0.39030643878504634, "ce": 1.273589240387082, "tv": 0.29216389870271087, "conf": 0.6407246398739517, "grad_norm": 7.720123291015625, "lr": 4.787351339789025e-05 }, { "step": 259, "T": 631, "positions": 16, "loss": 0.2676847738439392, "ce": 0.8059621575675919, "tv": 0.2078761702778138, "conf": 0.4845560808898881, "grad_norm": 5.6602582931518555, "lr": 4.785079828924566e-05 }, { "step": 260, "T": 625, "positions": 16, "loss": 0.5530821443535388, "ce": 2.2656142984051257, "tv": 0.3628007916267961, "conf": 0.7005806770175695, "grad_norm": 4.981041431427002, "lr": 4.782796795061469e-05 }, { "step": 261, "T": 621, "positions": 16, "loss": 0.6082613681210205, "ce": 2.211546499398537, "tv": 0.4301186070078984, "conf": 0.8003661817638204, "grad_norm": 5.3005218505859375, "lr": 4.780502249712414e-05 }, { "step": 262, "T": 551, "positions": 16, "loss": 0.6499065448297188, "ce": 2.969808939800714, "tv": 0.3921396220139286, "conf": 0.4876978542888537, "grad_norm": 3.336763620376587, "lr": 4.778196204448132e-05 }, { "step": 263, "T": 600, "positions": 16, "loss": 0.9297552816569805, "ce": 4.343653815798461, "tv": 0.5504332138225436, "conf": 0.8581324405968189, "grad_norm": 5.096346855163574, "lr": 4.7758786708973444e-05 }, { "step": 264, "T": 540, "positions": 16, "loss": 0.20734028039987606, "ce": 0.7238359434278436, "tv": 0.14995187405656907, "conf": 0.1996660323929973, "grad_norm": 1.957106351852417, "lr": 4.773549660746706e-05 }, { "step": 265, "T": 542, "positions": 16, "loss": 0.14673043470975244, "ce": 0.5227413057166359, "tv": 0.10495145631284686, "conf": 0.1506180063006468, "grad_norm": 1.7088171243667603, "lr": 4.7712091857407395e-05 }, { "step": 266, "T": 915, "positions": 16, "loss": 0.7079503361674142, "ce": 3.4775613594028982, "tv": 0.40021578862069873, "conf": 0.7628007296589203, "grad_norm": 4.0379133224487305, "lr": 4.768857257681789e-05 }, { "step": 267, "T": 620, "positions": 16, "loss": 0.858518997207284, "ce": 3.9970304891467094, "tv": 0.5097955050878227, "conf": 0.7870252700522542, "grad_norm": 6.261332035064697, "lr": 4.7664938884299485e-05 }, { "step": 268, "T": 673, "positions": 16, "loss": 0.7327953479543794, "ce": 3.054616359906504, "tv": 0.474815239518648, "conf": 0.8609482552856207, "grad_norm": 3.727945566177368, "lr": 4.764119089903008e-05 }, { "step": 269, "T": 621, "positions": 16, "loss": 0.44021874805912375, "ce": 1.384278534387704, "tv": 0.3353232118533924, "conf": 0.8457879815250635, "grad_norm": 2.80397891998291, "lr": 4.76173287407639e-05 }, { "step": 270, "T": 529, "positions": 16, "loss": 0.4915143954567611, "ce": 2.1030925688537536, "tv": 0.31245015759486705, "conf": 1.1440955031794147, "grad_norm": 3.1928257942199707, "lr": 4.759335252983094e-05 }, { "step": 271, "T": 1196, "positions": 16, "loss": 0.8263937132433057, "ce": 3.544344663619995, "tv": 0.5243991822935641, "conf": 0.763151194609236, "grad_norm": 5.100829124450684, "lr": 4.7569262387136315e-05 }, { "step": 272, "T": 1273, "positions": 16, "loss": 0.8262681928172242, "ce": 3.794763496352971, "tv": 0.49643538091913797, "conf": 1.0503981778165326, "grad_norm": 3.951857089996338, "lr": 4.754505843415968e-05 }, { "step": 273, "T": 524, "positions": 16, "loss": 0.7810960123315454, "ce": 3.0544967092573643, "tv": 0.5284959468990564, "conf": 1.192129012197256, "grad_norm": 3.5174202919006348, "lr": 4.752074079295457e-05 }, { "step": 274, "T": 624, "positions": 16, "loss": 0.41364666260778904, "ce": 1.4317762032151222, "tv": 0.30052116280421615, "conf": 1.1480367574840784, "grad_norm": 3.430241823196411, "lr": 4.7496309586147856e-05 }, { "step": 275, "T": 525, "positions": 16, "loss": 0.7309216923313215, "ce": 2.8311773049936164, "tv": 0.497559979732614, "conf": 0.7912781535123941, "grad_norm": 3.1290926933288574, "lr": 4.7471764936939054e-05 }, { "step": 276, "T": 535, "positions": 16, "loss": 0.3523701340891421, "ce": 1.3390308127272874, "tv": 0.24274118128232658, "conf": 0.4819467645138502, "grad_norm": 2.67090106010437, "lr": 4.744710696909976e-05 }, { "step": 277, "T": 631, "positions": 16, "loss": 0.2721413613471668, "ce": 1.1153253927695914, "tv": 0.1784542482928373, "conf": 0.7790125757455826, "grad_norm": 2.1440343856811523, "lr": 4.7422335806972996e-05 }, { "step": 278, "T": 872, "positions": 16, "loss": 0.8466391529655084, "ce": 3.7638324304425623, "tv": 0.5225065649283351, "conf": 1.1726716773264343, "grad_norm": 3.28509521484375, "lr": 4.739745157547258e-05 }, { "step": 279, "T": 1563, "positions": 16, "loss": 0.8238390734477434, "ce": 3.913731705455575, "tv": 0.4805176548252348, "conf": 1.2185438107699156, "grad_norm": 3.4842281341552734, "lr": 4.737245440008253e-05 }, { "step": 280, "T": 675, "positions": 16, "loss": 0.6064041174468002, "ce": 2.3534003616723567, "tv": 0.41229341416874377, "conf": 0.8104196513013449, "grad_norm": 2.3628945350646973, "lr": 4.734734440685638e-05 }, { "step": 281, "T": 534, "positions": 16, "loss": 0.23444749671398313, "ce": 0.9227652150875656, "tv": 0.15796774500995525, "conf": 0.7982673729547969, "grad_norm": 2.5113582611083984, "lr": 4.7322121722416596e-05 }, { "step": 282, "T": 673, "positions": 16, "loss": 0.539012588938931, "ce": 2.2231146186823025, "tv": 0.35189015019568615, "conf": 1.1091210842132568, "grad_norm": 3.56768798828125, "lr": 4.729678647395389e-05 }, { "step": 283, "T": 672, "positions": 16, "loss": 0.8376456005062209, "ce": 3.449134023467195, "tv": 0.5474802247335901, "conf": 0.4921082452055998, "grad_norm": 3.722539186477661, "lr": 4.727133878922663e-05 }, { "step": 284, "T": 532, "positions": 16, "loss": 0.7636046152911149, "ce": 3.2856530681601726, "tv": 0.48337702779099345, "conf": 0.6208418179303408, "grad_norm": 8.023844718933105, "lr": 4.724577879656014e-05 }, { "step": 285, "T": 594, "positions": 16, "loss": 0.20318128271901514, "ce": 0.6363188466930296, "tv": 0.15505488342023455, "conf": 0.5285610156133771, "grad_norm": 2.257763385772705, "lr": 4.72201066248461e-05 }, { "step": 286, "T": 539, "positions": 16, "loss": 0.30919390876351827, "ce": 1.035803977118121, "tv": 0.2284594629654748, "conf": 0.34488438119660714, "grad_norm": 4.020167350769043, "lr": 4.7194322403541854e-05 }, { "step": 287, "T": 540, "positions": 16, "loss": 0.1200757047481602, "ce": 0.30028679607005415, "tv": 0.10005225191707723, "conf": 0.2116994740208611, "grad_norm": 2.2768712043762207, "lr": 4.716842626266981e-05 }, { "step": 288, "T": 531, "positions": 16, "loss": 0.6441756039275788, "ce": 2.400903147943609, "tv": 0.44898366343113594, "conf": 0.8341188407503068, "grad_norm": 5.603632926940918, "lr": 4.7142418332816735e-05 }, { "step": 289, "T": 530, "positions": 16, "loss": 0.4953750914428383, "ce": 1.8360162700992078, "tv": 0.3464149716310203, "conf": 1.1124623198993504, "grad_norm": 7.086953639984131, "lr": 4.711629874513312e-05 }, { "step": 290, "T": 676, "positions": 16, "loss": 0.44078346714377403, "ce": 1.5911943973624147, "tv": 0.3129600442480296, "conf": 0.8190013412677217, "grad_norm": 2.5893025398254395, "lr": 4.709006763133251e-05 }, { "step": 291, "T": 529, "positions": 16, "loss": 0.5488018392788945, "ce": 2.011993157560937, "tv": 0.3862250289967051, "conf": 0.6463861205561443, "grad_norm": 3.233264446258545, "lr": 4.706372512369085e-05 }, { "step": 292, "T": 626, "positions": 16, "loss": 0.4655929654763895, "ce": 1.4471144968156295, "tv": 0.35653502564673545, "conf": 0.9122113839257509, "grad_norm": 4.653862476348877, "lr": 4.703727135504581e-05 }, { "step": 293, "T": 569, "positions": 16, "loss": 1.0089908838272095, "ce": 5.179131376557052, "tv": 0.5456419298425317, "conf": 1.010402299463749, "grad_norm": 8.52396297454834, "lr": 4.701070645879612e-05 }, { "step": 294, "T": 957, "positions": 16, "loss": 1.0435335878282785, "ce": 4.7164008393883705, "tv": 0.6354372296482325, "conf": 1.171168946661055, "grad_norm": 3.380704879760742, "lr": 4.698403056890088e-05 }, { "step": 295, "T": 630, "positions": 16, "loss": 0.5131645745423157, "ce": 1.8805459248251282, "tv": 0.3612333204364404, "conf": 0.8985086721622793, "grad_norm": 4.66869592666626, "lr": 4.6957243819878924e-05 }, { "step": 296, "T": 672, "positions": 16, "loss": 0.8746849717572331, "ce": 3.3352248864248395, "tv": 0.6012916606850922, "conf": 0.6616832059225999, "grad_norm": 3.990767478942871, "lr": 4.69303463468081e-05 }, { "step": 297, "T": 659, "positions": 16, "loss": 0.45724341410095803, "ce": 1.5743752868729644, "tv": 0.3331176690699067, "conf": 0.7575325400102884, "grad_norm": 3.9033854007720947, "lr": 4.6903338285324606e-05 }, { "step": 298, "T": 559, "positions": 16, "loss": 0.3762243686051079, "ce": 1.4878220467662686, "tv": 0.2527135010659549, "conf": 0.6601298362948, "grad_norm": 1.7275636196136475, "lr": 4.687621977162231e-05 }, { "step": 299, "T": 674, "positions": 16, "loss": 0.7719149207696319, "ce": 2.866471854969859, "tv": 0.5391864003613591, "conf": 0.6032696564216167, "grad_norm": 2.7845518589019775, "lr": 4.684899094245204e-05 }, { "step": 300, "T": 525, "positions": 16, "loss": 0.5300799942924641, "ce": 1.812088300473988, "tv": 0.3876346475444734, "conf": 0.7547827258240432, "grad_norm": 4.120368957519531, "lr": 4.682165193512096e-05 }, { "step": 301, "T": 535, "positions": 16, "loss": 0.4832125386892585, "ce": 1.688501951226499, "tv": 0.34929150124662556, "conf": 0.7889273721375503, "grad_norm": 3.8907785415649414, "lr": 4.679420288749178e-05 }, { "step": 302, "T": 533, "positions": 16, "loss": 0.613637129052222, "ce": 2.303279529176507, "tv": 0.4258991002316179, "conf": 0.7450613542232531, "grad_norm": 2.4648971557617188, "lr": 4.676664393798213e-05 }, { "step": 303, "T": 603, "positions": 16, "loss": 0.4953061355627142, "ce": 1.8419628624687903, "tv": 0.3456776092934888, "conf": 0.6061846806987887, "grad_norm": 5.310257434844971, "lr": 4.673897522556385e-05 }, { "step": 304, "T": 600, "positions": 16, "loss": 0.2944302222094848, "ce": 1.0802370985547896, "tv": 0.2071183513944561, "conf": 0.4713246930477908, "grad_norm": 1.7418129444122314, "lr": 4.6711196889762287e-05 }, { "step": 305, "T": 594, "positions": 16, "loss": 0.3944477437325986, "ce": 1.3606892010752745, "tv": 0.28708759874280076, "conf": 0.8148580148117617, "grad_norm": 3.1937031745910645, "lr": 4.668330907065557e-05 }, { "step": 306, "T": 533, "positions": 16, "loss": 0.7435305968683679, "ce": 2.5524473813711666, "tv": 0.5425398688530549, "conf": 1.075924028409645, "grad_norm": 5.690776348114014, "lr": 4.6655311908873925e-05 }, { "step": 307, "T": 533, "positions": 16, "loss": 0.4120321436203085, "ce": 1.5174364748145308, "tv": 0.289209431513882, "conf": 1.0984507463872433, "grad_norm": 4.904698371887207, "lr": 4.662720554559897e-05 }, { "step": 308, "T": 602, "positions": 16, "loss": 0.4164665699499892, "ce": 2.128879127332766, "tv": 0.22619850278715603, "conf": 0.8059893036261201, "grad_norm": 10.701120376586914, "lr": 4.6598990122562996e-05 }, { "step": 309, "T": 647, "positions": 16, "loss": 0.2028578856625245, "ce": 0.6333364665065346, "tv": 0.15502692441441468, "conf": 0.5134157700813375, "grad_norm": 1.375350832939148, "lr": 4.6570665782048236e-05 }, { "step": 310, "T": 576, "positions": 16, "loss": 0.2673473156428372, "ce": 1.213680945626038, "tv": 0.16219913301574707, "conf": 0.413432871340774, "grad_norm": 2.9268436431884766, "lr": 4.6542232666886174e-05 }, { "step": 311, "T": 531, "positions": 16, "loss": 0.38810055927024223, "ce": 1.5004472990999602, "tv": 0.26450648784521036, "conf": 0.5670219742460176, "grad_norm": 2.9504730701446533, "lr": 4.6513690920456807e-05 }, { "step": 312, "T": 643, "positions": 16, "loss": 0.3987461477663601, "ce": 1.6137606709235115, "tv": 0.2637445472573745, "conf": 1.114850754634972, "grad_norm": 4.195267677307129, "lr": 4.648504068668794e-05 }, { "step": 313, "T": 1335, "positions": 16, "loss": 0.7766610059188679, "ce": 3.520679251572801, "tv": 0.47177009188453667, "conf": 1.0176723024396779, "grad_norm": 3.37692928314209, "lr": 4.645628211005443e-05 }, { "step": 314, "T": 538, "positions": 16, "loss": 0.34878958971239626, "ce": 1.2806583784986287, "tv": 0.24524861271493137, "conf": 0.6449154842412099, "grad_norm": 5.539743423461914, "lr": 4.64274153355775e-05 }, { "step": 315, "T": 605, "positions": 16, "loss": 0.23055876758735394, "ce": 0.9391068729073595, "tv": 0.1518312106745725, "conf": 0.40203133356408216, "grad_norm": 0.9693098664283752, "lr": 4.639844050882396e-05 }, { "step": 316, "T": 554, "positions": 16, "loss": 0.5739188541701878, "ce": 2.144148940045852, "tv": 0.3994488565658685, "conf": 0.754095497308299, "grad_norm": 4.002387046813965, "lr": 4.636935777590551e-05 }, { "step": 317, "T": 912, "positions": 16, "loss": 0.6864449536660686, "ce": 3.2341040720930323, "tv": 0.4033716960111633, "conf": 0.9397197382058948, "grad_norm": 2.9484188556671143, "lr": 4.634016728347798e-05 }, { "step": 318, "T": 525, "positions": 16, "loss": 0.7824090234935284, "ce": 2.8382151126861572, "tv": 0.5539861340075731, "conf": 1.0756634008139372, "grad_norm": 2.491741895675659, "lr": 4.63108691787406e-05 }, { "step": 319, "T": 530, "positions": 16, "loss": 0.5597842656152352, "ce": 2.4300961644748895, "tv": 0.3519718390325579, "conf": 0.5285553273715777, "grad_norm": 3.3310751914978027, "lr": 4.6281463609435285e-05 }, { "step": 320, "T": 638, "positions": 16, "loss": 0.5226241867057979, "ce": 1.7816767129697837, "tv": 0.38272946869255975, "conf": 0.601015758831636, "grad_norm": 3.5404155254364014, "lr": 4.625195072384582e-05 }, { "step": 321, "T": 547, "positions": 16, "loss": 0.8224700577557087, "ce": 2.930555362254381, "tv": 0.5882383733987808, "conf": 1.2137319948524237, "grad_norm": 2.5239980220794678, "lr": 4.6222330670797194e-05 }, { "step": 322, "T": 675, "positions": 16, "loss": 0.7013337954413146, "ce": 2.5508608408854343, "tv": 0.4958308134227991, "conf": 0.7795297205448151, "grad_norm": 5.238683223724365, "lr": 4.619260359965478e-05 }, { "step": 323, "T": 563, "positions": 16, "loss": 0.6448844133410603, "ce": 2.507719788234681, "tv": 0.43790271040052176, "conf": 0.5556652927771211, "grad_norm": 3.6179070472717285, "lr": 4.616276966032363e-05 }, { "step": 324, "T": 626, "positions": 16, "loss": 0.6178507262375206, "ce": 2.0788310278148856, "tv": 0.45551958365831524, "conf": 1.0288583552464843, "grad_norm": 3.842985153198242, "lr": 4.613282900324769e-05 }, { "step": 325, "T": 673, "positions": 16, "loss": 0.6924974676803686, "ce": 2.9565709939342923, "tv": 0.44093378144316375, "conf": 0.530911561101675, "grad_norm": 6.012840747833252, "lr": 4.610278177940908e-05 }, { "step": 326, "T": 535, "positions": 16, "loss": 0.18798579521899228, "ce": 0.684825021600318, "tv": 0.1327814309270252, "conf": 0.41065749659901485, "grad_norm": 1.8532674312591553, "lr": 4.607262814032727e-05 }, { "step": 327, "T": 534, "positions": 16, "loss": 0.12975258615369967, "ce": 0.45966628128144293, "tv": 0.0930955028416065, "conf": 0.3065808237588499, "grad_norm": 1.5680514574050903, "lr": 4.604236823805836e-05 }, { "step": 328, "T": 546, "positions": 16, "loss": 0.4106618418591097, "ce": 1.5348409570287913, "tv": 0.2857530612673145, "conf": 0.7816667588113546, "grad_norm": 2.965900421142578, "lr": 4.6012002225194325e-05 }, { "step": 329, "T": 539, "positions": 16, "loss": 0.46298455178839504, "ce": 1.6595060429135629, "tv": 0.3300377274681523, "conf": 0.5813685598259326, "grad_norm": 2.036290168762207, "lr": 4.59815302548622e-05 }, { "step": 330, "T": 608, "positions": 16, "loss": 0.5495492257614387, "ce": 2.3570284001798427, "tv": 0.3487181879027048, "conf": 0.995367556810379, "grad_norm": 5.650441646575928, "lr": 4.595095248072336e-05 }, { "step": 331, "T": 648, "positions": 16, "loss": 0.45673677390732337, "ce": 1.5151805666246219, "tv": 0.3391319165821187, "conf": 1.0724680101848207, "grad_norm": 2.3364779949188232, "lr": 4.59202690569727e-05 }, { "step": 332, "T": 529, "positions": 16, "loss": 0.4914159722684417, "ce": 1.6682272067118902, "tv": 0.36065917687665205, "conf": 1.0893832216970623, "grad_norm": 3.2399702072143555, "lr": 4.588948013833786e-05 }, { "step": 333, "T": 580, "positions": 16, "loss": 0.6321642554539721, "ce": 2.2237017266106704, "tv": 0.45532675638241926, "conf": 0.5716718165203929, "grad_norm": 5.1030659675598145, "lr": 4.585858588007849e-05 }, { "step": 334, "T": 544, "positions": 16, "loss": 0.5443972020875663, "ce": 2.3055389949586242, "tv": 0.3487147950800136, "conf": 0.44706591725116596, "grad_norm": 11.587101936340332, "lr": 4.582758643798542e-05 }, { "step": 335, "T": 527, "positions": 16, "loss": 0.7711804118007421, "ce": 2.824719298630953, "tv": 0.5430094227194786, "conf": 0.7576565779745579, "grad_norm": 5.995885848999023, "lr": 4.5796481968379906e-05 }, { "step": 336, "T": 567, "positions": 16, "loss": 0.47723809489252744, "ce": 2.221454516722588, "tv": 0.2834362750436412, "conf": 0.3397048038896173, "grad_norm": 3.0920801162719727, "lr": 4.5765272628112814e-05 }, { "step": 337, "T": 672, "positions": 16, "loss": 0.6383094538468868, "ce": 2.305134218884632, "tv": 0.45310670230537653, "conf": 0.974112996365875, "grad_norm": 4.709658145904541, "lr": 4.573395857456385e-05 }, { "step": 338, "T": 589, "positions": 16, "loss": 0.5906965441099601, "ce": 2.091105241095647, "tv": 0.4239844691037433, "conf": 0.9005631897889543, "grad_norm": 4.7054595947265625, "lr": 4.570253996564075e-05 }, { "step": 339, "T": 620, "positions": 16, "loss": 0.5763899168669013, "ce": 2.193317416575155, "tv": 0.3967313163302606, "conf": 0.9662823057733476, "grad_norm": 3.6566057205200195, "lr": 4.5671016959778515e-05 }, { "step": 340, "T": 533, "positions": 16, "loss": 0.298335291146941, "ce": 1.1188416490485906, "tv": 0.20716792339953827, "conf": 0.4396072896197438, "grad_norm": 2.472132921218872, "lr": 4.563938971593855e-05 }, { "step": 341, "T": 1060, "positions": 16, "loss": 0.6630659636139171, "ce": 2.9067231141074217, "tv": 0.41377073461626424, "conf": 0.5446101443376392, "grad_norm": 2.2755818367004395, "lr": 4.560765839360794e-05 }, { "step": 342, "T": 538, "positions": 16, "loss": 0.5823757924008532, "ce": 2.015263020293787, "tv": 0.423166115047934, "conf": 0.7299611185444519, "grad_norm": 5.383785724639893, "lr": 4.557582315279859e-05 }, { "step": 343, "T": 529, "positions": 16, "loss": 0.19466952135553584, "ce": 0.6728067131189164, "tv": 0.14154317282373086, "conf": 0.6924319565296173, "grad_norm": 2.294349193572998, "lr": 4.554388415404644e-05 }, { "step": 344, "T": 635, "positions": 16, "loss": 0.6425714949145913, "ce": 2.2541093949694186, "tv": 0.46351176174357533, "conf": 0.7262925789691508, "grad_norm": 2.1490306854248047, "lr": 4.551184155841065e-05 }, { "step": 345, "T": 628, "positions": 16, "loss": 0.3614914553181734, "ce": 1.3608811048907228, "tv": 0.2504481583600864, "conf": 0.6146192885935307, "grad_norm": 8.302788734436035, "lr": 4.547969552747279e-05 }, { "step": 346, "T": 538, "positions": 16, "loss": 0.3502271555844345, "ce": 1.3087038712692447, "tv": 0.2437297573851538, "conf": 0.5519974641501904, "grad_norm": 4.7349371910095215, "lr": 4.544744622333602e-05 }, { "step": 347, "T": 574, "positions": 16, "loss": 0.5587694798596203, "ce": 2.1734807030734373, "tv": 0.37935710698366165, "conf": 1.0220800405368209, "grad_norm": 6.400606155395508, "lr": 4.5415093808624285e-05 }, { "step": 348, "T": 595, "positions": 16, "loss": 0.43495487817563117, "ce": 1.4603603335308435, "tv": 0.32102095444861334, "conf": 0.9046028808224946, "grad_norm": 4.862573146820068, "lr": 4.538263844648149e-05 }, { "step": 349, "T": 640, "positions": 16, "loss": 0.6020571701446897, "ce": 2.4288127054733195, "tv": 0.3990843364226748, "conf": 0.705237272195518, "grad_norm": 2.660292863845825, "lr": 4.5350080300570654e-05 }, { "step": 350, "T": 588, "positions": 16, "loss": 0.1634163839444227, "ce": 0.5050191024824642, "tv": 0.1254605301655829, "conf": 0.4090901098679751, "grad_norm": 1.7891916036605835, "lr": 4.531741953507313e-05 }, { "step": 351, "T": 533, "positions": 16, "loss": 0.2588139997351391, "ce": 1.047226076871084, "tv": 0.17121267277252628, "conf": 0.5196249494329095, "grad_norm": 1.6339894533157349, "lr": 4.528465631468773e-05 }, { "step": 352, "T": 675, "positions": 16, "loss": 0.8425568893871969, "ce": 3.5189491654309677, "tv": 0.545179986191215, "conf": 0.8240285830106586, "grad_norm": 4.200933456420898, "lr": 4.5251790804629934e-05 }, { "step": 353, "T": 532, "positions": 16, "loss": 0.41217605083511444, "ce": 1.381764477599745, "tv": 0.30444400515261805, "conf": 0.4742719135247171, "grad_norm": 1.3983758687973022, "lr": 4.521882317063103e-05 }, { "step": 354, "T": 586, "positions": 16, "loss": 0.6054227156564593, "ce": 2.689210577867925, "tv": 0.37389073660597205, "conf": 0.822700135409832, "grad_norm": 4.617712497711182, "lr": 4.5185753578937284e-05 }, { "step": 355, "T": 1048, "positions": 16, "loss": 0.7372009490500204, "ce": 3.5355156361183617, "tv": 0.42627710819942877, "conf": 0.5061712396563962, "grad_norm": 7.218335151672363, "lr": 4.5152582196309124e-05 }, { "step": 356, "T": 615, "positions": 16, "loss": 0.4213254479582247, "ce": 1.7891499598190421, "tv": 0.26934494508896023, "conf": 0.7399491654255144, "grad_norm": 3.4156460762023926, "lr": 4.511930919002025e-05 }, { "step": 357, "T": 672, "positions": 16, "loss": 0.552542434692441, "ce": 2.2243715868899017, "tv": 0.3667836403619731, "conf": 0.7783042208757252, "grad_norm": 4.741540908813477, "lr": 4.5085934727856856e-05 }, { "step": 358, "T": 579, "positions": 16, "loss": 0.6709207489620894, "ce": 2.718590820208192, "tv": 0.44340187637135386, "conf": 0.5114804483018816, "grad_norm": 2.4100096225738525, "lr": 4.505245897811672e-05 }, { "step": 359, "T": 588, "positions": 16, "loss": 0.6295196870341897, "ce": 2.2617241637781262, "tv": 0.4481636332347989, "conf": 0.8639764632680453, "grad_norm": 4.886844635009766, "lr": 4.501888210960839e-05 }, { "step": 360, "T": 527, "positions": 16, "loss": 0.1929479336777149, "ce": 0.7479000014791382, "tv": 0.13128660079564725, "conf": 0.6127133775153197, "grad_norm": 2.57783842086792, "lr": 4.498520429165034e-05 }, { "step": 361, "T": 1060, "positions": 16, "loss": 0.7558027759077959, "ce": 3.2679522956022993, "tv": 0.47667504695709795, "conf": 0.7891203672170377, "grad_norm": 3.233368158340454, "lr": 4.495142569407009e-05 }, { "step": 362, "T": 629, "positions": 16, "loss": 0.45993475268187467, "ce": 1.7550829329702538, "tv": 0.3160294019980938, "conf": 0.8828379712067544, "grad_norm": 2.8686885833740234, "lr": 4.4917546487203374e-05 }, { "step": 363, "T": 529, "positions": 16, "loss": 0.48770500237128545, "ce": 1.954331998565408, "tv": 0.3247464511132421, "conf": 0.648280165140477, "grad_norm": 3.116297960281372, "lr": 4.488356684189325e-05 }, { "step": 364, "T": 629, "positions": 16, "loss": 0.2391649451747071, "ce": 0.7810530707726571, "tv": 0.1789551516303618, "conf": 0.591187309473753, "grad_norm": 3.5542566776275635, "lr": 4.484948692948929e-05 }, { "step": 365, "T": 621, "positions": 16, "loss": 0.2291323455865495, "ce": 0.8894129205308445, "tv": 0.15576782741845818, "conf": 0.6652174163609743, "grad_norm": 2.5089335441589355, "lr": 4.4815306921846645e-05 }, { "step": 366, "T": 591, "positions": 16, "loss": 0.34376101616362575, "ce": 1.265005824079708, "tv": 0.24140048825574922, "conf": 0.6885951671138173, "grad_norm": 2.9846031665802, "lr": 4.478102699132525e-05 }, { "step": 367, "T": 604, "positions": 16, "loss": 0.417104973741516, "ce": 1.4628101178095676, "tv": 0.30091552488011075, "conf": 0.4349661034902965, "grad_norm": 3.6006810665130615, "lr": 4.474664731078889e-05 }, { "step": 368, "T": 673, "positions": 16, "loss": 0.25139511700035655, "ce": 0.968038217525077, "tv": 0.1717681069385435, "conf": 1.4123861864209175, "grad_norm": 3.3488569259643555, "lr": 4.4712168053604407e-05 }, { "step": 369, "T": 531, "positions": 16, "loss": 0.551806601928547, "ce": 2.1283371747704223, "tv": 0.3766365386545658, "conf": 0.6315698204562068, "grad_norm": 3.638645648956299, "lr": 4.467758939364074e-05 }, { "step": 370, "T": 539, "positions": 16, "loss": 0.17241994066171173, "ce": 0.4650670749719481, "tv": 0.1399035944032221, "conf": 0.33761042915284634, "grad_norm": 1.8391385078430176, "lr": 4.4642911505268105e-05 }, { "step": 371, "T": 676, "positions": 16, "loss": 0.6002270462486194, "ce": 2.297150188969681, "tv": 0.41168003203347325, "conf": 1.153099762974307, "grad_norm": 4.4955034255981445, "lr": 4.4608134563357084e-05 }, { "step": 372, "T": 966, "positions": 16, "loss": 0.8992827215297439, "ce": 4.140525314312669, "tv": 0.5391446455250843, "conf": 0.7672231729375198, "grad_norm": 5.390698432922363, "lr": 4.457325874327777e-05 }, { "step": 373, "T": 638, "positions": 16, "loss": 0.3198968949727714, "ce": 1.019450338324532, "tv": 0.2421687371097505, "conf": 0.9962526056915522, "grad_norm": 5.095595836639404, "lr": 4.4538284220898864e-05 }, { "step": 374, "T": 539, "positions": 16, "loss": 0.6926459951791912, "ce": 2.7233390090987086, "tv": 0.46701345569454134, "conf": 0.7123482320457697, "grad_norm": 6.168480396270752, "lr": 4.45032111725868e-05 }, { "step": 375, "T": 543, "positions": 16, "loss": 0.3211594057138427, "ce": 0.8599532672888017, "tv": 0.2612934275821317, "conf": 1.063128530044196, "grad_norm": 2.4309747219085693, "lr": 4.446803977520485e-05 }, { "step": 376, "T": 544, "positions": 16, "loss": 0.24390341049729614, "ce": 0.6236399942645221, "tv": 0.20171046319592278, "conf": 0.9082147315202747, "grad_norm": 3.898834228515625, "lr": 4.443277020611223e-05 }, { "step": 377, "T": 532, "positions": 16, "loss": 0.28017248182732146, "ce": 1.1777089911192888, "tv": 0.18044621174340136, "conf": 0.606370821595192, "grad_norm": 3.370074510574341, "lr": 4.43974026431632e-05 }, { "step": 378, "T": 536, "positions": 16, "loss": 0.23204798429242146, "ce": 0.8194418649397051, "tv": 0.16678199635884994, "conf": 0.5265259853726718, "grad_norm": 3.4872782230377197, "lr": 4.4361937264706186e-05 }, { "step": 379, "T": 553, "positions": 16, "loss": 0.4499231944864732, "ce": 1.8043439071298053, "tv": 0.29943201162677724, "conf": 0.625483192212414, "grad_norm": 4.425645351409912, "lr": 4.432637424958286e-05 }, { "step": 380, "T": 651, "positions": 16, "loss": 0.5602259745355695, "ce": 1.9523792445397703, "tv": 0.4055423028767109, "conf": 1.1206933464854956, "grad_norm": 2.8192858695983887, "lr": 4.4290713777127245e-05 }, { "step": 381, "T": 573, "positions": 16, "loss": 0.9958097822964191, "ce": 4.440370179712772, "tv": 0.613080870360136, "conf": 0.8764403043314815, "grad_norm": 4.591569423675537, "lr": 4.425495602716484e-05 }, { "step": 382, "T": 549, "positions": 16, "loss": 0.43359361910370353, "ce": 1.6256665925284324, "tv": 0.30114108315137855, "conf": 0.8928083923528902, "grad_norm": 5.822729587554932, "lr": 4.421910118001166e-05 }, { "step": 383, "T": 573, "positions": 16, "loss": 0.5025574041646905, "ce": 1.7784333463932853, "tv": 0.36079341810545884, "conf": 0.9930097403102991, "grad_norm": 7.299745559692383, "lr": 4.418314941647335e-05 }, { "step": 384, "T": 534, "positions": 16, "loss": 0.7439260128885508, "ce": 2.510740876197815, "tv": 0.5476132556796074, "conf": 1.31852174596861, "grad_norm": 2.936187505722046, "lr": 4.414710091784429e-05 }, { "step": 385, "T": 548, "positions": 16, "loss": 0.5441444252210204, "ce": 2.199539267212458, "tv": 0.36021165252896026, "conf": 0.7173873006249778, "grad_norm": 5.117702007293701, "lr": 4.4110955865906666e-05 }, { "step": 386, "T": 588, "positions": 16, "loss": 0.4881706075684633, "ce": 2.189820671657799, "tv": 0.29909836905426346, "conf": 1.0874808770604432, "grad_norm": 2.6838245391845703, "lr": 4.407471444292955e-05 }, { "step": 387, "T": 654, "positions": 16, "loss": 0.5520997741259634, "ce": 1.8263098332099617, "tv": 0.4105208874680102, "conf": 1.2067177211865783, "grad_norm": 4.093173027038574, "lr": 4.4038376831667996e-05 }, { "step": 388, "T": 536, "positions": 16, "loss": 0.44435718515887856, "ce": 1.5146249137469567, "tv": 0.32543854269897565, "conf": 1.2057310154195875, "grad_norm": 3.6596298217773438, "lr": 4.400194321536209e-05 }, { "step": 389, "T": 959, "positions": 16, "loss": 0.9262878689914942, "ce": 4.488907856866717, "tv": 0.5304412106052041, "conf": 1.5183086190372705, "grad_norm": 3.9937899112701416, "lr": 4.396541377773605e-05 }, { "step": 390, "T": 532, "positions": 16, "loss": 0.29849517352886323, "ce": 1.2374525147833992, "tv": 0.19416658323916636, "conf": 0.9393502525053918, "grad_norm": 5.579324722290039, "lr": 4.392878870299731e-05 }, { "step": 391, "T": 654, "positions": 16, "loss": 0.43969554803334177, "ce": 1.4540045361500233, "tv": 0.3269945590873249, "conf": 1.245684466790408, "grad_norm": 5.08801794052124, "lr": 4.389206817583554e-05 }, { "step": 392, "T": 627, "positions": 16, "loss": 0.29222246372592053, "ce": 1.189396153728012, "tv": 0.1925364978815196, "conf": 1.1353049147874117, "grad_norm": 3.7527682781219482, "lr": 4.38552523814218e-05 }, { "step": 393, "T": 531, "positions": 16, "loss": 0.4278893841765239, "ce": 1.2844970479345648, "tv": 0.3327107625154895, "conf": 1.1638473164539391, "grad_norm": 3.4215078353881836, "lr": 4.381834150540749e-05 }, { "step": 394, "T": 614, "positions": 16, "loss": 0.2726068284828216, "ce": 0.6702794475713745, "tv": 0.22842098830733448, "conf": 1.179402125068009, "grad_norm": 13.310689926147461, "lr": 4.378133573392353e-05 }, { "step": 395, "T": 530, "positions": 16, "loss": 0.7423881363938563, "ce": 3.0384739543515025, "tv": 0.48726751397771295, "conf": 0.7582802823599195, "grad_norm": 3.498652696609497, "lr": 4.374423525357936e-05 }, { "step": 396, "T": 545, "positions": 16, "loss": 0.38938652418437414, "ce": 1.4651797823316883, "tv": 0.2698539467091905, "conf": 0.45650500716874376, "grad_norm": 3.5395171642303467, "lr": 4.3707040251462e-05 }, { "step": 397, "T": 605, "positions": 16, "loss": 0.47091044252738357, "ce": 1.8541254810988903, "tv": 0.31721989484503865, "conf": 0.991220043040812, "grad_norm": 6.506549835205078, "lr": 4.366975091513512e-05 }, { "step": 398, "T": 529, "positions": 16, "loss": 0.6599584426730871, "ce": 2.623850111849606, "tv": 0.4417482651770115, "conf": 1.2904656920582056, "grad_norm": 3.795670509338379, "lr": 4.363236743263808e-05 }, { "step": 399, "T": 1075, "positions": 16, "loss": 0.6466223142488161, "ce": 3.1591346379073, "tv": 0.3674542599255801, "conf": 0.6186397177043546, "grad_norm": 3.2007153034210205, "lr": 4.359488999248501e-05 }, { "step": 400, "T": 1443, "positions": 16, "loss": 0.968896908685565, "ce": 4.856854289770126, "tv": 0.5369016714394093, "conf": 1.0059907969553024, "grad_norm": 6.043054103851318, "lr": 4.3557318783663844e-05 }, { "step": 401, "T": 535, "positions": 16, "loss": 0.8580101989209652, "ce": 3.3474774789065123, "tv": 0.5814027208834887, "conf": 0.8196326099568978, "grad_norm": 3.7057249546051025, "lr": 4.351965399563534e-05 }, { "step": 402, "T": 594, "positions": 16, "loss": 0.19905243604443967, "ce": 0.5293411530437879, "tv": 0.162353686770075, "conf": 0.5690491748973727, "grad_norm": 1.5184065103530884, "lr": 4.348189581833216e-05 }, { "step": 403, "T": 886, "positions": 16, "loss": 0.6808817996989092, "ce": 3.435571156485821, "tv": 0.3748052126975381, "conf": 0.5238039776595542, "grad_norm": 3.165422201156616, "lr": 4.3444044442157914e-05 }, { "step": 404, "T": 532, "positions": 16, "loss": 0.6426543610868976, "ce": 2.4090546818915755, "tv": 0.4463876783847809, "conf": 0.5755262262537144, "grad_norm": 1.6941373348236084, "lr": 4.340610005798618e-05 }, { "step": 405, "T": 560, "positions": 16, "loss": 0.4414871698827483, "ce": 1.8462680057273246, "tv": 0.2854004190885462, "conf": 0.9125512098617037, "grad_norm": 3.0678861141204834, "lr": 4.336806285715953e-05 }, { "step": 406, "T": 597, "positions": 16, "loss": 0.5896294040139765, "ce": 2.3901360774761997, "tv": 0.3895730990334414, "conf": 0.519777943205554, "grad_norm": 2.2990357875823975, "lr": 4.332993303148861e-05 }, { "step": 407, "T": 536, "positions": 16, "loss": 0.11201476940004795, "ce": 0.16604683995137748, "tv": 0.10601121104991762, "conf": 0.30270312656648457, "grad_norm": 1.0333747863769531, "lr": 4.329171077325114e-05 }, { "step": 408, "T": 578, "positions": 16, "loss": 0.25592378398869187, "ce": 0.6440179572673514, "tv": 0.21280221722554415, "conf": 0.77231515198946, "grad_norm": 5.162101745605469, "lr": 4.3253396275190926e-05 }, { "step": 409, "T": 675, "positions": 16, "loss": 0.32586782650105306, "ce": 1.147061347568524, "tv": 0.23462410364300013, "conf": 0.48668967284106657, "grad_norm": 2.291299343109131, "lr": 4.321498973051696e-05 }, { "step": 410, "T": 672, "positions": 16, "loss": 0.8283227793872356, "ce": 3.1930847242474556, "tv": 0.5655714571475983, "conf": 0.7950242040678859, "grad_norm": 3.8840203285217285, "lr": 4.317649133290236e-05 }, { "step": 411, "T": 529, "positions": 16, "loss": 0.3139810775392107, "ce": 0.9737260848414735, "tv": 0.2406760740050231, "conf": 0.742911379551515, "grad_norm": 5.0654425621032715, "lr": 4.3137901276483435e-05 }, { "step": 412, "T": 676, "positions": 16, "loss": 0.3550593893742189, "ce": 0.9327352216932923, "tv": 0.2908732006326318, "conf": 1.13193090574714, "grad_norm": 3.794126510620117, "lr": 4.309921975585872e-05 }, { "step": 413, "T": 555, "positions": 16, "loss": 0.25664951180260687, "ce": 0.9788606782306601, "tv": 0.17640383595244202, "conf": 0.6066129854880273, "grad_norm": 2.5444037914276123, "lr": 4.306044696608797e-05 }, { "step": 414, "T": 1244, "positions": 16, "loss": 0.7202913969522342, "ce": 3.579940566094592, "tv": 0.4025525980978273, "conf": 0.5681998338550329, "grad_norm": 3.753582239151001, "lr": 4.302158310269118e-05 }, { "step": 415, "T": 549, "positions": 16, "loss": 0.6617921991273761, "ce": 2.478111904449179, "tv": 0.4599789031781256, "conf": 0.4686707593500614, "grad_norm": 4.644536018371582, "lr": 4.29826283616476e-05 }, { "step": 416, "T": 599, "positions": 16, "loss": 0.457701486942824, "ce": 1.685134521118016, "tv": 0.3213200561440317, "conf": 1.0268672641832381, "grad_norm": 4.390504837036133, "lr": 4.294358293939478e-05 }, { "step": 417, "T": 526, "positions": 16, "loss": 0.4416733207181096, "ce": 1.6667446761857718, "tv": 0.3055542786605656, "conf": 0.5544481351971626, "grad_norm": 10.94820499420166, "lr": 4.29044470328275e-05 }, { "step": 418, "T": 576, "positions": 16, "loss": 0.7588833218778746, "ce": 3.2009030820227053, "tv": 0.48754779970113304, "conf": 0.6579858298573527, "grad_norm": 4.633713245391846, "lr": 4.286522083929686e-05 }, { "step": 419, "T": 542, "positions": 16, "loss": 0.4271612635238853, "ce": 1.6036712079421704, "tv": 0.2964379420627665, "conf": 0.7839939540135674, "grad_norm": 2.6188032627105713, "lr": 4.282590455660923e-05 }, { "step": 420, "T": 537, "positions": 16, "loss": 0.08989621075397736, "ce": 0.22828069581100863, "tv": 0.07452015874696372, "conf": 0.508821977232472, "grad_norm": 2.5088250637054443, "lr": 4.278649838302531e-05 }, { "step": 421, "T": 536, "positions": 16, "loss": 0.20871949633237818, "ce": 0.7258514169319028, "tv": 0.1512603932105776, "conf": 0.45140956596696924, "grad_norm": 2.6276001930236816, "lr": 4.274700251725906e-05 }, { "step": 422, "T": 543, "positions": 16, "loss": 0.187883491627872, "ce": 0.7370840659291389, "tv": 0.12686120904618292, "conf": 0.25510436011245474, "grad_norm": 1.4434820413589478, "lr": 4.270741715847674e-05 }, { "step": 423, "T": 573, "positions": 16, "loss": 0.6383527832222171, "ce": 2.639202591322828, "tv": 0.4160361367248697, "conf": 0.6559495857800357, "grad_norm": 3.4097092151641846, "lr": 4.266774250629589e-05 }, { "step": 424, "T": 928, "positions": 16, "loss": 0.6965844426304102, "ce": 3.152250819141045, "tv": 0.4237326253205538, "conf": 1.1570998698487074, "grad_norm": 3.292340040206909, "lr": 4.262797876078436e-05 }, { "step": 425, "T": 979, "positions": 16, "loss": 0.8432176906717359, "ce": 3.7166888545980328, "tv": 0.5239431334703113, "conf": 0.7778025220977725, "grad_norm": 3.0258235931396484, "lr": 4.258812612245924e-05 }, { "step": 426, "T": 542, "positions": 16, "loss": 0.1763203401787905, "ce": 0.6295857553977839, "tv": 0.12595751571643632, "conf": 0.31585328324581496, "grad_norm": 2.1655142307281494, "lr": 4.2548184792285884e-05 }, { "step": 427, "T": 525, "positions": 16, "loss": 0.3623341570546472, "ce": 1.302210122856195, "tv": 0.2579034828913791, "conf": 0.5445599602845057, "grad_norm": 5.579949378967285, "lr": 4.250815497167692e-05 }, { "step": 428, "T": 1286, "positions": 16, "loss": 0.6358826923788001, "ce": 2.8525882120925417, "tv": 0.38958208946860395, "conf": 0.8533748730551451, "grad_norm": 4.617562294006348, "lr": 4.2468036862491176e-05 }, { "step": 429, "T": 538, "positions": 16, "loss": 0.16346406868979102, "ce": 0.35587261581910923, "tv": 0.14208534393947048, "conf": 0.24846284103114158, "grad_norm": 3.0901219844818115, "lr": 4.242783066703273e-05 }, { "step": 430, "T": 541, "positions": 16, "loss": 0.4815760525580117, "ce": 2.1482329460809524, "tv": 0.2963919631367844, "conf": 0.7290548917273156, "grad_norm": 5.49978494644165, "lr": 4.2387536588049803e-05 }, { "step": 431, "T": 537, "positions": 16, "loss": 0.5282610398717225, "ce": 2.109796927179559, "tv": 0.3525348308030516, "conf": 1.0509113271546084, "grad_norm": 5.841348171234131, "lr": 4.234715482873384e-05 }, { "step": 432, "T": 550, "positions": 16, "loss": 0.33925368093605357, "ce": 1.0056187037823747, "tv": 0.26521311832675565, "conf": 0.8719936567358673, "grad_norm": 2.966637372970581, "lr": 4.2306685592718384e-05 }, { "step": 433, "T": 568, "positions": 16, "loss": 0.5652028038166463, "ce": 2.687153129838407, "tv": 0.3294305484741926, "conf": 1.4156349133700132, "grad_norm": 4.302952289581299, "lr": 4.226612908407814e-05 }, { "step": 434, "T": 524, "positions": 16, "loss": 0.7600802553351969, "ce": 2.995047205244191, "tv": 0.5117506014648825, "conf": 0.8943054479314014, "grad_norm": 3.337815523147583, "lr": 4.222548550732789e-05 }, { "step": 435, "T": 601, "positions": 16, "loss": 0.17780371469416423, "ce": 0.6089025910878263, "tv": 0.12990384338991134, "conf": 0.7078094546450302, "grad_norm": 2.581139326095581, "lr": 4.2184755067421464e-05 }, { "step": 436, "T": 530, "positions": 16, "loss": 0.361715229897527, "ce": 1.2146159415933653, "tv": 0.26694849076739047, "conf": 1.2304644760718588, "grad_norm": 4.341346740722656, "lr": 4.214393796975072e-05 }, { "step": 437, "T": 525, "positions": 16, "loss": 0.49676004200591706, "ce": 1.6672054345544893, "tv": 0.36671055541955866, "conf": 1.2617848870625785, "grad_norm": 3.1166088581085205, "lr": 4.2103034420144516e-05 }, { "step": 438, "T": 526, "positions": 16, "loss": 0.7253845273517072, "ce": 2.7878999112872407, "tv": 0.49621614813804626, "conf": 1.0003498572841636, "grad_norm": 5.534845352172852, "lr": 4.2062044624867656e-05 }, { "step": 439, "T": 584, "positions": 16, "loss": 0.20353428338557933, "ce": 0.7211295447224302, "tv": 0.1460236982902643, "conf": 0.7043813057243824, "grad_norm": 2.063307285308838, "lr": 4.2020968790619864e-05 }, { "step": 440, "T": 1681, "positions": 16, "loss": 0.8534512482583523, "ce": 4.152767177904025, "tv": 0.48686060844920576, "conf": 1.5741524119948735, "grad_norm": 3.6705167293548584, "lr": 4.197980712453471e-05 }, { "step": 441, "T": 532, "positions": 16, "loss": 0.475164769894036, "ce": 1.934458666138653, "tv": 0.3130209919763729, "conf": 0.7778518091258473, "grad_norm": 2.699068069458008, "lr": 4.193855983417863e-05 }, { "step": 442, "T": 535, "positions": 16, "loss": 0.3797866488894215, "ce": 1.6934242045799692, "tv": 0.233826921434229, "conf": 0.5433185483561829, "grad_norm": 1.9014556407928467, "lr": 4.1897227127549795e-05 }, { "step": 443, "T": 1305, "positions": 16, "loss": 0.9487467141389061, "ce": 4.500011494037608, "tv": 0.554161737094546, "conf": 0.7866365513764322, "grad_norm": 3.6409904956817627, "lr": 4.1855809213077146e-05 }, { "step": 444, "T": 545, "positions": 16, "loss": 0.4065401627376559, "ce": 1.3344071513784002, "tv": 0.3034438291870174, "conf": 0.5126832712430769, "grad_norm": 5.71279764175415, "lr": 4.1814306299619264e-05 }, { "step": 445, "T": 537, "positions": 16, "loss": 0.37143885673140176, "ce": 1.4709763993901106, "tv": 0.2492680202012707, "conf": 0.6054470336239319, "grad_norm": 3.5730350017547607, "lr": 4.177271859646338e-05 }, { "step": 446, "T": 963, "positions": 16, "loss": 0.6837269189127255, "ce": 3.35939559644612, "tv": 0.3864304056041874, "conf": 0.8894662984445745, "grad_norm": 5.939113616943359, "lr": 4.1731046313324294e-05 }, { "step": 447, "T": 539, "positions": 16, "loss": 0.31882536282910223, "ce": 1.1569058908755778, "tv": 0.22570531913697778, "conf": 0.9365373311585026, "grad_norm": 3.7471346855163574, "lr": 4.16892896603433e-05 }, { "step": 448, "T": 653, "positions": 16, "loss": 0.8597839591093361, "ce": 3.4054219375830144, "tv": 0.5769353262148798, "conf": 1.7079144862946123, "grad_norm": 4.606351375579834, "lr": 4.1647448848087166e-05 }, { "step": 449, "T": 528, "positions": 16, "loss": 0.32041580746590625, "ce": 1.111999748611197, "tv": 0.232462049665628, "conf": 0.567906394120655, "grad_norm": 2.912684440612793, "lr": 4.160552408754704e-05 }, { "step": 450, "T": 599, "positions": 16, "loss": 0.4692076159408316, "ce": 1.6485269550830708, "tv": 0.33817213063593954, "conf": 0.8262546912301332, "grad_norm": 4.203156471252441, "lr": 4.156351559013741e-05 }, { "step": 451, "T": 543, "positions": 16, "loss": 0.2924108193765278, "ce": 1.07253291469533, "tv": 0.20573058917761955, "conf": 0.5895088439087885, "grad_norm": 2.3934147357940674, "lr": 4.152142356769501e-05 }, { "step": 452, "T": 532, "positions": 16, "loss": 0.3862512632990729, "ce": 1.6370477030245638, "tv": 0.24727389669442346, "conf": 0.45521766628181126, "grad_norm": 3.7216694355010986, "lr": 4.147924823247775e-05 }, { "step": 453, "T": 1026, "positions": 16, "loss": 0.8545632921159267, "ce": 3.5564143834635615, "tv": 0.5543576355557889, "conf": 1.2351759681478143, "grad_norm": 4.754981994628906, "lr": 4.143698979716372e-05 }, { "step": 454, "T": 577, "positions": 16, "loss": 0.4628985530580394, "ce": 1.53246782644419, "tv": 0.34405752771999687, "conf": 0.9171295212581754, "grad_norm": 3.6353487968444824, "lr": 4.1394648474849994e-05 }, { "step": 455, "T": 1744, "positions": 16, "loss": 0.9397547049447894, "ce": 4.437799476087093, "tv": 0.5510830767452717, "conf": 1.2509547318331897, "grad_norm": 4.845308780670166, "lr": 4.135222447905166e-05 }, { "step": 456, "T": 643, "positions": 16, "loss": 0.20808768611459527, "ce": 0.6894799086039711, "tv": 0.15459966122580227, "conf": 0.785554425790906, "grad_norm": 2.7296526432037354, "lr": 4.1309718023700705e-05 }, { "step": 457, "T": 678, "positions": 16, "loss": 0.5955701700004283, "ce": 2.175147928064689, "tv": 0.4200615407899022, "conf": 1.369980527088046, "grad_norm": 3.3094892501831055, "lr": 4.1267129323144915e-05 }, { "step": 458, "T": 534, "positions": 16, "loss": 0.6208721839939244, "ce": 2.344364881137153, "tv": 0.4293729990313295, "conf": 2.105837784134991, "grad_norm": 4.044519424438477, "lr": 4.122445859214682e-05 }, { "step": 459, "T": 561, "positions": 16, "loss": 0.2215881876036292, "ce": 0.7564203908841591, "tv": 0.1621624032559339, "conf": 0.7609655184205621, "grad_norm": 4.430034160614014, "lr": 4.118170604588263e-05 }, { "step": 460, "T": 529, "positions": 16, "loss": 0.5555070209811674, "ce": 1.7570153544575078, "tv": 0.4220061042433372, "conf": 0.7697581209940836, "grad_norm": 2.7735788822174072, "lr": 4.113887189994109e-05 }, { "step": 461, "T": 618, "positions": 16, "loss": 0.13672387293627253, "ce": 0.4493315384286234, "tv": 0.10198968883923953, "conf": 0.3844920738483779, "grad_norm": 2.9154772758483887, "lr": 4.1095956370322466e-05 }, { "step": 462, "T": 592, "positions": 16, "loss": 0.5603979653678834, "ce": 2.124174455180764, "tv": 0.38664502231404185, "conf": 1.45039464533329, "grad_norm": 3.9222803115844727, "lr": 4.1052959673437376e-05 }, { "step": 463, "T": 538, "positions": 16, "loss": 0.30308075418020053, "ce": 1.0324610985504705, "tv": 0.22203849674633602, "conf": 0.7155870053444091, "grad_norm": 2.8128082752227783, "lr": 4.100988202610577e-05 }, { "step": 464, "T": 524, "positions": 16, "loss": 0.47401147238269914, "ce": 1.6074461315001827, "tv": 0.34807428147178143, "conf": 1.0191487858819528, "grad_norm": 4.28604793548584, "lr": 4.096672364555581e-05 }, { "step": 465, "T": 639, "positions": 16, "loss": 0.4889939865861379, "ce": 2.330583803348418, "tv": 0.2843729018532031, "conf": 0.9346166867053398, "grad_norm": 3.1909584999084473, "lr": 4.092348474942276e-05 }, { "step": 466, "T": 574, "positions": 16, "loss": 0.2105959305081342, "ce": 0.740388973721565, "tv": 0.15173003780364525, "conf": 0.7995476327923825, "grad_norm": 4.62415885925293, "lr": 4.0880165555747905e-05 }, { "step": 467, "T": 595, "positions": 16, "loss": 0.1833464373194147, "ce": 0.6668609719636152, "tv": 0.129622599502909, "conf": 0.6804467647452839, "grad_norm": 3.0667669773101807, "lr": 4.083676628297743e-05 }, { "step": 468, "T": 529, "positions": 16, "loss": 0.5699731020140462, "ce": 2.538698471150383, "tv": 0.3512258463597391, "conf": 1.0720434918618302, "grad_norm": 2.0656216144561768, "lr": 4.079328714996139e-05 }, { "step": 469, "T": 530, "positions": 16, "loss": 0.5535675254650414, "ce": 2.4183687325567007, "tv": 0.34636739175766706, "conf": 0.9031699174665846, "grad_norm": 7.7702484130859375, "lr": 4.074972837595248e-05 }, { "step": 470, "T": 526, "positions": 16, "loss": 0.5976037150248885, "ce": 2.010410981194582, "tv": 0.4406251395121217, "conf": 0.7933855624869466, "grad_norm": 2.7919063568115234, "lr": 4.070609018060506e-05 }, { "step": 471, "T": 541, "positions": 16, "loss": 0.5507284454070032, "ce": 2.7997997847560328, "tv": 0.3008316416890011, "conf": 0.6957531347870827, "grad_norm": 6.962092399597168, "lr": 4.066237278397395e-05 }, { "step": 472, "T": 672, "positions": 16, "loss": 0.5586273533554049, "ce": 2.2263559900284235, "tv": 0.3733242006128421, "conf": 0.7555261451052502, "grad_norm": 2.856224536895752, "lr": 4.061857640651337e-05 }, { "step": 473, "T": 1079, "positions": 16, "loss": 0.7363189375901129, "ce": 3.5802502400038065, "tv": 0.4203265944379382, "conf": 1.6174782666909664, "grad_norm": 6.486457824707031, "lr": 4.0574701269075844e-05 }, { "step": 474, "T": 529, "positions": 16, "loss": 0.53211415322221, "ce": 1.8748383600031957, "tv": 0.3829225842273445, "conf": 1.2132202151629485, "grad_norm": 2.3660888671875, "lr": 4.0530747592911025e-05 }, { "step": 475, "T": 544, "positions": 16, "loss": 0.437389372382313, "ce": 1.4660615149186924, "tv": 0.32309248368255794, "conf": 0.9407602575083729, "grad_norm": 3.690601348876953, "lr": 4.048671559966461e-05 }, { "step": 476, "T": 1567, "positions": 16, "loss": 0.7105953076970764, "ce": 3.3221590925713826, "tv": 0.4204215517747798, "conf": 1.2505092285573483, "grad_norm": 1.8313339948654175, "lr": 4.0442605511377275e-05 }, { "step": 477, "T": 533, "positions": 16, "loss": 0.6400810080740484, "ce": 2.452403751689417, "tv": 0.43871181652502855, "conf": 0.8393734588607913, "grad_norm": 3.463865280151367, "lr": 4.039841755048346e-05 }, { "step": 478, "T": 1190, "positions": 16, "loss": 0.8167053377255797, "ce": 3.8782362644560635, "tv": 0.4765352443791926, "conf": 1.326409920133301, "grad_norm": 2.476976156234741, "lr": 4.035415193981032e-05 }, { "step": 479, "T": 541, "positions": 16, "loss": 0.12429541765231988, "ce": 0.42305441567827984, "tv": 0.09109997435371042, "conf": 0.1809106914442964, "grad_norm": 1.444351315498352, "lr": 4.030980890257655e-05 }, { "step": 480, "T": 1679, "positions": 16, "loss": 0.8173649384625605, "ce": 3.8631187755745486, "tv": 0.478947858340689, "conf": 2.0423047223544017, "grad_norm": 2.257000207901001, "lr": 4.02653886623913e-05 }, { "step": 481, "T": 526, "positions": 16, "loss": 0.6123483697883785, "ce": 2.0519018075428903, "tv": 0.4523980114609003, "conf": 1.7572160452546086, "grad_norm": 2.782115936279297, "lr": 4.022089144325305e-05 }, { "step": 482, "T": 636, "positions": 16, "loss": 0.5638597896322608, "ce": 1.8298166556633078, "tv": 0.42319791531190276, "conf": 1.476464280858636, "grad_norm": 4.938227653503418, "lr": 4.017631746954842e-05 }, { "step": 483, "T": 545, "positions": 16, "loss": 0.4726575181230146, "ce": 1.8366388620584075, "tv": 0.32110404260311043, "conf": 0.2798377714352682, "grad_norm": 2.7349181175231934, "lr": 4.0131666966051127e-05 }, { "step": 484, "T": 531, "positions": 16, "loss": 0.41198187916597817, "ce": 1.539648911331824, "tv": 0.2866855459069484, "conf": 0.42563849105499685, "grad_norm": 1.7981173992156982, "lr": 4.008694015792077e-05 }, { "step": 485, "T": 528, "positions": 16, "loss": 0.2798035592859378, "ce": 0.8669408090545403, "tv": 0.21456609438246232, "conf": 0.7012849333696067, "grad_norm": 3.3917858600616455, "lr": 4.004213727070175e-05 }, { "step": 486, "T": 605, "positions": 16, "loss": 0.6283008260461429, "ce": 2.761338851117216, "tv": 0.3912965970548612, "conf": 0.771619763695071, "grad_norm": 2.4981517791748047, "lr": 3.999725853032212e-05 }, { "step": 487, "T": 527, "positions": 16, "loss": 0.28986589518899564, "ce": 1.2789028296363085, "tv": 0.17997290565108415, "conf": 0.58893231296679, "grad_norm": 1.7623192071914673, "lr": 3.9952304163092404e-05 }, { "step": 488, "T": 545, "positions": 16, "loss": 0.18372840742813423, "ce": 0.7147783048230849, "tv": 0.1247228556931077, "conf": 0.18338260966993403, "grad_norm": 2.222935676574707, "lr": 3.990727439570453e-05 }, { "step": 489, "T": 595, "positions": 16, "loss": 0.5637779235839844, "ce": 2.0042358487844467, "tv": 0.40372706204652786, "conf": 0.7753319207113236, "grad_norm": 4.259436130523682, "lr": 3.9862169455230615e-05 }, { "step": 490, "T": 525, "positions": 16, "loss": 0.5628839246928692, "ce": 1.8363366955891252, "tv": 0.4213891834951937, "conf": 0.5731406207196414, "grad_norm": 5.147116184234619, "lr": 3.9816989569121876e-05 }, { "step": 491, "T": 675, "positions": 16, "loss": 0.5692507512430893, "ce": 1.8539386229822412, "tv": 0.4265076664160006, "conf": 0.8999658103712136, "grad_norm": 2.9356532096862793, "lr": 3.977173496520745e-05 }, { "step": 492, "T": 593, "positions": 16, "loss": 0.21801811554541928, "ce": 0.8466933968002195, "tv": 0.1481653132232168, "conf": 0.6018946853437228, "grad_norm": 3.1884541511535645, "lr": 3.972640587169326e-05 }, { "step": 493, "T": 984, "positions": 16, "loss": 0.7116073226752633, "ce": 3.481374873682398, "tv": 0.40385535732366407, "conf": 0.9850641370285302, "grad_norm": 2.6023147106170654, "lr": 3.9681002517160845e-05 }, { "step": 494, "T": 672, "positions": 16, "loss": 0.4288440481759608, "ce": 1.4494784129783511, "tv": 0.3154402426443994, "conf": 1.054736401885748, "grad_norm": 2.2739827632904053, "lr": 3.963552513056623e-05 }, { "step": 495, "T": 675, "positions": 16, "loss": 0.45276981333154254, "ce": 1.6590849198400974, "tv": 0.31873480370268226, "conf": 1.2103640120476484, "grad_norm": 2.3275296688079834, "lr": 3.9589973941238764e-05 }, { "step": 496, "T": 1257, "positions": 16, "loss": 0.8863665212338674, "ce": 4.371078882927122, "tv": 0.4991762654426566, "conf": 0.9310872452915646, "grad_norm": 3.8559844493865967, "lr": 3.9544349178879956e-05 }, { "step": 497, "T": 673, "positions": 16, "loss": 0.7585364058613777, "ce": 2.8281679041683674, "tv": 0.5285773724317551, "conf": 1.0851855454966426, "grad_norm": 2.5794289112091064, "lr": 3.949865107356233e-05 }, { "step": 498, "T": 535, "positions": 16, "loss": 0.336673711018193, "ce": 1.1422557384259449, "tv": 0.24716461734487893, "conf": 0.5895215298514813, "grad_norm": 1.929549217224121, "lr": 3.945287985572826e-05 }, { "step": 499, "T": 608, "positions": 16, "loss": 0.5290519430418499, "ce": 1.780687251884956, "tv": 0.3899813728639856, "conf": 1.4225861806044122, "grad_norm": 2.871415376663208, "lr": 3.940703575618878e-05 }, { "step": 500, "T": 524, "positions": 16, "loss": 0.90361263230443, "ce": 3.4353799000382423, "tv": 0.6223051827400923, "conf": 1.01601492613554, "grad_norm": 3.0423083305358887, "lr": 3.936111900612248e-05 }, { "step": 501, "T": 629, "positions": 16, "loss": 0.5166093583393376, "ce": 1.6865021824378346, "tv": 0.3866212803259259, "conf": 1.1907040925671026, "grad_norm": 2.919882297515869, "lr": 3.9315129837074284e-05 }, { "step": 502, "T": 533, "positions": 16, "loss": 0.5096400530455867, "ce": 2.174691696431182, "tv": 0.3246343166902079, "conf": 1.0141022498345365, "grad_norm": 2.7139577865600586, "lr": 3.9269068480954305e-05 }, { "step": 503, "T": 524, "positions": 16, "loss": 0.7785983411595225, "ce": 2.8339329808950424, "tv": 0.5502278376370668, "conf": 1.4624414131976664, "grad_norm": 4.539449214935303, "lr": 3.922293517003668e-05 }, { "step": 504, "T": 672, "positions": 16, "loss": 0.7125893346965313, "ce": 2.6058581210672855, "tv": 0.5022261496633291, "conf": 0.9510242082178593, "grad_norm": 4.161253452301025, "lr": 3.9176730136958406e-05 }, { "step": 505, "T": 597, "positions": 16, "loss": 0.5321146679380035, "ce": 2.0119833043445396, "tv": 0.36768482299339666, "conf": 1.5338313474785537, "grad_norm": 4.169112682342529, "lr": 3.9130453614718135e-05 }, { "step": 506, "T": 605, "positions": 16, "loss": 0.5509048670064658, "ce": 1.6653405463148374, "tv": 0.4270786854904145, "conf": 1.7054396802559495, "grad_norm": 2.2907044887542725, "lr": 3.908410583667502e-05 }, { "step": 507, "T": 672, "positions": 16, "loss": 0.8332128562033176, "ce": 3.4043498933315277, "tv": 0.5475309677422047, "conf": 1.4518091088393703, "grad_norm": 2.320293426513672, "lr": 3.9037687036547546e-05 }, { "step": 508, "T": 524, "positions": 16, "loss": 0.7334348706062883, "ce": 2.711006969155278, "tv": 0.5137046470772475, "conf": 1.3986618872731924, "grad_norm": 3.992449998855591, "lr": 3.899119744841232e-05 }, { "step": 509, "T": 533, "positions": 16, "loss": 0.2021362678060541, "ce": 0.6534696778933267, "tv": 0.15198810865513224, "conf": 0.6222992922557751, "grad_norm": 1.7685768604278564, "lr": 3.894463730670294e-05 }, { "step": 510, "T": 609, "positions": 16, "loss": 0.49844934677821584, "ce": 1.5486857004580088, "tv": 0.3817564348864835, "conf": 1.0631997154795272, "grad_norm": 2.055544853210449, "lr": 3.8898006846208787e-05 }, { "step": 511, "T": 622, "positions": 16, "loss": 0.572724197874777, "ce": 2.2024884009733796, "tv": 0.39163927535992116, "conf": 1.0054002804681659, "grad_norm": 3.7139503955841064, "lr": 3.885130630207381e-05 }, { "step": 512, "T": 932, "positions": 16, "loss": 0.6309837497974513, "ce": 2.4379872324861935, "tv": 0.43020558010903187, "conf": 1.045343423174927, "grad_norm": 4.694517612457275, "lr": 3.880453590979539e-05 }, { "step": 513, "T": 529, "positions": 16, "loss": 0.6912927273660898, "ce": 3.0143140892032534, "tv": 0.43317923974245787, "conf": 1.0664649668615311, "grad_norm": 4.161386966705322, "lr": 3.875769590522314e-05 }, { "step": 514, "T": 677, "positions": 16, "loss": 0.7261417988775065, "ce": 2.6392324730295513, "tv": 0.5135761691926746, "conf": 1.3865478964188611, "grad_norm": 2.23649525642395, "lr": 3.871078652455767e-05 }, { "step": 515, "T": 1210, "positions": 16, "loss": 0.9218827112199506, "ce": 4.823457345253701, "tv": 0.4883744480102905, "conf": 1.1545112191124645, "grad_norm": 3.8024160861968994, "lr": 3.8663808004349476e-05 }, { "step": 516, "T": 533, "positions": 16, "loss": 0.25289082192466594, "ce": 1.0332241246142075, "tv": 0.16618713010757347, "conf": 0.520260705749024, "grad_norm": 2.5741982460021973, "lr": 3.86167605814977e-05 }, { "step": 517, "T": 618, "positions": 16, "loss": 0.45023293736630876, "ce": 1.8681062643081532, "tv": 0.2926914452809797, "conf": 1.3384490525621033, "grad_norm": 2.693167209625244, "lr": 3.8569644493248906e-05 }, { "step": 518, "T": 530, "positions": 16, "loss": 0.484087341450504, "ce": 2.006309907315881, "tv": 0.31495150475529954, "conf": 0.724558437243104, "grad_norm": 3.271125555038452, "lr": 3.8522459977195955e-05 }, { "step": 519, "T": 527, "positions": 16, "loss": 0.6036062925122678, "ce": 1.9787404919043183, "tv": 0.4508136110380292, "conf": 1.2773049275856465, "grad_norm": 2.5386745929718018, "lr": 3.847520727127676e-05 }, { "step": 520, "T": 924, "positions": 16, "loss": 0.6321503048966406, "ce": 2.5363613638910465, "tv": 0.42057131335604936, "conf": 1.7066328360554053, "grad_norm": 3.1963791847229004, "lr": 3.842788661377307e-05 }, { "step": 521, "T": 550, "positions": 16, "loss": 0.7654615174978971, "ce": 2.892948182299733, "tv": 0.5290741175413132, "conf": 1.4808127784635872, "grad_norm": 2.704075336456299, "lr": 3.8380498243309324e-05 }, { "step": 522, "T": 653, "positions": 16, "loss": 0.8200059905648232, "ce": 3.4039979175431654, "tv": 0.5328957687597722, "conf": 2.2635880757297855, "grad_norm": 2.7562999725341797, "lr": 3.8333042398851425e-05 }, { "step": 523, "T": 659, "positions": 16, "loss": 0.6227905055275187, "ce": 2.5752490356826456, "tv": 0.4058506808360107, "conf": 1.1981220151560308, "grad_norm": 2.405254602432251, "lr": 3.828551931970549e-05 }, { "step": 524, "T": 535, "positions": 16, "loss": 0.31473839587852126, "ce": 1.1542230807044689, "tv": 0.22146231995611743, "conf": 0.6673446063796291, "grad_norm": 4.0513410568237305, "lr": 3.8237929245516716e-05 }, { "step": 525, "T": 675, "positions": 16, "loss": 0.7248437469825149, "ce": 2.546910246834159, "tv": 0.5223919302225113, "conf": 1.3934242591785733, "grad_norm": 2.4589133262634277, "lr": 3.819027241626812e-05 }, { "step": 526, "T": 624, "positions": 16, "loss": 0.3352933368455524, "ce": 1.181903597209839, "tv": 0.2412255322146848, "conf": 1.8994650522928338, "grad_norm": 3.816176414489746, "lr": 3.8142549072279344e-05 }, { "step": 527, "T": 634, "positions": 16, "loss": 0.6853872379288077, "ce": 2.656821839045733, "tv": 0.46633896324783564, "conf": 0.9463656132284086, "grad_norm": 3.7458081245422363, "lr": 3.8094759454205454e-05 }, { "step": 528, "T": 547, "positions": 16, "loss": 0.5545602974016219, "ce": 2.0664823782863095, "tv": 0.3865689686499536, "conf": 1.6465589624131098, "grad_norm": 3.3899247646331787, "lr": 3.8046903803035716e-05 }, { "step": 529, "T": 536, "positions": 16, "loss": 0.5391522442805581, "ce": 2.072879797790847, "tv": 0.3687380673636653, "conf": 1.5785066056996584, "grad_norm": 3.607912540435791, "lr": 3.7998982360092385e-05 }, { "step": 530, "T": 672, "positions": 16, "loss": 0.2110787734418409, "ce": 0.6406904334342016, "tv": 0.1633441574376775, "conf": 1.182565196446376, "grad_norm": 2.4300906658172607, "lr": 3.795099536702946e-05 }, { "step": 531, "T": 673, "positions": 16, "loss": 0.47117950638016737, "ce": 1.77491536036041, "tv": 0.3263199742279994, "conf": 1.7281057998692404, "grad_norm": 4.268533229827881, "lr": 3.7902943065831526e-05 }, { "step": 532, "T": 574, "positions": 16, "loss": 0.7024294399889186, "ce": 2.90304112713784, "tv": 0.4579170191427693, "conf": 1.1076505275268573, "grad_norm": 3.109557867050171, "lr": 3.785482569881248e-05 }, { "step": 533, "T": 673, "positions": 16, "loss": 0.6767096421681345, "ce": 2.263716134475544, "tv": 0.5003756296355277, "conf": 1.462448171339929, "grad_norm": 4.502072811126709, "lr": 3.780664350861431e-05 }, { "step": 534, "T": 1093, "positions": 16, "loss": 0.6896915069955867, "ce": 2.941085108119296, "tv": 0.4395366434910102, "conf": 2.221339038180929, "grad_norm": 2.4268510341644287, "lr": 3.7758396738205934e-05 }, { "step": 535, "T": 572, "positions": 16, "loss": 0.8095294684171677, "ce": 3.3459115149453282, "tv": 0.5277092345058918, "conf": 1.7246233797632158, "grad_norm": 2.633190155029297, "lr": 3.7710085630881885e-05 }, { "step": 536, "T": 607, "positions": 16, "loss": 0.592136666737133, "ce": 2.41151832198193, "tv": 0.38998315644130344, "conf": 1.2872523416058357, "grad_norm": 6.994320869445801, "lr": 3.766171043026115e-05 }, { "step": 537, "T": 604, "positions": 16, "loss": 0.3323526113817934, "ce": 0.9924992336600553, "tv": 0.2590029881976079, "conf": 1.5962466602629148, "grad_norm": 4.1762471199035645, "lr": 3.761327138028592e-05 }, { "step": 538, "T": 674, "positions": 16, "loss": 0.5344141144851164, "ce": 2.3970285355944725, "tv": 0.3274569585482823, "conf": 1.1033105094088569, "grad_norm": 4.326852321624756, "lr": 3.756476872522035e-05 }, { "step": 539, "T": 529, "positions": 16, "loss": 0.5810174737125635, "ce": 2.0832218469586223, "tv": 0.4141058870591223, "conf": 1.7837482964387164, "grad_norm": 4.522335052490234, "lr": 3.7516202709649344e-05 }, { "step": 540, "T": 674, "positions": 16, "loss": 0.260636109049301, "ce": 0.8320165349332456, "tv": 0.19714938691686257, "conf": 1.6250659891411487, "grad_norm": 2.1561601161956787, "lr": 3.7467573578477316e-05 }, { "step": 541, "T": 983, "positions": 16, "loss": 0.6894665204745252, "ce": 3.384971723691706, "tv": 0.3899659268281539, "conf": 0.8699205391940268, "grad_norm": 2.393235445022583, "lr": 3.741888157692695e-05 }, { "step": 542, "T": 531, "positions": 16, "loss": 0.24615203528264828, "ce": 0.8255549560537929, "tv": 0.18177393276619114, "conf": 0.9892255652225685, "grad_norm": 3.092308282852173, "lr": 3.7370126950537976e-05 }, { "step": 543, "T": 643, "positions": 16, "loss": 0.7595396423712373, "ce": 2.741069497540593, "tv": 0.5393696622923017, "conf": 1.879662720952183, "grad_norm": 4.603034973144531, "lr": 3.7321309945165905e-05 }, { "step": 544, "T": 672, "positions": 16, "loss": 0.46624154204619117, "ce": 1.8077380440022353, "tv": 0.31718638332677074, "conf": 0.6960561852283718, "grad_norm": 3.221903085708618, "lr": 3.727243080698082e-05 }, { "step": 545, "T": 577, "positions": 16, "loss": 0.37748325325810583, "ce": 1.2800869495185907, "tv": 0.2771939598915196, "conf": 1.4221624190104194, "grad_norm": 6.85723876953125, "lr": 3.722348978246612e-05 }, { "step": 546, "T": 540, "positions": 16, "loss": 0.33637127275869716, "ce": 1.0679688699151484, "tv": 0.2550826535207307, "conf": 1.2442905617558608, "grad_norm": 2.500579595565796, "lr": 3.717448711841728e-05 }, { "step": 547, "T": 1793, "positions": 16, "loss": 0.8535061205475358, "ce": 3.9176232082172646, "tv": 0.5130486722118803, "conf": 1.2296580346735482, "grad_norm": 2.692924976348877, "lr": 3.712542306194061e-05 }, { "step": 548, "T": 1349, "positions": 16, "loss": 0.8615431624348275, "ce": 4.440741703481763, "tv": 0.4638544376794016, "conf": 0.7349462215206586, "grad_norm": 2.688136339187622, "lr": 3.707629786045198e-05 }, { "step": 549, "T": 658, "positions": 16, "loss": 0.6988633247092366, "ce": 2.5692641973728314, "tv": 0.4910410069860518, "conf": 0.9865106819197536, "grad_norm": 2.9473490715026855, "lr": 3.702711176167565e-05 }, { "step": 550, "T": 537, "positions": 16, "loss": 0.7631960902363062, "ce": 3.5367341227829456, "tv": 0.4550251953769475, "conf": 0.7786561336833984, "grad_norm": 3.3095946311950684, "lr": 3.69778650136429e-05 }, { "step": 551, "T": 532, "positions": 16, "loss": 0.36803844913265493, "ce": 1.0436075849429471, "tv": 0.2929752261898102, "conf": 0.719122588634491, "grad_norm": 2.55497670173645, "lr": 3.692855786469089e-05 }, { "step": 552, "T": 591, "positions": 16, "loss": 0.37529759097014903, "ce": 1.6495090062687723, "tv": 0.2337185531760042, "conf": 0.4320652268233971, "grad_norm": 2.868647336959839, "lr": 3.687919056346135e-05 }, { "step": 553, "T": 525, "positions": 16, "loss": 0.6903715403750539, "ce": 2.7803282011300325, "tv": 0.4581541335210204, "conf": 0.6931343355681747, "grad_norm": 3.044922113418579, "lr": 3.682976335889935e-05 }, { "step": 554, "T": 526, "positions": 16, "loss": 0.34369561805942794, "ce": 1.0758008179946046, "tv": 0.2623506174350041, "conf": 0.846702201360813, "grad_norm": 3.906418800354004, "lr": 3.678027650025203e-05 }, { "step": 555, "T": 588, "positions": 16, "loss": 0.19894642064900836, "ce": 0.6709183426210075, "tv": 0.14650510160936392, "conf": 0.5418282569153234, "grad_norm": 4.254439830780029, "lr": 3.673073023706734e-05 }, { "step": 556, "T": 605, "positions": 16, "loss": 0.2811883489484899, "ce": 1.374859940059423, "tv": 0.15966927977569867, "conf": 0.554889388004085, "grad_norm": 2.186521053314209, "lr": 3.668112481919281e-05 }, { "step": 557, "T": 532, "positions": 16, "loss": 0.3366166250416427, "ce": 1.2662933820465696, "tv": 0.2333192068927019, "conf": 0.7553227988114486, "grad_norm": 3.204119920730591, "lr": 3.663146049677425e-05 }, { "step": 558, "T": 980, "positions": 16, "loss": 0.7634087488986552, "ce": 3.136641639750451, "tv": 0.49971621471922845, "conf": 1.1763637093827128, "grad_norm": 2.3075027465820312, "lr": 3.658173752025452e-05 }, { "step": 559, "T": 533, "positions": 16, "loss": 0.8361196098849177, "ce": 3.3255309755913913, "tv": 0.5595183512195945, "conf": 0.8286199478898197, "grad_norm": 4.60019588470459, "lr": 3.6531956140372246e-05 }, { "step": 560, "T": 563, "positions": 16, "loss": 0.32695854292978765, "ce": 1.274051922754552, "tv": 0.22172594746007235, "conf": 0.5853984877949188, "grad_norm": 3.371432304382324, "lr": 3.6482116608160574e-05 }, { "step": 561, "T": 532, "positions": 16, "loss": 0.35934623098842167, "ce": 1.3566635710581636, "tv": 0.24853320068405083, "conf": 0.8664201916557204, "grad_norm": 2.051811695098877, "lr": 3.643221917494589e-05 }, { "step": 562, "T": 584, "positions": 16, "loss": 0.5924227052310016, "ce": 2.099468551692553, "tv": 0.4249731717573013, "conf": 1.229385893095241, "grad_norm": 2.780425786972046, "lr": 3.6382264092346575e-05 }, { "step": 563, "T": 534, "positions": 16, "loss": 0.4954830217156996, "ce": 2.3197340377839737, "tv": 0.2927884484633978, "conf": 0.6950172796841798, "grad_norm": 2.882883310317993, "lr": 3.633225161227169e-05 }, { "step": 564, "T": 528, "positions": 16, "loss": 0.5932754464447498, "ce": 2.0699965915991925, "tv": 0.42919531650841236, "conf": 0.9962387718260288, "grad_norm": 4.564088821411133, "lr": 3.628218198691976e-05 }, { "step": 565, "T": 539, "positions": 16, "loss": 0.29384273895266233, "ce": 1.1462165645334608, "tv": 0.19913454268680653, "conf": 0.5510860760696232, "grad_norm": 3.0707590579986572, "lr": 3.6232055468777454e-05 }, { "step": 566, "T": 650, "positions": 16, "loss": 0.20934560352543485, "ce": 0.9257446301024856, "tv": 0.12974571121958434, "conf": 0.8957840110811048, "grad_norm": 2.415900230407715, "lr": 3.6181872310618356e-05 }, { "step": 567, "T": 550, "positions": 16, "loss": 0.4339560230146162, "ce": 1.3532757217180915, "tv": 0.3318093931302428, "conf": 1.1481489114034389, "grad_norm": 5.8821120262146, "lr": 3.6131632765501675e-05 }, { "step": 568, "T": 676, "positions": 16, "loss": 0.5365568651759531, "ce": 1.9952635906120122, "tv": 0.3744783416186692, "conf": 1.6820660954335835, "grad_norm": 5.596973896026611, "lr": 3.608133708677093e-05 }, { "step": 569, "T": 676, "positions": 16, "loss": 0.5891556342985496, "ce": 2.346479374732553, "tv": 0.39389743905667274, "conf": 1.1480713632757649, "grad_norm": 2.8962838649749756, "lr": 3.603098552805273e-05 }, { "step": 570, "T": 530, "positions": 16, "loss": 0.7501089088327717, "ce": 3.0544081193220336, "tv": 0.49407567299203947, "conf": 0.643125787639292, "grad_norm": 4.372773170471191, "lr": 3.598057834325547e-05 }, { "step": 571, "T": 576, "positions": 16, "loss": 0.7916725436225533, "ce": 3.0753197157755494, "tv": 0.5379339619539678, "conf": 1.3979585426859558, "grad_norm": 3.665693521499634, "lr": 3.5930115786568054e-05 }, { "step": 572, "T": 598, "positions": 16, "loss": 0.43665308774870937, "ce": 1.5759314372244262, "tv": 0.31006660194907454, "conf": 1.455142108070504, "grad_norm": 6.721442222595215, "lr": 3.587959811245858e-05 }, { "step": 573, "T": 531, "positions": 16, "loss": 0.31434713820635807, "ce": 1.0034282970082131, "tv": 0.23778255955403438, "conf": 1.1075182562935737, "grad_norm": 2.9624128341674805, "lr": 3.5829025575673136e-05 }, { "step": 574, "T": 1060, "positions": 16, "loss": 0.7695070444133307, "ce": 3.5193873334828822, "tv": 0.46396479540271685, "conf": 1.2157741845929566, "grad_norm": 4.372456073760986, "lr": 3.577839843123443e-05 }, { "step": 575, "T": 563, "positions": 16, "loss": 0.5085562351159751, "ce": 1.8251506655942649, "tv": 0.36226796731352806, "conf": 1.1620052105281502, "grad_norm": 3.093451976776123, "lr": 3.572771693444055e-05 }, { "step": 576, "T": 644, "positions": 16, "loss": 0.15353467724344227, "ce": 0.4956525314228202, "tv": 0.11552158385893563, "conf": 0.5237493082913716, "grad_norm": 1.1833610534667969, "lr": 3.567698134086369e-05 }, { "step": 577, "T": 621, "positions": 16, "loss": 0.24268282546654518, "ce": 0.8946016096899712, "tv": 0.17024740609667788, "conf": 0.7903931636901689, "grad_norm": 3.7853052616119385, "lr": 3.5626191906348784e-05 }, { "step": 578, "T": 641, "positions": 16, "loss": 0.39608817453336087, "ce": 1.5420027886684693, "tv": 0.2687643351382576, "conf": 1.6520532705691744, "grad_norm": 2.4557838439941406, "lr": 3.5575348887012336e-05 }, { "step": 579, "T": 528, "positions": 16, "loss": 0.20156429032067535, "ce": 0.9805975285553359, "tv": 0.11500503591742017, "conf": 0.5669674789096462, "grad_norm": 1.559299349784851, "lr": 3.5524452539241005e-05 }, { "step": 580, "T": 602, "positions": 16, "loss": 0.562306473730132, "ce": 1.9866386990761384, "tv": 0.40404733188916, "conf": 1.0562766958028078, "grad_norm": 2.088501214981079, "lr": 3.547350311969041e-05 }, { "step": 581, "T": 538, "positions": 16, "loss": 0.6098838088500997, "ce": 2.8191345386344437, "tv": 0.3644115204433547, "conf": 1.028270565031562, "grad_norm": 3.8840255737304688, "lr": 3.542250088528379e-05 }, { "step": 582, "T": 538, "positions": 16, "loss": 0.10483795995787659, "ce": 0.29746489389617636, "tv": 0.08343496632915048, "conf": 0.346305129681431, "grad_norm": 1.6079970598220825, "lr": 3.537144609321069e-05 }, { "step": 583, "T": 542, "positions": 16, "loss": 0.23779990470939083, "ce": 0.5847732114016253, "tv": 0.19924732157232938, "conf": 0.7903582731655674, "grad_norm": 2.3004508018493652, "lr": 3.532033900092571e-05 }, { "step": 584, "T": 673, "positions": 16, "loss": 0.6885790172964334, "ce": 2.665592335164547, "tv": 0.46891085989773273, "conf": 1.0975793949328363, "grad_norm": 3.014089584350586, "lr": 3.526917986614717e-05 }, { "step": 585, "T": 597, "positions": 16, "loss": 0.566976907037315, "ce": 2.450773358375045, "tv": 0.3576662191771902, "conf": 1.1240494225036741, "grad_norm": 3.8572254180908203, "lr": 3.521796894685585e-05 }, { "step": 586, "T": 537, "positions": 16, "loss": 0.28403102409265557, "ce": 0.8649972885400246, "tv": 0.21947921409264382, "conf": 0.4001472261734307, "grad_norm": 2.6891186237335205, "lr": 3.516670650129365e-05 }, { "step": 587, "T": 533, "positions": 16, "loss": 0.12083485560197005, "ce": 0.3769201289882176, "tv": 0.09238093495332578, "conf": 0.38530541164800525, "grad_norm": 1.6776014566421509, "lr": 3.5115392787962276e-05 }, { "step": 588, "T": 674, "positions": 16, "loss": 0.5744172317427001, "ce": 2.2198839215316184, "tv": 0.3915876064202166, "conf": 1.0196078694855544, "grad_norm": 4.538568496704102, "lr": 3.506402806562202e-05 }, { "step": 589, "T": 550, "positions": 16, "loss": 0.10375642948565655, "ce": 0.3039356763051728, "tv": 0.0815142974279297, "conf": 0.38072051543713314, "grad_norm": 1.0243403911590576, "lr": 3.501261259329033e-05 }, { "step": 590, "T": 621, "positions": 16, "loss": 0.359252280301007, "ce": 1.3486784633323623, "tv": 0.2493160478916252, "conf": 0.7624004173430876, "grad_norm": 1.6221200227737427, "lr": 3.496114663024064e-05 }, { "step": 591, "T": 535, "positions": 16, "loss": 0.5977723633986898, "ce": 2.0091141953653278, "tv": 0.44095662218751386, "conf": 1.1075154359641601, "grad_norm": 2.0732614994049072, "lr": 3.490963043600094e-05 }, { "step": 592, "T": 526, "positions": 16, "loss": 0.49083806210546754, "ce": 2.37372966753901, "tv": 0.2816279093094636, "conf": 0.709177227843611, "grad_norm": 4.228431701660156, "lr": 3.485806427035254e-05 }, { "step": 593, "T": 987, "positions": 16, "loss": 0.6820310105540557, "ce": 2.8862077755038626, "tv": 0.4371225005015731, "conf": 0.9074767749170629, "grad_norm": 2.197835683822632, "lr": 3.480644839332876e-05 }, { "step": 594, "T": 673, "positions": 16, "loss": 0.773970996029675, "ce": 2.7109286561608315, "tv": 0.5587535053491592, "conf": 1.2260730881243944, "grad_norm": 3.3468422889709473, "lr": 3.475478306521357e-05 }, { "step": 595, "T": 532, "positions": 16, "loss": 0.5493076068814844, "ce": 2.467528586508706, "tv": 0.3361719446256757, "conf": 0.9249374251812696, "grad_norm": 3.571885108947754, "lr": 3.4703068546540334e-05 }, { "step": 596, "T": 675, "positions": 16, "loss": 0.6474580560716277, "ce": 2.712609284028076, "tv": 0.4179968367352558, "conf": 0.86598648450083, "grad_norm": 3.898268222808838, "lr": 3.465130509809044e-05 }, { "step": 597, "T": 553, "positions": 16, "loss": 0.505962769452708, "ce": 2.1052870704797897, "tv": 0.32826008453048416, "conf": 0.9258513496056366, "grad_norm": 4.576888084411621, "lr": 3.459949298089205e-05 }, { "step": 598, "T": 673, "positions": 16, "loss": 0.7073677331209183, "ce": 3.2739312276244164, "tv": 0.4221940105780959, "conf": 0.822009252384305, "grad_norm": 5.071561336517334, "lr": 3.454763245621871e-05 }, { "step": 599, "T": 526, "positions": 16, "loss": 0.412776278988531, "ce": 1.2897759096013033, "tv": 0.3153318930671958, "conf": 0.5118245624471456, "grad_norm": 2.7460296154022217, "lr": 3.4495723785588105e-05 }, { "step": 600, "T": 533, "positions": 16, "loss": 0.38792697369353846, "ce": 1.4803988189232768, "tv": 0.2665412209898932, "conf": 0.4173363304726081, "grad_norm": 4.68614387512207, "lr": 3.4443767230760696e-05 }, { "step": 601, "T": 1795, "positions": 16, "loss": 0.8566929277149029, "ce": 4.02000116428826, "tv": 0.5052142446511425, "conf": 1.3266303543932736, "grad_norm": 2.0003724098205566, "lr": 3.439176305373839e-05 }, { "step": 602, "T": 536, "positions": 16, "loss": 0.27524737905241636, "ce": 0.8523154585256663, "tv": 0.21112870742763334, "conf": 0.6608431311324239, "grad_norm": 2.594478130340576, "lr": 3.4339711516763264e-05 }, { "step": 603, "T": 676, "positions": 16, "loss": 0.6788499988615513, "ce": 2.629826529417187, "tv": 0.4620748106390238, "conf": 1.1281474207062274, "grad_norm": 5.118382930755615, "lr": 3.428761288231621e-05 }, { "step": 604, "T": 586, "positions": 16, "loss": 0.2606299018370919, "ce": 0.9453194772358984, "tv": 0.18455328610434663, "conf": 0.8805744543496985, "grad_norm": 2.7586307525634766, "lr": 3.4235467413115625e-05 }, { "step": 605, "T": 553, "positions": 16, "loss": 0.5498072881018743, "ce": 1.9983725095614773, "tv": 0.38885560270864516, "conf": 0.7030996359244455, "grad_norm": 4.482380390167236, "lr": 3.418327537211607e-05 }, { "step": 606, "T": 614, "positions": 16, "loss": 0.23238532878713158, "ce": 0.8120390594549463, "tv": 0.1679793665134639, "conf": 0.5122825046645971, "grad_norm": 3.3901207447052, "lr": 3.413103702250694e-05 }, { "step": 607, "T": 530, "positions": 16, "loss": 0.34371553168602986, "ce": 1.2097706378772273, "tv": 0.2474871915037511, "conf": 1.5053856736049056, "grad_norm": 2.2469170093536377, "lr": 3.40787526277112e-05 }, { "step": 608, "T": 597, "positions": 16, "loss": 0.14116175773415307, "ce": 0.36383164686822056, "tv": 0.11642066128842998, "conf": 0.5374547138198977, "grad_norm": 1.0233654975891113, "lr": 3.402642245138394e-05 }, { "step": 609, "T": 628, "positions": 16, "loss": 0.38516959827393293, "ce": 1.3623793921069591, "tv": 0.27659073285758495, "conf": 1.2299749222001992, "grad_norm": 3.4166650772094727, "lr": 3.397404675741118e-05 }, { "step": 610, "T": 672, "positions": 16, "loss": 0.8585684923455119, "ce": 3.4209293657913804, "tv": 0.5738617135211825, "conf": 0.9605602455558255, "grad_norm": 3.1485912799835205, "lr": 3.392162580990843e-05 }, { "step": 611, "T": 672, "positions": 16, "loss": 0.8607845338992774, "ce": 3.516547412611544, "tv": 0.5656997570767999, "conf": 1.1719408640637994, "grad_norm": 2.2771413326263428, "lr": 3.386915987321939e-05 }, { "step": 612, "T": 593, "positions": 16, "loss": 0.1607552920577291, "ce": 0.4951807225270386, "tv": 0.12359690539960866, "conf": 0.6455038924468681, "grad_norm": 1.0190575122833252, "lr": 3.381664921191469e-05 }, { "step": 613, "T": 576, "positions": 16, "loss": 0.6391251041786745, "ce": 2.42568298173137, "tv": 0.44061870174482465, "conf": 1.4300135094672441, "grad_norm": 3.193694829940796, "lr": 3.376409409079043e-05 }, { "step": 614, "T": 669, "positions": 16, "loss": 0.22245931671477592, "ce": 0.8017533903393996, "tv": 0.15809332325852665, "conf": 1.4099463828124499, "grad_norm": 3.3448843955993652, "lr": 3.371149477486693e-05 }, { "step": 615, "T": 530, "positions": 16, "loss": 0.8200707100331783, "ce": 2.9669084306806326, "tv": 0.5815332029014826, "conf": 1.6837122291326523, "grad_norm": 2.560114622116089, "lr": 3.365885152938737e-05 }, { "step": 616, "T": 678, "positions": 16, "loss": 0.47488204060027783, "ce": 1.635575219648672, "tv": 0.3459161537793989, "conf": 1.4847416291146551, "grad_norm": 3.344613790512085, "lr": 3.360616461981646e-05 }, { "step": 617, "T": 675, "positions": 16, "loss": 0.4742772045538004, "ce": 1.7984989438227785, "tv": 0.32714145960926544, "conf": 0.7105518141761422, "grad_norm": 2.7758495807647705, "lr": 3.3553434311839106e-05 }, { "step": 618, "T": 537, "positions": 16, "loss": 0.3127311341868335, "ce": 1.3203491975964425, "tv": 0.20077358209891827, "conf": 0.43867138901487124, "grad_norm": 2.6631968021392822, "lr": 3.350066087135903e-05 }, { "step": 619, "T": 577, "positions": 16, "loss": 0.42780884823878296, "ce": 1.4333900736019132, "tv": 0.31607761023769854, "conf": 0.9436406055094722, "grad_norm": 2.954592704772949, "lr": 3.344784456449747e-05 }, { "step": 620, "T": 613, "positions": 16, "loss": 0.26021330219919037, "ce": 0.8802318416180697, "tv": 0.19132235019606014, "conf": 0.5014492005422646, "grad_norm": 1.7599867582321167, "lr": 3.339498565759185e-05 }, { "step": 621, "T": 548, "positions": 16, "loss": 0.2949054786677152, "ce": 1.1140351822052708, "tv": 0.2038910703467991, "conf": 0.5663217178080231, "grad_norm": 2.896548271179199, "lr": 3.334208441719437e-05 }, { "step": 622, "T": 526, "positions": 16, "loss": 0.7825856804847717, "ce": 2.7141840942204, "tv": 0.5679636336863041, "conf": 1.565648963674903, "grad_norm": 3.1159565448760986, "lr": 3.328914111007076e-05 }, { "step": 623, "T": 607, "positions": 16, "loss": 0.1462225062532525, "ce": 0.6022876013576024, "tv": 0.09554860274874954, "conf": 0.5429458995258756, "grad_norm": 3.0984270572662354, "lr": 3.323615600319883e-05 }, { "step": 624, "T": 674, "positions": 16, "loss": 0.3914017522006361, "ce": 1.674946607475249, "tv": 0.24878566226925614, "conf": 0.9777911197733573, "grad_norm": 4.140708923339844, "lr": 3.318312936376719e-05 }, { "step": 625, "T": 1001, "positions": 16, "loss": 0.7203141748905182, "ce": 2.9726686510257423, "tv": 0.47005257476121187, "conf": 2.206781371620707, "grad_norm": 2.5004756450653076, "lr": 3.313006145917391e-05 }, { "step": 626, "T": 537, "positions": 16, "loss": 0.5189621737372363, "ce": 2.365336256750652, "tv": 0.3138094974710839, "conf": 0.4453213004162535, "grad_norm": 2.080355405807495, "lr": 3.3076952557025106e-05 }, { "step": 627, "T": 534, "positions": 16, "loss": 0.10058554697025102, "ce": 0.22575820176592742, "tv": 0.08667748192601721, "conf": 0.1571932030137475, "grad_norm": 1.301570177078247, "lr": 3.3023802925133694e-05 }, { "step": 628, "T": 542, "positions": 16, "loss": 0.36808297042080085, "ce": 1.0821194504387677, "tv": 0.28874557793096756, "conf": 0.9327555967030321, "grad_norm": 5.315221309661865, "lr": 3.297061283151791e-05 }, { "step": 629, "T": 676, "positions": 16, "loss": 0.8159864050103351, "ce": 2.817587149913834, "tv": 0.5935863170743687, "conf": 1.3384181794244796, "grad_norm": 4.254496097564697, "lr": 3.2917382544400064e-05 }, { "step": 630, "T": 536, "positions": 16, "loss": 0.28528100388302846, "ce": 1.161710866442263, "tv": 0.18789989646893446, "conf": 0.6917783216341153, "grad_norm": 2.7595813274383545, "lr": 3.286411233220515e-05 }, { "step": 631, "T": 524, "positions": 16, "loss": 0.7026743707829155, "ce": 2.8547864633001154, "tv": 0.46355080348439515, "conf": 0.998630068206694, "grad_norm": 2.774428129196167, "lr": 3.281080246355948e-05 }, { "step": 632, "T": 563, "positions": 16, "loss": 0.3855027499150765, "ce": 1.5338076315342732, "tv": 0.25791330642846333, "conf": 0.8670176778751624, "grad_norm": 3.0963003635406494, "lr": 3.275745320728936e-05 }, { "step": 633, "T": 617, "positions": 16, "loss": 0.386307531820421, "ce": 1.3425731977763462, "tv": 0.2800557954078613, "conf": 1.1812021380842568, "grad_norm": 3.523597240447998, "lr": 3.27040648324197e-05 }, { "step": 634, "T": 537, "positions": 16, "loss": 0.3450294630274584, "ce": 1.0008413548057433, "tv": 0.2721614750134904, "conf": 1.2475613572119713, "grad_norm": 3.455080270767212, "lr": 3.265063760817266e-05 }, { "step": 635, "T": 621, "positions": 16, "loss": 0.33827644848497584, "ce": 1.0095797670037427, "tv": 0.2636871970607899, "conf": 1.429903801992623, "grad_norm": 2.6983277797698975, "lr": 3.259717180396636e-05 }, { "step": 636, "T": 672, "positions": 16, "loss": 0.6142075140960515, "ce": 2.2619186025112867, "tv": 0.431128503754735, "conf": 1.590780094731599, "grad_norm": 2.210740566253662, "lr": 3.2543667689413405e-05 }, { "step": 637, "T": 631, "positions": 16, "loss": 0.2135465998617292, "ce": 0.5382515970832173, "tv": 0.17746826909933588, "conf": 1.1141272954833852, "grad_norm": 2.7533042430877686, "lr": 3.2490125534319636e-05 }, { "step": 638, "T": 528, "positions": 16, "loss": 0.3695811211364344, "ce": 1.5237050446266949, "tv": 0.24134513374883682, "conf": 1.0804046930311526, "grad_norm": 2.5195841789245605, "lr": 3.243654560868268e-05 }, { "step": 639, "T": 1043, "positions": 16, "loss": 0.7933405956136994, "ce": 3.531021287664771, "tv": 0.4891538647352718, "conf": 1.591679220829974, "grad_norm": 3.053316354751587, "lr": 3.2382928182690676e-05 }, { "step": 640, "T": 584, "positions": 16, "loss": 0.2971799120950891, "ce": 1.2173407825224558, "tv": 0.1949398129754627, "conf": 1.030886685475707, "grad_norm": 2.3021280765533447, "lr": 3.232927352672081e-05 }, { "step": 641, "T": 656, "positions": 16, "loss": 0.5790341752581298, "ce": 2.06122755818069, "tv": 0.4143460369668901, "conf": 1.45321844316868, "grad_norm": 2.829719066619873, "lr": 3.2275581911338084e-05 }, { "step": 642, "T": 673, "positions": 16, "loss": 0.6272576321407541, "ce": 2.37664869197215, "tv": 0.43288085578933533, "conf": 1.0386969687036007, "grad_norm": 2.7216315269470215, "lr": 3.222185360729379e-05 }, { "step": 643, "T": 532, "positions": 16, "loss": 0.2394908254082111, "ce": 1.0234642424995855, "tv": 0.15238268145913025, "conf": 0.723021156154573, "grad_norm": 1.8515230417251587, "lr": 3.216808888552429e-05 }, { "step": 644, "T": 629, "positions": 16, "loss": 0.278761496883817, "ce": 1.1430456554010107, "tv": 0.18272992372567387, "conf": 0.9102238682098687, "grad_norm": 2.046396493911743, "lr": 3.211428801714958e-05 }, { "step": 645, "T": 546, "positions": 16, "loss": 0.6100918759620981, "ce": 2.349326935283898, "tv": 0.4168435385727207, "conf": 0.5317139857490929, "grad_norm": 2.913774251937866, "lr": 3.20604512734719e-05 }, { "step": 646, "T": 576, "positions": 16, "loss": 0.4901834645716008, "ce": 1.9189534858160187, "tv": 0.33143123603076674, "conf": 1.1116607010771986, "grad_norm": 3.5494978427886963, "lr": 3.2006578925974445e-05 }, { "step": 647, "T": 607, "positions": 16, "loss": 0.2981212300801417, "ce": 1.0116227584439912, "tv": 0.21884329368913313, "conf": 0.9261016123346053, "grad_norm": 10.514735221862793, "lr": 3.195267124631992e-05 }, { "step": 648, "T": 582, "positions": 16, "loss": 0.5945632082875818, "ce": 2.1882697145920247, "tv": 0.41748470487073064, "conf": 0.9468005895023452, "grad_norm": 3.281895637512207, "lr": 3.189872850634922e-05 }, { "step": 649, "T": 919, "positions": 16, "loss": 0.6823074775747955, "ce": 2.7244730847887695, "tv": 0.45540019893087447, "conf": 0.6633296970976517, "grad_norm": 2.882612466812134, "lr": 3.184475097808001e-05 }, { "step": 650, "T": 531, "positions": 16, "loss": 0.2424247695971644, "ce": 0.9934877035657337, "tv": 0.15897332946587994, "conf": 0.5256256415018683, "grad_norm": 5.296032905578613, "lr": 3.179073893370541e-05 }, { "step": 651, "T": 650, "positions": 16, "loss": 0.26045891599824245, "ce": 0.7553696083973591, "tv": 0.205468843862036, "conf": 0.6328964902913867, "grad_norm": 7.955765724182129, "lr": 3.1736692645592583e-05 }, { "step": 652, "T": 1101, "positions": 16, "loss": 0.7655685227437061, "ce": 3.2828384589283814, "tv": 0.48587184221833013, "conf": 0.9116625127726365, "grad_norm": 2.5027737617492676, "lr": 3.168261238628137e-05 }, { "step": 653, "T": 618, "positions": 16, "loss": 0.3704152096124744, "ce": 1.1952431041108866, "tv": 0.2787676758252928, "conf": 0.8836637109343428, "grad_norm": 4.534707069396973, "lr": 3.162849842848294e-05 }, { "step": 654, "T": 1573, "positions": 16, "loss": 0.7051338970049983, "ce": 3.1384506033791695, "tv": 0.4347653858130798, "conf": 1.1918719866953325, "grad_norm": 3.9446775913238525, "lr": 3.157435104507837e-05 }, { "step": 655, "T": 675, "positions": 16, "loss": 0.43094794009812176, "ce": 1.6936596047598869, "tv": 0.29064664454199374, "conf": 0.9694726738380268, "grad_norm": 3.056347131729126, "lr": 3.152017050911727e-05 }, { "step": 656, "T": 677, "positions": 16, "loss": 0.2785477755824104, "ce": 0.9424799885000539, "tv": 0.20477754488092614, "conf": 0.5795233100435269, "grad_norm": 1.952000617980957, "lr": 3.1465957093816504e-05 }, { "step": 657, "T": 678, "positions": 16, "loss": 0.7609265220817178, "ce": 3.081418726782431, "tv": 0.5030940726355766, "conf": 0.9019912795629352, "grad_norm": 5.142168045043945, "lr": 3.141171107255867e-05 }, { "step": 658, "T": 618, "positions": 16, "loss": 0.14903545687093356, "ce": 0.20936505186318755, "tv": 0.14233216988895947, "conf": 0.6408973624916143, "grad_norm": 2.952324151992798, "lr": 3.1357432718890815e-05 }, { "step": 659, "T": 540, "positions": 16, "loss": 0.2291328574046929, "ce": 0.7923047626400148, "tv": 0.16655819524021354, "conf": 0.318173001294781, "grad_norm": 1.8138794898986816, "lr": 3.1303122306523025e-05 }, { "step": 660, "T": 537, "positions": 16, "loss": 0.22874459055447005, "ce": 0.756395901434928, "tv": 0.1701166739439941, "conf": 0.3951666181675364, "grad_norm": 2.959256172180176, "lr": 3.124878010932704e-05 }, { "step": 661, "T": 637, "positions": 16, "loss": 0.19385427030647406, "ce": 0.6253525254460328, "tv": 0.14591002107044915, "conf": 0.5090928880100023, "grad_norm": 2.5707879066467285, "lr": 3.1194406401334894e-05 }, { "step": 662, "T": 675, "positions": 16, "loss": 0.4057978259588708, "ce": 1.3901131750717468, "tv": 0.296429459565843, "conf": 0.7909006109621259, "grad_norm": 2.6595327854156494, "lr": 3.1140001456737525e-05 }, { "step": 663, "T": 670, "positions": 16, "loss": 0.5266761970706284, "ce": 2.165127898188075, "tv": 0.34462601924315095, "conf": 1.2232370679848827, "grad_norm": 7.05876350402832, "lr": 3.108556554988338e-05 }, { "step": 664, "T": 529, "positions": 16, "loss": 0.47031198133572616, "ce": 1.5260821995880178, "tv": 0.3530041862541111, "conf": 1.0503573747394692, "grad_norm": 3.6654295921325684, "lr": 3.103109895527703e-05 }, { "step": 665, "T": 615, "positions": 16, "loss": 0.3476401144034753, "ce": 1.2164190078210595, "tv": 0.251109142194764, "conf": 0.9546667867624592, "grad_norm": 4.104849338531494, "lr": 3.0976601947577805e-05 }, { "step": 666, "T": 529, "positions": 16, "loss": 0.7884069830179214, "ce": 2.9616056997328997, "tv": 0.5469404961913824, "conf": 1.435746813978767, "grad_norm": 5.492303371429443, "lr": 3.0922074801598414e-05 }, { "step": 667, "T": 889, "positions": 16, "loss": 0.44330848356185015, "ce": 1.7209809197229333, "tv": 0.30134488400653936, "conf": 1.0972555528357475, "grad_norm": 2.8110573291778564, "lr": 3.0867517792303515e-05 }, { "step": 668, "T": 676, "positions": 16, "loss": 0.513321803895451, "ce": 1.8507223085543956, "tv": 0.36472175649760175, "conf": 1.2921886515396181, "grad_norm": 2.3646936416625977, "lr": 3.081293119480838e-05 }, { "step": 669, "T": 545, "positions": 16, "loss": 0.16449217058379872, "ce": 0.6609205573865324, "tv": 0.10933346566889668, "conf": 0.624262408832692, "grad_norm": 1.0272321701049805, "lr": 3.075831528437748e-05 }, { "step": 670, "T": 669, "positions": 16, "loss": 0.5145563071127981, "ce": 1.7763080477016047, "tv": 0.3743616798892617, "conf": 1.2047237745136954, "grad_norm": 2.29659104347229, "lr": 3.0703670336423096e-05 }, { "step": 671, "T": 672, "positions": 16, "loss": 0.3109370312449755, "ce": 1.0383888376131836, "tv": 0.23010904780494457, "conf": 1.3507033395580947, "grad_norm": 4.72783088684082, "lr": 3.0648996626503955e-05 }, { "step": 672, "T": 546, "positions": 16, "loss": 0.39927611069288105, "ce": 1.057775518310791, "tv": 0.326109507936053, "conf": 0.7468934436328709, "grad_norm": 3.4680919647216797, "lr": 3.05942944303238e-05 }, { "step": 673, "T": 532, "positions": 16, "loss": 0.4764403443550691, "ce": 1.8625057777389884, "tv": 0.32243306771852076, "conf": 0.3528245132911252, "grad_norm": 4.646407604217529, "lr": 3.053956402373004e-05 }, { "step": 674, "T": 1190, "positions": 16, "loss": 0.7122291233536089, "ce": 3.122582661917477, "tv": 0.44441206822375534, "conf": 2.016209136973316, "grad_norm": 2.982987880706787, "lr": 3.0484805682712313e-05 }, { "step": 675, "T": 611, "positions": 16, "loss": 0.3933835335592448, "ce": 1.1689393227161418, "tv": 0.3072106715062546, "conf": 1.067587144661067, "grad_norm": 3.688098907470703, "lr": 3.0430019683401168e-05 }, { "step": 676, "T": 526, "positions": 16, "loss": 0.20396490974781045, "ce": 0.6814061950481118, "tv": 0.15091588837094605, "conf": 0.8216907707970478, "grad_norm": 2.4223334789276123, "lr": 3.0375206302066596e-05 }, { "step": 677, "T": 673, "positions": 16, "loss": 0.9018899016082287, "ce": 3.607040222734213, "tv": 0.6013176701962948, "conf": 1.9763718623435125, "grad_norm": 2.9934370517730713, "lr": 3.0320365815116654e-05 }, { "step": 678, "T": 539, "positions": 16, "loss": 0.20109080276142777, "ce": 0.5883053847601332, "tv": 0.15806696943036513, "conf": 0.4890600235717102, "grad_norm": 3.0272998809814453, "lr": 3.0265498499096127e-05 }, { "step": 679, "T": 526, "positions": 16, "loss": 0.4840837097290205, "ce": 2.3682229011392337, "tv": 0.27473489953990793, "conf": 1.5913198569230644, "grad_norm": 2.4243099689483643, "lr": 3.0210604630685048e-05 }, { "step": 680, "T": 675, "positions": 16, "loss": 0.37426085327751935, "ce": 1.2161145641002804, "tv": 0.2807215708307922, "conf": 1.9155586290125939, "grad_norm": 2.6973767280578613, "lr": 3.015568448669738e-05 }, { "step": 681, "T": 552, "positions": 16, "loss": 0.3650428406949686, "ce": 2.160548227859124, "tv": 0.16554224129049544, "conf": 0.4277395635533594, "grad_norm": 1.1364914178848267, "lr": 3.010073834407957e-05 }, { "step": 682, "T": 616, "positions": 16, "loss": 0.5953329112380743, "ce": 2.092842163692694, "tv": 0.42894300844636746, "conf": 1.8003222137341481, "grad_norm": 2.8414225578308105, "lr": 3.0045766479909166e-05 }, { "step": 683, "T": 528, "positions": 16, "loss": 0.7140660490258597, "ce": 2.695292007178068, "tv": 0.4939298197859898, "conf": 1.2309509365877602, "grad_norm": 3.8938071727752686, "lr": 2.9990769171393423e-05 }, { "step": 684, "T": 525, "positions": 16, "loss": 0.8988371519371867, "ce": 3.7033160384744406, "tv": 0.5872283773496747, "conf": 2.007402445189655, "grad_norm": 2.2905433177948, "lr": 2.993574669586792e-05 }, { "step": 685, "T": 675, "positions": 16, "loss": 0.5680237417109311, "ce": 2.1632308503612876, "tv": 0.390778532018885, "conf": 1.190844517230289, "grad_norm": 3.4586760997772217, "lr": 2.9880699330795135e-05 }, { "step": 686, "T": 611, "positions": 16, "loss": 0.40434813356841914, "ce": 1.6377228359524452, "tv": 0.2673065022172523, "conf": 1.0511406588320256, "grad_norm": 3.28908634185791, "lr": 2.9825627353763048e-05 }, { "step": 687, "T": 594, "positions": 16, "loss": 0.7849164831714006, "ce": 3.4594936783541925, "tv": 0.4877412504720269, "conf": 1.0131599600342724, "grad_norm": 5.471715927124023, "lr": 2.977053104248375e-05 }, { "step": 688, "T": 654, "positions": 16, "loss": 0.5539275802002521, "ce": 2.1582195795381267, "tv": 0.3756729253618687, "conf": 1.4058145122047914, "grad_norm": 4.728559970855713, "lr": 2.971541067479207e-05 }, { "step": 689, "T": 539, "positions": 16, "loss": 0.18219689982834097, "ce": 0.7065580941216467, "tv": 0.12393454490029399, "conf": 0.3675910895806691, "grad_norm": 2.5602800846099854, "lr": 2.966026652864412e-05 }, { "step": 690, "T": 529, "positions": 16, "loss": 0.4972985432323185, "ce": 2.268481623417756, "tv": 0.3005004341393942, "conf": 1.463195307052549, "grad_norm": 3.390573024749756, "lr": 2.9605098882115927e-05 }, { "step": 691, "T": 631, "positions": 16, "loss": 0.5772491054376587, "ce": 2.9402456718271424, "tv": 0.31469392138296826, "conf": 1.2818098351818321, "grad_norm": 4.254605293273926, "lr": 2.954990801340202e-05 }, { "step": 692, "T": 533, "positions": 16, "loss": 0.44625362528313417, "ce": 1.9264249000098062, "tv": 0.2817901605976658, "conf": 0.6201686309419724, "grad_norm": 2.1162214279174805, "lr": 2.9494694200814033e-05 }, { "step": 693, "T": 537, "positions": 16, "loss": 0.2665835423122189, "ce": 1.105379372908203, "tv": 0.1733840223041625, "conf": 0.4580229236328299, "grad_norm": 2.941364049911499, "lr": 2.9439457722779317e-05 }, { "step": 694, "T": 525, "positions": 16, "loss": 0.9139641970396042, "ce": 3.545512467622757, "tv": 0.6215699408203363, "conf": 1.9843741061922628, "grad_norm": 4.177820682525635, "lr": 2.938419885783949e-05 }, { "step": 695, "T": 851, "positions": 16, "loss": 0.5698432908975519, "ce": 2.101279085560236, "tv": 0.3996837747981772, "conf": 1.727806999348104, "grad_norm": 8.159337043762207, "lr": 2.9328917884649077e-05 }, { "step": 696, "T": 551, "positions": 16, "loss": 0.42857135825033765, "ce": 1.4957288145305938, "tv": 0.30999829595384654, "conf": 1.4609777868193508, "grad_norm": 2.1795074939727783, "lr": 2.9273615081974075e-05 }, { "step": 697, "T": 897, "positions": 16, "loss": 0.7494317882592441, "ce": 3.550867594212832, "tv": 0.43816115112713305, "conf": 1.8729840572923422, "grad_norm": 5.662908554077148, "lr": 2.9218290728690583e-05 }, { "step": 698, "T": 603, "positions": 16, "loss": 0.10998729123093653, "ce": 0.2557006015767911, "tv": 0.0937969219594379, "conf": 0.7257517638863646, "grad_norm": 1.587748408317566, "lr": 2.916294510378335e-05 }, { "step": 699, "T": 578, "positions": 16, "loss": 0.829642306547612, "ce": 3.776086861733347, "tv": 0.5022595543414354, "conf": 1.6997895777385565, "grad_norm": 3.3528847694396973, "lr": 2.9107578486344413e-05 }, { "step": 700, "T": 532, "positions": 16, "loss": 0.3106689058622578, "ce": 1.078673683497982, "tv": 0.22533503946760902, "conf": 1.6311842055363823, "grad_norm": 3.5794131755828857, "lr": 2.9052191155571663e-05 }, { "step": 701, "T": 675, "positions": 16, "loss": 0.3447178633305157, "ce": 1.166408012717966, "tv": 0.25341896587815427, "conf": 1.7072299571333462, "grad_norm": 2.519392490386963, "lr": 2.899678339076741e-05 }, { "step": 702, "T": 536, "positions": 16, "loss": 0.25299981061107246, "ce": 1.05620915513191, "tv": 0.16375433275425166, "conf": 0.3941887738171772, "grad_norm": 3.5872364044189453, "lr": 2.894135547133706e-05 }, { "step": 703, "T": 526, "positions": 16, "loss": 0.5866773540619761, "ce": 2.0782858059101272, "tv": 0.42094309523236006, "conf": 1.760534601242398, "grad_norm": 2.8666162490844727, "lr": 2.8885907676787622e-05 }, { "step": 704, "T": 530, "positions": 16, "loss": 0.4271719552707509, "ce": 1.4172753101411217, "tv": 0.31716048315138323, "conf": 2.602033068554925, "grad_norm": 4.681580543518066, "lr": 2.8830440286726328e-05 }, { "step": 705, "T": 527, "positions": 16, "loss": 0.880348979961127, "ce": 3.826507864985615, "tv": 0.55299800215289, "conf": 1.4651098558679223, "grad_norm": 3.6815826892852783, "lr": 2.8774953580859243e-05 }, { "step": 706, "T": 535, "positions": 16, "loss": 0.13096541506820358, "ce": 0.37964735405012107, "tv": 0.10333408837323077, "conf": 0.31487101244783844, "grad_norm": 2.7149221897125244, "lr": 2.87194478389898e-05 }, { "step": 707, "T": 531, "positions": 16, "loss": 0.6314496649429202, "ce": 2.299006085842848, "tv": 0.4461656315252185, "conf": 2.014233922585845, "grad_norm": 3.584641456604004, "lr": 2.866392334101748e-05 }, { "step": 708, "T": 536, "positions": 16, "loss": 0.20939853679632847, "ce": 0.6928338548673594, "tv": 0.15568350030096667, "conf": 0.24816441241160814, "grad_norm": 4.767368793487549, "lr": 2.8608380366936293e-05 }, { "step": 709, "T": 585, "positions": 16, "loss": 0.33550034381914884, "ce": 1.1041944783355575, "tv": 0.2500898871803656, "conf": 1.6681083783337223, "grad_norm": 4.684582233428955, "lr": 2.8552819196833453e-05 }, { "step": 710, "T": 524, "positions": 16, "loss": 0.7124899718910456, "ce": 2.3969038139330223, "tv": 0.5253328923135996, "conf": 1.6036921279446688, "grad_norm": 2.2433345317840576, "lr": 2.8497240110887914e-05 }, { "step": 711, "T": 562, "positions": 16, "loss": 0.3024862804750228, "ce": 0.9162016869413492, "tv": 0.23429569178006204, "conf": 2.8219794115803296, "grad_norm": 3.927220344543457, "lr": 2.8441643389368977e-05 }, { "step": 712, "T": 1482, "positions": 16, "loss": 0.7918049941895333, "ce": 3.644290108297355, "tv": 0.4748621987664592, "conf": 3.151002432590147, "grad_norm": 2.876028299331665, "lr": 2.8386029312634882e-05 }, { "step": 713, "T": 669, "positions": 16, "loss": 0.527454629263957, "ce": 1.6100675469642738, "tv": 0.40716429513850017, "conf": 1.3091154817381376, "grad_norm": 2.6326234340667725, "lr": 2.8330398161131376e-05 }, { "step": 714, "T": 591, "positions": 16, "loss": 0.20946786575950682, "ce": 0.5660841504286509, "tv": 0.1698438333114609, "conf": 1.4201601097101957, "grad_norm": 4.6898698806762695, "lr": 2.827475021539031e-05 }, { "step": 715, "T": 524, "positions": 16, "loss": 0.6737763909623027, "ce": 2.4998817570740357, "tv": 0.4708758071064949, "conf": 3.089457456022501, "grad_norm": 3.9288718700408936, "lr": 2.821908575602824e-05 }, { "step": 716, "T": 532, "positions": 16, "loss": 0.37107890319020953, "ce": 1.1854435334144, "tv": 0.2805939515528735, "conf": 1.7178407480208797, "grad_norm": 2.4891202449798584, "lr": 2.8163405063744975e-05 }, { "step": 717, "T": 672, "positions": 16, "loss": 0.6006679863494355, "ce": 2.301451330713462, "tv": 0.4116920736269094, "conf": 1.652108684880659, "grad_norm": 2.4347474575042725, "lr": 2.8107708419322194e-05 }, { "step": 718, "T": 534, "positions": 16, "loss": 0.5029997003730386, "ce": 1.624250330380164, "tv": 0.37841631879564375, "conf": 3.0825625268956856, "grad_norm": 3.6536483764648438, "lr": 2.8051996103622003e-05 }, { "step": 719, "T": 1014, "positions": 16, "loss": 0.697824428381864, "ce": 2.777876923850272, "tv": 0.4667074966127984, "conf": 2.756248833763493, "grad_norm": 2.443943977355957, "lr": 2.799626839758555e-05 }, { "step": 720, "T": 675, "positions": 16, "loss": 0.4146127339154191, "ce": 1.6670794666097208, "tv": 0.2754497628666286, "conf": 1.9241558565593095, "grad_norm": 2.5107061862945557, "lr": 2.7940525582231596e-05 }, { "step": 721, "T": 647, "positions": 16, "loss": 0.6022540219564689, "ce": 2.0849523670767667, "tv": 0.4375097701413324, "conf": 2.501814996536268, "grad_norm": 5.647247314453125, "lr": 2.7884767938655076e-05 }, { "step": 722, "T": 596, "positions": 16, "loss": 0.23225483906026057, "ce": 0.7311205269106722, "tv": 0.17682532320213795, "conf": 1.381312981247902, "grad_norm": 2.9237797260284424, "lr": 2.782899574802571e-05 }, { "step": 723, "T": 674, "positions": 16, "loss": 0.7009216565638781, "ce": 2.7295543868094683, "tv": 0.4755180338397622, "conf": 1.8096448853611946, "grad_norm": 3.421143054962158, "lr": 2.7773209291586567e-05 }, { "step": 724, "T": 858, "positions": 16, "loss": 0.42057552860933356, "ce": 1.4992417732137255, "tv": 0.3007237218262162, "conf": 1.0229285289463201, "grad_norm": 2.570253372192383, "lr": 2.7717408850652666e-05 }, { "step": 725, "T": 597, "positions": 16, "loss": 0.41571204279898666, "ce": 1.4280859045684338, "tv": 0.30322607324342243, "conf": 0.8101363339992531, "grad_norm": 4.566580772399902, "lr": 2.7661594706609538e-05 }, { "step": 726, "T": 604, "positions": 16, "loss": 0.3890816220209672, "ce": 1.3818540015181497, "tv": 0.27877358660043683, "conf": 1.4356219546358533, "grad_norm": 4.277795314788818, "lr": 2.760576714091182e-05 }, { "step": 727, "T": 589, "positions": 16, "loss": 0.14853985170339, "ce": 0.32520609346147467, "tv": 0.12891026801116823, "conf": 0.5577318591771155, "grad_norm": 2.5984973907470703, "lr": 2.7549926435081824e-05 }, { "step": 728, "T": 623, "positions": 16, "loss": 0.2703761379234493, "ce": 0.9208500396634918, "tv": 0.19810127522214316, "conf": 2.3709864877164364, "grad_norm": 6.464245319366455, "lr": 2.749407287070812e-05 }, { "step": 729, "T": 884, "positions": 16, "loss": 0.7690297459048452, "ce": 3.547477996429052, "tv": 0.46031326252978033, "conf": 0.8093277401785599, "grad_norm": 4.039319038391113, "lr": 2.7438206729444145e-05 }, { "step": 730, "T": 673, "positions": 16, "loss": 0.5415248110657558, "ce": 2.1149561626370996, "tv": 0.36669911436911207, "conf": 1.2562463101785397, "grad_norm": 4.0559258460998535, "lr": 2.738232829300672e-05 }, { "step": 731, "T": 540, "positions": 16, "loss": 0.13018907029436377, "ce": 0.4154814732437444, "tv": 0.09848991513354122, "conf": 0.30706282747689784, "grad_norm": 2.7537758350372314, "lr": 2.7326437843174696e-05 }, { "step": 732, "T": 1324, "positions": 16, "loss": 0.8387987739115488, "ce": 4.1469309281383175, "tv": 0.47122855426277965, "conf": 0.9061497158116936, "grad_norm": 2.8926048278808594, "lr": 2.72705356617875e-05 }, { "step": 733, "T": 677, "positions": 16, "loss": 0.5866366522132012, "ce": 2.307740555559576, "tv": 0.39540289568321896, "conf": 1.4140758710229784, "grad_norm": 1.779266119003296, "lr": 2.7214622030743693e-05 }, { "step": 734, "T": 615, "positions": 16, "loss": 0.6030583372339606, "ce": 2.1328575008083135, "tv": 0.4330806410871446, "conf": 2.04047248698771, "grad_norm": 2.9116270542144775, "lr": 2.7158697231999614e-05 }, { "step": 735, "T": 1500, "positions": 16, "loss": 0.715666021220386, "ce": 3.2057530358433723, "tv": 0.4389896937645972, "conf": 0.8465238651260734, "grad_norm": 3.1778171062469482, "lr": 2.710276154756789e-05 }, { "step": 736, "T": 672, "positions": 16, "loss": 0.855139959603548, "ce": 3.3180745225399733, "tv": 0.581480560824275, "conf": 1.7378720906563103, "grad_norm": 1.8909332752227783, "lr": 2.704681525951604e-05 }, { "step": 737, "T": 561, "positions": 16, "loss": 0.274629036664237, "ce": 1.0146314224984962, "tv": 0.1924065624832565, "conf": 0.7923422073927593, "grad_norm": 2.9632956981658936, "lr": 2.6990858649965068e-05 }, { "step": 738, "T": 610, "positions": 16, "loss": 0.49586141150211915, "ce": 1.6865218409802765, "tv": 0.36356580595020205, "conf": 0.6529047540734609, "grad_norm": 2.1622722148895264, "lr": 2.693489200108802e-05 }, { "step": 739, "T": 535, "positions": 16, "loss": 0.39604281930223806, "ce": 1.7836687502276618, "tv": 0.241862161528843, "conf": 0.5448471798721357, "grad_norm": 3.715893507003784, "lr": 2.6878915595108573e-05 }, { "step": 740, "T": 587, "positions": 16, "loss": 0.21173089546664414, "ce": 0.6356280323784631, "tv": 0.1646312112598025, "conf": 0.754336307307085, "grad_norm": 2.6089606285095215, "lr": 2.6822929714299587e-05 }, { "step": 741, "T": 633, "positions": 16, "loss": 0.12148448404605006, "ce": 0.22965855217495346, "tv": 0.10946514195984491, "conf": 0.5850091114321003, "grad_norm": 2.1878879070281982, "lr": 2.6766934640981723e-05 }, { "step": 742, "T": 677, "positions": 16, "loss": 0.3742586155422032, "ce": 1.4928216547705233, "tv": 0.24997384171001613, "conf": 1.2152643203735352, "grad_norm": 3.895106554031372, "lr": 2.6710930657521988e-05 }, { "step": 743, "T": 598, "positions": 16, "loss": 0.26305688604770694, "ce": 0.6162968645658111, "tv": 0.2238079996968736, "conf": 1.1543220608418778, "grad_norm": 2.814039707183838, "lr": 2.6654918046332323e-05 }, { "step": 744, "T": 654, "positions": 16, "loss": 0.6779837519861758, "ce": 2.4314263483975083, "tv": 0.4831568000372499, "conf": 1.3528941722615855, "grad_norm": 3.9238393306732178, "lr": 2.6598897089868175e-05 }, { "step": 745, "T": 546, "positions": 16, "loss": 0.20541283955753897, "ce": 1.1569224237926221, "tv": 0.09968953738098207, "conf": 0.6004514654632658, "grad_norm": 1.3378432989120483, "lr": 2.6542868070627063e-05 }, { "step": 746, "T": 617, "positions": 16, "loss": 0.3262814878253266, "ce": 0.9264362768544743, "tv": 0.25959763512946665, "conf": 1.182768425515448, "grad_norm": 5.6743950843811035, "lr": 2.648683127114718e-05 }, { "step": 747, "T": 537, "positions": 16, "loss": 0.10030435860790021, "ce": 0.2204272531241429, "tv": 0.08695737470407039, "conf": 0.31329593174046977, "grad_norm": 0.5454471111297607, "lr": 2.643078697400595e-05 }, { "step": 748, "T": 649, "positions": 16, "loss": 0.6479131579399109, "ce": 2.338425480760634, "tv": 0.4600784592330456, "conf": 1.2897684928029776, "grad_norm": 4.555239677429199, "lr": 2.63747354618186e-05 }, { "step": 749, "T": 676, "positions": 16, "loss": 0.5007040585278446, "ce": 1.8209885589903934, "tv": 0.35400578327335097, "conf": 1.1461585668405938, "grad_norm": 4.130442142486572, "lr": 2.6318677017236744e-05 }, { "step": 750, "T": 672, "positions": 16, "loss": 0.7313411682844162, "ce": 2.600365737453103, "tv": 0.5236717658117414, "conf": 1.7386472669895738, "grad_norm": 1.9435880184173584, "lr": 2.626261192294695e-05 }, { "step": 751, "T": 665, "positions": 16, "loss": 0.243498018317041, "ce": 0.8210871097617201, "tv": 0.17932144749647705, "conf": 0.6859118042133332, "grad_norm": 2.662580728530884, "lr": 2.620654046166932e-05 }, { "step": 752, "T": 542, "positions": 16, "loss": 0.21668503943874384, "ce": 0.8994927270796325, "tv": 0.14081752043239248, "conf": 0.733185759863602, "grad_norm": 2.7407188415527344, "lr": 2.6150462916156066e-05 }, { "step": 753, "T": 526, "positions": 16, "loss": 0.6375180094619282, "ce": 2.42682106277789, "tv": 0.4387065480113961, "conf": 0.9865931165404618, "grad_norm": 3.6338818073272705, "lr": 2.6094379569190082e-05 }, { "step": 754, "T": 570, "positions": 16, "loss": 0.19875453475606264, "ce": 0.7197322262619252, "tv": 0.14086812711684615, "conf": 0.7775902153569518, "grad_norm": 2.237971067428589, "lr": 2.6038290703583517e-05 }, { "step": 755, "T": 673, "positions": 16, "loss": 0.26577753991159625, "ce": 1.002243067931886, "tv": 0.18394804128774922, "conf": 1.0138646955097101, "grad_norm": 2.373605251312256, "lr": 2.5982196602176328e-05 }, { "step": 756, "T": 1148, "positions": 16, "loss": 0.7179386822763263, "ce": 3.191078301417974, "tv": 0.44314538153685135, "conf": 0.7930449045443225, "grad_norm": 2.9332540035247803, "lr": 2.592609754783492e-05 }, { "step": 757, "T": 675, "positions": 16, "loss": 0.45419312501326203, "ce": 1.5635324083232263, "tv": 0.33093322697095573, "conf": 0.778955738807781, "grad_norm": 5.162529945373535, "lr": 2.5869993823450627e-05 }, { "step": 758, "T": 609, "positions": 16, "loss": 0.3070186103532251, "ce": 0.9595931961509905, "tv": 0.23451032596631194, "conf": 1.136807598418897, "grad_norm": 2.6059463024139404, "lr": 2.5813885711938357e-05 }, { "step": 759, "T": 528, "positions": 16, "loss": 0.5657591220558515, "ce": 2.2004216938917125, "tv": 0.3841299686077946, "conf": 0.7106277838756796, "grad_norm": 4.58778190612793, "lr": 2.5757773496235142e-05 }, { "step": 760, "T": 547, "positions": 16, "loss": 0.32649280497389555, "ce": 0.9652612642876193, "tv": 0.2555185282981256, "conf": 0.623175355462656, "grad_norm": 3.6112136840820312, "lr": 2.5701657459298693e-05 }, { "step": 761, "T": 529, "positions": 16, "loss": 0.26539012565444864, "ce": 0.9728040776499256, "tv": 0.18678857793929637, "conf": 0.580120516002097, "grad_norm": 4.769775867462158, "lr": 2.5645537884106026e-05 }, { "step": 762, "T": 550, "positions": 16, "loss": 0.46634171236655675, "ce": 1.652143392868311, "tv": 0.33458597943536006, "conf": 0.5521147922709133, "grad_norm": 3.675138235092163, "lr": 2.5589415053651948e-05 }, { "step": 763, "T": 639, "positions": 16, "loss": 0.5107091572317586, "ce": 1.7850414638551229, "tv": 0.3691166951539344, "conf": 1.138117288168317, "grad_norm": 5.109763145446777, "lr": 2.553328925094773e-05 }, { "step": 764, "T": 615, "positions": 16, "loss": 0.3128605989454627, "ce": 1.2084320966211806, "tv": 0.21335266109690565, "conf": 1.6349672013893723, "grad_norm": 2.780799150466919, "lr": 2.547716075901959e-05 }, { "step": 765, "T": 536, "positions": 16, "loss": 0.06552395337712369, "ce": 0.1795455599185516, "tv": 0.05285488586741849, "conf": 0.15441268494532778, "grad_norm": 0.6733750700950623, "lr": 2.5421029860907342e-05 }, { "step": 766, "T": 534, "positions": 16, "loss": 0.23358553142224991, "ce": 0.9109415722793983, "tv": 0.15832375019908795, "conf": 0.2948313404340297, "grad_norm": 1.012115716934204, "lr": 2.5364896839662928e-05 }, { "step": 767, "T": 538, "positions": 16, "loss": 0.7926897779107094, "ce": 2.891725867986679, "tv": 0.5594635475426912, "conf": 1.2619135607674252, "grad_norm": 2.7796971797943115, "lr": 2.5308761978348967e-05 }, { "step": 768, "T": 525, "positions": 16, "loss": 0.441881229488331, "ce": 1.6087979317952659, "tv": 0.3122238224157172, "conf": 1.1838376027027468, "grad_norm": 2.423797845840454, "lr": 2.5252625560037386e-05 }, { "step": 769, "T": 671, "positions": 16, "loss": 0.9092018641531467, "ce": 3.664212927222252, "tv": 0.6030895113945007, "conf": 1.4688419629819691, "grad_norm": 3.5472207069396973, "lr": 2.5196487867807944e-05 }, { "step": 770, "T": 536, "positions": 16, "loss": 0.3130734344103985, "ce": 1.2265598677726075, "tv": 0.2115749437548402, "conf": 0.4880217848269677, "grad_norm": 3.228468179702759, "lr": 2.514034918474686e-05 }, { "step": 771, "T": 532, "positions": 16, "loss": 0.2952205164010593, "ce": 0.801584864415986, "tv": 0.2389578072461518, "conf": 1.112238537720259, "grad_norm": 4.308993816375732, "lr": 2.50842097939453e-05 }, { "step": 772, "T": 553, "positions": 16, "loss": 0.31974761810488417, "ce": 1.2533906918124558, "tv": 0.21600949153798865, "conf": 0.5655274227310656, "grad_norm": 1.5234702825546265, "lr": 2.5028069978498038e-05 }, { "step": 773, "T": 673, "positions": 16, "loss": 0.8546517072245479, "ce": 3.70807147026062, "tv": 0.5376050658524036, "conf": 1.0701959606958553, "grad_norm": 3.728828191757202, "lr": 2.4971930021501965e-05 }, { "step": 774, "T": 1324, "positions": 16, "loss": 0.9514826365702902, "ce": 4.774659473114298, "tv": 0.5266852068743901, "conf": 1.5334033733815886, "grad_norm": 2.4323410987854004, "lr": 2.4915790206054705e-05 }, { "step": 775, "T": 642, "positions": 16, "loss": 0.40301105146545524, "ce": 1.3768713466558893, "tv": 0.2948043610658715, "conf": 2.149226439741019, "grad_norm": 5.116608142852783, "lr": 2.485965081525315e-05 }, { "step": 776, "T": 548, "positions": 16, "loss": 0.24514453892470556, "ce": 1.146111536134356, "tv": 0.14503710680583026, "conf": 1.001300660295783, "grad_norm": 1.7782808542251587, "lr": 2.480351213219205e-05 }, { "step": 777, "T": 642, "positions": 16, "loss": 0.6402708580717444, "ce": 2.2404467510059476, "tv": 0.462473557330668, "conf": 1.4557304618647322, "grad_norm": 4.957493305206299, "lr": 2.4747374439962626e-05 }, { "step": 778, "T": 528, "positions": 16, "loss": 0.6869095768779516, "ce": 2.3524782117456198, "tv": 0.5018464257009327, "conf": 1.5842359429225326, "grad_norm": 2.8926687240600586, "lr": 2.4691238021651042e-05 }, { "step": 779, "T": 537, "positions": 16, "loss": 0.24381793159864174, "ce": 0.8741223999672911, "tv": 0.17378409349657886, "conf": 0.81651141330142, "grad_norm": 2.31516432762146, "lr": 2.4635103160337085e-05 }, { "step": 780, "T": 672, "positions": 16, "loss": 0.4486700406523596, "ce": 1.6167422385415193, "tv": 0.31888424645148916, "conf": 1.0737748813506869, "grad_norm": 2.825042724609375, "lr": 2.457897013909266e-05 }, { "step": 781, "T": 526, "positions": 16, "loss": 0.3228858856705301, "ce": 1.2120493609365894, "tv": 0.22408995040359514, "conf": 0.5723550191726194, "grad_norm": 3.2637875080108643, "lr": 2.452283924098041e-05 }, { "step": 782, "T": 538, "positions": 16, "loss": 0.5206898786127567, "ce": 1.6648805756121874, "tv": 0.3935575783252716, "conf": 2.262709631675534, "grad_norm": 2.2692487239837646, "lr": 2.4466710749052284e-05 }, { "step": 783, "T": 527, "positions": 16, "loss": 0.44766570688807406, "ce": 1.4827765629906935, "tv": 0.3326534005718713, "conf": 2.159951129912461, "grad_norm": 4.670868873596191, "lr": 2.4410584946348054e-05 }, { "step": 784, "T": 526, "positions": 16, "loss": 0.2852622593527485, "ce": 1.1280094268440735, "tv": 0.19162369484183728, "conf": 0.9574750571373443, "grad_norm": 3.025542974472046, "lr": 2.4354462115893977e-05 }, { "step": 785, "T": 529, "positions": 16, "loss": 0.2402932616325586, "ce": 0.7459044809388615, "tv": 0.18411423605947164, "conf": 0.9978878407067711, "grad_norm": 2.783693790435791, "lr": 2.429834254070131e-05 }, { "step": 786, "T": 968, "positions": 16, "loss": 0.7461699590785429, "ce": 3.135920574859483, "tv": 0.48064210027223453, "conf": 2.7379828958482904, "grad_norm": 2.960840940475464, "lr": 2.424222650376486e-05 }, { "step": 787, "T": 1360, "positions": 16, "loss": 0.9614813878433779, "ce": 4.630200001061894, "tv": 0.5538459935924038, "conf": 1.746673783114602, "grad_norm": 3.51423978805542, "lr": 2.418611428806165e-05 }, { "step": 788, "T": 672, "positions": 16, "loss": 0.8123783804476261, "ce": 3.1252153404057026, "tv": 0.5553965102881193, "conf": 2.0908732754178345, "grad_norm": 1.8178489208221436, "lr": 2.413000617654938e-05 }, { "step": 789, "T": 582, "positions": 16, "loss": 0.6649766666814685, "ce": 2.7724055647850037, "tv": 0.4308179197832942, "conf": 1.434814107555212, "grad_norm": 3.0600898265838623, "lr": 2.4073902452165078e-05 }, { "step": 790, "T": 613, "positions": 16, "loss": 0.4540388678724412, "ce": 1.4099714517087705, "tv": 0.347824138305441, "conf": 1.6276859649697144, "grad_norm": 4.378692626953125, "lr": 2.4017803397823675e-05 }, { "step": 791, "T": 646, "positions": 16, "loss": 0.6585919307544827, "ce": 2.245846461970359, "tv": 0.4822303280234337, "conf": 1.8091418303083628, "grad_norm": 2.3020544052124023, "lr": 2.396170929641649e-05 }, { "step": 792, "T": 672, "positions": 16, "loss": 0.7817930206656456, "ce": 2.9904548283666372, "tv": 0.5363861601799726, "conf": 2.142439881153223, "grad_norm": 2.9876646995544434, "lr": 2.390562043080992e-05 }, { "step": 793, "T": 535, "positions": 16, "loss": 0.6272359869435604, "ce": 2.805738535709679, "tv": 0.3851801586060901, "conf": 1.5487798250548863, "grad_norm": 5.488945484161377, "lr": 2.3849537083843936e-05 }, { "step": 794, "T": 537, "positions": 16, "loss": 0.35901950790866977, "ce": 1.3963049106428116, "tv": 0.24376556988136144, "conf": 0.8353205596058615, "grad_norm": 5.396600246429443, "lr": 2.3793459538330678e-05 }, { "step": 795, "T": 1797, "positions": 16, "loss": 0.764878022717312, "ce": 3.2762425914406776, "tv": 0.4858375149779022, "conf": 1.958694331769948, "grad_norm": 3.2732467651367188, "lr": 2.373738807705306e-05 }, { "step": 796, "T": 527, "positions": 16, "loss": 0.6602163878633291, "ce": 2.315251737699157, "tv": 0.4763235850914498, "conf": 1.0744678654430544, "grad_norm": 2.935354471206665, "lr": 2.368132298276326e-05 }, { "step": 797, "T": 544, "positions": 16, "loss": 0.14375407224679293, "ce": 0.5119585068879928, "tv": 0.10284247652998602, "conf": 0.3395680601388449, "grad_norm": 2.0652928352355957, "lr": 2.3625264538181408e-05 }, { "step": 798, "T": 675, "positions": 16, "loss": 0.3620077255000069, "ce": 1.4852474173603696, "tv": 0.23720331969070685, "conf": 1.453141527264961, "grad_norm": 3.573408365249634, "lr": 2.3569213025994056e-05 }, { "step": 799, "T": 525, "positions": 16, "loss": 0.6382128465920687, "ce": 2.0431396942585707, "tv": 0.48210986237972975, "conf": 1.6021016566082835, "grad_norm": 3.345679759979248, "lr": 2.3513168728852823e-05 }, { "step": 800, "T": 583, "positions": 16, "loss": 0.12227657630501199, "ce": 0.36586081407983784, "tv": 0.09521166596459807, "conf": 0.5917829416021227, "grad_norm": 2.00085711479187, "lr": 2.345713192937295e-05 }, { "step": 801, "T": 620, "positions": 16, "loss": 0.40564296988054593, "ce": 1.458655095589961, "tv": 0.2886416290539273, "conf": 1.5439182134027956, "grad_norm": 4.616909980773926, "lr": 2.3401102910131834e-05 }, { "step": 802, "T": 538, "positions": 16, "loss": 0.1655266805436213, "ce": 0.5064461961976576, "tv": 0.12764673217793643, "conf": 0.4708550622675228, "grad_norm": 1.915352463722229, "lr": 2.334508195366768e-05 }, { "step": 803, "T": 637, "positions": 16, "loss": 0.14271257254949887, "ce": 0.3796387268375838, "tv": 0.11638745188247412, "conf": 0.9400617172057082, "grad_norm": 1.350771188735962, "lr": 2.3289069342478018e-05 }, { "step": 804, "T": 658, "positions": 16, "loss": 0.5636445239651948, "ce": 1.8695847347844392, "tv": 0.41854006168432534, "conf": 1.5183780566439964, "grad_norm": 3.826854705810547, "lr": 2.3233065359018283e-05 }, { "step": 805, "T": 609, "positions": 16, "loss": 0.31205849206162384, "ce": 1.1703997646427524, "tv": 0.21668724762457714, "conf": 1.1724372314028813, "grad_norm": 2.8077964782714844, "lr": 2.3177070285700425e-05 }, { "step": 806, "T": 903, "positions": 16, "loss": 0.8179631867969874, "ce": 3.806724594789557, "tv": 0.4858785702381283, "conf": 1.8580251490784576, "grad_norm": 2.861584186553955, "lr": 2.3121084404891433e-05 }, { "step": 807, "T": 524, "positions": 16, "loss": 0.6592054011998698, "ce": 2.305745335208485, "tv": 0.4762565391138196, "conf": 1.2756591780344024, "grad_norm": 9.157987594604492, "lr": 2.3065107998911982e-05 }, { "step": 808, "T": 829, "positions": 16, "loss": 0.5306936428896734, "ce": 2.3465506321040266, "tv": 0.32893176126890467, "conf": 0.7173371503004091, "grad_norm": 2.528646945953369, "lr": 2.3009141350034937e-05 }, { "step": 809, "T": 875, "positions": 16, "loss": 0.5643052476589219, "ce": 2.4366346050446737, "tv": 0.35626864471123554, "conf": 0.6620795667222694, "grad_norm": 2.062352418899536, "lr": 2.2953184740483967e-05 }, { "step": 810, "T": 533, "positions": 16, "loss": 0.4407784161157906, "ce": 1.4804008825285564, "tv": 0.3252648263478477, "conf": 1.0702221881336982, "grad_norm": 3.228414535522461, "lr": 2.2897238452432124e-05 }, { "step": 811, "T": 675, "positions": 16, "loss": 0.4458970233099535, "ce": 1.6068794146412984, "tv": 0.3168989868136123, "conf": 1.1007831731112674, "grad_norm": 2.794337034225464, "lr": 2.284130276800039e-05 }, { "step": 812, "T": 672, "positions": 16, "loss": 0.5419542824984092, "ce": 2.274329028486136, "tv": 0.34946818660682766, "conf": 0.991894281032728, "grad_norm": 2.208375930786133, "lr": 2.278537796925631e-05 }, { "step": 813, "T": 672, "positions": 16, "loss": 0.19169394112850568, "ce": 0.4714687646319362, "tv": 0.16060785117770138, "conf": 1.209790750056115, "grad_norm": 2.598992109298706, "lr": 2.2729464338212515e-05 }, { "step": 814, "T": 1087, "positions": 16, "loss": 0.5327325720368208, "ce": 2.0042249157850165, "tv": 0.36923343305716116, "conf": 1.0727410934086612, "grad_norm": 3.091066837310791, "lr": 2.267356215682531e-05 }, { "step": 815, "T": 672, "positions": 16, "loss": 0.6287194397300482, "ce": 2.7991522904485464, "tv": 0.3875602283515036, "conf": 1.2235773392021656, "grad_norm": 2.6844065189361572, "lr": 2.2617671706993294e-05 }, { "step": 816, "T": 854, "positions": 16, "loss": 0.8560200440588233, "ce": 3.9703261215295242, "tv": 0.5099860524678661, "conf": 1.4901289602621546, "grad_norm": 2.8646678924560547, "lr": 2.256179327055586e-05 }, { "step": 817, "T": 609, "positions": 16, "loss": 0.24630398478984716, "ce": 0.7792063465834076, "tv": 0.18709262337142718, "conf": 1.0690512388128752, "grad_norm": 2.7832767963409424, "lr": 2.250592712929188e-05 }, { "step": 818, "T": 545, "positions": 16, "loss": 0.3910394105096202, "ce": 1.3964281436510646, "tv": 0.27932955447795393, "conf": 1.019411824192073, "grad_norm": 2.1317930221557617, "lr": 2.2450073564918185e-05 }, { "step": 819, "T": 634, "positions": 16, "loss": 0.5314421359216794, "ce": 1.5043837093398906, "tv": 0.42333751678233966, "conf": 1.2778742210939527, "grad_norm": 2.6241443157196045, "lr": 2.239423285908819e-05 }, { "step": 820, "T": 649, "positions": 16, "loss": 0.49619221689636106, "ce": 1.8219368489756107, "tv": 0.3488872618968344, "conf": 0.9622605661704123, "grad_norm": 4.466280460357666, "lr": 2.233840529339046e-05 }, { "step": 821, "T": 675, "positions": 16, "loss": 0.40098574751027627, "ce": 1.550190821700653, "tv": 0.2732963078087778, "conf": 1.3016640042333165, "grad_norm": 3.5806009769439697, "lr": 2.2282591149347336e-05 }, { "step": 822, "T": 560, "positions": 16, "loss": 0.19583301108468731, "ce": 0.6609910819389597, "tv": 0.1441487814017819, "conf": 0.6445535821058002, "grad_norm": 1.5359975099563599, "lr": 2.2226790708413435e-05 }, { "step": 823, "T": 535, "positions": 16, "loss": 0.11164987600750464, "ce": 0.29521543199888356, "tv": 0.09125369730281818, "conf": 0.30038604403671343, "grad_norm": 1.3860294818878174, "lr": 2.21710042519743e-05 }, { "step": 824, "T": 532, "positions": 16, "loss": 0.2365416669181286, "ce": 0.8187873016722733, "tv": 0.17184771503525553, "conf": 0.31789883187218493, "grad_norm": 1.1909407377243042, "lr": 2.211523206134493e-05 }, { "step": 825, "T": 903, "positions": 16, "loss": 0.48375606519402936, "ce": 1.7900844853138551, "tv": 0.338608463993296, "conf": 0.8541410502257349, "grad_norm": 6.444535255432129, "lr": 2.2059474417768403e-05 }, { "step": 826, "T": 621, "positions": 16, "loss": 0.33996491113794036, "ce": 1.187317869225808, "tv": 0.24581458950706292, "conf": 1.1540147692430764, "grad_norm": 4.093491554260254, "lr": 2.2003731602414452e-05 }, { "step": 827, "T": 903, "positions": 16, "loss": 0.518802730475727, "ce": 2.3729779737315084, "tv": 0.3127832597383531, "conf": 1.3373090678032895, "grad_norm": 2.5414469242095947, "lr": 2.1948003896378002e-05 }, { "step": 828, "T": 539, "positions": 16, "loss": 0.24075817887171524, "ce": 0.9137877702077049, "tv": 0.16597712074644733, "conf": 0.7225000053682149, "grad_norm": 5.199372291564941, "lr": 2.1892291580677822e-05 }, { "step": 829, "T": 553, "positions": 16, "loss": 0.24194688834359113, "ce": 0.948460910918584, "tv": 0.16344533241681347, "conf": 0.5032804629809107, "grad_norm": 2.643609046936035, "lr": 2.1836594936255035e-05 }, { "step": 830, "T": 546, "positions": 16, "loss": 0.7655791193246841, "ce": 2.6330914720892906, "tv": 0.5580777525901794, "conf": 3.035017527639866, "grad_norm": 3.7198634147644043, "lr": 2.178091424397176e-05 }, { "step": 831, "T": 597, "positions": 16, "loss": 0.40552652909536846, "ce": 1.5850359187752474, "tv": 0.274469932919601, "conf": 0.9485689475659456, "grad_norm": 11.035550117492676, "lr": 2.1725249784609693e-05 }, { "step": 832, "T": 942, "positions": 16, "loss": 0.6572016834870738, "ce": 3.0204763647925574, "tv": 0.3946155981175252, "conf": 1.4819368116792937, "grad_norm": 3.7507059574127197, "lr": 2.166960183886863e-05 }, { "step": 833, "T": 672, "positions": 16, "loss": 0.6243205950013362, "ce": 2.3808068290818483, "tv": 0.42915547342272475, "conf": 1.297902080536005, "grad_norm": 5.144027233123779, "lr": 2.1613970687365127e-05 }, { "step": 834, "T": 527, "positions": 16, "loss": 0.5130213321535848, "ce": 1.7762223898316734, "tv": 0.3726656686048955, "conf": 1.9139065306517296, "grad_norm": 3.7165002822875977, "lr": 2.1558356610631032e-05 }, { "step": 835, "T": 589, "positions": 16, "loss": 0.12600772336554655, "ce": 0.2545014113691195, "tv": 0.11173064973445435, "conf": 0.5878390017905701, "grad_norm": 1.9194930791854858, "lr": 2.1502759889112085e-05 }, { "step": 836, "T": 541, "positions": 16, "loss": 0.49485792161431164, "ce": 1.5732465433466132, "tv": 0.3750369769986719, "conf": 1.788434343078734, "grad_norm": 2.810013771057129, "lr": 2.144718080316655e-05 }, { "step": 837, "T": 581, "positions": 16, "loss": 0.16114333004225045, "ce": 0.5278113809617935, "tv": 0.12040242966759251, "conf": 0.8095797889400274, "grad_norm": 5.2108540534973145, "lr": 2.139161963306371e-05 }, { "step": 838, "T": 678, "positions": 16, "loss": 0.4212366858355381, "ce": 1.4565725643678888, "tv": 0.30619938363088295, "conf": 1.187754424961895, "grad_norm": 2.2051610946655273, "lr": 2.1336076658982524e-05 }, { "step": 839, "T": 673, "positions": 16, "loss": 0.8547344990074635, "ce": 3.3323101475834846, "tv": 0.579448314383626, "conf": 1.516280674841255, "grad_norm": 4.2961955070495605, "lr": 2.1280552161010202e-05 }, { "step": 840, "T": 628, "positions": 16, "loss": 0.2256832003058662, "ce": 0.7326389571389882, "tv": 0.1693547933246009, "conf": 1.2493685295630712, "grad_norm": 1.8351266384124756, "lr": 2.1225046419140766e-05 }, { "step": 841, "T": 661, "positions": 16, "loss": 0.6527179852128029, "ce": 2.4110411521978676, "tv": 0.4573487367015332, "conf": 1.5107235075702192, "grad_norm": 1.7736269235610962, "lr": 2.1169559713273678e-05 }, { "step": 842, "T": 544, "positions": 16, "loss": 0.3161604686101782, "ce": 0.833091087783032, "tv": 0.25872372834419366, "conf": 0.7613896441378395, "grad_norm": 4.249778747558594, "lr": 2.1114092323212384e-05 }, { "step": 843, "T": 538, "positions": 16, "loss": 0.13354719925337122, "ce": 0.3970638827763082, "tv": 0.104267563265239, "conf": 0.31242685347631316, "grad_norm": 1.5196106433868408, "lr": 2.1058644528662945e-05 }, { "step": 844, "T": 539, "positions": 16, "loss": 0.12500468800431008, "ce": 0.33416309040632086, "tv": 0.1017648632740702, "conf": 0.7601879526734927, "grad_norm": 2.770392656326294, "lr": 2.1003216609232597e-05 }, { "step": 845, "T": 1009, "positions": 16, "loss": 0.8096230278606527, "ce": 3.9942852556341677, "tv": 0.4557716809285921, "conf": 1.2886346217273967, "grad_norm": 2.6970291137695312, "lr": 2.0947808844428346e-05 }, { "step": 846, "T": 531, "positions": 16, "loss": 0.37812696275068447, "ce": 1.7079179962165654, "tv": 0.2303724106750451, "conf": 1.11989353178069, "grad_norm": 3.273716926574707, "lr": 2.0892421513655596e-05 }, { "step": 847, "T": 537, "positions": 16, "loss": 0.0560228515153085, "ce": 0.1751783070365036, "tv": 0.04278336165543806, "conf": 0.3678827364676204, "grad_norm": 1.2053688764572144, "lr": 2.0837054896216656e-05 }, { "step": 848, "T": 531, "positions": 16, "loss": 0.5518440842602104, "ce": 2.244147752424972, "tv": 0.3638103691064316, "conf": 0.6601789731284953, "grad_norm": 3.33042049407959, "lr": 2.0781709271309423e-05 }, { "step": 849, "T": 629, "positions": 16, "loss": 0.4950705789597123, "ce": 1.6872377501713345, "tv": 0.3626075679785572, "conf": 1.5445825017717425, "grad_norm": 3.6526920795440674, "lr": 2.0726384918025934e-05 }, { "step": 850, "T": 674, "positions": 16, "loss": 0.6061845817603171, "ce": 2.3288672864437103, "tv": 0.41477541299536824, "conf": 1.6219578423188068, "grad_norm": 3.2317123413085938, "lr": 2.067108211535093e-05 }, { "step": 851, "T": 674, "positions": 16, "loss": 0.20132709168228757, "ce": 0.5037566957144008, "tv": 0.16772381565715477, "conf": 0.8171881704640374, "grad_norm": 3.242424964904785, "lr": 2.061580114216052e-05 }, { "step": 852, "T": 548, "positions": 16, "loss": 0.1853159615915274, "ce": 0.4626734202998648, "tv": 0.15449846872417083, "conf": 0.7332409073860617, "grad_norm": 4.662702560424805, "lr": 2.0560542277220686e-05 }, { "step": 853, "T": 539, "positions": 16, "loss": 0.2861317007118487, "ce": 0.8485658451354539, "tv": 0.22363901721473667, "conf": 1.0481050503653613, "grad_norm": 2.239448308944702, "lr": 2.0505305799185966e-05 }, { "step": 854, "T": 672, "positions": 16, "loss": 0.7928639706224203, "ce": 2.9465925693511963, "tv": 0.5535608045756817, "conf": 1.6774851940572262, "grad_norm": 3.5301706790924072, "lr": 2.0450091986597988e-05 }, { "step": 855, "T": 639, "positions": 16, "loss": 0.33399219036800787, "ce": 0.9017061124322936, "tv": 0.27091287047369406, "conf": 1.3221973506733775, "grad_norm": 25.75678062438965, "lr": 2.039490111788408e-05 }, { "step": 856, "T": 672, "positions": 16, "loss": 0.6112253388491808, "ce": 1.8341563799185678, "tv": 0.4753441306966124, "conf": 1.7891585703282544, "grad_norm": 4.772465229034424, "lr": 2.0339733471355883e-05 }, { "step": 857, "T": 539, "positions": 16, "loss": 0.24660022037551244, "ce": 0.8561135058648688, "tv": 0.17887653463731112, "conf": 0.48502091836276795, "grad_norm": 2.9597175121307373, "lr": 2.0284589325207935e-05 }, { "step": 858, "T": 591, "positions": 16, "loss": 0.7291877577081323, "ce": 2.6872992468997836, "tv": 0.5116198193281889, "conf": 1.39089182747648, "grad_norm": 2.4594500064849854, "lr": 2.022946895751625e-05 }, { "step": 859, "T": 584, "positions": 16, "loss": 0.3226905147530488, "ce": 1.2944547416051364, "tv": 0.21471671537074144, "conf": 1.2342251698415794, "grad_norm": 3.8971006870269775, "lr": 2.0174372646236968e-05 }, { "step": 860, "T": 591, "positions": 16, "loss": 0.27279512539098505, "ce": 0.6503529421097483, "tv": 0.23084426371497102, "conf": 1.3662050119019113, "grad_norm": 3.452284097671509, "lr": 2.0119300669204867e-05 }, { "step": 861, "T": 568, "positions": 16, "loss": 0.7514742084313184, "ce": 3.06250869424548, "tv": 0.4946926007978618, "conf": 1.4435851099624415, "grad_norm": 4.787016868591309, "lr": 2.0064253304132077e-05 }, { "step": 862, "T": 538, "positions": 16, "loss": 0.4062996956245115, "ce": 1.480911323768396, "tv": 0.28689841208051803, "conf": 1.965113753066008, "grad_norm": 2.6404786109924316, "lr": 2.000923082860658e-05 }, { "step": 863, "T": 598, "positions": 16, "loss": 0.1951336731272022, "ce": 0.7190133580551219, "tv": 0.13692483008890122, "conf": 0.489193212684836, "grad_norm": 1.8874790668487549, "lr": 1.9954233520090843e-05 }, { "step": 864, "T": 528, "positions": 16, "loss": 0.24405508356176142, "ce": 0.9582622310818465, "tv": 0.16469873798632761, "conf": 0.5574391554018803, "grad_norm": 1.104416012763977, "lr": 1.989926165592044e-05 }, { "step": 865, "T": 527, "positions": 16, "loss": 0.3773689986710451, "ce": 1.1506905227893185, "tv": 0.2914443966174076, "conf": 2.4528979864209575, "grad_norm": 3.4586219787597656, "lr": 1.9844315513302627e-05 }, { "step": 866, "T": 559, "positions": 16, "loss": 0.42974370574302156, "ce": 2.1861356930738793, "tv": 0.23458904875769804, "conf": 0.5562762801200734, "grad_norm": 4.246399879455566, "lr": 1.9789395369314954e-05 }, { "step": 867, "T": 525, "positions": 16, "loss": 0.31021152359335247, "ce": 1.0893377138236247, "tv": 0.22364196005673875, "conf": 1.5701835922802871, "grad_norm": 5.332109451293945, "lr": 1.9734501500903885e-05 }, { "step": 868, "T": 539, "positions": 16, "loss": 0.36736338055379747, "ce": 1.2848503458549203, "tv": 0.2654203993788542, "conf": 0.8370603971161472, "grad_norm": 4.011054039001465, "lr": 1.967963418488335e-05 }, { "step": 869, "T": 1627, "positions": 16, "loss": 0.7437009366722123, "ce": 2.9443775626067463, "tv": 0.499181311879056, "conf": 2.3984136656669577, "grad_norm": 4.437506675720215, "lr": 1.962479369793342e-05 }, { "step": 870, "T": 565, "positions": 16, "loss": 0.3415339687089727, "ce": 1.1945032323292253, "tv": 0.24675960127024155, "conf": 1.1796510173317074, "grad_norm": 2.597386360168457, "lr": 1.9569980316598838e-05 }, { "step": 871, "T": 613, "positions": 16, "loss": 0.257448331132764, "ce": 0.8074216016984792, "tv": 0.19634019404475112, "conf": 1.2265570238232613, "grad_norm": 3.5141916275024414, "lr": 1.9515194317287686e-05 }, { "step": 872, "T": 573, "positions": 16, "loss": 0.103078576423286, "ce": 0.1724013613088573, "tv": 0.09537604386969178, "conf": 1.9023902765475214, "grad_norm": 1.8329298496246338, "lr": 1.946043597626997e-05 }, { "step": 873, "T": 676, "positions": 16, "loss": 0.38738783156804857, "ce": 1.2478221601722908, "tv": 0.29178402964316774, "conf": 2.2769880258980932, "grad_norm": 2.8289153575897217, "lr": 1.9405705569676206e-05 }, { "step": 874, "T": 870, "positions": 16, "loss": 0.49452150475553935, "ce": 2.3179438559163827, "tv": 0.2919190290522238, "conf": 2.745668367475446, "grad_norm": 2.331111431121826, "lr": 1.935100337349604e-05 }, { "step": 875, "T": 640, "positions": 16, "loss": 0.6914254918228835, "ce": 2.5885235166642815, "tv": 0.48063680343329906, "conf": 1.3059156048111618, "grad_norm": 2.9565281867980957, "lr": 1.9296329663576906e-05 }, { "step": 876, "T": 659, "positions": 16, "loss": 0.6045110226259567, "ce": 1.9893382572336122, "tv": 0.4506413421477191, "conf": 2.871263633656781, "grad_norm": 2.0844297409057617, "lr": 1.924168471562252e-05 }, { "step": 877, "T": 615, "positions": 16, "loss": 0.3692479822484529, "ce": 1.3445913802693212, "tv": 0.2608765050695183, "conf": 1.4846276771634077, "grad_norm": 4.447091579437256, "lr": 1.9187068805191624e-05 }, { "step": 878, "T": 526, "positions": 16, "loss": 0.6826562881469727, "ce": 2.5159790692850947, "tv": 0.4789537526667118, "conf": 1.584316795124323, "grad_norm": 2.433582305908203, "lr": 1.9132482207696488e-05 }, { "step": 879, "T": 1042, "positions": 16, "loss": 0.5749171863790252, "ce": 2.6684483604174147, "tv": 0.3423026261589257, "conf": 1.0215639536909293, "grad_norm": 1.9134539365768433, "lr": 1.9077925198401585e-05 }, { "step": 880, "T": 581, "positions": 16, "loss": 0.4249559714808129, "ce": 1.854942025966011, "tv": 0.26606864371569827, "conf": 0.9347687792032957, "grad_norm": 6.474224090576172, "lr": 1.9023398052422198e-05 }, { "step": 881, "T": 946, "positions": 16, "loss": 0.633128211600706, "ce": 2.7432086803310085, "tv": 0.39867481676628813, "conf": 2.2118651197231998, "grad_norm": 4.501487731933594, "lr": 1.8968901044722974e-05 }, { "step": 882, "T": 669, "positions": 16, "loss": 0.5439845994114876, "ce": 2.0148229643236846, "tv": 0.3805581205524504, "conf": 1.3836165369430091, "grad_norm": 4.084130764007568, "lr": 1.8914434450116625e-05 }, { "step": 883, "T": 959, "positions": 16, "loss": 0.6543728959659347, "ce": 2.5002240356989205, "tv": 0.4492783281893935, "conf": 2.3216262314905975, "grad_norm": 5.2760515213012695, "lr": 1.8859998543262474e-05 }, { "step": 884, "T": 538, "positions": 16, "loss": 0.5752701964170228, "ce": 2.0365910672512655, "tv": 0.412901210439486, "conf": 1.2017247787024417, "grad_norm": 3.263841390609741, "lr": 1.88055935986651e-05 }, { "step": 885, "T": 538, "positions": 16, "loss": 0.09190896307154617, "ce": 0.261996462341358, "tv": 0.0730103512378264, "conf": 0.4564651178661734, "grad_norm": 4.735814094543457, "lr": 1.8751219890672972e-05 }, { "step": 886, "T": 564, "positions": 16, "loss": 0.8183849882334471, "ce": 3.226383096072823, "tv": 0.5508296578191221, "conf": 1.903225532732904, "grad_norm": 4.976960182189941, "lr": 1.8696877693476985e-05 }, { "step": 887, "T": 581, "positions": 16, "loss": 0.1397488426300697, "ce": 0.4539995896629989, "tv": 0.10483209363883361, "conf": 0.8152273860177957, "grad_norm": 1.3968167304992676, "lr": 1.8642567281109195e-05 }, { "step": 888, "T": 545, "positions": 16, "loss": 0.6030458856839687, "ce": 1.9357908588135615, "tv": 0.4549631243571639, "conf": 2.1095454109599814, "grad_norm": 2.90143084526062, "lr": 1.8588288927441334e-05 }, { "step": 889, "T": 525, "positions": 16, "loss": 0.7804150003939867, "ce": 2.925562800373882, "tv": 0.5420652581378818, "conf": 2.0491980663500726, "grad_norm": 1.895507574081421, "lr": 1.85340429061835e-05 }, { "step": 890, "T": 565, "positions": 16, "loss": 0.39527685098732945, "ce": 1.9563214163385965, "tv": 0.22182746148155275, "conf": 0.9765361223895983, "grad_norm": 2.231282949447632, "lr": 1.8479829490882733e-05 }, { "step": 891, "T": 554, "positions": 16, "loss": 0.21656184190032945, "ce": 0.7129598850946195, "tv": 0.16140650330271455, "conf": 0.571921103585737, "grad_norm": 1.9014092683792114, "lr": 1.842564895492164e-05 }, { "step": 892, "T": 1014, "positions": 16, "loss": 0.5256970555128646, "ce": 2.2456417574351235, "tv": 0.3345920894644223, "conf": 1.8749372059373854, "grad_norm": 2.192777633666992, "lr": 1.837150157151706e-05 }, { "step": 893, "T": 541, "positions": 16, "loss": 0.18602626055144356, "ce": 0.7404683130420153, "tv": 0.1244215971819358, "conf": 0.22135013498336775, "grad_norm": 1.7072606086730957, "lr": 1.831738761371863e-05 }, { "step": 894, "T": 538, "positions": 16, "loss": 0.16749428120601806, "ce": 0.45238422849132576, "tv": 0.1358398448501248, "conf": 0.7115478456398705, "grad_norm": 3.886427640914917, "lr": 1.826330735440742e-05 }, { "step": 895, "T": 585, "positions": 16, "loss": 0.5110417919840984, "ce": 1.9339861555076823, "tv": 0.3529368752024311, "conf": 2.1906324116057334, "grad_norm": 4.770397186279297, "lr": 1.8209261066294598e-05 }, { "step": 896, "T": 535, "positions": 16, "loss": 0.34882412160777676, "ce": 1.275385194198293, "tv": 0.24587290706449494, "conf": 0.8077164868287423, "grad_norm": 4.553905963897705, "lr": 1.815524902191999e-05 }, { "step": 897, "T": 534, "positions": 16, "loss": 0.30796912338701077, "ce": 1.0989506218174938, "tv": 0.22008229585480876, "conf": 0.6873140626825034, "grad_norm": 3.327796697616577, "lr": 1.8101271493650784e-05 }, { "step": 898, "T": 674, "positions": 16, "loss": 0.7225056504830718, "ce": 2.6728183906525373, "tv": 0.5058042295277119, "conf": 2.0208685132674873, "grad_norm": 2.721087694168091, "lr": 1.8047328753680083e-05 }, { "step": 899, "T": 1629, "positions": 16, "loss": 0.7202721293906507, "ce": 3.706941207046839, "tv": 0.38841999053329346, "conf": 1.6267035190408734, "grad_norm": 1.6646593809127808, "lr": 1.799342107402556e-05 }, { "step": 900, "T": 529, "positions": 16, "loss": 0.5280923675745726, "ce": 1.632250438327901, "tv": 0.40540815237909555, "conf": 1.2266803784295917, "grad_norm": 1.6784534454345703, "lr": 1.793954872652811e-05 }, { "step": 901, "T": 604, "positions": 16, "loss": 0.26959031048340876, "ce": 0.8433125986969117, "tv": 0.20584339773267857, "conf": 1.1399762185070943, "grad_norm": 2.6136012077331543, "lr": 1.7885711982850432e-05 }, { "step": 902, "T": 615, "positions": 16, "loss": 0.25215474809465377, "ce": 0.8690709227839761, "tv": 0.1836085195382111, "conf": 1.2696718390116075, "grad_norm": 2.090580940246582, "lr": 1.7831911114475715e-05 }, { "step": 903, "T": 563, "positions": 16, "loss": 0.19454282341030193, "ce": 0.7494821704586343, "tv": 0.13288290484342724, "conf": 0.36236984401875816, "grad_norm": 1.166440486907959, "lr": 1.777814639270622e-05 }, { "step": 904, "T": 572, "positions": 16, "loss": 0.32334923853341024, "ce": 1.1132313454872929, "tv": 0.23558455290913116, "conf": 1.0263182294049784, "grad_norm": 3.796642780303955, "lr": 1.7724418088661925e-05 }, { "step": 905, "T": 591, "positions": 16, "loss": 0.27352962547638526, "ce": 0.7870154065732322, "tv": 0.21647565751209186, "conf": 1.2534785905515662, "grad_norm": 4.501675605773926, "lr": 1.7670726473279193e-05 }, { "step": 906, "T": 595, "positions": 16, "loss": 0.3110239764500875, "ce": 1.086504455357499, "tv": 0.2248594897100702, "conf": 0.7439503240748309, "grad_norm": 5.9515557289123535, "lr": 1.7617071817309333e-05 }, { "step": 907, "T": 541, "positions": 16, "loss": 0.3920064420358358, "ce": 1.1412136523786103, "tv": 0.3087611993362316, "conf": 0.926622628920198, "grad_norm": 2.5122835636138916, "lr": 1.756345439131732e-05 }, { "step": 908, "T": 543, "positions": 16, "loss": 0.5045504609588534, "ce": 1.6013479256653227, "tv": 0.38268408738076687, "conf": 1.3550310375867411, "grad_norm": 4.236466884613037, "lr": 1.7509874465680377e-05 }, { "step": 909, "T": 678, "positions": 16, "loss": 0.5679079200854176, "ce": 2.321087519880848, "tv": 0.37311019791377475, "conf": 1.3287891717627645, "grad_norm": 2.3939578533172607, "lr": 1.7456332310586597e-05 }, { "step": 910, "T": 592, "positions": 16, "loss": 0.11503946717311919, "ce": 0.2938710789539982, "tv": 0.09516928660104895, "conf": 1.1047814190387726, "grad_norm": 1.2399131059646606, "lr": 1.7402828196033645e-05 }, { "step": 911, "T": 633, "positions": 16, "loss": 0.6111225658096373, "ce": 2.2248722766526043, "tv": 0.4318170533515513, "conf": 1.2708287016139366, "grad_norm": 2.871213912963867, "lr": 1.734936239182734e-05 }, { "step": 912, "T": 531, "positions": 16, "loss": 0.2563933781329979, "ce": 0.6506368883428877, "tv": 0.2125885510831722, "conf": 0.4206057387527835, "grad_norm": 1.2578068971633911, "lr": 1.729593516758031e-05 }, { "step": 913, "T": 576, "positions": 16, "loss": 0.3022481369187062, "ce": 0.8767059718043129, "tv": 0.2384194975961691, "conf": 1.5784957824580488, "grad_norm": 2.5285234451293945, "lr": 1.724254679271065e-05 }, { "step": 914, "T": 535, "positions": 16, "loss": 0.37186174174712505, "ce": 1.6019020616986381, "tv": 0.23519060875696596, "conf": 0.5381380202594528, "grad_norm": 4.314060211181641, "lr": 1.7189197536440524e-05 }, { "step": 915, "T": 606, "positions": 16, "loss": 0.6059039549845693, "ce": 2.306332088875479, "tv": 0.4169674981949356, "conf": 1.3897733822816463, "grad_norm": 2.606238842010498, "lr": 1.713588766779485e-05 }, { "step": 916, "T": 577, "positions": 16, "loss": 0.24662006954804383, "ce": 0.5349021967425358, "tv": 0.21458873133269663, "conf": 1.6783935439548525, "grad_norm": 4.878187656402588, "lr": 1.708261745559994e-05 }, { "step": 917, "T": 531, "positions": 16, "loss": 0.3276810435072548, "ce": 1.23830197466782, "tv": 0.2265009502389148, "conf": 1.7479443932317031, "grad_norm": 6.320509433746338, "lr": 1.7029387168482098e-05 }, { "step": 918, "T": 667, "positions": 16, "loss": 0.30028589828771146, "ce": 1.0467532066499672, "tv": 0.2173450767616032, "conf": 0.6540352919814723, "grad_norm": 3.4161102771759033, "lr": 1.6976197074866315e-05 }, { "step": 919, "T": 639, "positions": 16, "loss": 0.6928204561118037, "ce": 2.6098030316643417, "tv": 0.4798223930411041, "conf": 1.3141267811879516, "grad_norm": 3.323446750640869, "lr": 1.6923047442974893e-05 }, { "step": 920, "T": 527, "positions": 16, "loss": 0.6638621124438941, "ce": 2.3782899187644944, "tv": 0.47337014600634575, "conf": 1.5672887470573187, "grad_norm": 2.308319091796875, "lr": 1.6869938540826098e-05 }, { "step": 921, "T": 673, "positions": 16, "loss": 0.8848968427628279, "ce": 3.4638056287076324, "tv": 0.598351439461112, "conf": 0.9492841311657685, "grad_norm": 2.5933749675750732, "lr": 1.6816870636232816e-05 }, { "step": 922, "T": 599, "positions": 16, "loss": 0.1495353406862705, "ce": 0.41165210638030203, "tv": 0.1204112591658486, "conf": 0.4869412512951996, "grad_norm": 1.4569536447525024, "lr": 1.676384399680118e-05 }, { "step": 923, "T": 533, "positions": 16, "loss": 0.17129953390303854, "ce": 0.585563281387806, "tv": 0.12527023224356526, "conf": 0.24847139418125153, "grad_norm": 0.8126992583274841, "lr": 1.6710858889929255e-05 }, { "step": 924, "T": 535, "positions": 16, "loss": 0.33163970732857706, "ce": 0.8995305579373962, "tv": 0.2685407281132939, "conf": 0.6440838972339407, "grad_norm": 1.8812705278396606, "lr": 1.6657915582805634e-05 }, { "step": 925, "T": 584, "positions": 16, "loss": 0.6316597554832697, "ce": 2.458145539276302, "tv": 0.42871690448373556, "conf": 1.970923156477511, "grad_norm": 8.116971015930176, "lr": 1.6605014342408156e-05 }, { "step": 926, "T": 568, "positions": 16, "loss": 0.5397671563841868, "ce": 1.9978472529473947, "tv": 0.377758264367003, "conf": 1.4572517620620715, "grad_norm": 2.8457794189453125, "lr": 1.655215543550253e-05 }, { "step": 927, "T": 1111, "positions": 16, "loss": 0.5029668063943973, "ce": 1.809390456034862, "tv": 0.3578086359630106, "conf": 1.898579340413562, "grad_norm": 2.7284278869628906, "lr": 1.6499339128640977e-05 }, { "step": 928, "T": 673, "positions": 16, "loss": 0.5786360469646752, "ce": 2.102061561308801, "tv": 0.4093665424734354, "conf": 1.512710258597508, "grad_norm": 4.914899826049805, "lr": 1.6446565688160897e-05 }, { "step": 929, "T": 614, "positions": 16, "loss": 0.5049655985931167, "ce": 1.8422648833163748, "tv": 0.3563767960608857, "conf": 1.8321871231312215, "grad_norm": 2.9350085258483887, "lr": 1.6393835380183543e-05 }, { "step": 930, "T": 526, "positions": 16, "loss": 0.45623282296583056, "ce": 1.376706133596599, "tv": 0.35395801393315196, "conf": 2.237009085714817, "grad_norm": 4.864664554595947, "lr": 1.6341148470612634e-05 }, { "step": 931, "T": 532, "positions": 16, "loss": 0.1300765871819749, "ce": 0.5181501088649156, "tv": 0.08695730939325585, "conf": 0.8315266862600879, "grad_norm": 1.7003211975097656, "lr": 1.6288505225133082e-05 }, { "step": 932, "T": 624, "positions": 16, "loss": 0.4083111455474864, "ce": 1.1649338387669559, "tv": 0.32424196238025615, "conf": 1.5359497997706058, "grad_norm": 5.538665294647217, "lr": 1.6235905909209582e-05 }, { "step": 933, "T": 530, "positions": 16, "loss": 0.575309253334126, "ce": 2.1538313949567964, "tv": 0.399917918475694, "conf": 1.0464426040596493, "grad_norm": 2.8597519397735596, "lr": 1.6183350788085317e-05 }, { "step": 934, "T": 546, "positions": 16, "loss": 0.3250294230256259, "ce": 1.120586742839123, "tv": 0.236634179855173, "conf": 0.6878878825527863, "grad_norm": 1.5330994129180908, "lr": 1.6130840126780613e-05 }, { "step": 935, "T": 672, "positions": 16, "loss": 0.6558556770905852, "ce": 2.781978003680706, "tv": 0.41961987130343914, "conf": 2.828047689050436, "grad_norm": 3.35626220703125, "lr": 1.6078374190091582e-05 }, { "step": 936, "T": 528, "positions": 16, "loss": 0.6502847775700502, "ce": 2.2744485271105077, "tv": 0.4698221343278419, "conf": 1.201349443057552, "grad_norm": 3.9431729316711426, "lr": 1.6025953242588834e-05 }, { "step": 937, "T": 536, "positions": 16, "loss": 0.09897145215381897, "ce": 0.2876774493826133, "tv": 0.07800412375218002, "conf": 0.19711639383604052, "grad_norm": 0.5873656272888184, "lr": 1.5973577548616063e-05 }, { "step": 938, "T": 1543, "positions": 16, "loss": 0.8214353052244405, "ce": 3.7959392833499805, "tv": 0.49093484054901637, "conf": 1.0962961252153036, "grad_norm": 2.878342628479004, "lr": 1.592124737228881e-05 }, { "step": 939, "T": 673, "positions": 16, "loss": 0.7306809122674167, "ce": 3.0053230449557304, "tv": 0.47794289514422417, "conf": 1.0996750377516946, "grad_norm": 5.002344608306885, "lr": 1.5868962977493066e-05 }, { "step": 940, "T": 587, "positions": 16, "loss": 0.16454224806875573, "ce": 0.43527687986306773, "tv": 0.1344606287402712, "conf": 1.0770209720358253, "grad_norm": 2.4540765285491943, "lr": 1.5816724627883938e-05 }, { "step": 941, "T": 529, "positions": 16, "loss": 0.3244390055369877, "ce": 0.8932865530123308, "tv": 0.2612337216723972, "conf": 1.1262477280312178, "grad_norm": 5.321770191192627, "lr": 1.5764532586884384e-05 }, { "step": 942, "T": 981, "positions": 16, "loss": 0.8853183872997761, "ce": 3.629137299954891, "tv": 0.580449627712369, "conf": 2.3522561448189663, "grad_norm": 3.066476821899414, "lr": 1.571238711768379e-05 }, { "step": 943, "T": 557, "positions": 16, "loss": 0.3558544437837554, "ce": 1.3290770120511297, "tv": 0.2477186083997367, "conf": 0.8059005292943766, "grad_norm": 3.072587013244629, "lr": 1.566028848323674e-05 }, { "step": 944, "T": 543, "positions": 16, "loss": 0.40095049075034694, "ce": 1.3720442800790806, "tv": 0.2930511866652523, "conf": 1.0553704203070904, "grad_norm": 5.328531742095947, "lr": 1.5608236946261623e-05 }, { "step": 945, "T": 955, "positions": 16, "loss": 0.4829652744192572, "ce": 1.8716846407360208, "tv": 0.3286631460323406, "conf": 1.8748119130614214, "grad_norm": 2.254035234451294, "lr": 1.5556232769239316e-05 }, { "step": 946, "T": 536, "positions": 16, "loss": 0.22699033126991708, "ce": 0.5677372274039953, "tv": 0.18912955724226777, "conf": 1.1530601308922428, "grad_norm": 2.6246516704559326, "lr": 1.5504276214411897e-05 }, { "step": 947, "T": 952, "positions": 16, "loss": 0.6684504920849577, "ce": 2.880431816204805, "tv": 0.42267480455484474, "conf": 1.1412750719523501, "grad_norm": 2.3385963439941406, "lr": 1.54523675437813e-05 }, { "step": 948, "T": 960, "positions": 16, "loss": 0.7265112947206944, "ce": 3.248592380201444, "tv": 0.44628007935534697, "conf": 1.814048891849552, "grad_norm": 2.4141156673431396, "lr": 1.540050701910796e-05 }, { "step": 949, "T": 525, "positions": 16, "loss": 0.77143987827003, "ce": 3.1151207222719677, "tv": 0.511030909314286, "conf": 2.107362350734917, "grad_norm": 3.480419874191284, "lr": 1.5348694901909567e-05 }, { "step": 950, "T": 1031, "positions": 16, "loss": 0.7649549469351768, "ce": 3.2405992597341537, "tv": 0.4898833530023694, "conf": 1.5119648539694026, "grad_norm": 2.1788930892944336, "lr": 1.5296931453459672e-05 }, { "step": 951, "T": 674, "positions": 16, "loss": 0.513401404954493, "ce": 1.6063991310074925, "tv": 0.3919572136364877, "conf": 1.7414028899511322, "grad_norm": 3.0481834411621094, "lr": 1.5245216934786428e-05 }, { "step": 952, "T": 1015, "positions": 16, "loss": 0.7654016022570431, "ce": 3.0157153666950762, "tv": 0.515366739127785, "conf": 2.111744059395278, "grad_norm": 2.947173833847046, "lr": 1.5193551606671244e-05 }, { "step": 953, "T": 540, "positions": 16, "loss": 0.6632138788700104, "ce": 2.034555787802674, "tv": 0.5108425617218018, "conf": 1.5671057023864705, "grad_norm": 2.0521316528320312, "lr": 1.5141935729647461e-05 }, { "step": 954, "T": 557, "positions": 16, "loss": 0.5702131011057645, "ce": 2.1365846019471064, "tv": 0.3961718196514994, "conf": 1.0798782766942168, "grad_norm": 2.3197009563446045, "lr": 1.5090369563999074e-05 }, { "step": 955, "T": 672, "positions": 16, "loss": 0.5208264620014234, "ce": 2.040207960523958, "tv": 0.35200629029986885, "conf": 1.634967562538236, "grad_norm": 2.9298648834228516, "lr": 1.5038853369759364e-05 }, { "step": 956, "T": 531, "positions": 16, "loss": 0.67098627705127, "ce": 2.1633055477868766, "tv": 0.5051730233244598, "conf": 1.5064888675988186, "grad_norm": 2.562685489654541, "lr": 1.4987387406709669e-05 }, { "step": 957, "T": 536, "positions": 16, "loss": 0.19098508026945638, "ce": 0.5522657095119712, "tv": 0.15084279397706268, "conf": 0.4595920476290303, "grad_norm": 3.0079855918884277, "lr": 1.4935971934377993e-05 }, { "step": 958, "T": 532, "positions": 16, "loss": 0.9067898653447628, "ce": 3.788446720689535, "tv": 0.5866057723760605, "conf": 1.3415778539965686, "grad_norm": 4.212545871734619, "lr": 1.4884607212037726e-05 }, { "step": 959, "T": 536, "positions": 16, "loss": 0.20262546998674225, "ce": 0.5866286697564647, "tv": 0.15995844429471617, "conf": 0.6678133884539834, "grad_norm": 2.6387875080108643, "lr": 1.4833293498706365e-05 }, { "step": 960, "T": 554, "positions": 16, "loss": 0.44391304571399814, "ce": 1.340329393600996, "tv": 0.3443112348777504, "conf": 1.5694649446668336, "grad_norm": 2.725255012512207, "lr": 1.4782031053144158e-05 }, { "step": 961, "T": 673, "positions": 16, "loss": 0.6723438734188676, "ce": 2.7110028248280287, "tv": 0.4458262184634805, "conf": 1.1670356404501945, "grad_norm": 7.056548595428467, "lr": 1.473082013385283e-05 }, { "step": 962, "T": 630, "positions": 16, "loss": 0.31953484998393833, "ce": 1.177783138318091, "tv": 0.2241739269984464, "conf": 1.8348900188220814, "grad_norm": 2.0132265090942383, "lr": 1.4679660999074302e-05 }, { "step": 963, "T": 636, "positions": 16, "loss": 0.42268250629422255, "ce": 1.9196637840177573, "tv": 0.2563512329434161, "conf": 0.8461209061572674, "grad_norm": 5.219188690185547, "lr": 1.4628553906789322e-05 }, { "step": 964, "T": 673, "positions": 16, "loss": 0.7431747298687696, "ce": 3.1328555308282375, "tv": 0.47765466663986444, "conf": 1.6992139262147248, "grad_norm": 2.650087356567383, "lr": 1.457749911471622e-05 }, { "step": 965, "T": 677, "positions": 16, "loss": 0.4330501899821684, "ce": 1.843260942667257, "tv": 0.2763601033948362, "conf": 0.9348194051999599, "grad_norm": 7.47775936126709, "lr": 1.4526496880309592e-05 }, { "step": 966, "T": 576, "positions": 16, "loss": 0.39561772422894137, "ce": 1.36217141191446, "tv": 0.28822287572256755, "conf": 1.5752072266199804, "grad_norm": 2.439319610595703, "lr": 1.4475547460758999e-05 }, { "step": 967, "T": 541, "positions": 16, "loss": 0.6862216233275831, "ce": 2.232273028930649, "tv": 0.5144381364807487, "conf": 2.133152424765285, "grad_norm": 4.640216827392578, "lr": 1.442465111298768e-05 }, { "step": 968, "T": 612, "positions": 16, "loss": 0.14696817379444838, "ce": 0.3457167066189868, "tv": 0.12488500963081606, "conf": 1.0428198999725282, "grad_norm": 4.205510139465332, "lr": 1.4373808093651215e-05 }, { "step": 969, "T": 566, "positions": 16, "loss": 0.8682416009251028, "ce": 4.039946551609319, "tv": 0.5158299282484222, "conf": 1.3491567187239326, "grad_norm": 3.2643301486968994, "lr": 1.4323018659136314e-05 }, { "step": 970, "T": 528, "positions": 16, "loss": 0.7885962519794703, "ce": 3.001859672367573, "tv": 0.5426780907437205, "conf": 2.24987955596589, "grad_norm": 2.534027576446533, "lr": 1.427228306555945e-05 }, { "step": 971, "T": 548, "positions": 16, "loss": 0.2606933551141992, "ce": 0.8056723894551396, "tv": 0.20014011696912348, "conf": 1.0661464566706953, "grad_norm": 3.0362026691436768, "lr": 1.4221601568765574e-05 }, { "step": 972, "T": 576, "positions": 16, "loss": 0.7236112430109642, "ce": 3.18360611364551, "tv": 0.45027848877361976, "conf": 1.0524700415597734, "grad_norm": 4.955811977386475, "lr": 1.4170974424326868e-05 }, { "step": 973, "T": 1222, "positions": 16, "loss": 0.6458020256322925, "ce": 3.053390255161503, "tv": 0.37829224400411476, "conf": 2.2615735106141113, "grad_norm": 3.173525333404541, "lr": 1.4120401887541423e-05 }, { "step": 974, "T": 869, "positions": 16, "loss": 0.700813794235728, "ce": 3.0699173578911996, "tv": 0.4375800810958026, "conf": 1.4758164805207343, "grad_norm": 2.570568799972534, "lr": 1.4069884213431955e-05 }, { "step": 975, "T": 526, "positions": 16, "loss": 0.5705174368340522, "ce": 1.8951612275559455, "tv": 0.42333479691296816, "conf": 2.4574953988649213, "grad_norm": 2.545832395553589, "lr": 1.401942165674453e-05 }, { "step": 976, "T": 850, "positions": 16, "loss": 0.5919401204391761, "ce": 2.5558150010401732, "tv": 0.3737318028579466, "conf": 0.9011091102138256, "grad_norm": 1.9132792949676514, "lr": 1.3969014471947273e-05 }, { "step": 977, "T": 907, "positions": 16, "loss": 0.48827229051039467, "ce": 1.9709137189358898, "tv": 0.32353436716357464, "conf": 1.1585553190348845, "grad_norm": 3.086113452911377, "lr": 1.3918662913229082e-05 }, { "step": 978, "T": 1184, "positions": 16, "loss": 0.8071488754358143, "ce": 3.638990444596857, "tv": 0.4924998313654214, "conf": 1.5611227450353908, "grad_norm": 3.5687859058380127, "lr": 1.3868367234498328e-05 }, { "step": 979, "T": 592, "positions": 16, "loss": 0.3390795522136614, "ce": 1.1526562171202386, "tv": 0.24868213683657814, "conf": 1.0168457818576488, "grad_norm": 3.84572696685791, "lr": 1.3818127689381638e-05 }, { "step": 980, "T": 531, "positions": 16, "loss": 0.35163418336742325, "ce": 1.3775689056665215, "tv": 0.23764145233417366, "conf": 1.8700710056402219, "grad_norm": 2.092632532119751, "lr": 1.3767944531222553e-05 }, { "step": 981, "T": 599, "positions": 16, "loss": 0.2028713028958009, "ce": 0.6253142502249318, "tv": 0.15593320068546745, "conf": 1.3619718568029953, "grad_norm": 4.152412414550781, "lr": 1.3717818013080252e-05 }, { "step": 982, "T": 527, "positions": 16, "loss": 0.553064298350364, "ce": 1.8344626056496054, "tv": 0.4106867266818881, "conf": 1.6060104229545686, "grad_norm": 2.064171314239502, "lr": 1.3667748387728308e-05 }, { "step": 983, "T": 616, "positions": 16, "loss": 0.49568960070610046, "ce": 1.9305166881531477, "tv": 0.33626437140628695, "conf": 1.6631727032363415, "grad_norm": 7.171928405761719, "lr": 1.3617735907653434e-05 }, { "step": 984, "T": 533, "positions": 16, "loss": 0.4518363546976616, "ce": 1.9362984662420786, "tv": 0.2868961410367774, "conf": 0.8372486797887717, "grad_norm": 2.019270420074463, "lr": 1.356778082505411e-05 }, { "step": 985, "T": 626, "positions": 16, "loss": 0.2762168162632861, "ce": 0.8710080038712817, "tv": 0.2101289112441691, "conf": 1.4933257156211743, "grad_norm": 2.7977793216705322, "lr": 1.351788339183943e-05 }, { "step": 986, "T": 638, "positions": 16, "loss": 0.3145216504772179, "ce": 1.1360326797298796, "tv": 0.223242656252296, "conf": 1.267004005421768, "grad_norm": 3.8210761547088623, "lr": 1.346804385962776e-05 }, { "step": 987, "T": 535, "positions": 16, "loss": 0.31977581080809614, "ce": 1.3102469364143872, "tv": 0.20972347231463573, "conf": 0.5939920820283078, "grad_norm": 3.3617970943450928, "lr": 1.3418262479745483e-05 }, { "step": 988, "T": 1429, "positions": 16, "loss": 0.7211282012831361, "ce": 3.1248124986204857, "tv": 0.4540521832968807, "conf": 2.220594997592343, "grad_norm": 1.7808338403701782, "lr": 1.3368539503225746e-05 }, { "step": 989, "T": 675, "positions": 16, "loss": 0.38961016574830865, "ce": 1.2847579818935628, "tv": 0.29014930033736164, "conf": 1.801063488452999, "grad_norm": 2.815635919570923, "lr": 1.3318875180807186e-05 }, { "step": 990, "T": 529, "positions": 16, "loss": 0.8279534098801378, "ce": 3.9840028721092153, "tv": 0.4772812377836999, "conf": 1.185784893008531, "grad_norm": 2.5108449459075928, "lr": 1.3269269762932662e-05 }, { "step": 991, "T": 673, "positions": 16, "loss": 0.5888206639647251, "ce": 2.3621864434244344, "tv": 0.3917800522103789, "conf": 1.1641647286448809, "grad_norm": 3.1816492080688477, "lr": 1.3219723499747976e-05 }, { "step": 992, "T": 622, "positions": 16, "loss": 0.2283089785323682, "ce": 0.7695569677359799, "tv": 0.1681703273916355, "conf": 1.443707823450726, "grad_norm": 2.9758050441741943, "lr": 1.3170236641100645e-05 }, { "step": 993, "T": 630, "positions": 16, "loss": 0.5812116715569573, "ce": 2.368832079017011, "tv": 0.3825871764711337, "conf": 2.352319408065547, "grad_norm": 3.493013858795166, "lr": 1.3120809436538656e-05 }, { "step": 994, "T": 586, "positions": 16, "loss": 0.5131682064311462, "ce": 1.7137393817101838, "tv": 0.37977141980081797, "conf": 2.5666012112291128, "grad_norm": 3.632387399673462, "lr": 1.3071442135309117e-05 }, { "step": 995, "T": 535, "positions": 16, "loss": 0.7719816321041435, "ce": 2.863092093469959, "tv": 0.5396360396407545, "conf": 2.0151210593467113, "grad_norm": 2.4801347255706787, "lr": 1.3022134986357104e-05 }, { "step": 996, "T": 625, "positions": 16, "loss": 0.4199623330569011, "ce": 1.7230784824605507, "tv": 0.27517166020697914, "conf": 1.7344345756318944, "grad_norm": 2.83506178855896, "lr": 1.2972888238324354e-05 }, { "step": 997, "T": 535, "positions": 16, "loss": 0.32272800221926445, "ce": 1.1221155839348285, "tv": 0.23390716585709015, "conf": 0.9935936045513913, "grad_norm": 4.872188091278076, "lr": 1.2923702139548014e-05 }, { "step": 998, "T": 553, "positions": 16, "loss": 0.4711734975799118, "ce": 1.7585717707725053, "tv": 0.32812924790960096, "conf": 2.7643951448338187, "grad_norm": 4.131604194641113, "lr": 1.2874576938059402e-05 }, { "step": 999, "T": 532, "positions": 16, "loss": 0.21492776924787904, "ce": 0.7381904413769007, "tv": 0.15678748111531604, "conf": 1.0714727838217186, "grad_norm": 1.0922367572784424, "lr": 1.2825512881582719e-05 }, { "step": 1000, "T": 536, "positions": 16, "loss": 0.11197906043435069, "ce": 0.33181749969665475, "tv": 0.0875525681894942, "conf": 0.849205820552194, "grad_norm": 2.087340831756592, "lr": 1.2776510217533883e-05 }, { "step": 1001, "T": 896, "positions": 16, "loss": 0.6050766850166838, "ce": 2.6651242652196743, "tv": 0.3761825083493022, "conf": 1.634858006108061, "grad_norm": 2.8005242347717285, "lr": 1.2727569193019193e-05 }, { "step": 1002, "T": 544, "positions": 16, "loss": 0.35600902886903896, "ce": 1.3475485919902894, "tv": 0.24583797093305293, "conf": 1.6705824167154208, "grad_norm": 4.568016052246094, "lr": 1.2678690054834096e-05 }, { "step": 1003, "T": 581, "positions": 16, "loss": 0.3411183209173032, "ce": 1.1413329743299983, "tv": 0.2522055975859985, "conf": 1.5765915508000212, "grad_norm": 2.3571901321411133, "lr": 1.2629873049462032e-05 }, { "step": 1004, "T": 672, "positions": 16, "loss": 0.7879785960540175, "ce": 3.0367836328223348, "tv": 0.5381113663315773, "conf": 1.8820420968113467, "grad_norm": 3.34287428855896, "lr": 1.2581118423073052e-05 }, { "step": 1005, "T": 587, "positions": 16, "loss": 0.1717486221168656, "ce": 0.4051773372630123, "tv": 0.1458120973838959, "conf": 1.6700939855358001, "grad_norm": 3.854687452316284, "lr": 1.2532426421522687e-05 }, { "step": 1006, "T": 674, "positions": 16, "loss": 0.5101939113810658, "ce": 1.7882207749935333, "tv": 0.3681909295846708, "conf": 1.94444192876108, "grad_norm": 3.0852813720703125, "lr": 1.248379729035066e-05 }, { "step": 1007, "T": 563, "positions": 16, "loss": 0.2256811512988861, "ce": 0.7012017381107398, "tv": 0.1728455328338896, "conf": 1.2366749970723632, "grad_norm": 2.168381690979004, "lr": 1.2435231274779654e-05 }, { "step": 1008, "T": 566, "positions": 16, "loss": 0.6007301709141757, "ce": 2.378315185491374, "tv": 0.40322074596952007, "conf": 1.0498041758207386, "grad_norm": 2.3669302463531494, "lr": 1.2386728619714091e-05 }, { "step": 1009, "T": 612, "positions": 16, "loss": 0.3744705173312468, "ce": 1.1513717109419304, "tv": 0.2881481702606834, "conf": 1.5361025219566684, "grad_norm": 3.208301544189453, "lr": 1.2338289569738848e-05 }, { "step": 1010, "T": 673, "positions": 16, "loss": 0.28950149396973757, "ce": 0.9116298121568889, "tv": 0.22037613753514051, "conf": 2.0157560934167122, "grad_norm": 1.8419896364212036, "lr": 1.2289914369118114e-05 }, { "step": 1011, "T": 537, "positions": 16, "loss": 0.7394559746608138, "ce": 2.7373008264694363, "tv": 0.5174732487648726, "conf": 2.274164168421521, "grad_norm": 3.7283918857574463, "lr": 1.2241603261794074e-05 }, { "step": 1012, "T": 657, "positions": 16, "loss": 0.2598190474677722, "ce": 0.7802889355725711, "tv": 0.20198905403594836, "conf": 1.5714203916164202, "grad_norm": 3.8528780937194824, "lr": 1.2193356491385682e-05 }, { "step": 1013, "T": 595, "positions": 16, "loss": 0.1844138333290175, "ce": 0.8349763692131091, "tv": 0.11212909913592739, "conf": 1.7270181141793728, "grad_norm": 2.152726650238037, "lr": 1.214517430118753e-05 }, { "step": 1014, "T": 970, "positions": 16, "loss": 0.7005663918789651, "ce": 3.0760066933228245, "tv": 0.4366285776341101, "conf": 1.2426132526059803, "grad_norm": 2.949167251586914, "lr": 1.209705693416848e-05 }, { "step": 1015, "T": 1741, "positions": 16, "loss": 1.0044346569484333, "ce": 4.719807989444234, "tv": 0.5916154120786814, "conf": 2.3422898471035296, "grad_norm": 4.6351470947265625, "lr": 1.2049004632970542e-05 }, { "step": 1016, "T": 574, "positions": 16, "loss": 0.7001579804345965, "ce": 2.815529078245163, "tv": 0.4651167632546276, "conf": 1.3885291752230842, "grad_norm": 2.7175979614257812, "lr": 1.2001017639907622e-05 }, { "step": 1017, "T": 581, "positions": 16, "loss": 0.4323569506614149, "ce": 1.5893094006098636, "tv": 0.303806688523764, "conf": 1.0753741206872292, "grad_norm": 2.517892360687256, "lr": 1.1953096196964283e-05 }, { "step": 1018, "T": 563, "positions": 16, "loss": 0.41496005478984443, "ce": 1.6109865742982947, "tv": 0.2820682174569811, "conf": 1.3885161774596941, "grad_norm": 5.008639335632324, "lr": 1.190524054579455e-05 }, { "step": 1019, "T": 534, "positions": 16, "loss": 0.14733765304026747, "ce": 0.44956205145942363, "tv": 0.11375715927988495, "conf": 0.7648850769619457, "grad_norm": 0.9729120135307312, "lr": 1.1857450927720662e-05 }, { "step": 1020, "T": 645, "positions": 16, "loss": 0.8722427757456899, "ce": 3.420319888740778, "tv": 0.5891230832785368, "conf": 1.7467868637759238, "grad_norm": 2.2585363388061523, "lr": 1.1809727583731887e-05 }, { "step": 1021, "T": 672, "positions": 16, "loss": 0.6426847043367161, "ce": 2.6540625101406476, "tv": 0.4191982795164222, "conf": 1.9304871583962722, "grad_norm": 4.679949760437012, "lr": 1.1762070754483298e-05 }, { "step": 1022, "T": 578, "positions": 16, "loss": 0.44147264025741606, "ce": 1.7323928365985921, "tv": 0.298037078510788, "conf": 1.3120346819991369, "grad_norm": 3.4935648441314697, "lr": 1.1714480680294521e-05 }, { "step": 1023, "T": 527, "positions": 16, "loss": 0.5407885005697608, "ce": 2.118640413507819, "tv": 0.36547162383794785, "conf": 1.5203248070320114, "grad_norm": 3.2347517013549805, "lr": 1.1666957601148576e-05 }, { "step": 1024, "T": 645, "positions": 16, "loss": 0.32700130450029974, "ce": 1.1809457363820002, "tv": 0.23211859067123441, "conf": 1.1040677804456953, "grad_norm": 2.8348710536956787, "lr": 1.1619501756690679e-05 }, { "step": 1025, "T": 1008, "positions": 16, "loss": 0.5526890620649283, "ce": 2.5323604425268513, "tv": 0.3327255793401491, "conf": 0.9517247995135047, "grad_norm": 2.0991015434265137, "lr": 1.1572113386226937e-05 }, { "step": 1026, "T": 676, "positions": 16, "loss": 0.4319489723507104, "ce": 1.3947908683122705, "tv": 0.32496653708585654, "conf": 1.1885974707838614, "grad_norm": 3.8110015392303467, "lr": 1.152479272872325e-05 }, { "step": 1027, "T": 674, "positions": 16, "loss": 0.6294170096516609, "ce": 2.4034399297088385, "tv": 0.4323033355176449, "conf": 1.0603337566471964, "grad_norm": 1.8733564615249634, "lr": 1.1477540022804046e-05 }, { "step": 1028, "T": 528, "positions": 16, "loss": 0.6615366684782202, "ce": 2.3331193714475376, "tv": 0.4758052779143327, "conf": 1.4497446658482964, "grad_norm": 2.522770404815674, "lr": 1.1430355506751095e-05 }, { "step": 1029, "T": 538, "positions": 16, "loss": 0.4948090196121484, "ce": 1.7053665408748202, "tv": 0.3603026360506192, "conf": 1.2222931887154118, "grad_norm": 3.4390430450439453, "lr": 1.1383239418502303e-05 }, { "step": 1030, "T": 572, "positions": 16, "loss": 0.6444278198887332, "ce": 2.5295151306081607, "tv": 0.4349736784670313, "conf": 1.3284230867123732, "grad_norm": 1.8686374425888062, "lr": 1.1336191995650521e-05 }, { "step": 1031, "T": 536, "positions": 16, "loss": 0.32401085399033036, "ce": 0.8244134120086528, "tv": 0.26841058063109813, "conf": 1.3953076923578027, "grad_norm": 5.313992500305176, "lr": 1.1289213475442342e-05 }, { "step": 1032, "T": 673, "positions": 16, "loss": 0.7215213966846932, "ce": 2.711326176460716, "tv": 0.5004319711297285, "conf": 1.1944537049634363, "grad_norm": 3.0193123817443848, "lr": 1.1242304094776878e-05 }, { "step": 1033, "T": 603, "positions": 16, "loss": 0.19274068407321465, "ce": 0.6979778082655912, "tv": 0.13660323156000231, "conf": 0.8988265487142773, "grad_norm": 1.3938730955123901, "lr": 1.119546409020461e-05 }, { "step": 1034, "T": 532, "positions": 16, "loss": 0.3760310971292711, "ce": 1.3914886402653792, "tv": 0.263202481688495, "conf": 0.5663492576211411, "grad_norm": 5.182271480560303, "lr": 1.1148693697926198e-05 }, { "step": 1035, "T": 618, "positions": 16, "loss": 0.3427774094743654, "ce": 1.0170096661895514, "tv": 0.26786272833123803, "conf": 1.6212209053337574, "grad_norm": 5.70198392868042, "lr": 1.1101993153791221e-05 }, { "step": 1036, "T": 539, "positions": 16, "loss": 0.31677514648254146, "ce": 1.1063486404163996, "tv": 0.22904475696850568, "conf": 1.1059897233895484, "grad_norm": 5.829536437988281, "lr": 1.1055362693297052e-05 }, { "step": 1037, "T": 569, "positions": 16, "loss": 0.38638547347363783, "ce": 1.4991783289927412, "tv": 0.26274182557608583, "conf": 1.6432329516649702, "grad_norm": 2.36411714553833, "lr": 1.1008802551587688e-05 }, { "step": 1038, "T": 589, "positions": 16, "loss": 0.6949224565178156, "ce": 2.580212207045406, "tv": 0.4854458272457123, "conf": 1.769069730420597, "grad_norm": 4.186391830444336, "lr": 1.0962312963452467e-05 }, { "step": 1039, "T": 653, "positions": 16, "loss": 0.7015447722515091, "ce": 2.703861148096621, "tv": 0.47906517854426056, "conf": 2.03773676243145, "grad_norm": 5.114706993103027, "lr": 1.0915894163324986e-05 }, { "step": 1040, "T": 675, "positions": 16, "loss": 0.6132905781269073, "ce": 2.3202164517715573, "tv": 0.4236321561038494, "conf": 1.9522880578879267, "grad_norm": 5.013914108276367, "lr": 1.0869546385281868e-05 }, { "step": 1041, "T": 599, "positions": 16, "loss": 0.7554267371888272, "ce": 3.483660488978785, "tv": 0.4522896450362168, "conf": 1.2958592369351218, "grad_norm": 4.102430820465088, "lr": 1.082326986304159e-05 }, { "step": 1042, "T": 901, "positions": 16, "loss": 0.7150992248971306, "ce": 3.101438670659718, "tv": 0.44995040700359823, "conf": 2.3526985166563392, "grad_norm": 4.924022197723389, "lr": 1.0777064829963323e-05 }, { "step": 1043, "T": 597, "positions": 16, "loss": 0.11315182043654204, "ce": 0.2737666152467284, "tv": 0.09530573474148696, "conf": 0.5844365855849674, "grad_norm": 1.6708368062973022, "lr": 1.0730931519045697e-05 }, { "step": 1044, "T": 613, "positions": 16, "loss": 0.1185707182630722, "ce": 0.21537342029751017, "tv": 0.1078148740598408, "conf": 1.5637055411934853, "grad_norm": 3.380735158920288, "lr": 1.0684870162925728e-05 }, { "step": 1045, "T": 546, "positions": 16, "loss": 0.2832438490590903, "ce": 0.7513521299256354, "tv": 0.23123181697337714, "conf": 1.6688965130737077, "grad_norm": 2.0824928283691406, "lr": 1.063888099387753e-05 }, { "step": 1046, "T": 535, "positions": 16, "loss": 0.12548770173543744, "ce": 0.30935097382473487, "tv": 0.10505844814542797, "conf": 0.41798537177822936, "grad_norm": 2.1194868087768555, "lr": 1.0592964243811227e-05 }, { "step": 1047, "T": 527, "positions": 16, "loss": 0.6963380165398121, "ce": 2.4259641016833484, "tv": 0.5041573401540518, "conf": 1.7072371032554656, "grad_norm": 2.9104278087615967, "lr": 1.054712014427175e-05 }, { "step": 1048, "T": 672, "positions": 16, "loss": 0.7412809510715306, "ce": 2.669702468207106, "tv": 0.5270119091146626, "conf": 1.4283960068132728, "grad_norm": 4.2866339683532715, "lr": 1.050134892643767e-05 }, { "step": 1049, "T": 535, "positions": 16, "loss": 0.7510312832891941, "ce": 2.8517600727500394, "tv": 0.517616979777813, "conf": 1.8380575760893407, "grad_norm": 4.352040767669678, "lr": 1.0455650821120045e-05 }, { "step": 1050, "T": 629, "positions": 16, "loss": 0.5427073631435633, "ce": 1.9131067773560062, "tv": 0.3904407531954348, "conf": 1.2088455132907256, "grad_norm": 2.101212501525879, "lr": 1.0410026058761238e-05 }, { "step": 1051, "T": 529, "positions": 16, "loss": 0.45347895537497607, "ce": 2.1315310206218783, "tv": 0.26702874962211354, "conf": 1.6386116261875259, "grad_norm": 5.1179375648498535, "lr": 1.0364474869433768e-05 }, { "step": 1052, "T": 544, "positions": 16, "loss": 0.15879880959619186, "ce": 0.5813433277761533, "tv": 0.11184942516410956, "conf": 0.18669494080586446, "grad_norm": 1.7086939811706543, "lr": 1.0318997482839162e-05 }, { "step": 1053, "T": 1433, "positions": 16, "loss": 0.859353000203555, "ce": 4.105017655121628, "tv": 0.4987236067390768, "conf": 2.467193201649934, "grad_norm": 3.502816677093506, "lr": 1.0273594128306738e-05 }, { "step": 1054, "T": 528, "positions": 16, "loss": 0.5030814387664577, "ce": 2.243677841141107, "tv": 0.3096818112426263, "conf": 1.7633323047966833, "grad_norm": 2.2475082874298096, "lr": 1.0228265034792545e-05 }, { "step": 1055, "T": 674, "positions": 16, "loss": 0.6068007165449671, "ce": 2.3265829402953386, "tv": 0.4157138148439117, "conf": 1.1868807893684243, "grad_norm": 4.058404922485352, "lr": 1.0183010430878128e-05 }, { "step": 1056, "T": 527, "positions": 16, "loss": 0.40758716877462575, "ce": 1.3518741236546248, "tv": 0.3026664098179026, "conf": 1.4626651351718465, "grad_norm": 2.9145190715789795, "lr": 1.0137830544769391e-05 }, { "step": 1057, "T": 642, "positions": 16, "loss": 0.5949213998392224, "ce": 2.1990215834230185, "tv": 0.4166880641132593, "conf": 2.2020033835433424, "grad_norm": 2.2523460388183594, "lr": 1.009272560429548e-05 }, { "step": 1058, "T": 598, "positions": 16, "loss": 0.23458493438738515, "ce": 0.8621150245688796, "tv": 0.16485937403922435, "conf": 0.8238087277350132, "grad_norm": 3.5436692237854004, "lr": 1.00476958369076e-05 }, { "step": 1059, "T": 933, "positions": 16, "loss": 0.6567809835105436, "ce": 2.5584936850591475, "tv": 0.4454795804194873, "conf": 1.8675327571108937, "grad_norm": 3.003878116607666, "lr": 1.0002741469677888e-05 }, { "step": 1060, "T": 653, "positions": 16, "loss": 0.7931092865765095, "ce": 3.0944714210927486, "tv": 0.537402369081974, "conf": 1.9384688697755337, "grad_norm": 8.960192680358887, "lr": 9.957862729298249e-06 }, { "step": 1061, "T": 655, "positions": 16, "loss": 0.5011423868127167, "ce": 1.8331672588828951, "tv": 0.35313963051885366, "conf": 1.2565019136527553, "grad_norm": 3.8788740634918213, "lr": 9.91305984207923e-06 }, { "step": 1062, "T": 585, "positions": 16, "loss": 0.2312817405017995, "ce": 0.8131751805611884, "tv": 0.1666269159568401, "conf": 0.9562126763481389, "grad_norm": 3.3840932846069336, "lr": 9.868333033948884e-06 }, { "step": 1063, "T": 641, "positions": 16, "loss": 0.7393592800945044, "ce": 2.7428705766797066, "tv": 0.5167469047009945, "conf": 1.8391723670065403, "grad_norm": 3.355664014816284, "lr": 9.82368253045158e-06 }, { "step": 1064, "T": 532, "positions": 16, "loss": 0.4664286137688123, "ce": 1.8791555005295777, "tv": 0.30945896922980864, "conf": 1.0016532467449206, "grad_norm": 2.6979968547821045, "lr": 9.779108556746955e-06 }, { "step": 1065, "T": 526, "positions": 16, "loss": 0.6793941725045443, "ce": 2.371417036280036, "tv": 0.4913916513323784, "conf": 1.5114159868098795, "grad_norm": 3.6205687522888184, "lr": 9.734611337608703e-06 }, { "step": 1066, "T": 590, "positions": 16, "loss": 0.33569145352885243, "ce": 1.1839059981875835, "tv": 0.24144539776216334, "conf": 1.521878297309513, "grad_norm": 4.647593975067139, "lr": 9.690191097423459e-06 }, { "step": 1067, "T": 527, "positions": 16, "loss": 0.5269422642886639, "ce": 1.7466754561755806, "tv": 0.3914163541048765, "conf": 1.8825495977944229, "grad_norm": 2.936160087585449, "lr": 9.645848060189688e-06 }, { "step": 1068, "T": 838, "positions": 16, "loss": 0.5702005676575936, "ce": 2.2222172382171266, "tv": 0.3866431543137878, "conf": 2.0103667490766384, "grad_norm": 4.2158122062683105, "lr": 9.601582449516538e-06 }, { "step": 1069, "T": 528, "positions": 16, "loss": 0.32658724534121575, "ce": 1.1779599437140007, "tv": 0.2319902726176224, "conf": 0.33181714979582466, "grad_norm": 1.8685827255249023, "lr": 9.557394488622726e-06 }, { "step": 1070, "T": 594, "positions": 16, "loss": 0.2337703538723872, "ce": 0.546898042637622, "tv": 0.19897839314944576, "conf": 0.9277311384420983, "grad_norm": 2.6973226070404053, "lr": 9.51328440033539e-06 }, { "step": 1071, "T": 634, "positions": 16, "loss": 0.2393525491352193, "ce": 0.7103021509246901, "tv": 0.18702482018852606, "conf": 1.6644093451591289, "grad_norm": 2.9616732597351074, "lr": 9.469252407088986e-06 }, { "step": 1072, "T": 558, "positions": 16, "loss": 0.4946611252598814, "ce": 1.822929177840706, "tv": 0.34707579181849724, "conf": 1.4071501528042063, "grad_norm": 7.139134883880615, "lr": 9.425298730924162e-06 }, { "step": 1073, "T": 649, "positions": 16, "loss": 0.35372966855356935, "ce": 1.2951746954204282, "tv": 0.24912467838294106, "conf": 1.1590053364730775, "grad_norm": 1.7824581861495972, "lr": 9.381423593486629e-06 }, { "step": 1074, "T": 1085, "positions": 16, "loss": 0.6037101449546753, "ce": 2.7430700046729726, "tv": 0.36600351084780414, "conf": 1.54891252114885, "grad_norm": 2.6664745807647705, "lr": 9.337627216026057e-06 }, { "step": 1075, "T": 575, "positions": 16, "loss": 0.38455031805415274, "ce": 1.5290309342553883, "tv": 0.2573858194764398, "conf": 1.0499978446919158, "grad_norm": 2.585846424102783, "lr": 9.293909819394952e-06 }, { "step": 1076, "T": 611, "positions": 16, "loss": 0.25176104822458, "ce": 0.8405120486277156, "tv": 0.18634426012431504, "conf": 1.0628363184068803, "grad_norm": 1.4408645629882812, "lr": 9.250271624047523e-06 }, { "step": 1077, "T": 529, "positions": 16, "loss": 0.7347242650575936, "ce": 2.593477518763393, "tv": 0.5281961392611265, "conf": 1.83621012076037, "grad_norm": 2.6709704399108887, "lr": 9.206712850038607e-06 }, { "step": 1078, "T": 526, "positions": 16, "loss": 0.453965091102873, "ce": 1.578931393163657, "tv": 0.32896885532682063, "conf": 1.0996253529927458, "grad_norm": 3.2302050590515137, "lr": 9.163233717022568e-06 }, { "step": 1079, "T": 576, "positions": 16, "loss": 0.33011696218272846, "ce": 1.2756321397208126, "tv": 0.22505972068802294, "conf": 1.4962397078220109, "grad_norm": 4.569040775299072, "lr": 9.119834444252104e-06 }, { "step": 1080, "T": 674, "positions": 16, "loss": 0.6312742043082835, "ce": 2.275308250915259, "tv": 0.4486037639726419, "conf": 1.3730664612376131, "grad_norm": 2.494082450866699, "lr": 9.076515250577245e-06 }, { "step": 1081, "T": 535, "positions": 16, "loss": 0.30510355919977883, "ce": 0.830518612785454, "tv": 0.2467241327467491, "conf": 0.8863868874211107, "grad_norm": 1.480756402015686, "lr": 9.033276354444193e-06 }, { "step": 1082, "T": 975, "positions": 16, "loss": 0.7322521763026089, "ce": 3.221545860018523, "tv": 0.4556640056634933, "conf": 1.9262786394597242, "grad_norm": 1.9930720329284668, "lr": 8.990117973894233e-06 }, { "step": 1083, "T": 1794, "positions": 16, "loss": 0.819731776122353, "ce": 3.531733696174342, "tv": 0.5183982581147575, "conf": 1.4362251726561226, "grad_norm": 3.303704261779785, "lr": 8.947040326562638e-06 }, { "step": 1084, "T": 532, "positions": 16, "loss": 0.5389913725666702, "ce": 1.6663432251662016, "tv": 0.4137300713919103, "conf": 1.49930450739339, "grad_norm": 7.234155178070068, "lr": 8.904043629677541e-06 }, { "step": 1085, "T": 552, "positions": 16, "loss": 0.5094188828461483, "ce": 2.475995770450197, "tv": 0.29091033928852994, "conf": 1.0632406852619738, "grad_norm": 4.3624653816223145, "lr": 8.861128100058915e-06 }, { "step": 1086, "T": 541, "positions": 16, "loss": 0.5259929275071045, "ce": 2.0111497428842995, "tv": 0.3609755055504138, "conf": 2.1866617260179453, "grad_norm": 5.199951171875, "lr": 8.818293954117379e-06 }, { "step": 1087, "T": 678, "positions": 16, "loss": 0.3767162348522106, "ce": 1.3376132845805841, "tv": 0.2699498956935713, "conf": 1.0712283621833194, "grad_norm": 2.71870756149292, "lr": 8.775541407853177e-06 }, { "step": 1088, "T": 527, "positions": 16, "loss": 0.8268387517891824, "ce": 3.519117961404845, "tv": 0.5276966206729412, "conf": 1.365824872860685, "grad_norm": 3.504638671875, "lr": 8.732870676855096e-06 }, { "step": 1089, "T": 648, "positions": 16, "loss": 0.6890978417359293, "ce": 2.796416345052421, "tv": 0.4549513552337885, "conf": 1.8913687765598297, "grad_norm": 2.014378070831299, "lr": 8.690281976299303e-06 }, { "step": 1090, "T": 543, "positions": 16, "loss": 0.15391334682135493, "ce": 0.40079005816232893, "tv": 0.12648260052924343, "conf": 0.688350723919859, "grad_norm": 3.2880916595458984, "lr": 8.647775520948343e-06 }, { "step": 1091, "T": 920, "positions": 16, "loss": 0.773940903134644, "ce": 3.530424276163103, "tv": 0.467664981726557, "conf": 1.5334754499599512, "grad_norm": 3.463899612426758, "lr": 8.60535152515001e-06 }, { "step": 1092, "T": 672, "positions": 16, "loss": 0.4989908724091947, "ce": 1.8737349091097713, "tv": 0.3462415501708165, "conf": 1.0959846071748416, "grad_norm": 2.4536306858062744, "lr": 8.563010202836286e-06 }, { "step": 1093, "T": 882, "positions": 16, "loss": 0.5869535166566493, "ce": 2.378847404250564, "tv": 0.3878541948797647, "conf": 1.0020309458247247, "grad_norm": 2.6376075744628906, "lr": 8.520751767522257e-06 }, { "step": 1094, "T": 528, "positions": 16, "loss": 0.25153078723087674, "ce": 0.8242635252463515, "tv": 0.18789381773603964, "conf": 1.4317285835495035, "grad_norm": 4.427151679992676, "lr": 8.478576432304996e-06 }, { "step": 1095, "T": 674, "positions": 16, "loss": 0.6269158496597811, "ce": 2.6388842752130586, "tv": 0.4033638321543549, "conf": 1.8086038566803615, "grad_norm": 5.080389976501465, "lr": 8.436484409862588e-06 }, { "step": 1096, "T": 533, "positions": 16, "loss": 0.15127114019287546, "ce": 0.42953294222616023, "tv": 0.1203531651767662, "conf": 0.6707260891749336, "grad_norm": 2.263784885406494, "lr": 8.394475912452962e-06 }, { "step": 1097, "T": 623, "positions": 16, "loss": 0.5297921184919687, "ce": 1.8630892389146538, "tv": 0.3816479947345215, "conf": 1.8874756299410889, "grad_norm": 2.6327059268951416, "lr": 8.352551151912833e-06 }, { "step": 1098, "T": 1190, "positions": 16, "loss": 0.6372426438210823, "ce": 2.7698709653104743, "tv": 0.40028393883403623, "conf": 2.401988999201054, "grad_norm": 2.78273606300354, "lr": 8.310710339656707e-06 }, { "step": 1099, "T": 627, "positions": 16, "loss": 0.4012663509442973, "ce": 1.3529268215925185, "tv": 0.2955263011074294, "conf": 1.7588868070722916, "grad_norm": 3.1984126567840576, "lr": 8.268953686675715e-06 }, { "step": 1100, "T": 638, "positions": 16, "loss": 0.6526837572455406, "ce": 2.228846044279635, "tv": 0.4775546155869961, "conf": 1.6443799657936324, "grad_norm": 2.694159507751465, "lr": 8.227281403536625e-06 }, { "step": 1101, "T": 605, "positions": 16, "loss": 0.6480235201306641, "ce": 2.512988976552151, "tv": 0.44080512947402894, "conf": 2.3066315981559455, "grad_norm": 3.5119264125823975, "lr": 8.185693700380739e-06 }, { "step": 1102, "T": 664, "positions": 16, "loss": 0.1689303994025977, "ce": 0.3923594134957966, "tv": 0.14410495978154358, "conf": 1.2552184930139667, "grad_norm": 2.352632522583008, "lr": 8.144190786922856e-06 }, { "step": 1103, "T": 530, "positions": 16, "loss": 0.5306659824773305, "ce": 2.2680546146327742, "tv": 0.33762280189966987, "conf": 1.0141270027266778, "grad_norm": 2.193439245223999, "lr": 8.102772872450209e-06 }, { "step": 1104, "T": 541, "positions": 16, "loss": 0.23617865648225234, "ce": 0.8289506172039864, "tv": 0.1703151078394285, "conf": 0.49156937762485364, "grad_norm": 2.1826491355895996, "lr": 8.061440165821372e-06 }, { "step": 1105, "T": 533, "positions": 16, "loss": 0.5487947819492547, "ce": 2.2571175325316517, "tv": 0.358981148165185, "conf": 1.4869230912299827, "grad_norm": 2.584538221359253, "lr": 8.020192875465288e-06 }, { "step": 1106, "T": 616, "positions": 16, "loss": 0.5497563295066357, "ce": 2.1276125396689167, "tv": 0.3744389915955253, "conf": 1.2055139568110462, "grad_norm": 2.8067264556884766, "lr": 7.97903120938015e-06 }, { "step": 1107, "T": 672, "positions": 16, "loss": 0.8283670023083687, "ce": 3.2034866232424974, "tv": 0.5644648345187306, "conf": 2.0494561761152, "grad_norm": 4.935320854187012, "lr": 7.937955375132352e-06 }, { "step": 1108, "T": 524, "positions": 16, "loss": 0.6830321505549364, "ce": 2.459607323835371, "tv": 0.48563491326058283, "conf": 1.7914884679485112, "grad_norm": 3.8002309799194336, "lr": 7.896965579855486e-06 }, { "step": 1109, "T": 1177, "positions": 16, "loss": 0.7358772471779957, "ce": 3.216087060121936, "tv": 0.46029837816604413, "conf": 2.012376217338897, "grad_norm": 4.077267646789551, "lr": 7.85606203024929e-06 }, { "step": 1110, "T": 543, "positions": 16, "loss": 0.3971454491111217, "ce": 1.3155124290369713, "tv": 0.2951046847992984, "conf": 1.0781455968383762, "grad_norm": 4.290838241577148, "lr": 7.815244932578545e-06 }, { "step": 1111, "T": 537, "positions": 16, "loss": 0.3422410332932486, "ce": 1.127904530865635, "tv": 0.25494508832343854, "conf": 2.2980616882302622, "grad_norm": 2.883728265762329, "lr": 7.774514492672114e-06 }, { "step": 1112, "T": 533, "positions": 16, "loss": 0.3374835704162251, "ce": 1.3824111974536208, "tv": 0.2213805100764148, "conf": 1.2974847317673266, "grad_norm": 8.549200057983398, "lr": 7.733870915921856e-06 }, { "step": 1113, "T": 532, "positions": 16, "loss": 0.6503592133522034, "ce": 2.5544548630714417, "tv": 0.4387930119410157, "conf": 2.218115607876676, "grad_norm": 2.8153634071350098, "lr": 7.693314407281615e-06 }, { "step": 1114, "T": 572, "positions": 16, "loss": 0.17777694686992618, "ce": 0.6090010101367227, "tv": 0.129863166850555, "conf": 0.7884236468580639, "grad_norm": 2.0013084411621094, "lr": 7.652845171266168e-06 }, { "step": 1115, "T": 531, "positions": 16, "loss": 0.8243560650735162, "ce": 3.212114955065772, "tv": 0.5590495324868243, "conf": 1.8066400468692336, "grad_norm": 3.4874565601348877, "lr": 7.612463411950196e-06 }, { "step": 1116, "T": 553, "positions": 16, "loss": 0.2802414116995351, "ce": 1.059865056170679, "tv": 0.1936165657298261, "conf": 0.6160269560032248, "grad_norm": 4.590856552124023, "lr": 7.572169332967283e-06 }, { "step": 1117, "T": 540, "positions": 16, "loss": 0.32476818590475887, "ce": 1.1000962786353057, "tv": 0.2386206211681383, "conf": 0.6233659557798603, "grad_norm": 2.220959424972534, "lr": 7.531963137508827e-06 }, { "step": 1118, "T": 532, "positions": 16, "loss": 0.4043892577974475, "ce": 1.760080965031193, "tv": 0.25375685789549607, "conf": 1.3880409473671702, "grad_norm": 3.9606735706329346, "lr": 7.49184502832308e-06 }, { "step": 1119, "T": 552, "positions": 16, "loss": 0.450483985064011, "ce": 1.822757638090934, "tv": 0.2980091313529556, "conf": 0.8700682070088374, "grad_norm": 1.9713265895843506, "lr": 7.451815207714122e-06 }, { "step": 1120, "T": 565, "positions": 16, "loss": 0.3548186503985562, "ce": 1.3270776271724571, "tv": 0.24678988200503227, "conf": 1.1034684696935377, "grad_norm": 2.6710383892059326, "lr": 7.411873877540773e-06 }, { "step": 1121, "T": 534, "positions": 16, "loss": 0.2917223286203807, "ce": 0.9971161659843801, "tv": 0.21334522895267582, "conf": 1.5754561750156468, "grad_norm": 3.259089469909668, "lr": 7.372021239215648e-06 }, { "step": 1122, "T": 1720, "positions": 16, "loss": 0.8366597732529044, "ce": 3.188591110520065, "tv": 0.5753340749070048, "conf": 2.406904545998259, "grad_norm": 3.0469541549682617, "lr": 7.332257493704111e-06 }, { "step": 1123, "T": 539, "positions": 16, "loss": 0.5131547534838319, "ce": 2.0698696174367797, "tv": 0.3401864403858781, "conf": 1.3162346918834373, "grad_norm": 5.115562915802002, "lr": 7.292582841523268e-06 }, { "step": 1124, "T": 1793, "positions": 16, "loss": 1.004045853856951, "ce": 5.218490106984973, "tv": 0.5357742798514664, "conf": 1.3836691443123357, "grad_norm": 2.9584720134735107, "lr": 7.252997482740942e-06 }, { "step": 1125, "T": 581, "positions": 16, "loss": 0.35468971644968406, "ce": 1.3497000785719138, "tv": 0.2441330193532849, "conf": 0.9054629392542448, "grad_norm": 1.8997284173965454, "lr": 7.213501616974688e-06 }, { "step": 1126, "T": 1033, "positions": 16, "loss": 0.980177229270339, "ce": 4.62857654679101, "tv": 0.5747995343990624, "conf": 1.6687138526140188, "grad_norm": 2.8262460231781006, "lr": 7.174095443390766e-06 }, { "step": 1127, "T": 544, "positions": 16, "loss": 0.30215368740846316, "ce": 1.1066773465224742, "tv": 0.21276217872400593, "conf": 0.7311773904657457, "grad_norm": 2.083934783935547, "lr": 7.134779160703153e-06 }, { "step": 1128, "T": 1570, "positions": 16, "loss": 0.7761593052127864, "ce": 3.4665029914758634, "tv": 0.4772322293720208, "conf": 2.179727452602492, "grad_norm": 3.44797945022583, "lr": 7.095552967172503e-06 }, { "step": 1129, "T": 1007, "positions": 16, "loss": 0.7354038650628354, "ce": 3.3429776942111857, "tv": 0.4456734543273342, "conf": 1.500033781309753, "grad_norm": 1.824276328086853, "lr": 7.056417060605231e-06 }, { "step": 1130, "T": 656, "positions": 16, "loss": 0.36360598971077707, "ce": 1.6863727718282462, "tv": 0.21663190840718016, "conf": 1.2606192424835185, "grad_norm": 3.5992541313171387, "lr": 7.017371638352399e-06 }, { "step": 1131, "T": 605, "positions": 16, "loss": 0.35282067771913717, "ce": 1.4263732457884544, "tv": 0.2335370552391396, "conf": 1.0315434434086228, "grad_norm": 2.098538875579834, "lr": 6.978416897308826e-06 }, { "step": 1132, "T": 1792, "positions": 16, "loss": 1.0109331372368615, "ce": 5.13219675354776, "tv": 0.5530149528640322, "conf": 1.5996743074243263, "grad_norm": 3.1086716651916504, "lr": 6.939553033912036e-06 }, { "step": 1133, "T": 535, "positions": 16, "loss": 0.1295814083005098, "ce": 0.5410512089983968, "tv": 0.08386254336619459, "conf": 2.5007700350494133, "grad_norm": 1.446110725402832, "lr": 6.900780244141286e-06 }, { "step": 1134, "T": 550, "positions": 16, "loss": 0.5146260347100906, "ce": 1.9460605312488042, "tv": 0.3555777663132176, "conf": 1.505258956761482, "grad_norm": 2.3452813625335693, "lr": 6.862098723516572e-06 }, { "step": 1135, "T": 608, "positions": 16, "loss": 0.42365986909135245, "ce": 1.6673646797717083, "tv": 0.2854704354831483, "conf": 1.8452922077849507, "grad_norm": 3.170872926712036, "lr": 6.823508667097647e-06 }, { "step": 1136, "T": 901, "positions": 16, "loss": 0.500625103004495, "ce": 2.0834795997161564, "tv": 0.3247523638965504, "conf": 1.928440793765663, "grad_norm": 2.842155694961548, "lr": 6.785010269483041e-06 }, { "step": 1137, "T": 527, "positions": 16, "loss": 0.7277674414217472, "ce": 2.789435748010874, "tv": 0.49869319796562195, "conf": 2.041656007990241, "grad_norm": 3.0366055965423584, "lr": 6.746603724809081e-06 }, { "step": 1138, "T": 1058, "positions": 16, "loss": 0.6423547472804785, "ce": 2.911858902996755, "tv": 0.3901876419840846, "conf": 1.6175420616259544, "grad_norm": 3.7673447132110596, "lr": 6.708289226748868e-06 }, { "step": 1139, "T": 629, "positions": 16, "loss": 0.36624820129782165, "ce": 1.153852922265287, "tv": 0.27873657603959145, "conf": 1.650374133605851, "grad_norm": 1.9276379346847534, "lr": 6.670066968511399e-06 }, { "step": 1140, "T": 532, "positions": 16, "loss": 0.28697677329182625, "ce": 1.0314620570861734, "tv": 0.204256173921749, "conf": 1.2937118236113747, "grad_norm": 6.684239864349365, "lr": 6.6319371428404805e-06 }, { "step": 1141, "T": 674, "positions": 16, "loss": 0.6044552181847394, "ce": 2.251753505785018, "tv": 0.4214220908470452, "conf": 1.2230610682308907, "grad_norm": 3.1768813133239746, "lr": 6.593899942013832e-06 }, { "step": 1142, "T": 546, "positions": 16, "loss": 0.20951153664645972, "ce": 0.622243896995542, "tv": 0.16365239744482096, "conf": 0.5390077896609853, "grad_norm": 1.8672820329666138, "lr": 6.555955557842091e-06 }, { "step": 1143, "T": 574, "positions": 16, "loss": 0.4542555325169815, "ce": 1.7596792333731628, "tv": 0.3092084630570753, "conf": 1.9762672632932663, "grad_norm": 4.612153053283691, "lr": 6.518104181667844e-06 }, { "step": 1144, "T": 675, "positions": 16, "loss": 0.4603761746548116, "ce": 1.7245699291815981, "tv": 0.3199102259823121, "conf": 2.1673571562833605, "grad_norm": 5.396734237670898, "lr": 6.480346004364668e-06 }, { "step": 1145, "T": 549, "positions": 16, "loss": 0.39340833731694147, "ce": 1.2596048779087141, "tv": 0.29716427373932675, "conf": 1.110498782928996, "grad_norm": 2.1861982345581055, "lr": 6.442681216336161e-06 }, { "step": 1146, "T": 671, "positions": 16, "loss": 0.6689005866646767, "ce": 2.279479341581464, "tv": 0.48994738422334194, "conf": 1.700655127177015, "grad_norm": 3.176558017730713, "lr": 6.405110007514989e-06 }, { "step": 1147, "T": 609, "positions": 16, "loss": 0.3755757298970366, "ce": 1.0578196875005688, "tv": 0.29977085248310686, "conf": 2.4892979411989735, "grad_norm": 3.8306612968444824, "lr": 6.367632567361931e-06 }, { "step": 1148, "T": 570, "positions": 16, "loss": 0.7273825191587093, "ce": 3.5299619074241946, "tv": 0.4159848034832976, "conf": 1.7870465639811264, "grad_norm": 4.234694957733154, "lr": 6.3302490848648864e-06 }, { "step": 1149, "T": 540, "positions": 16, "loss": 0.34138189287205023, "ce": 1.2039197502108436, "tv": 0.245544353427249, "conf": 0.9223084744205252, "grad_norm": 3.7891595363616943, "lr": 6.2929597485379996e-06 }, { "step": 1150, "T": 618, "positions": 16, "loss": 0.39117494807578623, "ce": 1.466692172223702, "tv": 0.27167302556335926, "conf": 1.5755396864842623, "grad_norm": 3.7279207706451416, "lr": 6.255764746420645e-06 }, { "step": 1151, "T": 524, "positions": 16, "loss": 0.7522825002670288, "ce": 2.825043279211968, "tv": 0.5219757417216897, "conf": 1.5287006543949246, "grad_norm": 3.8366477489471436, "lr": 6.218664266076471e-06 }, { "step": 1152, "T": 673, "positions": 16, "loss": 0.7222649157047272, "ce": 2.7530675064772367, "tv": 0.49662019684910774, "conf": 1.5491782373283058, "grad_norm": 2.3358845710754395, "lr": 6.181658494592516e-06 }, { "step": 1153, "T": 526, "positions": 16, "loss": 0.3115877341688247, "ce": 0.9771368551264459, "tv": 0.2376378445305818, "conf": 1.440514678442642, "grad_norm": 3.310317039489746, "lr": 6.144747618578209e-06 }, { "step": 1154, "T": 531, "positions": 16, "loss": 0.4200176066806307, "ce": 1.6100066361614154, "tv": 0.28779661876615137, "conf": 0.8052554538059553, "grad_norm": 3.1718780994415283, "lr": 6.107931824164459e-06 }, { "step": 1155, "T": 536, "positions": 16, "loss": 0.4159080236868249, "ce": 1.6237524560203838, "tv": 0.2817030643154794, "conf": 0.656077838145869, "grad_norm": 4.7008795738220215, "lr": 6.071211297002694e-06 }, { "step": 1156, "T": 601, "positions": 16, "loss": 0.2544460554959187, "ce": 0.8427082346092902, "tv": 0.18908359587476298, "conf": 1.2316007959003272, "grad_norm": 3.5352656841278076, "lr": 6.034586222263949e-06 }, { "step": 1157, "T": 647, "positions": 16, "loss": 0.6629851469770074, "ce": 2.444260500720702, "tv": 0.4650656534358859, "conf": 1.773502471856773, "grad_norm": 2.3584227561950684, "lr": 5.99805678463792e-06 }, { "step": 1158, "T": 532, "positions": 16, "loss": 0.35957277832130785, "ce": 1.1647563380153088, "tv": 0.27010793887166074, "conf": 0.6204884144499374, "grad_norm": 1.5831753015518188, "lr": 5.961623168332006e-06 }, { "step": 1159, "T": 544, "positions": 16, "loss": 0.34891288561084366, "ce": 1.3657499390010344, "tv": 0.23593099511708715, "conf": 1.9412630845069718, "grad_norm": 7.205724716186523, "lr": 5.925285557070448e-06 }, { "step": 1160, "T": 673, "positions": 16, "loss": 0.7637637332081795, "ce": 2.8923470773734152, "tv": 0.5272544483887032, "conf": 1.4128625610883319, "grad_norm": 1.9570990800857544, "lr": 5.889044134093344e-06 }, { "step": 1161, "T": 593, "positions": 16, "loss": 0.23103711331350496, "ce": 0.7283526317537508, "tv": 0.17577983369483263, "conf": 0.9291931373432192, "grad_norm": 2.6490352153778076, "lr": 5.852899082155722e-06 }, { "step": 1162, "T": 568, "positions": 16, "loss": 0.12595634214085294, "ce": 0.41433221608167514, "tv": 0.09391458514437545, "conf": 1.0017533774162075, "grad_norm": 1.4492554664611816, "lr": 5.8168505835266564e-06 }, { "step": 1163, "T": 589, "positions": 16, "loss": 0.13305402903097274, "ce": 0.285341851236808, "tv": 0.11613316393231798, "conf": 0.7695695404915313, "grad_norm": 3.536107301712036, "lr": 5.780898819988354e-06 }, { "step": 1164, "T": 526, "positions": 16, "loss": 0.3200461373489816, "ce": 1.1209417804639088, "tv": 0.23105773091083393, "conf": 1.556739810217124, "grad_norm": 2.5960729122161865, "lr": 5.745043972835165e-06 }, { "step": 1165, "T": 538, "positions": 16, "loss": 0.13354579041788384, "ce": 0.5053299211716666, "tv": 0.09223644931262243, "conf": 0.44872049776631684, "grad_norm": 1.0609019994735718, "lr": 5.709286222872756e-06 }, { "step": 1166, "T": 1365, "positions": 16, "loss": 0.8010523398406804, "ce": 3.6084173147100955, "tv": 0.4891228884225711, "conf": 2.0186871673795395, "grad_norm": 3.4583053588867188, "lr": 5.67362575041715e-06 }, { "step": 1167, "T": 533, "positions": 16, "loss": 0.44878527711352945, "ce": 1.7385592514585255, "tv": 0.30547704985178825, "conf": 1.1529942137421623, "grad_norm": 6.555440425872803, "lr": 5.6380627352938225e-06 }, { "step": 1168, "T": 526, "positions": 16, "loss": 0.721759544685483, "ce": 2.459514567628503, "tv": 0.5286756306886673, "conf": 2.920924697071314, "grad_norm": 4.493744850158691, "lr": 5.602597356836803e-06 }, { "step": 1169, "T": 625, "positions": 16, "loss": 0.11931592902965349, "ce": 0.2431719568396602, "tv": 0.1055541435835039, "conf": 1.0831050864904803, "grad_norm": 1.739598035812378, "lr": 5.567229793887771e-06 }, { "step": 1170, "T": 544, "positions": 16, "loss": 0.6369716838817112, "ce": 2.391615123175143, "tv": 0.44201131457521114, "conf": 1.6097295185900964, "grad_norm": 2.943828582763672, "lr": 5.531960224795155e-06 }, { "step": 1171, "T": 531, "positions": 16, "loss": 0.6947546147275716, "ce": 2.7743169110035524, "tv": 0.46369214565493166, "conf": 1.582831886042868, "grad_norm": 2.846353769302368, "lr": 5.4967888274132045e-06 }, { "step": 1172, "T": 1057, "positions": 16, "loss": 0.6850572462753917, "ce": 3.063340280095872, "tv": 0.42080358913517557, "conf": 1.6044960618455661, "grad_norm": 3.0069851875305176, "lr": 5.461715779101137e-06 }, { "step": 1173, "T": 1004, "positions": 16, "loss": 0.7023704810999334, "ce": 2.8730255784466863, "tv": 0.4611865661572665, "conf": 1.5519022697262699, "grad_norm": 2.7102110385894775, "lr": 5.426741256722239e-06 }, { "step": 1174, "T": 673, "positions": 16, "loss": 0.6307148677296937, "ce": 2.296544256969355, "tv": 0.4456227128393948, "conf": 1.5202527083602035, "grad_norm": 3.7252275943756104, "lr": 5.391865436642923e-06 }, { "step": 1175, "T": 541, "positions": 16, "loss": 0.18372434049797448, "ce": 0.617996382260344, "tv": 0.135471892467649, "conf": 0.7493550400643351, "grad_norm": 1.9502137899398804, "lr": 5.357088494731902e-06 }, { "step": 1176, "T": 531, "positions": 16, "loss": 0.18250210985934245, "ce": 0.561573817947874, "tv": 0.14038304237055854, "conf": 1.013802783758214, "grad_norm": 2.7508373260498047, "lr": 5.32241060635926e-06 }, { "step": 1177, "T": 561, "positions": 16, "loss": 0.22220056409423705, "ce": 0.6807229417609051, "tv": 0.17125363856030162, "conf": 0.7291336717543268, "grad_norm": 3.6279184818267822, "lr": 5.287831946395591e-06 }, { "step": 1178, "T": 535, "positions": 16, "loss": 0.280004119646037, "ce": 0.8123451219724416, "tv": 0.2208551338117104, "conf": 0.976834831521046, "grad_norm": 2.3755273818969727, "lr": 5.253352689211114e-06 }, { "step": 1179, "T": 571, "positions": 16, "loss": 0.28219356198678724, "ce": 0.8361801063001622, "tv": 0.22063949951552786, "conf": 1.478411231107657, "grad_norm": 3.6112618446350098, "lr": 5.218973008674757e-06 }, { "step": 1180, "T": 672, "positions": 16, "loss": 0.8138198964297771, "ce": 2.9804060831665993, "tv": 0.5730881132185459, "conf": 1.8059677412966266, "grad_norm": 3.086906909942627, "lr": 5.184693078153361e-06 }, { "step": 1181, "T": 540, "positions": 16, "loss": 0.1490277429826392, "ce": 0.47657176640291254, "tv": 0.11263396054687291, "conf": 0.667065517747595, "grad_norm": 2.627484083175659, "lr": 5.1505130705107205e-06 }, { "step": 1182, "T": 532, "positions": 16, "loss": 0.11585106439781612, "ce": 0.35815695003725523, "tv": 0.08892818991353124, "conf": 0.2689440015100181, "grad_norm": 2.9881136417388916, "lr": 5.116433158106748e-06 }, { "step": 1183, "T": 676, "positions": 16, "loss": 0.6787058734335005, "ce": 2.4731793246464804, "tv": 0.4793199533596635, "conf": 1.1978909762110561, "grad_norm": 2.5108141899108887, "lr": 5.082453512796634e-06 }, { "step": 1184, "T": 838, "positions": 16, "loss": 0.5020227296517987, "ce": 1.944367187714306, "tv": 0.34176224829752755, "conf": 2.2147177283186466, "grad_norm": 2.13973069190979, "lr": 5.048574305929915e-06 }, { "step": 1185, "T": 583, "positions": 16, "loss": 0.46988377602610854, "ce": 1.703063027925964, "tv": 0.3328638617786055, "conf": 1.1703040689241142, "grad_norm": 2.712193727493286, "lr": 5.014795708349665e-06 }, { "step": 1186, "T": 527, "positions": 16, "loss": 0.2994884161516893, "ce": 0.9685067843172419, "tv": 0.22515305255183193, "conf": 1.6373481852576794, "grad_norm": 2.1504266262054443, "lr": 4.981117890391615e-06 }, { "step": 1187, "T": 675, "positions": 16, "loss": 0.6128963602532167, "ce": 2.6368660769148846, "tv": 0.3880108535377076, "conf": 1.3239977554762703, "grad_norm": 3.108010768890381, "lr": 4.947541021883287e-06 }, { "step": 1188, "T": 542, "positions": 16, "loss": 0.4050581620217031, "ce": 1.5053926362149355, "tv": 0.2827987706732529, "conf": 1.811998763408883, "grad_norm": 4.541426658630371, "lr": 4.914065272143153e-06 }, { "step": 1189, "T": 1211, "positions": 16, "loss": 0.6608839285327122, "ce": 3.0981428849190706, "tv": 0.39007740003580693, "conf": 1.3663033228604036, "grad_norm": 1.6941282749176025, "lr": 4.880690809979749e-06 }, { "step": 1190, "T": 527, "positions": 16, "loss": 0.12971195343743602, "ce": 0.4499221948956347, "tv": 0.09413304428926494, "conf": 0.2735234971696343, "grad_norm": 1.5370243787765503, "lr": 4.847417803690876e-06 }, { "step": 1191, "T": 1106, "positions": 16, "loss": 0.9181804342606483, "ce": 4.513670725773409, "tv": 0.5186815134456992, "conf": 0.9356588877283798, "grad_norm": 3.244919776916504, "lr": 4.814246421062718e-06 }, { "step": 1192, "T": 533, "positions": 16, "loss": 0.1946157597240017, "ce": 0.6665117176310957, "tv": 0.14218288288157055, "conf": 0.6451632677054704, "grad_norm": 3.5715057849884033, "lr": 4.781176829368969e-06 }, { "step": 1193, "T": 1721, "positions": 16, "loss": 0.8659273684024811, "ce": 4.064132630825043, "tv": 0.5105712153017521, "conf": 2.7546633665115223, "grad_norm": 2.566240072250366, "lr": 4.7482091953700705e-06 }, { "step": 1194, "T": 616, "positions": 16, "loss": 0.16651851946880925, "ce": 0.4712261161718061, "tv": 0.13266212657981669, "conf": 0.8868369598180834, "grad_norm": 3.0645928382873535, "lr": 4.7153436853122735e-06 }, { "step": 1195, "T": 672, "positions": 16, "loss": 0.5883389031514525, "ce": 2.2343607149086893, "tv": 0.4054476167075336, "conf": 1.646760466042906, "grad_norm": 3.4239022731781006, "lr": 4.682580464926875e-06 }, { "step": 1196, "T": 529, "positions": 16, "loss": 0.6344772027805448, "ce": 2.214248437434435, "tv": 0.45894706435501575, "conf": 2.167328841984272, "grad_norm": 2.6590054035186768, "lr": 4.649919699429347e-06 }, { "step": 1197, "T": 536, "positions": 16, "loss": 0.19323380171044846, "ce": 0.7803264859285264, "tv": 0.12800128023081925, "conf": 1.4217237776092588, "grad_norm": 3.8972110748291016, "lr": 4.617361553518515e-06 }, { "step": 1198, "T": 550, "positions": 16, "loss": 0.449657674140326, "ce": 1.4778269808921323, "tv": 0.33541665251368613, "conf": 1.2973333247387187, "grad_norm": 5.102912425994873, "lr": 4.584906191375715e-06 }, { "step": 1199, "T": 1223, "positions": 16, "loss": 0.8910271059721708, "ce": 4.14028088375926, "tv": 0.5299989040941, "conf": 2.6126687470714387, "grad_norm": 3.575639009475708, "lr": 4.552553776663982e-06 }, { "step": 1200, "T": 528, "positions": 16, "loss": 0.4342821816535434, "ce": 1.6065541313428184, "tv": 0.30402974365642876, "conf": 1.7533048996701837, "grad_norm": 11.23376750946045, "lr": 4.520304472527215e-06 }, { "step": 1201, "T": 535, "positions": 16, "loss": 0.24308764567729213, "ce": 0.6602345305295714, "tv": 0.19673799060910824, "conf": 1.0030656336941524, "grad_norm": 3.120288133621216, "lr": 4.4881584415893604e-06 }, { "step": 1202, "T": 580, "positions": 16, "loss": 0.35177602537896746, "ce": 0.9839072778677291, "tv": 0.28153923813260917, "conf": 1.3651608455733424, "grad_norm": 3.1797518730163574, "lr": 4.456115845953568e-06 }, { "step": 1203, "T": 533, "positions": 16, "loss": 0.06329002459096955, "ce": 0.23587669529604227, "tv": 0.04411372836693772, "conf": 0.1456653350533088, "grad_norm": 1.3719931840896606, "lr": 4.424176847201411e-06 }, { "step": 1204, "T": 531, "positions": 16, "loss": 0.3751773169860826, "ce": 1.3048262302327203, "tv": 0.2718829938166891, "conf": 0.8097548776445636, "grad_norm": 5.302657604217529, "lr": 4.392341606392067e-06 }, { "step": 1205, "T": 577, "positions": 16, "loss": 0.5224181803059764, "ce": 1.8544542680610903, "tv": 0.3744141756906174, "conf": 1.8064230842289817, "grad_norm": 3.597151517868042, "lr": 4.360610284061456e-06 }, { "step": 1206, "T": 583, "positions": 16, "loss": 0.5476283250227425, "ce": 2.2935571845030154, "tv": 0.35363623005605405, "conf": 1.8922041854635125, "grad_norm": 7.118447303771973, "lr": 4.328983040221493e-06 }, { "step": 1207, "T": 673, "positions": 16, "loss": 0.751063596457243, "ce": 3.082688618451357, "tv": 0.491994159296155, "conf": 1.474545356570161, "grad_norm": 2.6344759464263916, "lr": 4.297460034359249e-06 }, { "step": 1208, "T": 672, "positions": 16, "loss": 0.4223213531076908, "ce": 1.5481285701971501, "tv": 0.2972316537052393, "conf": 2.0072536105290055, "grad_norm": 3.962240695953369, "lr": 4.266041425436151e-06 }, { "step": 1209, "T": 1189, "positions": 16, "loss": 0.693299607373774, "ce": 2.902051053533796, "tv": 0.44788278464693576, "conf": 2.199914588651154, "grad_norm": 5.377158164978027, "lr": 4.234727371887187e-06 }, { "step": 1210, "T": 534, "positions": 16, "loss": 0.7226834511384368, "ce": 2.8915387801826, "tv": 0.4816995281726122, "conf": 1.7760835691151442, "grad_norm": 3.1690776348114014, "lr": 4.203518031620096e-06 }, { "step": 1211, "T": 531, "positions": 16, "loss": 0.35004832818958675, "ce": 1.3242358187344507, "tv": 0.24180527481075842, "conf": 1.0668958388378087, "grad_norm": 4.51986837387085, "lr": 4.1724135620145875e-06 }, { "step": 1212, "T": 678, "positions": 16, "loss": 0.49336038262117654, "ce": 1.6498682075377928, "tv": 0.3648595194099471, "conf": 1.5101137066594674, "grad_norm": 2.9324631690979004, "lr": 4.141414119921522e-06 }, { "step": 1213, "T": 525, "positions": 16, "loss": 0.5980618266039528, "ce": 2.2004096701275557, "tv": 0.4200231625000015, "conf": 1.8423237178940326, "grad_norm": 1.9220080375671387, "lr": 4.110519861662143e-06 }, { "step": 1214, "T": 674, "positions": 16, "loss": 0.2636423452404415, "ce": 0.7432679803514475, "tv": 0.2103506084804394, "conf": 1.6458522958597495, "grad_norm": 4.606310844421387, "lr": 4.07973094302731e-06 }, { "step": 1215, "T": 550, "positions": 16, "loss": 0.23559371141345764, "ce": 0.7406723386302474, "tv": 0.1794738631524524, "conf": 0.8080184018988055, "grad_norm": 2.6971914768218994, "lr": 4.049047519276642e-06 }, { "step": 1216, "T": 534, "positions": 16, "loss": 0.49922761018387973, "ce": 1.6931999332737178, "tv": 0.36656402074731886, "conf": 1.9359974679537117, "grad_norm": 2.934201717376709, "lr": 4.01846974513779e-06 }, { "step": 1217, "T": 607, "positions": 16, "loss": 0.3988213083575829, "ce": 1.7699528198354528, "tv": 0.24647337141504977, "conf": 1.3143785333777487, "grad_norm": 4.592804431915283, "lr": 3.987997774805677e-06 }, { "step": 1218, "T": 529, "positions": 16, "loss": 0.7270174212753773, "ce": 3.148019421787467, "tv": 0.458017167635262, "conf": 1.5092119156179251, "grad_norm": 4.480489730834961, "lr": 3.957631761941641e-06 }, { "step": 1219, "T": 538, "positions": 16, "loss": 0.7104349751025438, "ce": 2.4413652853108943, "tv": 0.5181093877181411, "conf": 1.9328958187143144, "grad_norm": 2.2169442176818848, "lr": 3.927371859672735e-06 }, { "step": 1220, "T": 538, "positions": 16, "loss": 0.30596126865839324, "ce": 1.0765810153826578, "tv": 0.22033685997917019, "conf": 0.8610273234403533, "grad_norm": 3.0199074745178223, "lr": 3.897218220590923e-06 }, { "step": 1221, "T": 538, "positions": 16, "loss": 0.20322901499048385, "ce": 0.7551452056368362, "tv": 0.14190500041695486, "conf": 0.4392906246742676, "grad_norm": 2.5697648525238037, "lr": 3.867170996752306e-06 }, { "step": 1222, "T": 851, "positions": 16, "loss": 0.8937280991303851, "ce": 3.6187154928504697, "tv": 0.590951703110477, "conf": 2.558313844107033, "grad_norm": 4.034590244293213, "lr": 3.837230339676379e-06 }, { "step": 1223, "T": 533, "positions": 16, "loss": 0.30346103975898586, "ce": 1.0907175499014556, "tv": 0.2159881007464719, "conf": 1.451201575890991, "grad_norm": 2.4765219688415527, "lr": 3.807396400345223e-06 }, { "step": 1224, "T": 1623, "positions": 16, "loss": 0.8223993474457529, "ce": 3.8683151272607574, "tv": 0.48396425840837765, "conf": 1.461249680214678, "grad_norm": 2.631096601486206, "lr": 3.7776693292028126e-06 }, { "step": 1225, "T": 547, "positions": 16, "loss": 0.3576845412753755, "ce": 1.2061119952049921, "tv": 0.2634148261458904, "conf": 1.3158856731770356, "grad_norm": 2.2596874237060547, "lr": 3.7480492761541837e-06 }, { "step": 1226, "T": 876, "positions": 16, "loss": 0.7660333041058038, "ce": 3.8149935503533925, "tv": 0.4272599354153499, "conf": 1.3066461396569338, "grad_norm": 3.9945449829101562, "lr": 3.7185363905647225e-06 }, { "step": 1227, "T": 528, "positions": 16, "loss": 0.498666787948423, "ce": 2.4009187722191996, "tv": 0.28730545194866863, "conf": 0.42535260864905666, "grad_norm": 3.123284339904785, "lr": 3.689130821259401e-06 }, { "step": 1228, "T": 652, "positions": 16, "loss": 0.7522436454892159, "ce": 2.9585117548704147, "tv": 0.5071027539670467, "conf": 1.8735300979897147, "grad_norm": 2.777338743209839, "lr": 3.6598327165220296e-06 }, { "step": 1229, "T": 611, "positions": 16, "loss": 0.22366561939452367, "ce": 0.59684914393074, "tv": 0.18220078986360022, "conf": 1.1413629118233075, "grad_norm": 3.66312313079834, "lr": 3.6306422240944966e-06 }, { "step": 1230, "T": 588, "positions": 16, "loss": 0.1511238037746807, "ce": 0.5819984695848461, "tv": 0.1032488423788891, "conf": 1.05550451191084, "grad_norm": 4.535802841186523, "lr": 3.601559491176043e-06 }, { "step": 1231, "T": 554, "positions": 16, "loss": 0.40616673545810045, "ce": 1.4287978074789862, "tv": 0.2925410695843311, "conf": 1.6287820601223757, "grad_norm": 3.0817675590515137, "lr": 3.5725846644225015e-06 }, { "step": 1232, "T": 592, "positions": 16, "loss": 0.4905063967453316, "ce": 1.8765262429660652, "tv": 0.3365042059449479, "conf": 1.3511292983618546, "grad_norm": 2.4315195083618164, "lr": 3.543717889945572e-06 }, { "step": 1233, "T": 540, "positions": 16, "loss": 0.31211273700682796, "ce": 1.5084616258068309, "tv": 0.17918508480761375, "conf": 0.6400473947345442, "grad_norm": 2.3203632831573486, "lr": 3.514959313312061e-06 }, { "step": 1234, "T": 537, "positions": 16, "loss": 0.31665417174372124, "ce": 0.9148603004409779, "tv": 0.2501868206782092, "conf": 1.062552654190199, "grad_norm": 2.758702039718628, "lr": 3.486309079543193e-06 }, { "step": 1235, "T": 673, "positions": 16, "loss": 0.3548515455004235, "ce": 1.267863382460007, "tv": 0.25340580605006835, "conf": 1.1836681514178053, "grad_norm": 3.3687868118286133, "lr": 3.457767333113834e-06 }, { "step": 1236, "T": 575, "positions": 16, "loss": 0.1662278937001247, "ce": 0.42214154726389097, "tv": 0.13779304856143426, "conf": 1.3226587232202291, "grad_norm": 2.5969583988189697, "lr": 3.4293342179517714e-06 }, { "step": 1237, "T": 545, "positions": 16, "loss": 0.6554726921240217, "ce": 2.6141423000226496, "tv": 0.4378427337287576, "conf": 1.8907311144284904, "grad_norm": 4.72629976272583, "lr": 3.4010098774370087e-06 }, { "step": 1238, "T": 540, "positions": 16, "loss": 0.32014759853427677, "ce": 1.037235239939946, "tv": 0.24047120580598857, "conf": 1.2091266123461537, "grad_norm": 2.263869285583496, "lr": 3.372794454401032e-06 }, { "step": 1239, "T": 525, "positions": 16, "loss": 0.43848679409347824, "ce": 1.5479635900619542, "tv": 0.31521161054934055, "conf": 1.3926041794984048, "grad_norm": 1.7806702852249146, "lr": 3.3446880911260775e-06 }, { "step": 1240, "T": 1002, "positions": 16, "loss": 0.8587324099498801, "ce": 3.7165224498457974, "tv": 0.5412001874065027, "conf": 1.4861004447566462, "grad_norm": 2.818704605102539, "lr": 3.316690929344435e-06 }, { "step": 1241, "T": 599, "positions": 16, "loss": 0.3292634245699446, "ce": 1.41345008793067, "tv": 0.20879823814175325, "conf": 0.6629357885867648, "grad_norm": 1.7093842029571533, "lr": 3.288803110237715e-06 }, { "step": 1242, "T": 566, "positions": 16, "loss": 0.2815848890604684, "ce": 0.9279338945434574, "tv": 0.20976833093482128, "conf": 0.9869000678924067, "grad_norm": 4.869225978851318, "lr": 3.2610247744361526e-06 }, { "step": 1243, "T": 542, "positions": 16, "loss": 0.4873262697365135, "ce": 1.6985883877496235, "tv": 0.35274159209802747, "conf": 1.714394842274487, "grad_norm": 2.9028565883636475, "lr": 3.2333560620178727e-06 }, { "step": 1244, "T": 571, "positions": 16, "loss": 0.7724968050606549, "ce": 3.4060008442029357, "tv": 0.47988524357788265, "conf": 1.0577726827032166, "grad_norm": 4.245540618896484, "lr": 3.2057971125082244e-06 }, { "step": 1245, "T": 551, "positions": 16, "loss": 0.6905473103106488, "ce": 3.3780642695091956, "tv": 0.39193431675084867, "conf": 1.6609570559202353, "grad_norm": 2.5576491355895996, "lr": 3.178348064879047e-06 }, { "step": 1246, "T": 557, "positions": 16, "loss": 0.48456461261957884, "ce": 1.6912462702021003, "tv": 0.3504888731986284, "conf": 1.6418324598344043, "grad_norm": 2.794179916381836, "lr": 3.15100905754796e-06 }, { "step": 1247, "T": 1795, "positions": 16, "loss": 0.9237948046065867, "ce": 4.233635179232806, "tv": 0.5560347465798259, "conf": 2.012873587773356, "grad_norm": 8.433623313903809, "lr": 3.1237802283776983e-06 }, { "step": 1248, "T": 654, "positions": 16, "loss": 0.5098330562614137, "ce": 1.7706829261151142, "tv": 0.36973863519961014, "conf": 0.9374079884437378, "grad_norm": 3.5916833877563477, "lr": 3.096661714675397e-06 }, { "step": 1249, "T": 1011, "positions": 16, "loss": 0.6632425875468471, "ce": 2.8074511695335786, "tv": 0.424997203728708, "conf": 1.6246811900155276, "grad_norm": 1.7919386625289917, "lr": 3.0696536531919007e-06 }, { "step": 1250, "T": 673, "positions": 16, "loss": 0.5949487872421741, "ce": 2.202759262174368, "tv": 0.4163031857460737, "conf": 2.0455339192412794, "grad_norm": 3.190870523452759, "lr": 3.042756180121073e-06 }, { "step": 1251, "T": 588, "positions": 16, "loss": 0.556362433261711, "ce": 2.3722300469808033, "tv": 0.3545993799561984, "conf": 1.722416402597986, "grad_norm": 4.105607509613037, "lr": 3.0159694310991166e-06 }, { "step": 1252, "T": 540, "positions": 16, "loss": 0.46258124822179525, "ce": 1.7427303268596006, "tv": 0.32034246379589604, "conf": 2.517124418639642, "grad_norm": 5.428915023803711, "lr": 2.989293541203883e-06 }, { "step": 1253, "T": 667, "positions": 16, "loss": 0.21362587828116375, "ce": 0.5896345105533101, "tv": 0.17184714088580222, "conf": 1.1033207430443484, "grad_norm": 2.8839097023010254, "lr": 2.962728644954191e-06 }, { "step": 1254, "T": 952, "positions": 16, "loss": 0.6317840198171325, "ce": 2.386967934377026, "tv": 0.43676359922392294, "conf": 2.9677887867874233, "grad_norm": 4.390307903289795, "lr": 2.93627487630915e-06 }, { "step": 1255, "T": 862, "positions": 16, "loss": 0.6715757576748729, "ce": 2.5992023712024093, "tv": 0.4573950441554189, "conf": 1.8706247223672108, "grad_norm": 4.947389602661133, "lr": 2.909932368667495e-06 }, { "step": 1256, "T": 945, "positions": 16, "loss": 0.9442110965028405, "ce": 4.6885592504404485, "tv": 0.528172429651022, "conf": 1.9381270703460345, "grad_norm": 4.028873443603516, "lr": 2.883701254866886e-06 }, { "step": 1257, "T": 545, "positions": 16, "loss": 0.5937123773619533, "ce": 2.2094450561271515, "tv": 0.414186519337818, "conf": 1.466009644080204, "grad_norm": 2.8345253467559814, "lr": 2.857581667183265e-06 }, { "step": 1258, "T": 677, "positions": 16, "loss": 0.4674019127560314, "ce": 1.714323834181414, "tv": 0.32885504214209504, "conf": 1.6915514107234344, "grad_norm": 1.8563206195831299, "lr": 2.8315737373301955e-06 }, { "step": 1259, "T": 659, "positions": 16, "loss": 0.6444509942957666, "ce": 2.775917345948983, "tv": 0.4076214004890062, "conf": 0.9524664726372167, "grad_norm": 3.8662476539611816, "lr": 2.8056775964581505e-06 }, { "step": 1260, "T": 676, "positions": 16, "loss": 0.392484499301645, "ce": 1.310291161717032, "tv": 0.29050598526373506, "conf": 1.5728675649576567, "grad_norm": 3.703993082046509, "lr": 2.7798933751539106e-06 }, { "step": 1261, "T": 525, "positions": 16, "loss": 0.627388424705714, "ce": 2.068719358649105, "tv": 0.46724054543301463, "conf": 2.284282205204363, "grad_norm": 3.0228466987609863, "lr": 2.7542212034398635e-06 }, { "step": 1262, "T": 568, "positions": 16, "loss": 0.4241114545147866, "ce": 1.6369197780732065, "tv": 0.2893549809232354, "conf": 1.2484991890619312, "grad_norm": 3.5367491245269775, "lr": 2.7286612107733728e-06 }, { "step": 1263, "T": 613, "positions": 16, "loss": 0.330363739762106, "ce": 1.1805932553761522, "tv": 0.23589379580516834, "conf": 1.3429636755331826, "grad_norm": 3.218217611312866, "lr": 2.703213526046108e-06 }, { "step": 1264, "T": 535, "positions": 16, "loss": 0.32773340485937297, "ce": 1.0365511569973478, "tv": 0.24897588933708903, "conf": 1.9413122742459774, "grad_norm": 3.6974382400512695, "lr": 2.6778782775834067e-06 }, { "step": 1265, "T": 524, "positions": 16, "loss": 0.4110735195572488, "ce": 1.366518966286094, "tv": 0.3049129242426716, "conf": 1.5976343808321474, "grad_norm": 2.4359145164489746, "lr": 2.652655593143627e-06 }, { "step": 1266, "T": 529, "positions": 16, "loss": 0.5163655980177282, "ce": 1.770367623299535, "tv": 0.377032045345004, "conf": 1.4996738287663902, "grad_norm": 3.919119358062744, "lr": 2.6275455999174775e-06 }, { "step": 1267, "T": 549, "positions": 16, "loss": 0.37380724265813114, "ce": 1.2755521222271398, "tv": 0.2736133728885761, "conf": 1.5696879489929738, "grad_norm": 2.33994460105896, "lr": 2.6025484245274207e-06 }, { "step": 1268, "T": 878, "positions": 16, "loss": 0.6901372912179795, "ce": 2.790806202829117, "tv": 0.45672964649565984, "conf": 2.112523225635335, "grad_norm": 3.5527937412261963, "lr": 2.577664193027013e-06 }, { "step": 1269, "T": 570, "positions": 16, "loss": 0.4228130272713315, "ce": 1.4574259364671889, "tv": 0.3078560383328295, "conf": 1.0635212643295655, "grad_norm": 3.846290111541748, "lr": 2.552893030900247e-06 }, { "step": 1270, "T": 675, "positions": 16, "loss": 0.444377260748297, "ce": 1.7190171719703358, "tv": 0.30275059648556635, "conf": 1.2696258060313994, "grad_norm": 5.046192169189453, "lr": 2.528235063060952e-06 }, { "step": 1271, "T": 590, "positions": 16, "loss": 0.5550776163581759, "ce": 1.8166538445475453, "tv": 0.4149024917714996, "conf": 1.997552690539974, "grad_norm": 3.190276861190796, "lr": 2.50369041385215e-06 }, { "step": 1272, "T": 619, "positions": 16, "loss": 0.25012249969586264, "ce": 0.745234216582503, "tv": 0.19511008134577423, "conf": 1.394807095755823, "grad_norm": 1.4016579389572144, "lr": 2.479259207045431e-06 }, { "step": 1273, "T": 540, "positions": 16, "loss": 0.23033853631022794, "ce": 0.8260272258839905, "tv": 0.16415088822395774, "conf": 0.5246190492075611, "grad_norm": 3.177346706390381, "lr": 2.45494156584033e-06 }, { "step": 1274, "T": 557, "positions": 16, "loss": 0.6715475697383226, "ce": 2.67790082789179, "tv": 0.44861943588693975, "conf": 1.5016024552631535, "grad_norm": 4.586958408355713, "lr": 2.4307376128636843e-06 }, { "step": 1275, "T": 674, "positions": 16, "loss": 0.7116253552958369, "ce": 2.651727977208793, "tv": 0.4960584044456482, "conf": 1.3192483917810023, "grad_norm": 2.514854669570923, "lr": 2.4066474701690627e-06 }, { "step": 1276, "T": 551, "positions": 16, "loss": 0.33571518610597195, "ce": 1.5538863130959157, "tv": 0.20036285694220624, "conf": 1.3129589919149365, "grad_norm": 3.4912302494049072, "lr": 2.3826712592361105e-06 }, { "step": 1277, "T": 536, "positions": 16, "loss": 0.4901680149268941, "ce": 1.5905814412617474, "tv": 0.36789986373332795, "conf": 1.9281088322022697, "grad_norm": 3.523179054260254, "lr": 2.3588091009699274e-06 }, { "step": 1278, "T": 562, "positions": 16, "loss": 0.3119765783249022, "ce": 0.9841173275204937, "tv": 0.23729427442594897, "conf": 1.3256826751071484, "grad_norm": 1.7260839939117432, "lr": 2.3350611157005182e-06 }, { "step": 1279, "T": 914, "positions": 16, "loss": 0.6873956667041057, "ce": 3.0552264096240833, "tv": 0.42430336056713713, "conf": 1.5639438082490873, "grad_norm": 3.411593437194824, "lr": 2.311427423182111e-06 }, { "step": 1280, "T": 603, "positions": 16, "loss": 0.4126334180982667, "ce": 1.5429383482915, "tv": 0.2870439889629779, "conf": 1.2230570103067748, "grad_norm": 2.4981682300567627, "lr": 2.287908142592604e-06 }, { "step": 1281, "T": 637, "positions": 16, "loss": 0.4386472126861918, "ce": 1.6964191428924096, "tv": 0.2988947759004077, "conf": 1.9619526290419316, "grad_norm": 3.8151512145996094, "lr": 2.2645033925329515e-06 }, { "step": 1282, "T": 678, "positions": 16, "loss": 0.6076713305374142, "ce": 2.253636911385911, "tv": 0.4247862614574842, "conf": 1.9039191639021738, "grad_norm": 2.2914342880249023, "lr": 2.241213291026553e-06 }, { "step": 1283, "T": 1167, "positions": 16, "loss": 0.8133536868263036, "ce": 3.4902222608579905, "tv": 0.5159238773630932, "conf": 2.1380281554884277, "grad_norm": 4.846731185913086, "lr": 2.2180379555186844e-06 }, { "step": 1284, "T": 590, "positions": 16, "loss": 0.13224416239609127, "ce": 0.3461824784454848, "tv": 0.10847324556561944, "conf": 0.5431436949086983, "grad_norm": 1.6368472576141357, "lr": 2.1949775028758614e-06 }, { "step": 1285, "T": 539, "positions": 16, "loss": 0.3113267201988492, "ce": 1.1422208883668645, "tv": 0.21900515715242364, "conf": 0.9310621530626652, "grad_norm": 2.474282741546631, "lr": 2.1720320493853143e-06 }, { "step": 1286, "T": 1006, "positions": 16, "loss": 0.6340453557204455, "ce": 2.5430900184437633, "tv": 0.42192928050644696, "conf": 2.4789683747585514, "grad_norm": 2.9903109073638916, "lr": 2.149201710754348e-06 }, { "step": 1287, "T": 528, "positions": 16, "loss": 0.9067225286562461, "ce": 3.8972340743930545, "tv": 0.574443474994041, "conf": 1.0931397628446575, "grad_norm": 4.780813217163086, "lr": 2.126486602109759e-06 }, { "step": 1288, "T": 535, "positions": 16, "loss": 0.4759875703603029, "ce": 1.5191924392420333, "tv": 0.36007592489477247, "conf": 1.9005134414869644, "grad_norm": 2.1235511302948, "lr": 2.103886837997307e-06 }, { "step": 1289, "T": 607, "positions": 16, "loss": 0.3046506021971709, "ce": 1.0522841434798806, "tv": 0.22158021120731064, "conf": 1.255547586395445, "grad_norm": 2.554248571395874, "lr": 2.0814025323810856e-06 }, { "step": 1290, "T": 540, "positions": 16, "loss": 0.12073998796904561, "ce": 0.28695081780682585, "tv": 0.10227211901838018, "conf": 0.5040578398905862, "grad_norm": 2.1914398670196533, "lr": 2.0590337986429536e-06 }, { "step": 1291, "T": 1532, "positions": 16, "loss": 0.6850589781861345, "ce": 3.1222147170065, "tv": 0.41426391782442806, "conf": 1.5641227813766818, "grad_norm": 2.606445789337158, "lr": 2.036780749581993e-06 }, { "step": 1292, "T": 532, "positions": 16, "loss": 0.40420939546993395, "ce": 1.7268506823410235, "tv": 0.2572492588701607, "conf": 0.6891027519292265, "grad_norm": 3.698800802230835, "lr": 2.0146434974139068e-06 }, { "step": 1293, "T": 1716, "positions": 16, "loss": 0.8730372756253928, "ce": 4.034493924002163, "tv": 0.5217643083306029, "conf": 1.8829142151516862, "grad_norm": 1.8649672269821167, "lr": 1.9926221537704794e-06 }, { "step": 1294, "T": 545, "positions": 16, "loss": 0.4382272017392097, "ce": 1.6172633869429092, "tv": 0.30722317813888367, "conf": 1.45475668955067, "grad_norm": 2.6525063514709473, "lr": 1.9707168296989957e-06 }, { "step": 1295, "T": 606, "positions": 16, "loss": 0.42953202507032984, "ce": 1.5902778046518051, "tv": 0.30056028594572126, "conf": 1.4407472365435297, "grad_norm": 4.18870210647583, "lr": 1.9489276356616874e-06 }, { "step": 1296, "T": 582, "positions": 16, "loss": 0.2849927684001159, "ce": 1.0060579643686651, "tv": 0.20487443921774684, "conf": 0.9057947664440462, "grad_norm": 2.3288230895996094, "lr": 1.9272546815351817e-06 }, { "step": 1297, "T": 615, "positions": 16, "loss": 0.5567254032557685, "ce": 2.2779357354355625, "tv": 0.3654798228999425, "conf": 1.304137279646966, "grad_norm": 2.394080877304077, "lr": 1.9056980766099269e-06 }, { "step": 1298, "T": 534, "positions": 16, "loss": 0.3987264256393246, "ce": 1.3447922609575045, "tv": 0.29360800456561265, "conf": 0.8574204032734087, "grad_norm": 2.797790050506592, "lr": 1.884257929589664e-06 }, { "step": 1299, "T": 542, "positions": 16, "loss": 0.3418408353900304, "ce": 1.1697121503329981, "tv": 0.2498551408189087, "conf": 1.133191202588975, "grad_norm": 4.980839252471924, "lr": 1.862934348590889e-06 }, { "step": 1300, "T": 626, "positions": 16, "loss": 0.32821775087722926, "ce": 1.0302106673989329, "tv": 0.2502185411794926, "conf": 1.6162630098867905, "grad_norm": 2.8659374713897705, "lr": 1.8417274411422575e-06 }, { "step": 1301, "T": 525, "positions": 16, "loss": 0.5749356054839154, "ce": 2.0796059193271503, "tv": 0.407750020869571, "conf": 1.2323032074765568, "grad_norm": 3.4970829486846924, "lr": 1.82063731418409e-06 }, { "step": 1302, "T": 627, "positions": 16, "loss": 0.22313086722351727, "ce": 0.5247332651955503, "tv": 0.189619493086866, "conf": 2.620516334276772, "grad_norm": 3.8585550785064697, "lr": 1.7996640740678167e-06 }, { "step": 1303, "T": 641, "positions": 16, "loss": 0.43976509506842376, "ce": 1.740834782320917, "tv": 0.29520179518976875, "conf": 1.4237769665508069, "grad_norm": 2.8962910175323486, "lr": 1.7788078265554398e-06 }, { "step": 1304, "T": 605, "positions": 16, "loss": 0.253629302626905, "ce": 0.7465352242783467, "tv": 0.19886198190761206, "conf": 1.3011561241368956, "grad_norm": 2.535637617111206, "lr": 1.7580686768190025e-06 }, { "step": 1305, "T": 532, "positions": 16, "loss": 0.3576208998074435, "ce": 1.4702147685347882, "tv": 0.23399935271845607, "conf": 0.7904784587783308, "grad_norm": 1.889372706413269, "lr": 1.7374467294400566e-06 }, { "step": 1306, "T": 526, "positions": 16, "loss": 0.6783004039898515, "ce": 2.3704718966037035, "tv": 0.49028134904801846, "conf": 1.8722595921790344, "grad_norm": 2.4893429279327393, "lr": 1.716942088409132e-06 }, { "step": 1307, "T": 545, "positions": 16, "loss": 0.2559288087920777, "ce": 0.7440597291872564, "tv": 0.2016920426683555, "conf": 0.9975848817144833, "grad_norm": 3.4026706218719482, "lr": 1.6965548571252294e-06 }, { "step": 1308, "T": 561, "positions": 16, "loss": 0.5112230828381144, "ce": 2.2151417655913974, "tv": 0.3218987851869315, "conf": 1.2059940171893686, "grad_norm": 3.1544885635375977, "lr": 1.6762851383952616e-06 }, { "step": 1309, "T": 540, "positions": 16, "loss": 0.3030110105928543, "ce": 1.067124888883427, "tv": 0.2181094641055097, "conf": 0.8672968114675426, "grad_norm": 3.8823771476745605, "lr": 1.6561330344335935e-06 }, { "step": 1310, "T": 1025, "positions": 16, "loss": 0.7885285930369719, "ce": 3.495192641606991, "tv": 0.48778814350953326, "conf": 2.6877873822685214, "grad_norm": 2.9532673358917236, "lr": 1.6360986468614614e-06 }, { "step": 1311, "T": 1316, "positions": 16, "loss": 0.7134118796675466, "ce": 3.2546546361255366, "tv": 0.4310515577890328, "conf": 1.3717508377390004, "grad_norm": 1.4244188070297241, "lr": 1.6161820767065045e-06 }, { "step": 1312, "T": 593, "positions": 16, "loss": 0.8272894956171513, "ce": 3.6837839521467686, "tv": 0.5099012218415737, "conf": 2.4858479760587215, "grad_norm": 3.8709969520568848, "lr": 1.5963834244022535e-06 }, { "step": 1313, "T": 1364, "positions": 16, "loss": 0.9213610184087884, "ce": 4.419323846937459, "tv": 0.5326984955681837, "conf": 1.0751094228662623, "grad_norm": 2.523613214492798, "lr": 1.5767027897875957e-06 }, { "step": 1314, "T": 563, "positions": 16, "loss": 0.5345822620947729, "ce": 2.2467383039197557, "tv": 0.34434271827376506, "conf": 1.207656074759793, "grad_norm": 2.608604669570923, "lr": 1.5571402721062944e-06 }, { "step": 1315, "T": 621, "positions": 16, "loss": 0.5568107590243017, "ce": 2.1382672368725935, "tv": 0.38109337483501804, "conf": 1.9277956154133744, "grad_norm": 4.405494213104248, "lr": 1.5376959700064863e-06 }, { "step": 1316, "T": 1798, "positions": 16, "loss": 0.7352071111090481, "ce": 3.3526808151509613, "tv": 0.4443766949698329, "conf": 2.3018674668855965, "grad_norm": 3.083189010620117, "lr": 1.5183699815401826e-06 }, { "step": 1317, "T": 565, "positions": 16, "loss": 0.6611776797612947, "ce": 2.69309912581798, "tv": 0.43540866021157854, "conf": 1.8576968032290404, "grad_norm": 5.952272415161133, "lr": 1.4991624041627711e-06 }, { "step": 1318, "T": 592, "positions": 16, "loss": 0.34953452376066707, "ce": 1.0940068227482698, "tv": 0.26681537745753303, "conf": 1.292860374383963, "grad_norm": 5.669597148895264, "lr": 1.4800733347325152e-06 }, { "step": 1319, "T": 559, "positions": 16, "loss": 0.18398696157873928, "ce": 0.666515745325114, "tv": 0.1303726533697045, "conf": 1.071647357539959, "grad_norm": 2.362575054168701, "lr": 1.4611028695101032e-06 }, { "step": 1320, "T": 538, "positions": 16, "loss": 0.08331461847410537, "ce": 0.23935874522430822, "tv": 0.06597638258244842, "conf": 0.32710846787267656, "grad_norm": 3.0379669666290283, "lr": 1.44225110415811e-06 }, { "step": 1321, "T": 554, "positions": 16, "loss": 0.5788643799824058, "ce": 2.345156094070262, "tv": 0.38260974521290336, "conf": 1.0897350684981575, "grad_norm": 3.1925320625305176, "lr": 1.4235181337405535e-06 }, { "step": 1322, "T": 577, "positions": 16, "loss": 0.7389976242557168, "ce": 3.0495237964205444, "tv": 0.4822724936529994, "conf": 1.7187909673157264, "grad_norm": 3.496079683303833, "lr": 1.4049040527224028e-06 }, { "step": 1323, "T": 630, "positions": 16, "loss": 0.2157803875124955, "ce": 0.7299504678321682, "tv": 0.1586503763537621, "conf": 0.9410049844776436, "grad_norm": 4.395636558532715, "lr": 1.3864089549691012e-06 }, { "step": 1324, "T": 672, "positions": 16, "loss": 0.5654377158480202, "ce": 2.050132908813339, "tv": 0.4004715935461718, "conf": 1.0060768627075667, "grad_norm": 3.300370693206787, "lr": 1.368032933746094e-06 }, { "step": 1325, "T": 536, "positions": 16, "loss": 0.20001863559537014, "ce": 0.6165792412698465, "tv": 0.1537341269213357, "conf": 0.7122853432569837, "grad_norm": 2.671929359436035, "lr": 1.349776081718357e-06 }, { "step": 1326, "T": 528, "positions": 16, "loss": 0.3420420333313814, "ce": 1.2531595679247403, "tv": 0.24080676158700953, "conf": 1.3612448564838957, "grad_norm": 2.951481580734253, "lr": 1.3316384909499269e-06 }, { "step": 1327, "T": 547, "positions": 16, "loss": 0.500007895519957, "ce": 1.746154274791479, "tv": 0.36154719395563006, "conf": 0.7772764536489376, "grad_norm": 3.2133450508117676, "lr": 1.3136202529034525e-06 }, { "step": 1328, "T": 639, "positions": 16, "loss": 0.2056921381408756, "ce": 0.7037474644891972, "tv": 0.1503526486485498, "conf": 0.9291574292634159, "grad_norm": 3.3357317447662354, "lr": 1.2957214584396997e-06 }, { "step": 1329, "T": 602, "positions": 16, "loss": 0.3572312010492169, "ce": 1.572926041606479, "tv": 0.2221539935671899, "conf": 1.0922927600331604, "grad_norm": 4.621433258056641, "lr": 1.2779421978171362e-06 }, { "step": 1330, "T": 665, "positions": 16, "loss": 0.19785778750338068, "ce": 0.6461875173981753, "tv": 0.14804337339955964, "conf": 1.2521113026014064, "grad_norm": 2.0523149967193604, "lr": 1.2602825606914475e-06 }, { "step": 1331, "T": 600, "positions": 16, "loss": 0.6019231902901083, "ce": 2.416053827153519, "tv": 0.40035312925465405, "conf": 1.5605458584905136, "grad_norm": 3.137321949005127, "lr": 1.242742636115088e-06 }, { "step": 1332, "T": 1649, "positions": 16, "loss": 0.961312742670998, "ce": 4.679292366839945, "tv": 0.5482039013877511, "conf": 1.7454785018926486, "grad_norm": 3.308865785598755, "lr": 1.2253225125368362e-06 }, { "step": 1333, "T": 1232, "positions": 16, "loss": 0.8264846826677967, "ce": 3.1539041074738634, "tv": 0.567882518560964, "conf": 2.4848985870208553, "grad_norm": 2.4967780113220215, "lr": 1.2080222778013573e-06 }, { "step": 1334, "T": 528, "positions": 16, "loss": 0.5778510831496533, "ce": 1.805721251410887, "tv": 0.4414210756858665, "conf": 1.2513032980205026, "grad_norm": 3.4177699089050293, "lr": 1.1908420191487413e-06 }, { "step": 1335, "T": 1067, "positions": 16, "loss": 0.8347478378564119, "ce": 3.450706450268626, "tv": 0.5440857792273164, "conf": 2.062666726538737, "grad_norm": 3.4593076705932617, "lr": 1.1737818232140845e-06 }, { "step": 1336, "T": 524, "positions": 16, "loss": 0.8215979924425483, "ce": 3.656299671391025, "tv": 0.5066311256960034, "conf": 1.579361911630258, "grad_norm": 2.9861645698547363, "lr": 1.1568417760270318e-06 }, { "step": 1337, "T": 634, "positions": 16, "loss": 0.3954171084915288, "ce": 1.1492589998524636, "tv": 0.3116569003323093, "conf": 1.5992323771351948, "grad_norm": 3.4055349826812744, "lr": 1.140021963011359e-06 }, { "step": 1338, "T": 593, "positions": 16, "loss": 0.2430524351511849, "ce": 0.8987658918449597, "tv": 0.17019538579552318, "conf": 1.2404923362191767, "grad_norm": 5.414515972137451, "lr": 1.1233224689845251e-06 }, { "step": 1339, "T": 524, "positions": 16, "loss": 0.5427128813935269, "ce": 1.8058727944990096, "tv": 0.40236176993130357, "conf": 1.4853488928026366, "grad_norm": 4.161301612854004, "lr": 1.106743378157271e-06 }, { "step": 1340, "T": 574, "positions": 16, "loss": 0.351082963461522, "ce": 1.2149809614638798, "tv": 0.25509430281817913, "conf": 1.1515106329475202, "grad_norm": 3.1692450046539307, "lr": 1.0902847741331712e-06 }, { "step": 1341, "T": 602, "positions": 16, "loss": 0.3928176970457571, "ce": 1.2393586530602079, "tv": 0.29875759674928304, "conf": 1.1725906654178635, "grad_norm": 3.0894126892089844, "lr": 1.0739467399082159e-06 }, { "step": 1342, "T": 573, "positions": 16, "loss": 0.23328806975041516, "ce": 0.8091519804428629, "tv": 0.1693031950671866, "conf": 0.5782721392615713, "grad_norm": 2.440384864807129, "lr": 1.0577293578703962e-06 }, { "step": 1343, "T": 677, "positions": 16, "loss": 0.47665613745630253, "ce": 1.869305651470313, "tv": 0.32191730466729496, "conf": 0.9300159830805228, "grad_norm": 2.060943365097046, "lr": 1.041632709799306e-06 }, { "step": 1344, "T": 568, "positions": 16, "loss": 0.8019957542419434, "ce": 3.0999247385188937, "tv": 0.5466703251004219, "conf": 1.6583843639818951, "grad_norm": 3.7723796367645264, "lr": 1.0256568768656882e-06 }, { "step": 1345, "T": 961, "positions": 16, "loss": 0.6608691439032555, "ce": 2.592832127586007, "tv": 0.44620659900829196, "conf": 2.5646031132200733, "grad_norm": 5.495153427124023, "lr": 1.0098019396310637e-06 }, { "step": 1346, "T": 569, "positions": 16, "loss": 0.6967251561582088, "ce": 2.6756995408795774, "tv": 0.47683913353830576, "conf": 1.2062350646447157, "grad_norm": 4.366162300109863, "lr": 9.94067978047311e-07 }, { "step": 1347, "T": 533, "positions": 16, "loss": 0.3873721952622873, "ce": 1.3747818002266285, "tv": 0.2776600193028571, "conf": 2.3326405425309513, "grad_norm": 3.3744850158691406, "lr": 9.784550714562634e-07 }, { "step": 1348, "T": 539, "positions": 16, "loss": 0.523498943541199, "ce": 1.640027069952339, "tv": 0.39944027457386255, "conf": 1.3811534851411125, "grad_norm": 2.118760347366333, "lr": 9.629632985893033e-07 }, { "step": 1349, "T": 562, "positions": 16, "loss": 0.38454888548585586, "ce": 1.5105768862231344, "tv": 0.25943464945885353, "conf": 0.7077796741177735, "grad_norm": 3.1137471199035645, "lr": 9.475927375669741e-07 }, { "step": 1350, "T": 531, "positions": 16, "loss": 0.21832645665381278, "ce": 0.7933207961998505, "tv": 0.15443820126711216, "conf": 0.4797239902873116, "grad_norm": 2.2769935131073, "lr": 9.323434658985858e-07 }, { "step": 1351, "T": 539, "positions": 16, "loss": 0.6125578606734052, "ce": 2.630360469902371, "tv": 0.3883575723739341, "conf": 0.9640967466170878, "grad_norm": 4.305639266967773, "lr": 9.172155604818128e-07 }, { "step": 1352, "T": 618, "positions": 16, "loss": 0.43413843266898766, "ce": 1.5679690346942152, "tv": 0.30815725616685086, "conf": 1.3851084287415762, "grad_norm": 4.794898986816406, "lr": 9.02209097602319e-07 }, { "step": 1353, "T": 535, "positions": 16, "loss": 0.07735644797458008, "ce": 0.22024982275845062, "tv": 0.061479404339479515, "conf": 0.5091864357470405, "grad_norm": 2.1804661750793457, "lr": 8.873241529333776e-07 }, { "step": 1354, "T": 675, "positions": 16, "loss": 0.6936723226681352, "ce": 2.5529553769156337, "tv": 0.48708533123135567, "conf": 2.4604504310991615, "grad_norm": 2.598315477371216, "lr": 8.725608015354686e-07 }, { "step": 1355, "T": 606, "positions": 16, "loss": 0.13777102567837574, "ce": 0.45942674395757876, "tv": 0.10203150360030122, "conf": 0.7201071858518091, "grad_norm": 1.4300020933151245, "lr": 8.579191178559126e-07 }, { "step": 1356, "T": 529, "positions": 16, "loss": 0.5761774540878832, "ce": 2.030436197092058, "tv": 0.4145931521197781, "conf": 1.246530931352197, "grad_norm": 2.2109572887420654, "lr": 8.433991757284986e-07 }, { "step": 1357, "T": 529, "positions": 16, "loss": 0.9136035777628422, "ce": 3.6068764813244343, "tv": 0.6143510183319449, "conf": 2.0969655142107513, "grad_norm": 2.00809383392334, "lr": 8.290010483731125e-07 }, { "step": 1358, "T": 530, "positions": 16, "loss": 0.6078294809194631, "ce": 2.180779203576094, "tv": 0.4330572998005664, "conf": 1.4911749268329686, "grad_norm": 2.6745171546936035, "lr": 8.147248083953562e-07 }, { "step": 1359, "T": 1003, "positions": 16, "loss": 0.5665045387322607, "ce": 2.488114797589276, "tv": 0.3529922855559562, "conf": 1.051145116925909, "grad_norm": 1.4558802843093872, "lr": 8.005705277861874e-07 }, { "step": 1360, "T": 551, "positions": 16, "loss": 0.45722498960094526, "ce": 1.9818785114002822, "tv": 0.2878190475748852, "conf": 1.8184956457698718, "grad_norm": 4.0106587409973145, "lr": 7.865382779215641e-07 }, { "step": 1361, "T": 810, "positions": 16, "loss": 0.421307695318319, "ce": 1.7729059498301467, "tv": 0.271130105600605, "conf": 1.2540531637219732, "grad_norm": 1.6723641157150269, "lr": 7.726281295620835e-07 }, { "step": 1362, "T": 533, "positions": 16, "loss": 0.31826633211039734, "ce": 1.0705027254034576, "tv": 0.23468450737709645, "conf": 0.8806994913641667, "grad_norm": 3.4816946983337402, "lr": 7.588401528525968e-07 }, { "step": 1363, "T": 672, "positions": 16, "loss": 0.7086706832051277, "ce": 2.415406484855339, "tv": 0.5190333742648363, "conf": 1.470931820338592, "grad_norm": 2.3267858028411865, "lr": 7.451744173219116e-07 }, { "step": 1364, "T": 1041, "positions": 16, "loss": 0.8540232655045656, "ce": 4.4116321040733055, "tv": 0.458733400560277, "conf": 1.420378851604255, "grad_norm": 4.541460990905762, "lr": 7.316309918823788e-07 }, { "step": 1365, "T": 599, "positions": 16, "loss": 0.34892021594714606, "ce": 1.1821216600119442, "tv": 0.2563422881394217, "conf": 1.5509028051781182, "grad_norm": 2.863093376159668, "lr": 7.182099448295926e-07 }, { "step": 1366, "T": 678, "positions": 16, "loss": 0.22608062223662273, "ce": 0.5481004673056304, "tv": 0.19030063885475101, "conf": 1.3280920196383548, "grad_norm": 2.447845697402954, "lr": 7.049113438420212e-07 }, { "step": 1367, "T": 672, "positions": 16, "loss": 0.6387135150143877, "ce": 2.598574220086448, "tv": 0.420951216656249, "conf": 0.9576050124014728, "grad_norm": 4.37203311920166, "lr": 6.917352559806772e-07 }, { "step": 1368, "T": 570, "positions": 16, "loss": 0.8039219407364726, "ce": 3.140282080581528, "tv": 0.5443263822235167, "conf": 1.80593238491565, "grad_norm": 5.28290319442749, "lr": 6.786817476887725e-07 }, { "step": 1369, "T": 1218, "positions": 16, "loss": 0.9779249182902277, "ce": 4.669787060469389, "tv": 0.5677180113270879, "conf": 1.7339687160419999, "grad_norm": 3.98014760017395, "lr": 6.657508847913801e-07 }, { "step": 1370, "T": 529, "positions": 16, "loss": 0.3261018989533113, "ce": 1.039199301298595, "tv": 0.2468688538592687, "conf": 0.8243788761243828, "grad_norm": 2.9239542484283447, "lr": 6.529427324951154e-07 }, { "step": 1371, "T": 532, "positions": 16, "loss": 0.3662923152260191, "ce": 1.242418183742302, "tv": 0.26894499076479406, "conf": 1.1846198884964618, "grad_norm": 2.6815345287323, "lr": 6.402573553878022e-07 }, { "step": 1372, "T": 541, "positions": 16, "loss": 0.2568731539996634, "ce": 0.8105563197625543, "tv": 0.19535280066571659, "conf": 0.6946943266943885, "grad_norm": 3.426892042160034, "lr": 6.27694817438132e-07 }, { "step": 1373, "T": 531, "positions": 16, "loss": 0.38915664491651114, "ce": 1.4338999180617975, "tv": 0.27307406495674513, "conf": 1.3983406539663168, "grad_norm": 3.159423351287842, "lr": 6.152551819953667e-07 }, { "step": 1374, "T": 630, "positions": 16, "loss": 0.3192265586221765, "ce": 1.2803588124515954, "tv": 0.2124340870323067, "conf": 0.7492949064674299, "grad_norm": 1.8888734579086304, "lr": 6.029385117889946e-07 }, { "step": 1375, "T": 1225, "positions": 16, "loss": 0.8129932829760946, "ce": 3.68085291859461, "tv": 0.4943422262731474, "conf": 2.57584311518076, "grad_norm": 3.0094072818756104, "lr": 5.907448689284361e-07 }, { "step": 1376, "T": 622, "positions": 16, "loss": 0.5431699150976783, "ce": 2.028994179725487, "tv": 0.3780783372276346, "conf": 1.6868189961852356, "grad_norm": 2.8042452335357666, "lr": 5.786743149027135e-07 }, { "step": 1377, "T": 950, "positions": 16, "loss": 0.5595700795338416, "ce": 2.1485029428529288, "tv": 0.3830220094368997, "conf": 2.1928414328969184, "grad_norm": 2.21563458442688, "lr": 5.667269105801537e-07 }, { "step": 1378, "T": 537, "positions": 16, "loss": 0.16518097570587997, "ce": 0.6363149051061896, "tv": 0.11283277416623605, "conf": 0.24460147791705822, "grad_norm": 1.466214895248413, "lr": 5.549027162080666e-07 }, { "step": 1379, "T": 616, "positions": 16, "loss": 0.1444751341623487, "ce": 0.3815755222567532, "tv": 0.11813064878151636, "conf": 1.1211972944438457, "grad_norm": 1.798103928565979, "lr": 5.432017914124593e-07 }, { "step": 1380, "T": 532, "positions": 16, "loss": 0.1260855177652047, "ce": 0.3024972821604024, "tv": 0.10648421252608387, "conf": 1.3670764637437947, "grad_norm": 1.7907907962799072, "lr": 5.316241951977191e-07 }, { "step": 1381, "T": 587, "positions": 16, "loss": 0.22826984566654573, "ce": 0.5812620565980069, "tv": 0.1890484870527871, "conf": 0.9948174129616714, "grad_norm": 3.2528157234191895, "lr": 5.201699859463283e-07 }, { "step": 1382, "T": 611, "positions": 16, "loss": 0.5443939625911298, "ce": 2.1586811551132996, "tv": 0.3650287185009802, "conf": 1.5534179259210124, "grad_norm": 2.736449956893921, "lr": 5.088392214185556e-07 }, { "step": 1383, "T": 534, "positions": 16, "loss": 0.24599529360193628, "ce": 0.7564560113460743, "tv": 0.189277442290404, "conf": 0.6469115080823112, "grad_norm": 2.605602264404297, "lr": 4.976319587521788e-07 }, { "step": 1384, "T": 532, "positions": 16, "loss": 0.09460572763600794, "ce": 0.18598889524037077, "tv": 0.08445204155214014, "conf": 0.12921953179102275, "grad_norm": 1.0757803916931152, "lr": 4.865482544621991e-07 }, { "step": 1385, "T": 525, "positions": 16, "loss": 0.29875031008850783, "ce": 1.2884013853763463, "tv": 0.1887890723592136, "conf": 0.9661238199696527, "grad_norm": 1.6090339422225952, "lr": 4.755881644405352e-07 }, { "step": 1386, "T": 628, "positions": 16, "loss": 0.5006580656990991, "ce": 1.916410364836338, "tv": 0.3433522628911305, "conf": 1.562176304630606, "grad_norm": 1.4924249649047852, "lr": 4.64751743955763e-07 }, { "step": 1387, "T": 940, "positions": 16, "loss": 0.6963636733707972, "ce": 3.194141599892646, "tv": 0.4188327808496979, "conf": 3.2770044371976654, "grad_norm": 2.5542678833007812, "lr": 4.540390476528239e-07 }, { "step": 1388, "T": 672, "positions": 16, "loss": 0.31989437788661235, "ce": 0.9780544596994787, "tv": 0.24676550153799326, "conf": 1.1582405042410322, "grad_norm": 3.196180582046509, "lr": 4.434501295527582e-07 }, { "step": 1389, "T": 642, "positions": 16, "loss": 0.7045742415357381, "ce": 2.7151715777581558, "tv": 0.481174539134372, "conf": 2.383767822990194, "grad_norm": 2.8787827491760254, "lr": 4.329850430524252e-07 }, { "step": 1390, "T": 673, "positions": 16, "loss": 0.5567433116957545, "ce": 1.7249403893947601, "tv": 0.42694367095828056, "conf": 2.7540433793328702, "grad_norm": 2.6515402793884277, "lr": 4.226438409242389e-07 }, { "step": 1391, "T": 532, "positions": 16, "loss": 0.42580571670441714, "ce": 1.6389799085209233, "tv": 0.29100857864978025, "conf": 1.899168690970177, "grad_norm": 3.8799221515655518, "lr": 4.124265753159023e-07 }, { "step": 1392, "T": 528, "positions": 16, "loss": 0.7060778141021729, "ce": 2.6595435878261924, "tv": 0.4890260826796293, "conf": 2.142836606130004, "grad_norm": 2.982240676879883, "lr": 4.0233329775013484e-07 }, { "step": 1393, "T": 585, "positions": 16, "loss": 0.5845260816172413, "ce": 2.2354812421035604, "tv": 0.40108662942748197, "conf": 1.4412997625607535, "grad_norm": 3.333622694015503, "lr": 3.9236405912442544e-07 }, { "step": 1394, "T": 544, "positions": 16, "loss": 0.264361153434038, "ce": 0.8680619246815695, "tv": 0.19728328635801518, "conf": 1.0360017678038957, "grad_norm": 4.059370994567871, "lr": 3.8251890971077174e-07 }, { "step": 1395, "T": 544, "positions": 16, "loss": 0.2457871010483359, "ce": 0.7034554198436496, "tv": 0.19493507712468272, "conf": 0.832837652506889, "grad_norm": 2.772946357727051, "lr": 3.7279789915542483e-07 }, { "step": 1396, "T": 674, "positions": 16, "loss": 0.6112042415479664, "ce": 2.243789719432243, "tv": 0.4298058541753562, "conf": 1.7598580067529976, "grad_norm": 3.159332036972046, "lr": 3.632010764786309e-07 }, { "step": 1397, "T": 558, "positions": 16, "loss": 0.24606053001480177, "ce": 0.7326369714755856, "tv": 0.19199648951325798, "conf": 1.219666166137607, "grad_norm": 2.346780776977539, "lr": 3.5372849007440645e-07 }, { "step": 1398, "T": 596, "positions": 16, "loss": 0.39628321758027596, "ce": 1.2122268951998194, "tv": 0.30562282344362757, "conf": 1.4853732958411676, "grad_norm": 5.706543922424316, "lr": 3.44380187710272e-07 }, { "step": 1399, "T": 644, "positions": 16, "loss": 0.5843330398201942, "ce": 1.9094162742840126, "tv": 0.4371015715878457, "conf": 1.9822756698122248, "grad_norm": 3.979228973388672, "lr": 3.351562165270161e-07 }, { "step": 1400, "T": 676, "positions": 16, "loss": 0.36878986030933447, "ce": 1.2994422775846033, "tv": 0.26538403212907724, "conf": 1.4000888825394213, "grad_norm": 4.21524715423584, "lr": 3.260566230384704e-07 }, { "step": 1401, "T": 643, "positions": 16, "loss": 0.5320312170078978, "ce": 1.6897249398753047, "tv": 0.40339860290987417, "conf": 1.6429136809892952, "grad_norm": 2.257890462875366, "lr": 3.1708145313125713e-07 }, { "step": 1402, "T": 526, "positions": 16, "loss": 0.14401478498075448, "ce": 0.5155698814251082, "tv": 0.10273089327529306, "conf": 0.2857581240423315, "grad_norm": 1.0807805061340332, "lr": 3.0823075206457554e-07 }, { "step": 1403, "T": 540, "positions": 16, "loss": 0.1314990824994311, "ce": 0.3346032976570257, "tv": 0.10893195484641183, "conf": 0.5854785352594263, "grad_norm": 0.7905945777893066, "lr": 2.995045644699518e-07 }, { "step": 1404, "T": 665, "positions": 16, "loss": 0.7256838482571766, "ce": 2.9708711751154624, "tv": 0.47621860797517, "conf": 1.7578065465204418, "grad_norm": 3.0097434520721436, "lr": 2.9090293435103946e-07 }, { "step": 1405, "T": 530, "positions": 16, "loss": 0.5746653395617614, "ce": 2.3036247677373467, "tv": 0.3825587444662233, "conf": 1.9334724663058296, "grad_norm": 6.428211212158203, "lr": 2.82425905083375e-07 }, { "step": 1406, "T": 526, "positions": 16, "loss": 0.47188002537586726, "ce": 1.9795188857533503, "tv": 0.30436460419150535, "conf": 0.7829578548598874, "grad_norm": 3.9885048866271973, "lr": 2.7407351941417526e-07 }, { "step": 1407, "T": 539, "positions": 16, "loss": 0.23850079074964015, "ce": 0.7767289607418206, "tv": 0.17869767428419436, "conf": 0.6131941325271848, "grad_norm": 4.6230573654174805, "lr": 2.6584581946211e-07 }, { "step": 1408, "T": 548, "positions": 16, "loss": 0.5282010885021009, "ce": 2.1215022635624337, "tv": 0.3511676035595883, "conf": 2.0560061764842232, "grad_norm": 4.6240949630737305, "lr": 2.577428467170989e-07 }, { "step": 1409, "T": 545, "positions": 16, "loss": 0.42163386556421756, "ce": 2.029989143236193, "tv": 0.24292774752757396, "conf": 0.7347561365811259, "grad_norm": 3.0601682662963867, "lr": 2.4976464204009567e-07 }, { "step": 1410, "T": 889, "positions": 16, "loss": 0.7472257132540108, "ce": 3.374778571727802, "tv": 0.45527541944466066, "conf": 0.6577919493964522, "grad_norm": 2.522801160812378, "lr": 2.4191124566289025e-07 }, { "step": 1411, "T": 673, "positions": 16, "loss": 0.4513410174085948, "ce": 1.7286505972651867, "tv": 0.3094177423208748, "conf": 1.35319538410281, "grad_norm": 3.346902847290039, "lr": 2.341826971878902e-07 }, { "step": 1412, "T": 597, "positions": 16, "loss": 0.16752652035029314, "ce": 0.597546606247306, "tv": 0.1197465215682314, "conf": 0.6750122213779832, "grad_norm": 1.988788366317749, "lr": 2.2657903558793992e-07 }, { "step": 1413, "T": 601, "positions": 16, "loss": 0.4039953362007509, "ce": 1.5969171945744165, "tv": 0.2714484775751771, "conf": 1.0122436687331726, "grad_norm": 3.7248902320861816, "lr": 2.1910029920610974e-07 }, { "step": 1414, "T": 525, "positions": 16, "loss": 0.7026864122599363, "ce": 2.6272792061790824, "tv": 0.488842791877687, "conf": 2.072687036852585, "grad_norm": 3.683591365814209, "lr": 2.117465257555157e-07 }, { "step": 1415, "T": 541, "positions": 16, "loss": 0.15902051894045144, "ce": 0.35958341759396717, "tv": 0.13673574956555967, "conf": 0.5307832469875393, "grad_norm": 1.3902156352996826, "lr": 2.0451775231911385e-07 }, { "step": 1416, "T": 561, "positions": 16, "loss": 0.3225683951905012, "ce": 1.0723468821202005, "tv": 0.23925968432558875, "conf": 1.6530515905717713, "grad_norm": 2.3860256671905518, "lr": 1.974140153495202e-07 }, { "step": 1417, "T": 556, "positions": 16, "loss": 0.3511094097048044, "ce": 1.590950726982328, "tv": 0.21334926984127378, "conf": 0.8080562151607182, "grad_norm": 2.2902631759643555, "lr": 1.9043535066883833e-07 }, { "step": 1418, "T": 541, "positions": 16, "loss": 0.15529317538130272, "ce": 0.43026015618625024, "tv": 0.12474129065230954, "conf": 0.7523324506468307, "grad_norm": 2.8890364170074463, "lr": 1.8358179346845694e-07 }, { "step": 1419, "T": 582, "positions": 16, "loss": 0.663256403448031, "ce": 2.6737771821681235, "tv": 0.43986521198166884, "conf": 0.7219708979052264, "grad_norm": 3.085357904434204, "lr": 1.768533783088916e-07 }, { "step": 1420, "T": 543, "positions": 16, "loss": 0.49585982098437853, "ce": 2.018926271000936, "tv": 0.3266302205486227, "conf": 1.115913836237921, "grad_norm": 3.437324285507202, "lr": 1.7025013911959876e-07 }, { "step": 1421, "T": 636, "positions": 16, "loss": 0.39750200920389034, "ce": 1.0884068732775631, "tv": 0.32073481536644977, "conf": 1.2886097684785227, "grad_norm": 2.828590154647827, "lr": 1.6377210919880648e-07 }, { "step": 1422, "T": 530, "positions": 16, "loss": 0.6444926680705976, "ce": 2.1147771390355956, "tv": 0.4811277249827981, "conf": 2.198874597809464, "grad_norm": 2.653425693511963, "lr": 1.57419321213359e-07 }, { "step": 1423, "T": 581, "positions": 16, "loss": 0.37045499067608034, "ce": 1.218459188210545, "tv": 0.27623231209872756, "conf": 0.9171264180282037, "grad_norm": 3.940319061279297, "lr": 1.51191807198528e-07 }, { "step": 1424, "T": 569, "positions": 16, "loss": 0.3681696105195442, "ce": 1.3414347301331873, "tv": 0.26002905600034865, "conf": 1.398559032338, "grad_norm": 3.161468982696533, "lr": 1.4508959855787097e-07 }, { "step": 1425, "T": 573, "positions": 16, "loss": 0.43463184659503895, "ce": 1.7291192189813955, "tv": 0.29079992557853984, "conf": 1.706704773605452, "grad_norm": 3.080125331878662, "lr": 1.3911272606307036e-07 }, { "step": 1426, "T": 543, "positions": 16, "loss": 0.6886894097551703, "ce": 2.7555836951360106, "tv": 0.45903449691832066, "conf": 1.7820216286927462, "grad_norm": 4.066409587860107, "lr": 1.332612198537697e-07 }, { "step": 1427, "T": 546, "positions": 16, "loss": 0.2228632298276807, "ce": 0.8298141755526558, "tv": 0.15542423183887877, "conf": 0.7338478737360106, "grad_norm": 2.5591800212860107, "lr": 1.2753510943742386e-07 }, { "step": 1428, "T": 672, "positions": 16, "loss": 0.7449418241158128, "ce": 2.556678906083107, "tv": 0.5436377157457173, "conf": 2.3483014062512666, "grad_norm": 2.571697473526001, "lr": 1.2193442368915732e-07 }, { "step": 1429, "T": 604, "positions": 16, "loss": 0.6539486376568675, "ce": 2.981157222762704, "tv": 0.3953699069097638, "conf": 1.283149882525322, "grad_norm": 5.728784084320068, "lr": 1.1645919085161717e-07 }, { "step": 1430, "T": 622, "positions": 16, "loss": 0.43118471169691475, "ce": 1.6366553809539255, "tv": 0.29724353188066743, "conf": 1.1953300739856836, "grad_norm": 1.5393742322921753, "lr": 1.1110943853482047e-07 }, { "step": 1431, "T": 542, "positions": 16, "loss": 0.473977276145888, "ce": 1.9720564935832954, "tv": 0.30752403795850114, "conf": 0.7504242390692255, "grad_norm": 4.779374122619629, "lr": 1.0588519371602657e-07 }, { "step": 1432, "T": 539, "positions": 16, "loss": 0.2522634046035819, "ce": 0.7474198312847875, "tv": 0.1972460287506692, "conf": 1.6566470563411713, "grad_norm": 2.787431001663208, "lr": 1.0078648273959268e-07 }, { "step": 1433, "T": 636, "positions": 16, "loss": 0.32985723653109744, "ce": 1.1653409036807716, "tv": 0.23702572303591296, "conf": 1.3272847201442346, "grad_norm": 2.782348871231079, "lr": 9.581333131685467e-08 }, { "step": 1434, "T": 587, "positions": 16, "loss": 0.18271047204962088, "ce": 0.8347769660123845, "tv": 0.1102586376673571, "conf": 1.0635460482355086, "grad_norm": 1.378968358039856, "lr": 9.096576452597427e-08 }, { "step": 1435, "T": 633, "positions": 16, "loss": 0.49427844413366984, "ce": 1.8199229671090507, "tv": 0.34698463348468067, "conf": 1.366318000932381, "grad_norm": 2.9909591674804688, "lr": 8.624380681183653e-08 }, { "step": 1436, "T": 536, "positions": 16, "loss": 0.12962921611779166, "ce": 0.3005448071932051, "tv": 0.11063859799844522, "conf": 0.7581519434835584, "grad_norm": 4.206466197967529, "lr": 8.164748198591088e-08 }, { "step": 1437, "T": 529, "positions": 16, "loss": 0.24991470723762177, "ce": 0.9077850674657384, "tv": 0.17681799989077263, "conf": 1.729051529429853, "grad_norm": 5.767305850982666, "lr": 7.717681322613745e-08 }, { "step": 1438, "T": 580, "positions": 16, "loss": 0.16398944304455654, "ce": 0.4207664693517472, "tv": 0.13545865981177485, "conf": 1.0443958432381102, "grad_norm": 2.403186559677124, "lr": 7.283182307681324e-08 }, { "step": 1439, "T": 1549, "positions": 16, "loss": 0.7430567695409991, "ce": 3.1063058642102988, "tv": 0.48047355574090034, "conf": 3.614244196331433, "grad_norm": 3.3434643745422363, "lr": 6.861253344846719e-08 }, { "step": 1440, "T": 636, "positions": 16, "loss": 0.2699535967258271, "ce": 0.7052193532053934, "tv": 0.22159073085640557, "conf": 1.218800623895902, "grad_norm": 4.569448947906494, "lr": 6.451896561776305e-08 }, { "step": 1441, "T": 899, "positions": 16, "loss": 0.45136042856029235, "ce": 1.9692144442466315, "tv": 0.28270998608059017, "conf": 1.255079424230189, "grad_norm": 1.6832478046417236, "lr": 6.055114022738007e-08 }, { "step": 1442, "T": 611, "positions": 16, "loss": 0.13630895356891415, "ce": 0.42215990954022686, "tv": 0.10454772877892538, "conf": 0.8285169775317627, "grad_norm": 1.2091988325119019, "lr": 5.670907728591857e-08 }, { "step": 1443, "T": 533, "positions": 16, "loss": 0.2674841147454572, "ce": 0.970686871978387, "tv": 0.18935048897947127, "conf": 1.1503682388219636, "grad_norm": 1.8962655067443848, "lr": 5.299279616779174e-08 }, { "step": 1444, "T": 567, "positions": 16, "loss": 0.6597699656686018, "ce": 2.74990869031717, "tv": 0.4275323264191684, "conf": 1.5703926566784503, "grad_norm": 2.8308725357055664, "lr": 4.9402315613139575e-08 }, { "step": 1445, "T": 539, "positions": 16, "loss": 0.09796427045057499, "ce": 0.30396355926080787, "tv": 0.07507545497901447, "conf": 0.37208307436594623, "grad_norm": 0.9230567216873169, "lr": 4.593765372771508e-08 }, { "step": 1446, "T": 539, "positions": 16, "loss": 0.07894483599557134, "ce": 0.1595933708439361, "tv": 0.0699838949203695, "conf": 0.3566954682784882, "grad_norm": 1.5299242734909058, "lr": 4.259882798280934e-08 }, { "step": 1447, "T": 663, "positions": 16, "loss": 0.3397901367134182, "ce": 1.1089379172491363, "tv": 0.25432928414738853, "conf": 1.3601794758425683, "grad_norm": 3.675330400466919, "lr": 3.938585521515992e-08 }, { "step": 1448, "T": 601, "positions": 16, "loss": 0.48440531658798136, "ce": 1.8079127843023457, "tv": 0.3373489228024482, "conf": 1.1051262129257038, "grad_norm": 3.2207536697387695, "lr": 3.629875162686203e-08 }, { "step": 1449, "T": 532, "positions": 16, "loss": 0.20091106109794055, "ce": 0.6471227319074728, "tv": 0.15133198157627703, "conf": 0.5572082757626418, "grad_norm": 1.9588603973388672, "lr": 3.3337532785285285e-08 }, { "step": 1450, "T": 575, "positions": 16, "loss": 0.544797461188864, "ce": 1.8000499921226947, "tv": 0.4053249579446856, "conf": 1.4267425451544113, "grad_norm": 2.0384113788604736, "lr": 3.0502213623007095e-08 }, { "step": 1451, "T": 542, "positions": 16, "loss": 0.19833999492584553, "ce": 0.5196291686805239, "tv": 0.16264120635469226, "conf": 0.8603213098513163, "grad_norm": 1.8075727224349976, "lr": 2.779280843772103e-08 }, { "step": 1452, "T": 1125, "positions": 16, "loss": 0.6334450066497084, "ce": 2.895637375052047, "tv": 0.38209031311816943, "conf": 1.6593043470546434, "grad_norm": 2.5887110233306885, "lr": 2.5209330892178583e-08 }, { "step": 1453, "T": 586, "positions": 16, "loss": 0.4274321227494511, "ce": 1.5871944473747135, "tv": 0.29856965212820796, "conf": 1.2366395551446203, "grad_norm": 5.229408264160156, "lr": 2.2751794014111428e-08 }, { "step": 1454, "T": 576, "positions": 16, "loss": 0.2759639465766668, "ce": 0.9007264540987308, "tv": 0.20654589676269097, "conf": 1.4750314718345217, "grad_norm": 1.2342987060546875, "lr": 2.0420210196167577e-08 }, { "step": 1455, "T": 534, "positions": 16, "loss": 0.2967603208253422, "ce": 0.8923053093981252, "tv": 0.23058865906932624, "conf": 1.343701626463833, "grad_norm": 4.196194171905518, "lr": 1.8214591195853115e-08 }, { "step": 1456, "T": 535, "positions": 16, "loss": 0.3576356878329534, "ce": 1.471531645845971, "tv": 0.23386947727703955, "conf": 0.6485848946031183, "grad_norm": 2.2722387313842773, "lr": 1.613494813546834e-08 }, { "step": 1457, "T": 663, "positions": 16, "loss": 0.6157893631607294, "ce": 2.004514687694609, "tv": 0.46148654632270336, "conf": 2.5086123682558537, "grad_norm": 2.906865119934082, "lr": 1.4181291502052275e-08 }, { "step": 1458, "T": 593, "positions": 16, "loss": 0.5065930525514659, "ce": 2.3578948356768024, "tv": 0.3008928522677934, "conf": 1.0401635377129668, "grad_norm": 4.380695819854736, "lr": 1.2353631147335454e-08 }, { "step": 1459, "T": 672, "positions": 16, "loss": 0.8021348721231334, "ce": 3.300219665892655, "tv": 0.524569878631155, "conf": 1.5916401569106142, "grad_norm": 2.417882204055786, "lr": 1.0651976287678889e-08 }, { "step": 1460, "T": 572, "positions": 16, "loss": 0.6314567514345981, "ce": 2.5107219906058162, "tv": 0.42264951445395127, "conf": 1.165644305452588, "grad_norm": 3.4224863052368164, "lr": 9.076335504037969e-09 }, { "step": 1461, "T": 672, "positions": 16, "loss": 0.5740950496401638, "ce": 1.9592126002535224, "tv": 0.42019312852062285, "conf": 2.1732174828648567, "grad_norm": 3.0703022480010986, "lr": 7.626716741918061e-09 }, { "step": 1462, "T": 557, "positions": 16, "loss": 0.2718714058428304, "ce": 0.8701840753674333, "tv": 0.2053922256018268, "conf": 1.869872895756527, "grad_norm": 3.484792470932007, "lr": 6.303127311321766e-09 }, { "step": 1463, "T": 676, "positions": 16, "loss": 0.5743361449567601, "ce": 2.247755054559093, "tv": 0.3884007240412757, "conf": 1.546956653827877, "grad_norm": 2.3013737201690674, "lr": 5.105573886735049e-09 }, { "step": 1464, "T": 676, "positions": 16, "loss": 0.6575049310922623, "ce": 2.4238358647562563, "tv": 0.46124594658613205, "conf": 1.7409226342570037, "grad_norm": 1.7983351945877075, "lr": 4.034062507071723e-09 }, { "step": 1465, "T": 526, "positions": 16, "loss": 0.6668510008603334, "ce": 2.2682792532723397, "tv": 0.4889145391061902, "conf": 2.2644704687409103, "grad_norm": 4.4813971519470215, "lr": 3.088598575656798e-09 }, { "step": 1466, "T": 585, "positions": 16, "loss": 0.3450293638234143, "ce": 1.1614549970506687, "tv": 0.25431541679381553, "conf": 0.9266539641516488, "grad_norm": 2.291334390640259, "lr": 2.2691868601903977e-09 }, { "step": 1467, "T": 542, "positions": 16, "loss": 0.3860126223298721, "ce": 1.130263227217256, "tv": 0.3033181221690029, "conf": 1.1934681352215115, "grad_norm": 4.588921070098877, "lr": 1.5758314927283303e-09 }, { "step": 1468, "T": 572, "positions": 16, "loss": 0.36538490253224154, "ce": 1.473408685988943, "tv": 0.24227115233588847, "conf": 1.2027628189861816, "grad_norm": 3.2935938835144043, "lr": 1.0085359696654362e-09 }, { "step": 1469, "T": 567, "positions": 16, "loss": 0.6524407151155174, "ce": 2.6874668784439564, "tv": 0.4263266804628074, "conf": 1.2955643524546758, "grad_norm": 3.9391708374023438, "lr": 5.673031517050564e-10 }, { "step": 1470, "T": 591, "positions": 16, "loss": 0.4391441051266156, "ce": 1.9759818929887842, "tv": 0.26838436629623175, "conf": 1.2238494425546378, "grad_norm": 6.779214859008789, "lr": 2.521352638590324e-10 }, { "step": 1471, "T": 536, "positions": 16, "loss": 0.04045040856999549, "ce": 0.10928969697118873, "tv": 0.03280160060967319, "conf": 0.1864870432536918, "grad_norm": 0.348247766494751, "lr": 6.303389543105276e-11 } ], "chunks": 3, "chunk_sources": [ "/home/patrickd/s5-capture/s3/c0/train_metrics.json", "/home/patrickd/s5-capture/s3/c1/train_metrics.json", "/home/patrickd/s5-capture/s3/c2/train_metrics.json" ] }