GSA_Forge_checkpoints / checkpoint-5000 /trainer_state.json
rwlinno's picture
Upload folder using huggingface_hub
062e558 verified
Raw
History Blame Contribute Delete
228 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.03814508100107951,
"eval_steps": 200.0,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 7.6290162002159015e-06,
"grad_norm": 19.858008765227513,
"learning_rate": 7.628928898382667e-10,
"loss": 0.6787815093994141,
"step": 1,
"token_acc": 0.8377192982456141
},
{
"epoch": 3.81450810010795e-05,
"grad_norm": 26.05024103541224,
"learning_rate": 3.814464449191334e-09,
"loss": 0.7937984466552734,
"step": 5,
"token_acc": 0.8203161105462612
},
{
"epoch": 7.6290162002159e-05,
"grad_norm": 30.045629754543953,
"learning_rate": 7.628928898382667e-09,
"loss": 0.722916555404663,
"step": 10,
"token_acc": 0.8339535988597672
},
{
"epoch": 0.00011443524300323851,
"grad_norm": 41.77187019490099,
"learning_rate": 1.1443393347574e-08,
"loss": 0.7019518852233887,
"step": 15,
"token_acc": 0.8500863778872608
},
{
"epoch": 0.000152580324004318,
"grad_norm": 13.031898089287347,
"learning_rate": 1.5257857796765334e-08,
"loss": 0.7568827629089355,
"step": 20,
"token_acc": 0.8252126836813611
},
{
"epoch": 0.00019072540500539753,
"grad_norm": 26.463184090371293,
"learning_rate": 1.907232224595667e-08,
"loss": 0.7129611968994141,
"step": 25,
"token_acc": 0.8335900354918636
},
{
"epoch": 0.00022887048600647703,
"grad_norm": 16.079871921811126,
"learning_rate": 2.2886786695148e-08,
"loss": 0.7088653564453125,
"step": 30,
"token_acc": 0.8355570839064649
},
{
"epoch": 0.0002670155670075565,
"grad_norm": 33.687562516956945,
"learning_rate": 2.6701251144339337e-08,
"loss": 0.7810484886169433,
"step": 35,
"token_acc": 0.8183418693982074
},
{
"epoch": 0.000305160648008636,
"grad_norm": 17.19646111197456,
"learning_rate": 3.051571559353067e-08,
"loss": 0.6474244117736816,
"step": 40,
"token_acc": 0.8510296480210311
},
{
"epoch": 0.00034330572900971557,
"grad_norm": 24.770380370056916,
"learning_rate": 3.433018004272201e-08,
"loss": 0.66593337059021,
"step": 45,
"token_acc": 0.8447153957925327
},
{
"epoch": 0.00038145081001079507,
"grad_norm": 14.223809744043987,
"learning_rate": 3.814464449191334e-08,
"loss": 0.6652703762054444,
"step": 50,
"token_acc": 0.8436879598271634
},
{
"epoch": 0.00041959589101187456,
"grad_norm": 19.829855717229282,
"learning_rate": 4.195910894110467e-08,
"loss": 0.6511832237243652,
"step": 55,
"token_acc": 0.8471829658941021
},
{
"epoch": 0.00045774097201295406,
"grad_norm": 15.85798445068367,
"learning_rate": 4.5773573390296e-08,
"loss": 0.6936227798461914,
"step": 60,
"token_acc": 0.8382894155748565
},
{
"epoch": 0.0004958860530140336,
"grad_norm": 19.984224693650123,
"learning_rate": 4.958803783948734e-08,
"loss": 0.6485838890075684,
"step": 65,
"token_acc": 0.8432868352223191
},
{
"epoch": 0.000534031134015113,
"grad_norm": 13.967273121899467,
"learning_rate": 5.3402502288678674e-08,
"loss": 0.655977439880371,
"step": 70,
"token_acc": 0.8395167678901505
},
{
"epoch": 0.0005721762150161925,
"grad_norm": 23.9354021403603,
"learning_rate": 5.7216966737870006e-08,
"loss": 0.7054988861083984,
"step": 75,
"token_acc": 0.8299424184261036
},
{
"epoch": 0.000610321296017272,
"grad_norm": 14.366371084466175,
"learning_rate": 6.103143118706134e-08,
"loss": 0.6200805187225342,
"step": 80,
"token_acc": 0.8412296564195298
},
{
"epoch": 0.0006484663770183516,
"grad_norm": 13.927963325668722,
"learning_rate": 6.484589563625268e-08,
"loss": 0.5447507858276367,
"step": 85,
"token_acc": 0.8580637086175605
},
{
"epoch": 0.0006866114580194311,
"grad_norm": 17.084378717846253,
"learning_rate": 6.866036008544401e-08,
"loss": 0.5300666809082031,
"step": 90,
"token_acc": 0.8586914189276905
},
{
"epoch": 0.0007247565390205106,
"grad_norm": 12.831399981259963,
"learning_rate": 7.247482453463534e-08,
"loss": 0.5949723720550537,
"step": 95,
"token_acc": 0.8456714481366684
},
{
"epoch": 0.0007629016200215901,
"grad_norm": 9.654604645266131,
"learning_rate": 7.628928898382668e-08,
"loss": 0.5178804397583008,
"step": 100,
"token_acc": 0.8577652050919378
},
{
"epoch": 0.0008010467010226696,
"grad_norm": 9.419771464141803,
"learning_rate": 8.0103753433018e-08,
"loss": 0.5627875328063965,
"step": 105,
"token_acc": 0.8502238502238503
},
{
"epoch": 0.0008391917820237491,
"grad_norm": 10.700536352923347,
"learning_rate": 8.391821788220934e-08,
"loss": 0.5639850616455078,
"step": 110,
"token_acc": 0.8486978643636607
},
{
"epoch": 0.0008773368630248286,
"grad_norm": 18.083522443391814,
"learning_rate": 8.773268233140068e-08,
"loss": 0.5201422691345214,
"step": 115,
"token_acc": 0.8569876409072389
},
{
"epoch": 0.0009154819440259081,
"grad_norm": 8.157978934525882,
"learning_rate": 9.1547146780592e-08,
"loss": 0.5080629348754883,
"step": 120,
"token_acc": 0.8670629304651383
},
{
"epoch": 0.0009536270250269876,
"grad_norm": 7.8633829016429715,
"learning_rate": 9.536161122978335e-08,
"loss": 0.47640438079833985,
"step": 125,
"token_acc": 0.8592945806842265
},
{
"epoch": 0.000991772106028067,
"grad_norm": 10.996375473342454,
"learning_rate": 9.917607567897468e-08,
"loss": 0.4695584297180176,
"step": 130,
"token_acc": 0.870125663946107
},
{
"epoch": 0.0010299171870291467,
"grad_norm": 9.48145780304072,
"learning_rate": 1.0299054012816601e-07,
"loss": 0.41779556274414065,
"step": 135,
"token_acc": 0.8781450100252247
},
{
"epoch": 0.001068062268030226,
"grad_norm": 7.285733898226943,
"learning_rate": 1.0680500457735735e-07,
"loss": 0.42159037590026854,
"step": 140,
"token_acc": 0.8773419773095624
},
{
"epoch": 0.0011062073490313057,
"grad_norm": 6.059780464247734,
"learning_rate": 1.1061946902654869e-07,
"loss": 0.40407743453979494,
"step": 145,
"token_acc": 0.8766230324871482
},
{
"epoch": 0.001144352430032385,
"grad_norm": 7.452093118966406,
"learning_rate": 1.1443393347574001e-07,
"loss": 0.4567539215087891,
"step": 150,
"token_acc": 0.866771030732471
},
{
"epoch": 0.0011824975110334647,
"grad_norm": 6.407966242309223,
"learning_rate": 1.1824839792493135e-07,
"loss": 0.4267581939697266,
"step": 155,
"token_acc": 0.8669217106162044
},
{
"epoch": 0.001220642592034544,
"grad_norm": 6.675792771068159,
"learning_rate": 1.2206286237412268e-07,
"loss": 0.4057455062866211,
"step": 160,
"token_acc": 0.8701632511288642
},
{
"epoch": 0.0012587876730356237,
"grad_norm": 5.152432896859598,
"learning_rate": 1.2587732682331401e-07,
"loss": 0.417818546295166,
"step": 165,
"token_acc": 0.8704265211239135
},
{
"epoch": 0.0012969327540367033,
"grad_norm": 7.574436681211629,
"learning_rate": 1.2969179127250535e-07,
"loss": 0.3847885847091675,
"step": 170,
"token_acc": 0.8806957211824615
},
{
"epoch": 0.0013350778350377827,
"grad_norm": 5.227824773349837,
"learning_rate": 1.335062557216967e-07,
"loss": 0.39106528759002684,
"step": 175,
"token_acc": 0.8794553166158197
},
{
"epoch": 0.0013732229160388623,
"grad_norm": 5.370348537465429,
"learning_rate": 1.3732072017088803e-07,
"loss": 0.40700087547302244,
"step": 180,
"token_acc": 0.8752121614640986
},
{
"epoch": 0.0014113679970399417,
"grad_norm": 3.899114838688643,
"learning_rate": 1.4113518462007934e-07,
"loss": 0.33117434978485105,
"step": 185,
"token_acc": 0.8873824197210509
},
{
"epoch": 0.0014495130780410213,
"grad_norm": 7.517334555705535,
"learning_rate": 1.4494964906927068e-07,
"loss": 0.3739504337310791,
"step": 190,
"token_acc": 0.8766823414950543
},
{
"epoch": 0.0014876581590421007,
"grad_norm": 3.715493293332821,
"learning_rate": 1.4876411351846202e-07,
"loss": 0.36161179542541505,
"step": 195,
"token_acc": 0.880270056740645
},
{
"epoch": 0.0015258032400431803,
"grad_norm": 3.2584704553113855,
"learning_rate": 1.5257857796765336e-07,
"loss": 0.3406350612640381,
"step": 200,
"token_acc": 0.8852274327559498
},
{
"epoch": 0.0015639483210442597,
"grad_norm": 3.0327317389406394,
"learning_rate": 1.563930424168447e-07,
"loss": 0.33145263195037844,
"step": 205,
"token_acc": 0.8919872447248796
},
{
"epoch": 0.0016020934020453393,
"grad_norm": 3.651520865012892,
"learning_rate": 1.60207506866036e-07,
"loss": 0.3220701217651367,
"step": 210,
"token_acc": 0.8937093275488069
},
{
"epoch": 0.0016402384830464186,
"grad_norm": 4.110264197154364,
"learning_rate": 1.6402197131522735e-07,
"loss": 0.34341437816619874,
"step": 215,
"token_acc": 0.8845404225461916
},
{
"epoch": 0.0016783835640474982,
"grad_norm": 3.673651186053855,
"learning_rate": 1.6783643576441869e-07,
"loss": 0.31849493980407717,
"step": 220,
"token_acc": 0.8914095343856587
},
{
"epoch": 0.0017165286450485779,
"grad_norm": 3.4701434837015306,
"learning_rate": 1.7165090021361002e-07,
"loss": 0.29149889945983887,
"step": 225,
"token_acc": 0.8938066697402797
},
{
"epoch": 0.0017546737260496572,
"grad_norm": 4.501326153361515,
"learning_rate": 1.7546536466280136e-07,
"loss": 0.2862607479095459,
"step": 230,
"token_acc": 0.9036284845740896
},
{
"epoch": 0.0017928188070507368,
"grad_norm": 3.8208490412219533,
"learning_rate": 1.792798291119927e-07,
"loss": 0.27682464122772216,
"step": 235,
"token_acc": 0.9005648948478867
},
{
"epoch": 0.0018309638880518162,
"grad_norm": 3.1051066643962617,
"learning_rate": 1.83094293561184e-07,
"loss": 0.30829055309295655,
"step": 240,
"token_acc": 0.8974551460100593
},
{
"epoch": 0.0018691089690528958,
"grad_norm": 2.6304367853356623,
"learning_rate": 1.8690875801037535e-07,
"loss": 0.2855536937713623,
"step": 245,
"token_acc": 0.8998719762819217
},
{
"epoch": 0.0019072540500539752,
"grad_norm": 3.0160339634063056,
"learning_rate": 1.907232224595667e-07,
"loss": 0.2821728467941284,
"step": 250,
"token_acc": 0.9010830871771567
},
{
"epoch": 0.0019453991310550548,
"grad_norm": 3.3623406713328077,
"learning_rate": 1.9453768690875803e-07,
"loss": 0.28752238750457765,
"step": 255,
"token_acc": 0.8981417756366139
},
{
"epoch": 0.001983544212056134,
"grad_norm": 2.858724195529313,
"learning_rate": 1.9835215135794937e-07,
"loss": 0.2832742691040039,
"step": 260,
"token_acc": 0.9023319106822539
},
{
"epoch": 0.0020216892930572136,
"grad_norm": 3.7239371602902143,
"learning_rate": 2.0216661580714068e-07,
"loss": 0.2549095869064331,
"step": 265,
"token_acc": 0.9071066362901448
},
{
"epoch": 0.0020598343740582934,
"grad_norm": 3.481525766253563,
"learning_rate": 2.0598108025633202e-07,
"loss": 0.30463471412658694,
"step": 270,
"token_acc": 0.8938873626373627
},
{
"epoch": 0.002097979455059373,
"grad_norm": 2.8701813267844356,
"learning_rate": 2.0979554470552336e-07,
"loss": 0.27567908763885496,
"step": 275,
"token_acc": 0.9044994143181975
},
{
"epoch": 0.002136124536060452,
"grad_norm": 2.1624912096505766,
"learning_rate": 2.136100091547147e-07,
"loss": 0.27143154144287107,
"step": 280,
"token_acc": 0.9036544850498339
},
{
"epoch": 0.002174269617061532,
"grad_norm": 4.728168119300998,
"learning_rate": 2.1742447360390603e-07,
"loss": 0.26180579662323,
"step": 285,
"token_acc": 0.9078116254897257
},
{
"epoch": 0.0022124146980626114,
"grad_norm": 3.0085172830041564,
"learning_rate": 2.2123893805309737e-07,
"loss": 0.27187042236328124,
"step": 290,
"token_acc": 0.9030493094014813
},
{
"epoch": 0.002250559779063691,
"grad_norm": 2.7990717081011027,
"learning_rate": 2.2505340250228868e-07,
"loss": 0.27347536087036134,
"step": 295,
"token_acc": 0.9003181985069147
},
{
"epoch": 0.00228870486006477,
"grad_norm": 3.2087776560137002,
"learning_rate": 2.2886786695148002e-07,
"loss": 0.2732613325119019,
"step": 300,
"token_acc": 0.9009136735979836
},
{
"epoch": 0.00232684994106585,
"grad_norm": 2.7855962147313456,
"learning_rate": 2.3268233140067136e-07,
"loss": 0.25121102333068845,
"step": 305,
"token_acc": 0.9103824400754107
},
{
"epoch": 0.0023649950220669294,
"grad_norm": 23.00066788095994,
"learning_rate": 2.364967958498627e-07,
"loss": 0.29934229850769045,
"step": 310,
"token_acc": 0.8942279519484324
},
{
"epoch": 0.0024031401030680088,
"grad_norm": 4.913378638209299,
"learning_rate": 2.40311260299054e-07,
"loss": 0.2760341644287109,
"step": 315,
"token_acc": 0.9069470572426767
},
{
"epoch": 0.002441285184069088,
"grad_norm": 3.7230064334029422,
"learning_rate": 2.4412572474824535e-07,
"loss": 0.316583514213562,
"step": 320,
"token_acc": 0.8880706921944035
},
{
"epoch": 0.002479430265070168,
"grad_norm": 3.0630673787948273,
"learning_rate": 2.479401891974367e-07,
"loss": 0.28699023723602296,
"step": 325,
"token_acc": 0.8988292744005509
},
{
"epoch": 0.0025175753460712474,
"grad_norm": 2.7791052705585235,
"learning_rate": 2.5175465364662803e-07,
"loss": 0.2764303207397461,
"step": 330,
"token_acc": 0.8995172244497177
},
{
"epoch": 0.0025557204270723268,
"grad_norm": 4.808529142134823,
"learning_rate": 2.5556911809581937e-07,
"loss": 0.27858314514160154,
"step": 335,
"token_acc": 0.9007542460252862
},
{
"epoch": 0.0025938655080734066,
"grad_norm": 2.646253308159165,
"learning_rate": 2.593835825450107e-07,
"loss": 0.25584604740142824,
"step": 340,
"token_acc": 0.9082719082719083
},
{
"epoch": 0.002632010589074486,
"grad_norm": 5.050994193531453,
"learning_rate": 2.6319804699420204e-07,
"loss": 0.3128068447113037,
"step": 345,
"token_acc": 0.8978692138133725
},
{
"epoch": 0.0026701556700755654,
"grad_norm": 2.922439667410607,
"learning_rate": 2.670125114433934e-07,
"loss": 0.2790635108947754,
"step": 350,
"token_acc": 0.9022727272727272
},
{
"epoch": 0.0027083007510766447,
"grad_norm": 2.606552490620242,
"learning_rate": 2.708269758925847e-07,
"loss": 0.22434093952178955,
"step": 355,
"token_acc": 0.9184911328057305
},
{
"epoch": 0.0027464458320777246,
"grad_norm": 2.2464414779398516,
"learning_rate": 2.7464144034177606e-07,
"loss": 0.2485731601715088,
"step": 360,
"token_acc": 0.913417007852876
},
{
"epoch": 0.002784590913078804,
"grad_norm": 2.232285266114994,
"learning_rate": 2.7845590479096734e-07,
"loss": 0.2663278579711914,
"step": 365,
"token_acc": 0.9048203868590727
},
{
"epoch": 0.0028227359940798833,
"grad_norm": 4.178002931261007,
"learning_rate": 2.822703692401587e-07,
"loss": 0.2515146255493164,
"step": 370,
"token_acc": 0.9104171998976197
},
{
"epoch": 0.0028608810750809627,
"grad_norm": 2.890332520635348,
"learning_rate": 2.8608483368935e-07,
"loss": 0.25286097526550294,
"step": 375,
"token_acc": 0.9091865357643759
},
{
"epoch": 0.0028990261560820425,
"grad_norm": 3.0083368175320047,
"learning_rate": 2.8989929813854136e-07,
"loss": 0.25886220932006837,
"step": 380,
"token_acc": 0.907826982492276
},
{
"epoch": 0.002937171237083122,
"grad_norm": 4.890289775308246,
"learning_rate": 2.937137625877327e-07,
"loss": 0.27796077728271484,
"step": 385,
"token_acc": 0.902413515687852
},
{
"epoch": 0.0029753163180842013,
"grad_norm": 2.4007652310291543,
"learning_rate": 2.9752822703692404e-07,
"loss": 0.22965149879455565,
"step": 390,
"token_acc": 0.9193692149468632
},
{
"epoch": 0.003013461399085281,
"grad_norm": 2.703034695380086,
"learning_rate": 3.013426914861154e-07,
"loss": 0.26880793571472167,
"step": 395,
"token_acc": 0.9011542497376706
},
{
"epoch": 0.0030516064800863605,
"grad_norm": 2.7471412083206705,
"learning_rate": 3.051571559353067e-07,
"loss": 0.22871689796447753,
"step": 400,
"token_acc": 0.9171862392723539
},
{
"epoch": 0.00308975156108744,
"grad_norm": 2.8979547332295907,
"learning_rate": 3.0897162038449805e-07,
"loss": 0.2612307071685791,
"step": 405,
"token_acc": 0.904684700059732
},
{
"epoch": 0.0031278966420885193,
"grad_norm": 2.656438254711223,
"learning_rate": 3.127860848336894e-07,
"loss": 0.25812110900878904,
"step": 410,
"token_acc": 0.9117283950617284
},
{
"epoch": 0.003166041723089599,
"grad_norm": 2.80054119033248,
"learning_rate": 3.1660054928288073e-07,
"loss": 0.25269036293029784,
"step": 415,
"token_acc": 0.907069209515466
},
{
"epoch": 0.0032041868040906785,
"grad_norm": 7.1585542884169495,
"learning_rate": 3.20415013732072e-07,
"loss": 0.25886504650115966,
"step": 420,
"token_acc": 0.9069200430261742
},
{
"epoch": 0.003242331885091758,
"grad_norm": 2.340069613916146,
"learning_rate": 3.242294781812634e-07,
"loss": 0.27744815349578855,
"step": 425,
"token_acc": 0.8992904554818036
},
{
"epoch": 0.0032804769660928373,
"grad_norm": 3.0643890118779264,
"learning_rate": 3.280439426304547e-07,
"loss": 0.2664801120758057,
"step": 430,
"token_acc": 0.9005569456776172
},
{
"epoch": 0.003318622047093917,
"grad_norm": 2.708739491897395,
"learning_rate": 3.318584070796461e-07,
"loss": 0.26904611587524413,
"step": 435,
"token_acc": 0.9000482082596818
},
{
"epoch": 0.0033567671280949965,
"grad_norm": 4.501313817843271,
"learning_rate": 3.3567287152883737e-07,
"loss": 0.25872418880462644,
"step": 440,
"token_acc": 0.9131074027787988
},
{
"epoch": 0.003394912209096076,
"grad_norm": 2.9891978108449226,
"learning_rate": 3.3948733597802876e-07,
"loss": 0.27869994640350343,
"step": 445,
"token_acc": 0.8993040035757615
},
{
"epoch": 0.0034330572900971557,
"grad_norm": 4.01992715262689,
"learning_rate": 3.4330180042722005e-07,
"loss": 0.24958600997924804,
"step": 450,
"token_acc": 0.9094727018198787
},
{
"epoch": 0.003471202371098235,
"grad_norm": 2.542148367897184,
"learning_rate": 3.471162648764114e-07,
"loss": 0.2729614734649658,
"step": 455,
"token_acc": 0.8986539984164688
},
{
"epoch": 0.0035093474520993145,
"grad_norm": 2.3102049515467575,
"learning_rate": 3.509307293256027e-07,
"loss": 0.25960521697998046,
"step": 460,
"token_acc": 0.9041880532619208
},
{
"epoch": 0.003547492533100394,
"grad_norm": 4.215860115991377,
"learning_rate": 3.5474519377479406e-07,
"loss": 0.2580498218536377,
"step": 465,
"token_acc": 0.908362295198241
},
{
"epoch": 0.0035856376141014737,
"grad_norm": 2.5873046950881164,
"learning_rate": 3.585596582239854e-07,
"loss": 0.25073535442352296,
"step": 470,
"token_acc": 0.9081341984491217
},
{
"epoch": 0.003623782695102553,
"grad_norm": 9.841945468118112,
"learning_rate": 3.6237412267317674e-07,
"loss": 0.2642046928405762,
"step": 475,
"token_acc": 0.9030493273542601
},
{
"epoch": 0.0036619277761036325,
"grad_norm": 6.882617587596199,
"learning_rate": 3.66188587122368e-07,
"loss": 0.23246357440948487,
"step": 480,
"token_acc": 0.9140461215932913
},
{
"epoch": 0.003700072857104712,
"grad_norm": 2.0635484069380623,
"learning_rate": 3.700030515715594e-07,
"loss": 0.23335707187652588,
"step": 485,
"token_acc": 0.9135682885030921
},
{
"epoch": 0.0037382179381057917,
"grad_norm": 3.4740093175635174,
"learning_rate": 3.738175160207507e-07,
"loss": 0.2697558879852295,
"step": 490,
"token_acc": 0.9013763271726307
},
{
"epoch": 0.003776363019106871,
"grad_norm": 3.431739461613957,
"learning_rate": 3.776319804699421e-07,
"loss": 0.2169642448425293,
"step": 495,
"token_acc": 0.919410362785771
},
{
"epoch": 0.0038145081001079504,
"grad_norm": 1.5264757240403999,
"learning_rate": 3.814464449191334e-07,
"loss": 0.25090970993041994,
"step": 500,
"token_acc": 0.9121357432681667
},
{
"epoch": 0.0038526531811090303,
"grad_norm": 2.1883666824963797,
"learning_rate": 3.8526090936832477e-07,
"loss": 0.2483682632446289,
"step": 505,
"token_acc": 0.9091301932218038
},
{
"epoch": 0.0038907982621101097,
"grad_norm": 2.083496301448179,
"learning_rate": 3.8907537381751606e-07,
"loss": 0.2711325645446777,
"step": 510,
"token_acc": 0.9028123626776037
},
{
"epoch": 0.0039289433431111895,
"grad_norm": 2.641802435876891,
"learning_rate": 3.928898382667074e-07,
"loss": 0.27882430553436277,
"step": 515,
"token_acc": 0.9007821552723059
},
{
"epoch": 0.003967088424112268,
"grad_norm": 2.394684311531086,
"learning_rate": 3.9670430271589873e-07,
"loss": 0.2641146183013916,
"step": 520,
"token_acc": 0.9018478727975935
},
{
"epoch": 0.004005233505113348,
"grad_norm": 2.481739144309257,
"learning_rate": 4.0051876716509007e-07,
"loss": 0.26821279525756836,
"step": 525,
"token_acc": 0.9070000771783592
},
{
"epoch": 0.004043378586114427,
"grad_norm": 3.889296103853822,
"learning_rate": 4.0433323161428136e-07,
"loss": 0.22691609859466552,
"step": 530,
"token_acc": 0.9154143728611813
},
{
"epoch": 0.004081523667115507,
"grad_norm": 2.6368865169230022,
"learning_rate": 4.0814769606347275e-07,
"loss": 0.2556874752044678,
"step": 535,
"token_acc": 0.9065021725636251
},
{
"epoch": 0.004119668748116587,
"grad_norm": 2.6155085833020206,
"learning_rate": 4.1196216051266404e-07,
"loss": 0.21920876502990722,
"step": 540,
"token_acc": 0.92066594281578
},
{
"epoch": 0.004157813829117666,
"grad_norm": 1.9328229536953851,
"learning_rate": 4.1577662496185543e-07,
"loss": 0.23413827419281005,
"step": 545,
"token_acc": 0.909433962264151
},
{
"epoch": 0.004195958910118746,
"grad_norm": 2.9117172838218615,
"learning_rate": 4.195910894110467e-07,
"loss": 0.24360220432281493,
"step": 550,
"token_acc": 0.9103947742118716
},
{
"epoch": 0.0042341039911198254,
"grad_norm": 2.1250080249331256,
"learning_rate": 4.234055538602381e-07,
"loss": 0.24965639114379884,
"step": 555,
"token_acc": 0.9056687992858206
},
{
"epoch": 0.004272249072120904,
"grad_norm": 2.546842077158916,
"learning_rate": 4.272200183094294e-07,
"loss": 0.25338068008422854,
"step": 560,
"token_acc": 0.9083823425210799
},
{
"epoch": 0.004310394153121984,
"grad_norm": 2.055493648942245,
"learning_rate": 4.3103448275862073e-07,
"loss": 0.2616416931152344,
"step": 565,
"token_acc": 0.9030194104960461
},
{
"epoch": 0.004348539234123064,
"grad_norm": 3.081547453069715,
"learning_rate": 4.3484894720781207e-07,
"loss": 0.3007126569747925,
"step": 570,
"token_acc": 0.8967875647668394
},
{
"epoch": 0.004386684315124143,
"grad_norm": 2.6310833763661448,
"learning_rate": 4.386634116570034e-07,
"loss": 0.26825015544891356,
"step": 575,
"token_acc": 0.8971716563863397
},
{
"epoch": 0.004424829396125223,
"grad_norm": 1.9747613572324976,
"learning_rate": 4.4247787610619474e-07,
"loss": 0.24232356548309325,
"step": 580,
"token_acc": 0.9104139104139104
},
{
"epoch": 0.004462974477126302,
"grad_norm": 2.9507299451994164,
"learning_rate": 4.462923405553861e-07,
"loss": 0.2625190019607544,
"step": 585,
"token_acc": 0.9043351131451818
},
{
"epoch": 0.004501119558127382,
"grad_norm": 2.041702229587946,
"learning_rate": 4.5010680500457737e-07,
"loss": 0.27087769508361814,
"step": 590,
"token_acc": 0.8977926041592732
},
{
"epoch": 0.004539264639128461,
"grad_norm": 2.0495563743442253,
"learning_rate": 4.5392126945376876e-07,
"loss": 0.23960542678833008,
"step": 595,
"token_acc": 0.9126592532587083
},
{
"epoch": 0.00457740972012954,
"grad_norm": 2.436198073944371,
"learning_rate": 4.5773573390296005e-07,
"loss": 0.24157366752624512,
"step": 600,
"token_acc": 0.9103791174642635
},
{
"epoch": 0.00461555480113062,
"grad_norm": 2.5317430593867423,
"learning_rate": 4.6155019835215144e-07,
"loss": 0.271079421043396,
"step": 605,
"token_acc": 0.899757139268068
},
{
"epoch": 0.0046536998821317,
"grad_norm": 2.5553307431531627,
"learning_rate": 4.653646628013427e-07,
"loss": 0.24459531307220458,
"step": 610,
"token_acc": 0.9086593883501729
},
{
"epoch": 0.004691844963132779,
"grad_norm": 2.8304214669453165,
"learning_rate": 4.691791272505341e-07,
"loss": 0.2665693283081055,
"step": 615,
"token_acc": 0.9054301403294692
},
{
"epoch": 0.004729990044133859,
"grad_norm": 2.0461521335651165,
"learning_rate": 4.729935916997254e-07,
"loss": 0.2225412130355835,
"step": 620,
"token_acc": 0.9145840584754612
},
{
"epoch": 0.004768135125134939,
"grad_norm": 2.2406641298382355,
"learning_rate": 4.7680805614891674e-07,
"loss": 0.22277250289916992,
"step": 625,
"token_acc": 0.9159432870370371
},
{
"epoch": 0.0048062802061360176,
"grad_norm": 2.546827343906107,
"learning_rate": 4.80622520598108e-07,
"loss": 0.1908212661743164,
"step": 630,
"token_acc": 0.9257950530035336
},
{
"epoch": 0.004844425287137097,
"grad_norm": 2.1841664235416802,
"learning_rate": 4.844369850472995e-07,
"loss": 0.25454857349395754,
"step": 635,
"token_acc": 0.9049563764618526
},
{
"epoch": 0.004882570368138176,
"grad_norm": 2.572298033816989,
"learning_rate": 4.882514494964907e-07,
"loss": 0.23939483165740966,
"step": 640,
"token_acc": 0.9100228579344739
},
{
"epoch": 0.004920715449139256,
"grad_norm": 3.4160927766057685,
"learning_rate": 4.920659139456821e-07,
"loss": 0.25036001205444336,
"step": 645,
"token_acc": 0.9049896907216495
},
{
"epoch": 0.004958860530140336,
"grad_norm": 3.1113323822613976,
"learning_rate": 4.958803783948734e-07,
"loss": 0.22428619861602783,
"step": 650,
"token_acc": 0.9159893601300428
},
{
"epoch": 0.004997005611141415,
"grad_norm": 1.8929422066777826,
"learning_rate": 4.996948428440647e-07,
"loss": 0.2436244249343872,
"step": 655,
"token_acc": 0.9110026722435227
},
{
"epoch": 0.005035150692142495,
"grad_norm": 2.311586558615121,
"learning_rate": 5.035093072932561e-07,
"loss": 0.2434394359588623,
"step": 660,
"token_acc": 0.9068422934795566
},
{
"epoch": 0.005073295773143575,
"grad_norm": 1.9488749634400302,
"learning_rate": 5.073237717424474e-07,
"loss": 0.22594902515411378,
"step": 665,
"token_acc": 0.9147198366464476
},
{
"epoch": 0.0051114408541446535,
"grad_norm": 3.571265015516316,
"learning_rate": 5.111382361916387e-07,
"loss": 0.22313566207885743,
"step": 670,
"token_acc": 0.9155100735529261
},
{
"epoch": 0.005149585935145733,
"grad_norm": 8.81853964601659,
"learning_rate": 5.149527006408301e-07,
"loss": 0.23009917736053467,
"step": 675,
"token_acc": 0.91564106558654
},
{
"epoch": 0.005187731016146813,
"grad_norm": 2.3832019428324736,
"learning_rate": 5.187671650900214e-07,
"loss": 0.2520387411117554,
"step": 680,
"token_acc": 0.9045643153526971
},
{
"epoch": 0.005225876097147892,
"grad_norm": 2.773453447881878,
"learning_rate": 5.225816295392127e-07,
"loss": 0.22851839065551757,
"step": 685,
"token_acc": 0.9147792706333974
},
{
"epoch": 0.005264021178148972,
"grad_norm": 2.188161059769218,
"learning_rate": 5.263960939884041e-07,
"loss": 0.21708831787109376,
"step": 690,
"token_acc": 0.9175445873081709
},
{
"epoch": 0.005302166259150051,
"grad_norm": 2.120373057447545,
"learning_rate": 5.302105584375954e-07,
"loss": 0.240578031539917,
"step": 695,
"token_acc": 0.9101432498914773
},
{
"epoch": 0.005340311340151131,
"grad_norm": 7.143314253272402,
"learning_rate": 5.340250228867868e-07,
"loss": 0.24711151123046876,
"step": 700,
"token_acc": 0.905367231638418
},
{
"epoch": 0.0053784564211522105,
"grad_norm": 2.509437280811482,
"learning_rate": 5.378394873359781e-07,
"loss": 0.2288766860961914,
"step": 705,
"token_acc": 0.9190667983889353
},
{
"epoch": 0.0054166015021532895,
"grad_norm": 4.435980751184006,
"learning_rate": 5.416539517851694e-07,
"loss": 0.24404430389404297,
"step": 710,
"token_acc": 0.91390625
},
{
"epoch": 0.005454746583154369,
"grad_norm": 2.1864120237187454,
"learning_rate": 5.454684162343608e-07,
"loss": 0.22964107990264893,
"step": 715,
"token_acc": 0.9149337004564887
},
{
"epoch": 0.005492891664155449,
"grad_norm": 2.168263736034898,
"learning_rate": 5.492828806835521e-07,
"loss": 0.24856891632080078,
"step": 720,
"token_acc": 0.9097530347425701
},
{
"epoch": 0.005531036745156528,
"grad_norm": 2.9088664787654213,
"learning_rate": 5.530973451327435e-07,
"loss": 0.2512509346008301,
"step": 725,
"token_acc": 0.9060345353590823
},
{
"epoch": 0.005569181826157608,
"grad_norm": 3.3320808612830874,
"learning_rate": 5.569118095819347e-07,
"loss": 0.2525000810623169,
"step": 730,
"token_acc": 0.9031252030407381
},
{
"epoch": 0.005607326907158688,
"grad_norm": 1.6179035871243548,
"learning_rate": 5.607262740311261e-07,
"loss": 0.2348179340362549,
"step": 735,
"token_acc": 0.9133819951338199
},
{
"epoch": 0.005645471988159767,
"grad_norm": 2.85667421103228,
"learning_rate": 5.645407384803174e-07,
"loss": 0.23236339092254638,
"step": 740,
"token_acc": 0.914976065648507
},
{
"epoch": 0.0056836170691608465,
"grad_norm": 2.6909733482462914,
"learning_rate": 5.683552029295088e-07,
"loss": 0.23209877014160157,
"step": 745,
"token_acc": 0.9132922854038392
},
{
"epoch": 0.0057217621501619255,
"grad_norm": 1.5951096324195422,
"learning_rate": 5.721696673787e-07,
"loss": 0.22777156829833983,
"step": 750,
"token_acc": 0.9133550488599349
},
{
"epoch": 0.005759907231163005,
"grad_norm": 2.3881038308988383,
"learning_rate": 5.759841318278915e-07,
"loss": 0.2042149543762207,
"step": 755,
"token_acc": 0.9194363390506014
},
{
"epoch": 0.005798052312164085,
"grad_norm": 3.325840053407865,
"learning_rate": 5.797985962770827e-07,
"loss": 0.24493722915649413,
"step": 760,
"token_acc": 0.9089112637621154
},
{
"epoch": 0.005836197393165164,
"grad_norm": 1.8938911522743977,
"learning_rate": 5.836130607262741e-07,
"loss": 0.21641106605529786,
"step": 765,
"token_acc": 0.9197261978842564
},
{
"epoch": 0.005874342474166244,
"grad_norm": 1.988679833448486,
"learning_rate": 5.874275251754654e-07,
"loss": 0.21464207172393798,
"step": 770,
"token_acc": 0.9175137584978957
},
{
"epoch": 0.005912487555167324,
"grad_norm": 1.8627516018080663,
"learning_rate": 5.912419896246567e-07,
"loss": 0.2498110294342041,
"step": 775,
"token_acc": 0.9056705670567057
},
{
"epoch": 0.005950632636168403,
"grad_norm": 1.2851350756074684,
"learning_rate": 5.950564540738481e-07,
"loss": 0.242911434173584,
"step": 780,
"token_acc": 0.9100091183278389
},
{
"epoch": 0.0059887777171694825,
"grad_norm": 1.7215295791488319,
"learning_rate": 5.988709185230394e-07,
"loss": 0.19092849493026734,
"step": 785,
"token_acc": 0.9254417816509319
},
{
"epoch": 0.006026922798170562,
"grad_norm": 2.0995252488631713,
"learning_rate": 6.026853829722308e-07,
"loss": 0.21847922801971437,
"step": 790,
"token_acc": 0.9216517010244087
},
{
"epoch": 0.006065067879171641,
"grad_norm": 2.814177734711112,
"learning_rate": 6.064998474214221e-07,
"loss": 0.24257607460021974,
"step": 795,
"token_acc": 0.9114211550940947
},
{
"epoch": 0.006103212960172721,
"grad_norm": 2.2026833559693153,
"learning_rate": 6.103143118706134e-07,
"loss": 0.23102183341979982,
"step": 800,
"token_acc": 0.9107298271789134
},
{
"epoch": 0.0061413580411738,
"grad_norm": 2.0108750944280174,
"learning_rate": 6.141287763198048e-07,
"loss": 0.24538402557373046,
"step": 805,
"token_acc": 0.9061550420464984
},
{
"epoch": 0.00617950312217488,
"grad_norm": 2.304744993055787,
"learning_rate": 6.179432407689961e-07,
"loss": 0.2557771921157837,
"step": 810,
"token_acc": 0.9081625797783003
},
{
"epoch": 0.00621764820317596,
"grad_norm": 8.924798548774792,
"learning_rate": 6.217577052181874e-07,
"loss": 0.24581964015960694,
"step": 815,
"token_acc": 0.90890815429074
},
{
"epoch": 0.006255793284177039,
"grad_norm": 1.4409669493856745,
"learning_rate": 6.255721696673788e-07,
"loss": 0.20607192516326905,
"step": 820,
"token_acc": 0.9191860465116279
},
{
"epoch": 0.0062939383651781184,
"grad_norm": 2.925558681675864,
"learning_rate": 6.293866341165701e-07,
"loss": 0.2149512767791748,
"step": 825,
"token_acc": 0.9190668029583309
},
{
"epoch": 0.006332083446179198,
"grad_norm": 1.7110063722551234,
"learning_rate": 6.332010985657615e-07,
"loss": 0.21946215629577637,
"step": 830,
"token_acc": 0.9161647628937348
},
{
"epoch": 0.006370228527180277,
"grad_norm": 2.444989400700814,
"learning_rate": 6.370155630149527e-07,
"loss": 0.21233253479003905,
"step": 835,
"token_acc": 0.9200024520321216
},
{
"epoch": 0.006408373608181357,
"grad_norm": 3.2838014012433985,
"learning_rate": 6.40830027464144e-07,
"loss": 0.22280681133270264,
"step": 840,
"token_acc": 0.9153393695063116
},
{
"epoch": 0.006446518689182437,
"grad_norm": 1.9215417073538057,
"learning_rate": 6.446444919133355e-07,
"loss": 0.2385124683380127,
"step": 845,
"token_acc": 0.9088968579234973
},
{
"epoch": 0.006484663770183516,
"grad_norm": 2.319220162643885,
"learning_rate": 6.484589563625268e-07,
"loss": 0.2364635944366455,
"step": 850,
"token_acc": 0.909977141320425
},
{
"epoch": 0.006522808851184596,
"grad_norm": 1.9237714003957875,
"learning_rate": 6.52273420811718e-07,
"loss": 0.23012151718139648,
"step": 855,
"token_acc": 0.9148593082318793
},
{
"epoch": 0.006560953932185675,
"grad_norm": 2.046769328041715,
"learning_rate": 6.560878852609094e-07,
"loss": 0.2327641487121582,
"step": 860,
"token_acc": 0.9113459879206213
},
{
"epoch": 0.006599099013186754,
"grad_norm": 1.6115036705671675,
"learning_rate": 6.599023497101008e-07,
"loss": 0.2237847089767456,
"step": 865,
"token_acc": 0.9167464114832536
},
{
"epoch": 0.006637244094187834,
"grad_norm": 3.3051233128989015,
"learning_rate": 6.637168141592922e-07,
"loss": 0.19322073459625244,
"step": 870,
"token_acc": 0.9275590551181102
},
{
"epoch": 0.006675389175188913,
"grad_norm": 2.4262761074290506,
"learning_rate": 6.675312786084834e-07,
"loss": 0.22160861492156983,
"step": 875,
"token_acc": 0.9151369669584863
},
{
"epoch": 0.006713534256189993,
"grad_norm": 1.9677832165578242,
"learning_rate": 6.713457430576747e-07,
"loss": 0.225478196144104,
"step": 880,
"token_acc": 0.9115168539325843
},
{
"epoch": 0.006751679337191073,
"grad_norm": 5.096878328776845,
"learning_rate": 6.751602075068661e-07,
"loss": 0.24097964763641358,
"step": 885,
"token_acc": 0.9146459747817652
},
{
"epoch": 0.006789824418192152,
"grad_norm": 1.688691754760285,
"learning_rate": 6.789746719560575e-07,
"loss": 0.2104334592819214,
"step": 890,
"token_acc": 0.9222782379066139
},
{
"epoch": 0.006827969499193232,
"grad_norm": 3.120436618558742,
"learning_rate": 6.827891364052488e-07,
"loss": 0.23382256031036378,
"step": 895,
"token_acc": 0.9112284069097889
},
{
"epoch": 0.006866114580194311,
"grad_norm": 2.2142725887946937,
"learning_rate": 6.866036008544401e-07,
"loss": 0.25021958351135254,
"step": 900,
"token_acc": 0.9044054747647562
},
{
"epoch": 0.00690425966119539,
"grad_norm": 4.399873851436453,
"learning_rate": 6.904180653036314e-07,
"loss": 0.2350818157196045,
"step": 905,
"token_acc": 0.9120438237790294
},
{
"epoch": 0.00694240474219647,
"grad_norm": 2.5238462069718905,
"learning_rate": 6.942325297528228e-07,
"loss": 0.21383306980133057,
"step": 910,
"token_acc": 0.9169596314704813
},
{
"epoch": 0.006980549823197549,
"grad_norm": 1.8196001493704197,
"learning_rate": 6.98046994202014e-07,
"loss": 0.20101499557495117,
"step": 915,
"token_acc": 0.9233510235026535
},
{
"epoch": 0.007018694904198629,
"grad_norm": 1.8331955196154692,
"learning_rate": 7.018614586512054e-07,
"loss": 0.2478102207183838,
"step": 920,
"token_acc": 0.9081678497618834
},
{
"epoch": 0.007056839985199709,
"grad_norm": 1.6893601989964881,
"learning_rate": 7.056759231003968e-07,
"loss": 0.2171755313873291,
"step": 925,
"token_acc": 0.9169847990560145
},
{
"epoch": 0.007094985066200788,
"grad_norm": 2.7750882619687682,
"learning_rate": 7.094903875495881e-07,
"loss": 0.2517444610595703,
"step": 930,
"token_acc": 0.905949540952794
},
{
"epoch": 0.0071331301472018676,
"grad_norm": 2.295636343184957,
"learning_rate": 7.133048519987794e-07,
"loss": 0.21493184566497803,
"step": 935,
"token_acc": 0.919338959212377
},
{
"epoch": 0.007171275228202947,
"grad_norm": 5.274781334339023,
"learning_rate": 7.171193164479708e-07,
"loss": 0.2255638360977173,
"step": 940,
"token_acc": 0.9191055958791852
},
{
"epoch": 0.007209420309204026,
"grad_norm": 4.249507608934204,
"learning_rate": 7.209337808971621e-07,
"loss": 0.23890252113342286,
"step": 945,
"token_acc": 0.9082201572551822
},
{
"epoch": 0.007247565390205106,
"grad_norm": 1.8318728175930554,
"learning_rate": 7.247482453463535e-07,
"loss": 0.23565545082092285,
"step": 950,
"token_acc": 0.9120185682164358
},
{
"epoch": 0.007285710471206186,
"grad_norm": 2.3153771765190125,
"learning_rate": 7.285627097955447e-07,
"loss": 0.22924485206604003,
"step": 955,
"token_acc": 0.9118163094057518
},
{
"epoch": 0.007323855552207265,
"grad_norm": 2.1444734209902667,
"learning_rate": 7.32377174244736e-07,
"loss": 0.2102113723754883,
"step": 960,
"token_acc": 0.9237033861980283
},
{
"epoch": 0.007362000633208345,
"grad_norm": 2.313230627888173,
"learning_rate": 7.361916386939275e-07,
"loss": 0.22157104015350343,
"step": 965,
"token_acc": 0.9158624102944816
},
{
"epoch": 0.007400145714209424,
"grad_norm": 1.8991530777901926,
"learning_rate": 7.400061031431188e-07,
"loss": 0.21107559204101561,
"step": 970,
"token_acc": 0.9221950473861205
},
{
"epoch": 0.0074382907952105035,
"grad_norm": 1.9998413654401805,
"learning_rate": 7.438205675923101e-07,
"loss": 0.21765711307525634,
"step": 975,
"token_acc": 0.9134157944814463
},
{
"epoch": 0.007476435876211583,
"grad_norm": 2.3376032357783734,
"learning_rate": 7.476350320415014e-07,
"loss": 0.22147510051727295,
"step": 980,
"token_acc": 0.9163215117784106
},
{
"epoch": 0.007514580957212662,
"grad_norm": 1.5533529464193574,
"learning_rate": 7.514494964906927e-07,
"loss": 0.1793646216392517,
"step": 985,
"token_acc": 0.9294284173631706
},
{
"epoch": 0.007552726038213742,
"grad_norm": 2.325698269527801,
"learning_rate": 7.552639609398842e-07,
"loss": 0.22358329296112062,
"step": 990,
"token_acc": 0.915683082349749
},
{
"epoch": 0.007590871119214822,
"grad_norm": 2.2527992518345625,
"learning_rate": 7.590784253890754e-07,
"loss": 0.22756524085998536,
"step": 995,
"token_acc": 0.9161717164122998
},
{
"epoch": 0.007629016200215901,
"grad_norm": 2.026830062262338,
"learning_rate": 7.628928898382668e-07,
"loss": 0.22516098022460937,
"step": 1000,
"token_acc": 0.9157646132651053
},
{
"epoch": 0.007667161281216981,
"grad_norm": 2.2410727876260443,
"learning_rate": 7.667073542874581e-07,
"loss": 0.22787818908691407,
"step": 1005,
"token_acc": 0.9150561797752809
},
{
"epoch": 0.0077053063622180605,
"grad_norm": 5.3919391901547975,
"learning_rate": 7.705218187366495e-07,
"loss": 0.21928162574768068,
"step": 1010,
"token_acc": 0.919819946546631
},
{
"epoch": 0.0077434514432191395,
"grad_norm": 1.9669989639073129,
"learning_rate": 7.743362831858408e-07,
"loss": 0.23177022933959962,
"step": 1015,
"token_acc": 0.9152079453755432
},
{
"epoch": 0.007781596524220219,
"grad_norm": 2.5524717405659754,
"learning_rate": 7.781507476350321e-07,
"loss": 0.24666337966918944,
"step": 1020,
"token_acc": 0.9083541147132169
},
{
"epoch": 0.007819741605221298,
"grad_norm": 1.4083926404283291,
"learning_rate": 7.819652120842235e-07,
"loss": 0.21447858810424805,
"step": 1025,
"token_acc": 0.9166475841538814
},
{
"epoch": 0.007857886686222379,
"grad_norm": 1.5641199684666467,
"learning_rate": 7.857796765334148e-07,
"loss": 0.2286088466644287,
"step": 1030,
"token_acc": 0.9125537455338218
},
{
"epoch": 0.007896031767223458,
"grad_norm": 2.0673685817206033,
"learning_rate": 7.89594140982606e-07,
"loss": 0.20569605827331544,
"step": 1035,
"token_acc": 0.9217044637605385
},
{
"epoch": 0.007934176848224537,
"grad_norm": 2.8085525303032366,
"learning_rate": 7.934086054317975e-07,
"loss": 0.22242367267608643,
"step": 1040,
"token_acc": 0.917860620392266
},
{
"epoch": 0.007972321929225618,
"grad_norm": 4.060492391104231,
"learning_rate": 7.972230698809888e-07,
"loss": 0.23639769554138185,
"step": 1045,
"token_acc": 0.909541895977056
},
{
"epoch": 0.008010467010226697,
"grad_norm": 2.2698459873627517,
"learning_rate": 8.010375343301801e-07,
"loss": 0.20035903453826903,
"step": 1050,
"token_acc": 0.9249095769775122
},
{
"epoch": 0.008048612091227775,
"grad_norm": 1.7377397563130297,
"learning_rate": 8.048519987793714e-07,
"loss": 0.23490912914276124,
"step": 1055,
"token_acc": 0.9155097913027489
},
{
"epoch": 0.008086757172228854,
"grad_norm": 2.4126315196396213,
"learning_rate": 8.086664632285627e-07,
"loss": 0.2245621919631958,
"step": 1060,
"token_acc": 0.9150576261038766
},
{
"epoch": 0.008124902253229935,
"grad_norm": 2.1739004983045525,
"learning_rate": 8.124809276777542e-07,
"loss": 0.2533282995223999,
"step": 1065,
"token_acc": 0.9060859580052494
},
{
"epoch": 0.008163047334231014,
"grad_norm": 2.573169957629046,
"learning_rate": 8.162953921269455e-07,
"loss": 0.19038509130477904,
"step": 1070,
"token_acc": 0.9273980544997745
},
{
"epoch": 0.008201192415232093,
"grad_norm": 2.4332706882659543,
"learning_rate": 8.201098565761367e-07,
"loss": 0.17688820362091065,
"step": 1075,
"token_acc": 0.9309796196681567
},
{
"epoch": 0.008239337496233174,
"grad_norm": 2.2160397064060273,
"learning_rate": 8.239243210253281e-07,
"loss": 0.20977408885955812,
"step": 1080,
"token_acc": 0.922009012937927
},
{
"epoch": 0.008277482577234253,
"grad_norm": 1.4060743450729623,
"learning_rate": 8.277387854745195e-07,
"loss": 0.2200855016708374,
"step": 1085,
"token_acc": 0.9165340839554522
},
{
"epoch": 0.008315627658235332,
"grad_norm": 3.7213469366155834,
"learning_rate": 8.315532499237109e-07,
"loss": 0.2282848834991455,
"step": 1090,
"token_acc": 0.9172981341155382
},
{
"epoch": 0.008353772739236412,
"grad_norm": 2.8097142428748616,
"learning_rate": 8.353677143729021e-07,
"loss": 0.22783021926879882,
"step": 1095,
"token_acc": 0.9166942422236929
},
{
"epoch": 0.008391917820237491,
"grad_norm": 2.6392708839458203,
"learning_rate": 8.391821788220934e-07,
"loss": 0.18801560401916503,
"step": 1100,
"token_acc": 0.9303247340240596
},
{
"epoch": 0.00843006290123857,
"grad_norm": 2.2915610285231143,
"learning_rate": 8.429966432712848e-07,
"loss": 0.24878976345062256,
"step": 1105,
"token_acc": 0.9083246307623453
},
{
"epoch": 0.008468207982239651,
"grad_norm": 2.3415436884365097,
"learning_rate": 8.468111077204762e-07,
"loss": 0.2238091230392456,
"step": 1110,
"token_acc": 0.9185648354920352
},
{
"epoch": 0.00850635306324073,
"grad_norm": 2.2070835990485183,
"learning_rate": 8.506255721696674e-07,
"loss": 0.2474008560180664,
"step": 1115,
"token_acc": 0.9109291016124904
},
{
"epoch": 0.008544498144241809,
"grad_norm": 2.379225822204149,
"learning_rate": 8.544400366188588e-07,
"loss": 0.22392101287841798,
"step": 1120,
"token_acc": 0.9141245453870704
},
{
"epoch": 0.00858264322524289,
"grad_norm": 1.6785840155005407,
"learning_rate": 8.582545010680501e-07,
"loss": 0.2184514284133911,
"step": 1125,
"token_acc": 0.9168314305759296
},
{
"epoch": 0.008620788306243968,
"grad_norm": 1.626194260949076,
"learning_rate": 8.620689655172415e-07,
"loss": 0.21134648323059083,
"step": 1130,
"token_acc": 0.9182502467917077
},
{
"epoch": 0.008658933387245047,
"grad_norm": 4.459135117797278,
"learning_rate": 8.658834299664327e-07,
"loss": 0.19004669189453124,
"step": 1135,
"token_acc": 0.9273110024147112
},
{
"epoch": 0.008697078468246128,
"grad_norm": 6.993418125749847,
"learning_rate": 8.696978944156241e-07,
"loss": 0.21319918632507323,
"step": 1140,
"token_acc": 0.9190795618578586
},
{
"epoch": 0.008735223549247207,
"grad_norm": 2.6862929658080548,
"learning_rate": 8.735123588648155e-07,
"loss": 0.2073223829269409,
"step": 1145,
"token_acc": 0.9199803933582501
},
{
"epoch": 0.008773368630248286,
"grad_norm": 2.600638404331704,
"learning_rate": 8.773268233140068e-07,
"loss": 0.2453620433807373,
"step": 1150,
"token_acc": 0.9045744838853409
},
{
"epoch": 0.008811513711249367,
"grad_norm": 2.862925916358926,
"learning_rate": 8.81141287763198e-07,
"loss": 0.1902993440628052,
"step": 1155,
"token_acc": 0.9253010005087332
},
{
"epoch": 0.008849658792250446,
"grad_norm": 3.3281600708321615,
"learning_rate": 8.849557522123895e-07,
"loss": 0.21226978302001953,
"step": 1160,
"token_acc": 0.9180122145068184
},
{
"epoch": 0.008887803873251525,
"grad_norm": 2.6248829980165858,
"learning_rate": 8.887702166615808e-07,
"loss": 0.2133929967880249,
"step": 1165,
"token_acc": 0.9191057257445174
},
{
"epoch": 0.008925948954252604,
"grad_norm": 4.1588717166571705,
"learning_rate": 8.925846811107722e-07,
"loss": 0.2274726390838623,
"step": 1170,
"token_acc": 0.9149301973729455
},
{
"epoch": 0.008964094035253684,
"grad_norm": 3.288603302949556,
"learning_rate": 8.963991455599634e-07,
"loss": 0.2082827568054199,
"step": 1175,
"token_acc": 0.9207027157708115
},
{
"epoch": 0.009002239116254763,
"grad_norm": 3.560818627965942,
"learning_rate": 9.002136100091547e-07,
"loss": 0.1922168493270874,
"step": 1180,
"token_acc": 0.9274755927475593
},
{
"epoch": 0.009040384197255842,
"grad_norm": 2.8114340926065458,
"learning_rate": 9.040280744583462e-07,
"loss": 0.21101346015930175,
"step": 1185,
"token_acc": 0.9208867060463264
},
{
"epoch": 0.009078529278256923,
"grad_norm": 5.29748879050654,
"learning_rate": 9.078425389075375e-07,
"loss": 0.20406670570373536,
"step": 1190,
"token_acc": 0.9256397875422501
},
{
"epoch": 0.009116674359258002,
"grad_norm": 3.407825389334515,
"learning_rate": 9.116570033567288e-07,
"loss": 0.23571975231170655,
"step": 1195,
"token_acc": 0.9117741831539602
},
{
"epoch": 0.00915481944025908,
"grad_norm": 1.8345373720766989,
"learning_rate": 9.154714678059201e-07,
"loss": 0.22133150100708007,
"step": 1200,
"token_acc": 0.9154685735792247
},
{
"epoch": 0.009192964521260161,
"grad_norm": 8.357598934299423,
"learning_rate": 9.192859322551114e-07,
"loss": 0.230814790725708,
"step": 1205,
"token_acc": 0.9129955947136564
},
{
"epoch": 0.00923110960226124,
"grad_norm": 1.0828018930510888,
"learning_rate": 9.231003967043029e-07,
"loss": 0.20155847072601318,
"step": 1210,
"token_acc": 0.9257648703214785
},
{
"epoch": 0.00926925468326232,
"grad_norm": 1.4624609828046475,
"learning_rate": 9.269148611534941e-07,
"loss": 0.2023165702819824,
"step": 1215,
"token_acc": 0.9199700822737472
},
{
"epoch": 0.0093073997642634,
"grad_norm": 2.246915803213318,
"learning_rate": 9.307293256026854e-07,
"loss": 0.25096836090087893,
"step": 1220,
"token_acc": 0.9058573853989813
},
{
"epoch": 0.009345544845264479,
"grad_norm": 5.6048533369336795,
"learning_rate": 9.345437900518768e-07,
"loss": 0.22780070304870606,
"step": 1225,
"token_acc": 0.917071711657818
},
{
"epoch": 0.009383689926265558,
"grad_norm": 1.7763360937154278,
"learning_rate": 9.383582545010682e-07,
"loss": 0.219808030128479,
"step": 1230,
"token_acc": 0.9139512082588385
},
{
"epoch": 0.009421835007266639,
"grad_norm": 1.408454862734737,
"learning_rate": 9.421727189502595e-07,
"loss": 0.20850255489349365,
"step": 1235,
"token_acc": 0.9166717058716817
},
{
"epoch": 0.009459980088267718,
"grad_norm": 1.9607027388800182,
"learning_rate": 9.459871833994508e-07,
"loss": 0.197813880443573,
"step": 1240,
"token_acc": 0.9229813664596274
},
{
"epoch": 0.009498125169268797,
"grad_norm": 1.2090480998576494,
"learning_rate": 9.498016478486421e-07,
"loss": 0.2130037784576416,
"step": 1245,
"token_acc": 0.9221602787456445
},
{
"epoch": 0.009536270250269877,
"grad_norm": 1.9585489367417885,
"learning_rate": 9.536161122978335e-07,
"loss": 0.1951514959335327,
"step": 1250,
"token_acc": 0.9262824739548992
},
{
"epoch": 0.009574415331270956,
"grad_norm": 2.3126236850622535,
"learning_rate": 9.574305767470247e-07,
"loss": 0.20185506343841553,
"step": 1255,
"token_acc": 0.9235834720312613
},
{
"epoch": 0.009612560412272035,
"grad_norm": 1.8684366593016006,
"learning_rate": 9.61245041196216e-07,
"loss": 0.17978219985961913,
"step": 1260,
"token_acc": 0.9310434387377443
},
{
"epoch": 0.009650705493273116,
"grad_norm": 4.84147314756343,
"learning_rate": 9.650595056454074e-07,
"loss": 0.2505626678466797,
"step": 1265,
"token_acc": 0.9056459622583849
},
{
"epoch": 0.009688850574274195,
"grad_norm": 2.533686896206201,
"learning_rate": 9.68873970094599e-07,
"loss": 0.19906680583953856,
"step": 1270,
"token_acc": 0.9242752055387278
},
{
"epoch": 0.009726995655275274,
"grad_norm": 1.6265068669882792,
"learning_rate": 9.7268843454379e-07,
"loss": 0.1907017469406128,
"step": 1275,
"token_acc": 0.926312005751258
},
{
"epoch": 0.009765140736276353,
"grad_norm": 1.8093501110086971,
"learning_rate": 9.765028989929814e-07,
"loss": 0.2400496244430542,
"step": 1280,
"token_acc": 0.914736550705915
},
{
"epoch": 0.009803285817277433,
"grad_norm": 3.4094145839010666,
"learning_rate": 9.803173634421727e-07,
"loss": 0.23024744987487794,
"step": 1285,
"token_acc": 0.9157664433201148
},
{
"epoch": 0.009841430898278512,
"grad_norm": 2.3012455374124245,
"learning_rate": 9.841318278913643e-07,
"loss": 0.20666346549987794,
"step": 1290,
"token_acc": 0.9177931733693816
},
{
"epoch": 0.009879575979279591,
"grad_norm": 2.0286080743049686,
"learning_rate": 9.879462923405554e-07,
"loss": 0.2061232089996338,
"step": 1295,
"token_acc": 0.920604099244876
},
{
"epoch": 0.009917721060280672,
"grad_norm": 5.253123569167488,
"learning_rate": 9.917607567897468e-07,
"loss": 0.21556572914123534,
"step": 1300,
"token_acc": 0.9155959540732641
},
{
"epoch": 0.009955866141281751,
"grad_norm": 2.2843980861589874,
"learning_rate": 9.95575221238938e-07,
"loss": 0.2044231414794922,
"step": 1305,
"token_acc": 0.9234021988643228
},
{
"epoch": 0.00999401122228283,
"grad_norm": 3.1724891622517775,
"learning_rate": 9.993896856881294e-07,
"loss": 0.20736510753631593,
"step": 1310,
"token_acc": 0.9207219556774047
},
{
"epoch": 0.01003215630328391,
"grad_norm": 1.8567214087110475,
"learning_rate": 1.0032041501373208e-06,
"loss": 0.18762776851654053,
"step": 1315,
"token_acc": 0.9282498184458968
},
{
"epoch": 0.01007030138428499,
"grad_norm": 3.111715216837282,
"learning_rate": 1.0070186145865121e-06,
"loss": 0.21242661476135255,
"step": 1320,
"token_acc": 0.9205221561140219
},
{
"epoch": 0.010108446465286068,
"grad_norm": 2.502017462919091,
"learning_rate": 1.0108330790357034e-06,
"loss": 0.2024921417236328,
"step": 1325,
"token_acc": 0.9229410820356397
},
{
"epoch": 0.01014659154628715,
"grad_norm": 3.6186633058231403,
"learning_rate": 1.0146475434848948e-06,
"loss": 0.22059335708618164,
"step": 1330,
"token_acc": 0.9182776135243726
},
{
"epoch": 0.010184736627288228,
"grad_norm": 1.36270127671034,
"learning_rate": 1.0184620079340861e-06,
"loss": 0.2161959409713745,
"step": 1335,
"token_acc": 0.9173091270828152
},
{
"epoch": 0.010222881708289307,
"grad_norm": 2.9515316975515082,
"learning_rate": 1.0222764723832775e-06,
"loss": 0.21810717582702638,
"step": 1340,
"token_acc": 0.9188632619439868
},
{
"epoch": 0.010261026789290388,
"grad_norm": 2.4753465021884855,
"learning_rate": 1.0260909368324688e-06,
"loss": 0.24070491790771484,
"step": 1345,
"token_acc": 0.9079069767441861
},
{
"epoch": 0.010299171870291467,
"grad_norm": 2.2391195597119307,
"learning_rate": 1.0299054012816601e-06,
"loss": 0.19637494087219237,
"step": 1350,
"token_acc": 0.9266595727265287
},
{
"epoch": 0.010337316951292546,
"grad_norm": 1.501874801423555,
"learning_rate": 1.0337198657308515e-06,
"loss": 0.1850638747215271,
"step": 1355,
"token_acc": 0.9331290052939537
},
{
"epoch": 0.010375462032293626,
"grad_norm": 1.9215092692671498,
"learning_rate": 1.0375343301800428e-06,
"loss": 0.19690556526184083,
"step": 1360,
"token_acc": 0.9229206265976273
},
{
"epoch": 0.010413607113294705,
"grad_norm": 2.0351688007630298,
"learning_rate": 1.0413487946292342e-06,
"loss": 0.21026864051818847,
"step": 1365,
"token_acc": 0.9231841526045488
},
{
"epoch": 0.010451752194295784,
"grad_norm": 2.858713663197202,
"learning_rate": 1.0451632590784255e-06,
"loss": 0.20908102989196778,
"step": 1370,
"token_acc": 0.9198349284397226
},
{
"epoch": 0.010489897275296865,
"grad_norm": 2.009676428191389,
"learning_rate": 1.0489777235276168e-06,
"loss": 0.21531152725219727,
"step": 1375,
"token_acc": 0.9155263157894736
},
{
"epoch": 0.010528042356297944,
"grad_norm": 1.3183693610019735,
"learning_rate": 1.0527921879768082e-06,
"loss": 0.21780643463134766,
"step": 1380,
"token_acc": 0.9146623908783182
},
{
"epoch": 0.010566187437299023,
"grad_norm": 1.5891701689854854,
"learning_rate": 1.0566066524259995e-06,
"loss": 0.20951955318450927,
"step": 1385,
"token_acc": 0.9208152585553949
},
{
"epoch": 0.010604332518300102,
"grad_norm": 1.6393600889696769,
"learning_rate": 1.0604211168751909e-06,
"loss": 0.18614367246627808,
"step": 1390,
"token_acc": 0.9293369055592766
},
{
"epoch": 0.010642477599301182,
"grad_norm": 2.4931821771997993,
"learning_rate": 1.0642355813243822e-06,
"loss": 0.20893714427947999,
"step": 1395,
"token_acc": 0.9197709923664122
},
{
"epoch": 0.010680622680302261,
"grad_norm": 2.429247732987426,
"learning_rate": 1.0680500457735735e-06,
"loss": 0.2338862657546997,
"step": 1400,
"token_acc": 0.9117171420170538
},
{
"epoch": 0.01071876776130334,
"grad_norm": 2.4946059298594654,
"learning_rate": 1.0718645102227649e-06,
"loss": 0.2015028715133667,
"step": 1405,
"token_acc": 0.9202334630350194
},
{
"epoch": 0.010756912842304421,
"grad_norm": 2.3911240671166683,
"learning_rate": 1.0756789746719562e-06,
"loss": 0.21061160564422607,
"step": 1410,
"token_acc": 0.9214954892435808
},
{
"epoch": 0.0107950579233055,
"grad_norm": 1.4060079854508327,
"learning_rate": 1.0794934391211473e-06,
"loss": 0.199173903465271,
"step": 1415,
"token_acc": 0.9200819672131147
},
{
"epoch": 0.010833203004306579,
"grad_norm": 2.833313044537221,
"learning_rate": 1.0833079035703389e-06,
"loss": 0.1999805450439453,
"step": 1420,
"token_acc": 0.9246463780540077
},
{
"epoch": 0.01087134808530766,
"grad_norm": 2.327047671663845,
"learning_rate": 1.0871223680195302e-06,
"loss": 0.19843223094940185,
"step": 1425,
"token_acc": 0.921313265226546
},
{
"epoch": 0.010909493166308739,
"grad_norm": 1.5269905367181886,
"learning_rate": 1.0909368324687216e-06,
"loss": 0.22138309478759766,
"step": 1430,
"token_acc": 0.9168360583296509
},
{
"epoch": 0.010947638247309818,
"grad_norm": 2.2655424389651038,
"learning_rate": 1.0947512969179127e-06,
"loss": 0.210813307762146,
"step": 1435,
"token_acc": 0.9181891673403395
},
{
"epoch": 0.010985783328310898,
"grad_norm": 4.497225579379361,
"learning_rate": 1.0985657613671042e-06,
"loss": 0.19079312086105346,
"step": 1440,
"token_acc": 0.9275054979579014
},
{
"epoch": 0.011023928409311977,
"grad_norm": 2.52478899534166,
"learning_rate": 1.1023802258162956e-06,
"loss": 0.21445963382720948,
"step": 1445,
"token_acc": 0.9170598290598291
},
{
"epoch": 0.011062073490313056,
"grad_norm": 1.7316361926001056,
"learning_rate": 1.106194690265487e-06,
"loss": 0.22772903442382814,
"step": 1450,
"token_acc": 0.9130503768345894
},
{
"epoch": 0.011100218571314137,
"grad_norm": 2.057758459946669,
"learning_rate": 1.110009154714678e-06,
"loss": 0.19978053569793702,
"step": 1455,
"token_acc": 0.9198540343289634
},
{
"epoch": 0.011138363652315216,
"grad_norm": 3.32799514044511,
"learning_rate": 1.1138236191638694e-06,
"loss": 0.22225940227508545,
"step": 1460,
"token_acc": 0.9191302223308087
},
{
"epoch": 0.011176508733316295,
"grad_norm": 1.9671084973721287,
"learning_rate": 1.117638083613061e-06,
"loss": 0.17709168195724487,
"step": 1465,
"token_acc": 0.9289165446559298
},
{
"epoch": 0.011214653814317375,
"grad_norm": 2.843623860937252,
"learning_rate": 1.1214525480622523e-06,
"loss": 0.23144807815551757,
"step": 1470,
"token_acc": 0.9132642899820753
},
{
"epoch": 0.011252798895318454,
"grad_norm": 3.981730306815325,
"learning_rate": 1.1252670125114434e-06,
"loss": 0.19101991653442382,
"step": 1475,
"token_acc": 0.9289430197686517
},
{
"epoch": 0.011290943976319533,
"grad_norm": 2.2732967439308673,
"learning_rate": 1.1290814769606347e-06,
"loss": 0.2014781951904297,
"step": 1480,
"token_acc": 0.9204331450094162
},
{
"epoch": 0.011329089057320614,
"grad_norm": 1.7358051933519565,
"learning_rate": 1.132895941409826e-06,
"loss": 0.19597008228302001,
"step": 1485,
"token_acc": 0.9246412345857071
},
{
"epoch": 0.011367234138321693,
"grad_norm": 4.386958295840563,
"learning_rate": 1.1367104058590176e-06,
"loss": 0.21251888275146485,
"step": 1490,
"token_acc": 0.9164938106390097
},
{
"epoch": 0.011405379219322772,
"grad_norm": 1.3495754872247667,
"learning_rate": 1.1405248703082087e-06,
"loss": 0.18572702407836914,
"step": 1495,
"token_acc": 0.9293478260869565
},
{
"epoch": 0.011443524300323851,
"grad_norm": 2.131401890501634,
"learning_rate": 1.1443393347574e-06,
"loss": 0.18674194812774658,
"step": 1500,
"token_acc": 0.9313525311553897
},
{
"epoch": 0.011481669381324932,
"grad_norm": 2.2681546770419434,
"learning_rate": 1.1481537992065914e-06,
"loss": 0.20573811531066893,
"step": 1505,
"token_acc": 0.9230769230769231
},
{
"epoch": 0.01151981446232601,
"grad_norm": 2.561760072507073,
"learning_rate": 1.151968263655783e-06,
"loss": 0.19019944667816163,
"step": 1510,
"token_acc": 0.9296650273731094
},
{
"epoch": 0.01155795954332709,
"grad_norm": 1.7188274484662993,
"learning_rate": 1.155782728104974e-06,
"loss": 0.2178044557571411,
"step": 1515,
"token_acc": 0.9196603086877401
},
{
"epoch": 0.01159610462432817,
"grad_norm": 3.235463225723752,
"learning_rate": 1.1595971925541654e-06,
"loss": 0.1984017610549927,
"step": 1520,
"token_acc": 0.9209360180362279
},
{
"epoch": 0.01163424970532925,
"grad_norm": 2.5692116939673726,
"learning_rate": 1.1634116570033568e-06,
"loss": 0.19555585384368895,
"step": 1525,
"token_acc": 0.9253631097305012
},
{
"epoch": 0.011672394786330328,
"grad_norm": 1.8854921961626494,
"learning_rate": 1.1672261214525481e-06,
"loss": 0.2116323232650757,
"step": 1530,
"token_acc": 0.9217062992647546
},
{
"epoch": 0.011710539867331409,
"grad_norm": 2.669500633856558,
"learning_rate": 1.1710405859017395e-06,
"loss": 0.1701640009880066,
"step": 1535,
"token_acc": 0.9318149364558047
},
{
"epoch": 0.011748684948332488,
"grad_norm": 1.592838217336148,
"learning_rate": 1.1748550503509308e-06,
"loss": 0.2282194137573242,
"step": 1540,
"token_acc": 0.9147561648152525
},
{
"epoch": 0.011786830029333567,
"grad_norm": 2.2282897677548674,
"learning_rate": 1.1786695148001221e-06,
"loss": 0.1956816077232361,
"step": 1545,
"token_acc": 0.9246307558644656
},
{
"epoch": 0.011824975110334647,
"grad_norm": 1.8061329002643027,
"learning_rate": 1.1824839792493135e-06,
"loss": 0.2154175043106079,
"step": 1550,
"token_acc": 0.9203967446592065
},
{
"epoch": 0.011863120191335726,
"grad_norm": 1.5764186693317264,
"learning_rate": 1.1862984436985048e-06,
"loss": 0.19561152458190917,
"step": 1555,
"token_acc": 0.9265923806655624
},
{
"epoch": 0.011901265272336805,
"grad_norm": 1.659935290547986,
"learning_rate": 1.1901129081476962e-06,
"loss": 0.2137079954147339,
"step": 1560,
"token_acc": 0.9195372750642673
},
{
"epoch": 0.011939410353337886,
"grad_norm": 2.898445032788292,
"learning_rate": 1.1939273725968875e-06,
"loss": 0.2289292812347412,
"step": 1565,
"token_acc": 0.909896987873256
},
{
"epoch": 0.011977555434338965,
"grad_norm": 6.243199952546219,
"learning_rate": 1.1977418370460788e-06,
"loss": 0.21851663589477538,
"step": 1570,
"token_acc": 0.9196128291362481
},
{
"epoch": 0.012015700515340044,
"grad_norm": 1.7571699733255268,
"learning_rate": 1.2015563014952702e-06,
"loss": 0.1980261206626892,
"step": 1575,
"token_acc": 0.9212472270150357
},
{
"epoch": 0.012053845596341125,
"grad_norm": 2.1699285446581724,
"learning_rate": 1.2053707659444615e-06,
"loss": 0.22580332756042482,
"step": 1580,
"token_acc": 0.9135954135954136
},
{
"epoch": 0.012091990677342204,
"grad_norm": 1.732514950996368,
"learning_rate": 1.2091852303936528e-06,
"loss": 0.22654333114624023,
"step": 1585,
"token_acc": 0.914253172795314
},
{
"epoch": 0.012130135758343282,
"grad_norm": 1.661795196305072,
"learning_rate": 1.2129996948428442e-06,
"loss": 0.22451865673065186,
"step": 1590,
"token_acc": 0.9126977553217924
},
{
"epoch": 0.012168280839344363,
"grad_norm": 1.4590083737815314,
"learning_rate": 1.2168141592920355e-06,
"loss": 0.20143070220947265,
"step": 1595,
"token_acc": 0.9204172876304024
},
{
"epoch": 0.012206425920345442,
"grad_norm": 2.46864342423822,
"learning_rate": 1.2206286237412269e-06,
"loss": 0.21819610595703126,
"step": 1600,
"token_acc": 0.912496524881846
},
{
"epoch": 0.012244571001346521,
"grad_norm": 1.6025830230121634,
"learning_rate": 1.2244430881904182e-06,
"loss": 0.19731900691986085,
"step": 1605,
"token_acc": 0.9258449883449883
},
{
"epoch": 0.0122827160823476,
"grad_norm": 2.483067245428723,
"learning_rate": 1.2282575526396095e-06,
"loss": 0.21335134506225586,
"step": 1610,
"token_acc": 0.9173732202847544
},
{
"epoch": 0.01232086116334868,
"grad_norm": 1.9517715554284745,
"learning_rate": 1.2320720170888009e-06,
"loss": 0.17272624969482422,
"step": 1615,
"token_acc": 0.9347711383722465
},
{
"epoch": 0.01235900624434976,
"grad_norm": 1.6399033856232865,
"learning_rate": 1.2358864815379922e-06,
"loss": 0.19949382543563843,
"step": 1620,
"token_acc": 0.9236289082521784
},
{
"epoch": 0.012397151325350839,
"grad_norm": 2.151792387736221,
"learning_rate": 1.2397009459871836e-06,
"loss": 0.19206552505493163,
"step": 1625,
"token_acc": 0.9238697587256969
},
{
"epoch": 0.01243529640635192,
"grad_norm": 2.534029311567973,
"learning_rate": 1.2435154104363749e-06,
"loss": 0.19873465299606324,
"step": 1630,
"token_acc": 0.9245222159503266
},
{
"epoch": 0.012473441487352998,
"grad_norm": 1.9559510294056606,
"learning_rate": 1.247329874885566e-06,
"loss": 0.19946811199188233,
"step": 1635,
"token_acc": 0.9246297797097178
},
{
"epoch": 0.012511586568354077,
"grad_norm": 1.453141466392558,
"learning_rate": 1.2511443393347576e-06,
"loss": 0.20536503791809083,
"step": 1640,
"token_acc": 0.9232313095835424
},
{
"epoch": 0.012549731649355158,
"grad_norm": 4.520555396158313,
"learning_rate": 1.254958803783949e-06,
"loss": 0.21300430297851564,
"step": 1645,
"token_acc": 0.9190416442612177
},
{
"epoch": 0.012587876730356237,
"grad_norm": 2.0316442890376547,
"learning_rate": 1.2587732682331402e-06,
"loss": 0.21102609634399414,
"step": 1650,
"token_acc": 0.9197116057678847
},
{
"epoch": 0.012626021811357316,
"grad_norm": 2.0285353022866737,
"learning_rate": 1.2625877326823316e-06,
"loss": 0.2073594808578491,
"step": 1655,
"token_acc": 0.921292941005943
},
{
"epoch": 0.012664166892358397,
"grad_norm": 1.4427285612491225,
"learning_rate": 1.266402197131523e-06,
"loss": 0.20829238891601562,
"step": 1660,
"token_acc": 0.9174583659327149
},
{
"epoch": 0.012702311973359475,
"grad_norm": 1.649636196217041,
"learning_rate": 1.270216661580714e-06,
"loss": 0.20897669792175294,
"step": 1665,
"token_acc": 0.921950494313536
},
{
"epoch": 0.012740457054360554,
"grad_norm": 2.209278640340622,
"learning_rate": 1.2740311260299054e-06,
"loss": 0.2076167106628418,
"step": 1670,
"token_acc": 0.922119140625
},
{
"epoch": 0.012778602135361635,
"grad_norm": 1.2443598751648501,
"learning_rate": 1.2778455904790967e-06,
"loss": 0.18085036277770997,
"step": 1675,
"token_acc": 0.9292649098474342
},
{
"epoch": 0.012816747216362714,
"grad_norm": 1.3277199006322167,
"learning_rate": 1.281660054928288e-06,
"loss": 0.21973075866699218,
"step": 1680,
"token_acc": 0.9168342874825932
},
{
"epoch": 0.012854892297363793,
"grad_norm": 1.0668106374888908,
"learning_rate": 1.2854745193774796e-06,
"loss": 0.22137694358825682,
"step": 1685,
"token_acc": 0.9113379558294813
},
{
"epoch": 0.012893037378364874,
"grad_norm": 1.6310639096921686,
"learning_rate": 1.289288983826671e-06,
"loss": 0.22667374610900878,
"step": 1690,
"token_acc": 0.914311828775992
},
{
"epoch": 0.012931182459365953,
"grad_norm": 1.4214766228498597,
"learning_rate": 1.2931034482758623e-06,
"loss": 0.22266836166381837,
"step": 1695,
"token_acc": 0.9131068282667726
},
{
"epoch": 0.012969327540367032,
"grad_norm": 2.2992991104958405,
"learning_rate": 1.2969179127250536e-06,
"loss": 0.21235120296478271,
"step": 1700,
"token_acc": 0.9218462670099301
},
{
"epoch": 0.013007472621368112,
"grad_norm": 3.026295951876485,
"learning_rate": 1.3007323771742448e-06,
"loss": 0.21378889083862304,
"step": 1705,
"token_acc": 0.9199392492948579
},
{
"epoch": 0.013045617702369191,
"grad_norm": 1.6312519864851474,
"learning_rate": 1.304546841623436e-06,
"loss": 0.20472803115844726,
"step": 1710,
"token_acc": 0.9179938744257274
},
{
"epoch": 0.01308376278337027,
"grad_norm": 1.0726640254455453,
"learning_rate": 1.3083613060726274e-06,
"loss": 0.1912538766860962,
"step": 1715,
"token_acc": 0.9259595730183966
},
{
"epoch": 0.01312190786437135,
"grad_norm": 2.266790398469438,
"learning_rate": 1.3121757705218188e-06,
"loss": 0.21682517528533934,
"step": 1720,
"token_acc": 0.9173711480774475
},
{
"epoch": 0.01316005294537243,
"grad_norm": 2.3714828374438666,
"learning_rate": 1.3159902349710101e-06,
"loss": 0.20280778408050537,
"step": 1725,
"token_acc": 0.9212552192066806
},
{
"epoch": 0.013198198026373509,
"grad_norm": 1.3909624807900989,
"learning_rate": 1.3198046994202017e-06,
"loss": 0.2027653217315674,
"step": 1730,
"token_acc": 0.9255072979708081
},
{
"epoch": 0.013236343107374588,
"grad_norm": 2.4521139899018713,
"learning_rate": 1.323619163869393e-06,
"loss": 0.20808277130126954,
"step": 1735,
"token_acc": 0.9186080288154302
},
{
"epoch": 0.013274488188375668,
"grad_norm": 1.555470218647065,
"learning_rate": 1.3274336283185843e-06,
"loss": 0.22431426048278807,
"step": 1740,
"token_acc": 0.9134779068906816
},
{
"epoch": 0.013312633269376747,
"grad_norm": 1.441248702503238,
"learning_rate": 1.3312480927677755e-06,
"loss": 0.2018601417541504,
"step": 1745,
"token_acc": 0.9196772074692691
},
{
"epoch": 0.013350778350377826,
"grad_norm": 2.3386709753315347,
"learning_rate": 1.3350625572169668e-06,
"loss": 0.2292163372039795,
"step": 1750,
"token_acc": 0.9121559209310918
},
{
"epoch": 0.013388923431378907,
"grad_norm": 5.024093663791158,
"learning_rate": 1.3388770216661581e-06,
"loss": 0.22123022079467775,
"step": 1755,
"token_acc": 0.9130696345315085
},
{
"epoch": 0.013427068512379986,
"grad_norm": 2.263089571142426,
"learning_rate": 1.3426914861153495e-06,
"loss": 0.1904762029647827,
"step": 1760,
"token_acc": 0.9303740972165868
},
{
"epoch": 0.013465213593381065,
"grad_norm": 1.8853510889061462,
"learning_rate": 1.3465059505645408e-06,
"loss": 0.19381885528564452,
"step": 1765,
"token_acc": 0.9230888110447146
},
{
"epoch": 0.013503358674382146,
"grad_norm": 2.577275452411215,
"learning_rate": 1.3503204150137322e-06,
"loss": 0.23314962387084961,
"step": 1770,
"token_acc": 0.9113711034538023
},
{
"epoch": 0.013541503755383225,
"grad_norm": 1.866841694870885,
"learning_rate": 1.3541348794629235e-06,
"loss": 0.19566987752914428,
"step": 1775,
"token_acc": 0.9231478377472343
},
{
"epoch": 0.013579648836384304,
"grad_norm": 2.865649558539662,
"learning_rate": 1.357949343912115e-06,
"loss": 0.21033058166503907,
"step": 1780,
"token_acc": 0.9180656303972367
},
{
"epoch": 0.013617793917385384,
"grad_norm": 1.4950377275294398,
"learning_rate": 1.361763808361306e-06,
"loss": 0.18714826107025145,
"step": 1785,
"token_acc": 0.9264397120575885
},
{
"epoch": 0.013655938998386463,
"grad_norm": 2.033983781848697,
"learning_rate": 1.3655782728104975e-06,
"loss": 0.19366238117218018,
"step": 1790,
"token_acc": 0.9255272269436575
},
{
"epoch": 0.013694084079387542,
"grad_norm": 2.351824380201931,
"learning_rate": 1.3693927372596889e-06,
"loss": 0.20130431652069092,
"step": 1795,
"token_acc": 0.9194219400043793
},
{
"epoch": 0.013732229160388623,
"grad_norm": 1.8251128685360127,
"learning_rate": 1.3732072017088802e-06,
"loss": 0.22281627655029296,
"step": 1800,
"token_acc": 0.9153758356432414
},
{
"epoch": 0.013770374241389702,
"grad_norm": 2.401499329696542,
"learning_rate": 1.3770216661580715e-06,
"loss": 0.19353103637695312,
"step": 1805,
"token_acc": 0.9220260613703236
},
{
"epoch": 0.01380851932239078,
"grad_norm": 1.89519583573284,
"learning_rate": 1.3808361306072629e-06,
"loss": 0.1866304874420166,
"step": 1810,
"token_acc": 0.9277313305937367
},
{
"epoch": 0.013846664403391861,
"grad_norm": 1.7027204169917363,
"learning_rate": 1.3846505950564542e-06,
"loss": 0.20289239883422852,
"step": 1815,
"token_acc": 0.9226911826159633
},
{
"epoch": 0.01388480948439294,
"grad_norm": 2.9215876420648828,
"learning_rate": 1.3884650595056455e-06,
"loss": 0.20495715141296386,
"step": 1820,
"token_acc": 0.9194336283185841
},
{
"epoch": 0.01392295456539402,
"grad_norm": 1.5497451431707094,
"learning_rate": 1.3922795239548367e-06,
"loss": 0.1831570863723755,
"step": 1825,
"token_acc": 0.9275883110719371
},
{
"epoch": 0.013961099646395098,
"grad_norm": 2.7476133544792245,
"learning_rate": 1.396093988404028e-06,
"loss": 0.20632634162902833,
"step": 1830,
"token_acc": 0.9196860693263571
},
{
"epoch": 0.013999244727396179,
"grad_norm": 2.1027894548859214,
"learning_rate": 1.3999084528532196e-06,
"loss": 0.1743299961090088,
"step": 1835,
"token_acc": 0.930046678176354
},
{
"epoch": 0.014037389808397258,
"grad_norm": 1.847447633093144,
"learning_rate": 1.403722917302411e-06,
"loss": 0.1987192988395691,
"step": 1840,
"token_acc": 0.9233847065797274
},
{
"epoch": 0.014075534889398337,
"grad_norm": 2.013443824830499,
"learning_rate": 1.4075373817516022e-06,
"loss": 0.199137544631958,
"step": 1845,
"token_acc": 0.9229957805907173
},
{
"epoch": 0.014113679970399418,
"grad_norm": 1.7077196742648677,
"learning_rate": 1.4113518462007936e-06,
"loss": 0.19426697492599487,
"step": 1850,
"token_acc": 0.9262576981524434
},
{
"epoch": 0.014151825051400497,
"grad_norm": 1.51307975231394,
"learning_rate": 1.415166310649985e-06,
"loss": 0.21650102138519287,
"step": 1855,
"token_acc": 0.918355566957414
},
{
"epoch": 0.014189970132401575,
"grad_norm": 2.0886482235510377,
"learning_rate": 1.4189807750991763e-06,
"loss": 0.17356677055358888,
"step": 1860,
"token_acc": 0.9324702854114619
},
{
"epoch": 0.014228115213402656,
"grad_norm": 2.624774670628059,
"learning_rate": 1.4227952395483674e-06,
"loss": 0.19122252464294434,
"step": 1865,
"token_acc": 0.9251751286343066
},
{
"epoch": 0.014266260294403735,
"grad_norm": 1.5968001323097016,
"learning_rate": 1.4266097039975587e-06,
"loss": 0.18822126388549804,
"step": 1870,
"token_acc": 0.92570427674792
},
{
"epoch": 0.014304405375404814,
"grad_norm": 1.489665691365724,
"learning_rate": 1.43042416844675e-06,
"loss": 0.21942522525787353,
"step": 1875,
"token_acc": 0.9155977935801156
},
{
"epoch": 0.014342550456405895,
"grad_norm": 1.7515166817469345,
"learning_rate": 1.4342386328959416e-06,
"loss": 0.22170019149780273,
"step": 1880,
"token_acc": 0.9121032438088594
},
{
"epoch": 0.014380695537406974,
"grad_norm": 1.3700110658679234,
"learning_rate": 1.438053097345133e-06,
"loss": 0.18121532201766968,
"step": 1885,
"token_acc": 0.9314189927847825
},
{
"epoch": 0.014418840618408053,
"grad_norm": 2.00518484564903,
"learning_rate": 1.4418675617943243e-06,
"loss": 0.2080434799194336,
"step": 1890,
"token_acc": 0.9164961346066394
},
{
"epoch": 0.014456985699409133,
"grad_norm": 1.8884675066206056,
"learning_rate": 1.4456820262435156e-06,
"loss": 0.2121840238571167,
"step": 1895,
"token_acc": 0.9162307878007564
},
{
"epoch": 0.014495130780410212,
"grad_norm": 2.1824345919968478,
"learning_rate": 1.449496490692707e-06,
"loss": 0.20350403785705568,
"step": 1900,
"token_acc": 0.918848167539267
},
{
"epoch": 0.014533275861411291,
"grad_norm": 1.5316253448884665,
"learning_rate": 1.453310955141898e-06,
"loss": 0.20921869277954103,
"step": 1905,
"token_acc": 0.925898806472871
},
{
"epoch": 0.014571420942412372,
"grad_norm": 2.3441893476363935,
"learning_rate": 1.4571254195910894e-06,
"loss": 0.21341464519500733,
"step": 1910,
"token_acc": 0.9150692469962396
},
{
"epoch": 0.014609566023413451,
"grad_norm": 1.651816542894586,
"learning_rate": 1.4609398840402808e-06,
"loss": 0.20017762184143068,
"step": 1915,
"token_acc": 0.9229973268948866
},
{
"epoch": 0.01464771110441453,
"grad_norm": 2.585044349822408,
"learning_rate": 1.464754348489472e-06,
"loss": 0.1945590376853943,
"step": 1920,
"token_acc": 0.9227607599844901
},
{
"epoch": 0.01468585618541561,
"grad_norm": 1.5688531871448455,
"learning_rate": 1.4685688129386634e-06,
"loss": 0.18920326232910156,
"step": 1925,
"token_acc": 0.9262596160510084
},
{
"epoch": 0.01472400126641669,
"grad_norm": 2.905887412529429,
"learning_rate": 1.472383277387855e-06,
"loss": 0.22509899139404296,
"step": 1930,
"token_acc": 0.9149105017194251
},
{
"epoch": 0.014762146347417768,
"grad_norm": 1.5804581926671402,
"learning_rate": 1.4761977418370463e-06,
"loss": 0.2163076162338257,
"step": 1935,
"token_acc": 0.9134156503295205
},
{
"epoch": 0.014800291428418847,
"grad_norm": 1.0077040747608188,
"learning_rate": 1.4800122062862377e-06,
"loss": 0.20311894416809081,
"step": 1940,
"token_acc": 0.9220379312758066
},
{
"epoch": 0.014838436509419928,
"grad_norm": 2.2725789711639903,
"learning_rate": 1.4838266707354288e-06,
"loss": 0.21538724899291992,
"step": 1945,
"token_acc": 0.9189055546398549
},
{
"epoch": 0.014876581590421007,
"grad_norm": 1.6928053116041304,
"learning_rate": 1.4876411351846201e-06,
"loss": 0.21162450313568115,
"step": 1950,
"token_acc": 0.9175043078690408
},
{
"epoch": 0.014914726671422086,
"grad_norm": 3.3298784060402453,
"learning_rate": 1.4914555996338115e-06,
"loss": 0.22272152900695802,
"step": 1955,
"token_acc": 0.9126584003270201
},
{
"epoch": 0.014952871752423167,
"grad_norm": 1.831741482498097,
"learning_rate": 1.4952700640830028e-06,
"loss": 0.20517733097076415,
"step": 1960,
"token_acc": 0.9192071957335032
},
{
"epoch": 0.014991016833424246,
"grad_norm": 7.278346277301633,
"learning_rate": 1.4990845285321942e-06,
"loss": 0.20747642517089843,
"step": 1965,
"token_acc": 0.9185028384489561
},
{
"epoch": 0.015029161914425325,
"grad_norm": 1.547573297008924,
"learning_rate": 1.5028989929813855e-06,
"loss": 0.19480135440826415,
"step": 1970,
"token_acc": 0.920563863225069
},
{
"epoch": 0.015067306995426405,
"grad_norm": 1.3681622514724754,
"learning_rate": 1.506713457430577e-06,
"loss": 0.22907204627990724,
"step": 1975,
"token_acc": 0.9120563674321504
},
{
"epoch": 0.015105452076427484,
"grad_norm": 1.5249488630801071,
"learning_rate": 1.5105279218797684e-06,
"loss": 0.20479087829589843,
"step": 1980,
"token_acc": 0.9222253760999148
},
{
"epoch": 0.015143597157428563,
"grad_norm": 4.081627528589628,
"learning_rate": 1.5143423863289595e-06,
"loss": 0.2137747049331665,
"step": 1985,
"token_acc": 0.9177954950073535
},
{
"epoch": 0.015181742238429644,
"grad_norm": 1.6894128702132674,
"learning_rate": 1.5181568507781508e-06,
"loss": 0.19908016920089722,
"step": 1990,
"token_acc": 0.9214322514167954
},
{
"epoch": 0.015219887319430723,
"grad_norm": 2.4590399060956325,
"learning_rate": 1.5219713152273422e-06,
"loss": 0.20733392238616943,
"step": 1995,
"token_acc": 0.9258067248242567
},
{
"epoch": 0.015258032400431802,
"grad_norm": 1.68965750579938,
"learning_rate": 1.5257857796765335e-06,
"loss": 0.20816540718078613,
"step": 2000,
"token_acc": 0.9182299182299183
},
{
"epoch": 0.015296177481432882,
"grad_norm": 1.4890599804501008,
"learning_rate": 1.5296002441257249e-06,
"loss": 0.22224776744842528,
"step": 2005,
"token_acc": 0.9116207951070336
},
{
"epoch": 0.015334322562433961,
"grad_norm": 1.4654195879323395,
"learning_rate": 1.5334147085749162e-06,
"loss": 0.1821568489074707,
"step": 2010,
"token_acc": 0.9261594202898551
},
{
"epoch": 0.01537246764343504,
"grad_norm": 3.9871249443928125,
"learning_rate": 1.5372291730241075e-06,
"loss": 0.20052199363708495,
"step": 2015,
"token_acc": 0.9212617679422912
},
{
"epoch": 0.015410612724436121,
"grad_norm": 2.48012415664987,
"learning_rate": 1.541043637473299e-06,
"loss": 0.21919448375701905,
"step": 2020,
"token_acc": 0.9165497896213184
},
{
"epoch": 0.0154487578054372,
"grad_norm": 1.7109052888079344,
"learning_rate": 1.54485810192249e-06,
"loss": 0.199247145652771,
"step": 2025,
"token_acc": 0.9253078418664938
},
{
"epoch": 0.015486902886438279,
"grad_norm": 3.5421624543668107,
"learning_rate": 1.5486725663716816e-06,
"loss": 0.21096715927124024,
"step": 2030,
"token_acc": 0.9187313529640679
},
{
"epoch": 0.01552504796743936,
"grad_norm": 1.2433426150008775,
"learning_rate": 1.5524870308208729e-06,
"loss": 0.19352424144744873,
"step": 2035,
"token_acc": 0.9225436440410916
},
{
"epoch": 0.015563193048440439,
"grad_norm": 14.14899670749441,
"learning_rate": 1.5563014952700642e-06,
"loss": 0.23587605953216553,
"step": 2040,
"token_acc": 0.9093443354148082
},
{
"epoch": 0.015601338129441518,
"grad_norm": 2.1847543749245206,
"learning_rate": 1.5601159597192556e-06,
"loss": 0.19699397087097167,
"step": 2045,
"token_acc": 0.9213544511196068
},
{
"epoch": 0.015639483210442597,
"grad_norm": 1.0825414924262626,
"learning_rate": 1.563930424168447e-06,
"loss": 0.2287691831588745,
"step": 2050,
"token_acc": 0.9102622404123086
},
{
"epoch": 0.015677628291443677,
"grad_norm": 1.2217548710377804,
"learning_rate": 1.5677448886176382e-06,
"loss": 0.1877738356590271,
"step": 2055,
"token_acc": 0.9250274122807017
},
{
"epoch": 0.015715773372444758,
"grad_norm": 1.4763958582892123,
"learning_rate": 1.5715593530668296e-06,
"loss": 0.21463603973388673,
"step": 2060,
"token_acc": 0.9184824403998975
},
{
"epoch": 0.015753918453445835,
"grad_norm": 2.010529434998551,
"learning_rate": 1.5753738175160207e-06,
"loss": 0.1940779447555542,
"step": 2065,
"token_acc": 0.922939794181995
},
{
"epoch": 0.015792063534446916,
"grad_norm": 1.8848475201601689,
"learning_rate": 1.579188281965212e-06,
"loss": 0.20267610549926757,
"step": 2070,
"token_acc": 0.9193775571802267
},
{
"epoch": 0.015830208615447997,
"grad_norm": 1.391910086694349,
"learning_rate": 1.5830027464144034e-06,
"loss": 0.19234485626220704,
"step": 2075,
"token_acc": 0.924791086350975
},
{
"epoch": 0.015868353696449074,
"grad_norm": 1.6338044213078782,
"learning_rate": 1.586817210863595e-06,
"loss": 0.21937899589538573,
"step": 2080,
"token_acc": 0.9133715446564609
},
{
"epoch": 0.015906498777450154,
"grad_norm": 1.3327925001873593,
"learning_rate": 1.5906316753127863e-06,
"loss": 0.2007765531539917,
"step": 2085,
"token_acc": 0.9207003089598352
},
{
"epoch": 0.015944643858451235,
"grad_norm": 2.7961299496635994,
"learning_rate": 1.5944461397619776e-06,
"loss": 0.2202695846557617,
"step": 2090,
"token_acc": 0.9101036995515696
},
{
"epoch": 0.015982788939452312,
"grad_norm": 1.3213942912429475,
"learning_rate": 1.598260604211169e-06,
"loss": 0.1904911518096924,
"step": 2095,
"token_acc": 0.923354853995143
},
{
"epoch": 0.016020934020453393,
"grad_norm": 1.270581827462333,
"learning_rate": 1.6020750686603603e-06,
"loss": 0.2092984676361084,
"step": 2100,
"token_acc": 0.9161683339853574
},
{
"epoch": 0.01605907910145447,
"grad_norm": 2.016523003744791,
"learning_rate": 1.6058895331095514e-06,
"loss": 0.19184248447418212,
"step": 2105,
"token_acc": 0.9254820528033224
},
{
"epoch": 0.01609722418245555,
"grad_norm": 1.9177239785447313,
"learning_rate": 1.6097039975587428e-06,
"loss": 0.18534836769104004,
"step": 2110,
"token_acc": 0.9286262471220261
},
{
"epoch": 0.01613536926345663,
"grad_norm": 1.526611911347366,
"learning_rate": 1.613518462007934e-06,
"loss": 0.2118476390838623,
"step": 2115,
"token_acc": 0.9185791550942844
},
{
"epoch": 0.01617351434445771,
"grad_norm": 2.0722214330373854,
"learning_rate": 1.6173329264571254e-06,
"loss": 0.2075021505355835,
"step": 2120,
"token_acc": 0.9181445055370557
},
{
"epoch": 0.01621165942545879,
"grad_norm": 1.965900050100034,
"learning_rate": 1.621147390906317e-06,
"loss": 0.20493097305297853,
"step": 2125,
"token_acc": 0.9216409650999778
},
{
"epoch": 0.01624980450645987,
"grad_norm": 1.917856027323587,
"learning_rate": 1.6249618553555083e-06,
"loss": 0.20435960292816163,
"step": 2130,
"token_acc": 0.9195026999874419
},
{
"epoch": 0.016287949587460947,
"grad_norm": 1.385163101231075,
"learning_rate": 1.6287763198046997e-06,
"loss": 0.20995621681213378,
"step": 2135,
"token_acc": 0.9176576172313216
},
{
"epoch": 0.016326094668462028,
"grad_norm": 1.945062065998079,
"learning_rate": 1.632590784253891e-06,
"loss": 0.21435158252716063,
"step": 2140,
"token_acc": 0.9158542432475627
},
{
"epoch": 0.01636423974946311,
"grad_norm": 4.140816821864153,
"learning_rate": 1.6364052487030821e-06,
"loss": 0.2033250331878662,
"step": 2145,
"token_acc": 0.9200258212595036
},
{
"epoch": 0.016402384830464186,
"grad_norm": 1.580632021634232,
"learning_rate": 1.6402197131522735e-06,
"loss": 0.22437751293182373,
"step": 2150,
"token_acc": 0.9107025607353907
},
{
"epoch": 0.016440529911465267,
"grad_norm": 1.4682457425877335,
"learning_rate": 1.6440341776014648e-06,
"loss": 0.2036360263824463,
"step": 2155,
"token_acc": 0.9229340204949961
},
{
"epoch": 0.016478674992466347,
"grad_norm": 2.181365591111052,
"learning_rate": 1.6478486420506561e-06,
"loss": 0.22185654640197755,
"step": 2160,
"token_acc": 0.9125705076551168
},
{
"epoch": 0.016516820073467425,
"grad_norm": 2.6018612009629405,
"learning_rate": 1.6516631064998475e-06,
"loss": 0.21987247467041016,
"step": 2165,
"token_acc": 0.9132124352331606
},
{
"epoch": 0.016554965154468505,
"grad_norm": 1.817463644080717,
"learning_rate": 1.655477570949039e-06,
"loss": 0.2186957836151123,
"step": 2170,
"token_acc": 0.9181847441261667
},
{
"epoch": 0.016593110235469586,
"grad_norm": 1.0487582193593108,
"learning_rate": 1.6592920353982304e-06,
"loss": 0.19036821126937867,
"step": 2175,
"token_acc": 0.9253980288097043
},
{
"epoch": 0.016631255316470663,
"grad_norm": 1.4395960580502125,
"learning_rate": 1.6631064998474217e-06,
"loss": 0.20958418846130372,
"step": 2180,
"token_acc": 0.9220691059670083
},
{
"epoch": 0.016669400397471744,
"grad_norm": 2.7196478309203327,
"learning_rate": 1.6669209642966128e-06,
"loss": 0.2141490697860718,
"step": 2185,
"token_acc": 0.9148921412221682
},
{
"epoch": 0.016707545478472825,
"grad_norm": 1.4163543434827537,
"learning_rate": 1.6707354287458042e-06,
"loss": 0.19764747619628906,
"step": 2190,
"token_acc": 0.921038650551202
},
{
"epoch": 0.016745690559473902,
"grad_norm": 1.573961994238318,
"learning_rate": 1.6745498931949955e-06,
"loss": 0.187377405166626,
"step": 2195,
"token_acc": 0.9240839610912016
},
{
"epoch": 0.016783835640474982,
"grad_norm": 1.8705630187812095,
"learning_rate": 1.6783643576441869e-06,
"loss": 0.2050321102142334,
"step": 2200,
"token_acc": 0.9181295176279687
},
{
"epoch": 0.016821980721476063,
"grad_norm": 7.584455491138986,
"learning_rate": 1.6821788220933782e-06,
"loss": 0.20648694038391113,
"step": 2205,
"token_acc": 0.9215998115096896
},
{
"epoch": 0.01686012580247714,
"grad_norm": 1.4860470855551684,
"learning_rate": 1.6859932865425695e-06,
"loss": 0.202569580078125,
"step": 2210,
"token_acc": 0.9209997283346917
},
{
"epoch": 0.01689827088347822,
"grad_norm": 2.0163443137633004,
"learning_rate": 1.689807750991761e-06,
"loss": 0.20501141548156737,
"step": 2215,
"token_acc": 0.9197914079960268
},
{
"epoch": 0.016936415964479302,
"grad_norm": 1.2106613356260814,
"learning_rate": 1.6936222154409524e-06,
"loss": 0.179191255569458,
"step": 2220,
"token_acc": 0.928799672265465
},
{
"epoch": 0.01697456104548038,
"grad_norm": 1.2834943848493638,
"learning_rate": 1.6974366798901433e-06,
"loss": 0.2387387990951538,
"step": 2225,
"token_acc": 0.9104448866919176
},
{
"epoch": 0.01701270612648146,
"grad_norm": 1.448602672062971,
"learning_rate": 1.7012511443393349e-06,
"loss": 0.200026273727417,
"step": 2230,
"token_acc": 0.9214507632041065
},
{
"epoch": 0.01705085120748254,
"grad_norm": 1.6006266368622954,
"learning_rate": 1.7050656087885262e-06,
"loss": 0.22236857414245606,
"step": 2235,
"token_acc": 0.9151908396946565
},
{
"epoch": 0.017088996288483618,
"grad_norm": 1.2637036317909855,
"learning_rate": 1.7088800732377176e-06,
"loss": 0.20451066493988038,
"step": 2240,
"token_acc": 0.9191728138200618
},
{
"epoch": 0.0171271413694847,
"grad_norm": 1.6226471993173868,
"learning_rate": 1.712694537686909e-06,
"loss": 0.2219872236251831,
"step": 2245,
"token_acc": 0.9169106197446278
},
{
"epoch": 0.01716528645048578,
"grad_norm": 1.735201346802431,
"learning_rate": 1.7165090021361002e-06,
"loss": 0.18203834295272828,
"step": 2250,
"token_acc": 0.9317848084367908
},
{
"epoch": 0.017203431531486856,
"grad_norm": 2.435904939892559,
"learning_rate": 1.7203234665852916e-06,
"loss": 0.19181541204452515,
"step": 2255,
"token_acc": 0.92481860302474
},
{
"epoch": 0.017241576612487937,
"grad_norm": 1.65661036588874,
"learning_rate": 1.724137931034483e-06,
"loss": 0.1956552505493164,
"step": 2260,
"token_acc": 0.9260287393860223
},
{
"epoch": 0.017279721693489018,
"grad_norm": 1.0819906617380828,
"learning_rate": 1.727952395483674e-06,
"loss": 0.18059015274047852,
"step": 2265,
"token_acc": 0.9292368223109414
},
{
"epoch": 0.017317866774490095,
"grad_norm": 1.6002356566426421,
"learning_rate": 1.7317668599328654e-06,
"loss": 0.19932678937911988,
"step": 2270,
"token_acc": 0.9214941716707877
},
{
"epoch": 0.017356011855491175,
"grad_norm": 1.1079381537246162,
"learning_rate": 1.735581324382057e-06,
"loss": 0.19120435714721679,
"step": 2275,
"token_acc": 0.9273379494452323
},
{
"epoch": 0.017394156936492256,
"grad_norm": 1.53785753183134,
"learning_rate": 1.7393957888312483e-06,
"loss": 0.19630740880966185,
"step": 2280,
"token_acc": 0.9227402057342964
},
{
"epoch": 0.017432302017493333,
"grad_norm": 1.4712308568417343,
"learning_rate": 1.7432102532804396e-06,
"loss": 0.19461945295333863,
"step": 2285,
"token_acc": 0.9220398183723367
},
{
"epoch": 0.017470447098494414,
"grad_norm": 2.5791162103829417,
"learning_rate": 1.747024717729631e-06,
"loss": 0.21393980979919433,
"step": 2290,
"token_acc": 0.9130397086873591
},
{
"epoch": 0.017508592179495495,
"grad_norm": 43.800078653780155,
"learning_rate": 1.7508391821788223e-06,
"loss": 0.19713973999023438,
"step": 2295,
"token_acc": 0.9214104025962635
},
{
"epoch": 0.017546737260496572,
"grad_norm": 1.9654500648264297,
"learning_rate": 1.7546536466280136e-06,
"loss": 0.2350461483001709,
"step": 2300,
"token_acc": 0.9086113381380488
},
{
"epoch": 0.017584882341497653,
"grad_norm": 1.9134236841261363,
"learning_rate": 1.7584681110772047e-06,
"loss": 0.2019649028778076,
"step": 2305,
"token_acc": 0.9232497192062897
},
{
"epoch": 0.017623027422498733,
"grad_norm": 2.361715548003919,
"learning_rate": 1.762282575526396e-06,
"loss": 0.2066159725189209,
"step": 2310,
"token_acc": 0.9187770743085638
},
{
"epoch": 0.01766117250349981,
"grad_norm": 1.2092740748048816,
"learning_rate": 1.7660970399755874e-06,
"loss": 0.20651938915252685,
"step": 2315,
"token_acc": 0.9184702505624078
},
{
"epoch": 0.01769931758450089,
"grad_norm": 1.6128451958371073,
"learning_rate": 1.769911504424779e-06,
"loss": 0.1882943868637085,
"step": 2320,
"token_acc": 0.9270037784387812
},
{
"epoch": 0.01773746266550197,
"grad_norm": 1.946971608596617,
"learning_rate": 1.7737259688739703e-06,
"loss": 0.20328311920166015,
"step": 2325,
"token_acc": 0.9200323696020011
},
{
"epoch": 0.01777560774650305,
"grad_norm": 1.632542140732511,
"learning_rate": 1.7775404333231617e-06,
"loss": 0.20440187454223632,
"step": 2330,
"token_acc": 0.9213475126686175
},
{
"epoch": 0.01781375282750413,
"grad_norm": 1.2145575179580637,
"learning_rate": 1.781354897772353e-06,
"loss": 0.2012385129928589,
"step": 2335,
"token_acc": 0.9209455364100788
},
{
"epoch": 0.017851897908505207,
"grad_norm": 1.3591655504286844,
"learning_rate": 1.7851693622215443e-06,
"loss": 0.19382941722869873,
"step": 2340,
"token_acc": 0.9212500910614119
},
{
"epoch": 0.017890042989506288,
"grad_norm": 1.0717228824551364,
"learning_rate": 1.7889838266707355e-06,
"loss": 0.21175992488861084,
"step": 2345,
"token_acc": 0.922087005175549
},
{
"epoch": 0.01792818807050737,
"grad_norm": 2.325392821747694,
"learning_rate": 1.7927982911199268e-06,
"loss": 0.23594927787780762,
"step": 2350,
"token_acc": 0.9093370408312601
},
{
"epoch": 0.017966333151508446,
"grad_norm": 1.239319152483903,
"learning_rate": 1.7966127555691181e-06,
"loss": 0.21045536994934083,
"step": 2355,
"token_acc": 0.9215314849893806
},
{
"epoch": 0.018004478232509526,
"grad_norm": 3.1691778551920597,
"learning_rate": 1.8004272200183095e-06,
"loss": 0.20625803470611573,
"step": 2360,
"token_acc": 0.9206715181209021
},
{
"epoch": 0.018042623313510607,
"grad_norm": 1.411321351181866,
"learning_rate": 1.804241684467501e-06,
"loss": 0.19805383682250977,
"step": 2365,
"token_acc": 0.9235854253561029
},
{
"epoch": 0.018080768394511684,
"grad_norm": 1.1880825933465795,
"learning_rate": 1.8080561489166924e-06,
"loss": 0.20730855464935302,
"step": 2370,
"token_acc": 0.9196210707621645
},
{
"epoch": 0.018118913475512765,
"grad_norm": 1.7944546891048616,
"learning_rate": 1.8118706133658837e-06,
"loss": 0.2249981641769409,
"step": 2375,
"token_acc": 0.9104702022963368
},
{
"epoch": 0.018157058556513846,
"grad_norm": 1.2583919759088358,
"learning_rate": 1.815685077815075e-06,
"loss": 0.19177207946777344,
"step": 2380,
"token_acc": 0.9237647967061245
},
{
"epoch": 0.018195203637514923,
"grad_norm": 1.0946831287441978,
"learning_rate": 1.8194995422642662e-06,
"loss": 0.2124457836151123,
"step": 2385,
"token_acc": 0.9188567276250454
},
{
"epoch": 0.018233348718516004,
"grad_norm": 1.8383446187231192,
"learning_rate": 1.8233140067134575e-06,
"loss": 0.19349327087402343,
"step": 2390,
"token_acc": 0.9261338454720047
},
{
"epoch": 0.018271493799517084,
"grad_norm": 1.941119152091989,
"learning_rate": 1.8271284711626488e-06,
"loss": 0.19319617748260498,
"step": 2395,
"token_acc": 0.9235645584377733
},
{
"epoch": 0.01830963888051816,
"grad_norm": 1.7019295318182834,
"learning_rate": 1.8309429356118402e-06,
"loss": 0.19932930469512938,
"step": 2400,
"token_acc": 0.9212983548243664
},
{
"epoch": 0.018347783961519242,
"grad_norm": 1.3368502297109144,
"learning_rate": 1.8347574000610315e-06,
"loss": 0.18543851375579834,
"step": 2405,
"token_acc": 0.9270160684787505
},
{
"epoch": 0.018385929042520323,
"grad_norm": 1.2547834713592887,
"learning_rate": 1.8385718645102229e-06,
"loss": 0.20437846183776856,
"step": 2410,
"token_acc": 0.9186669311644515
},
{
"epoch": 0.0184240741235214,
"grad_norm": 0.9967423599143065,
"learning_rate": 1.8423863289594144e-06,
"loss": 0.18953574895858766,
"step": 2415,
"token_acc": 0.9250604351329573
},
{
"epoch": 0.01846221920452248,
"grad_norm": 1.390685551905582,
"learning_rate": 1.8462007934086057e-06,
"loss": 0.19771461486816405,
"step": 2420,
"token_acc": 0.9227697408077155
},
{
"epoch": 0.01850036428552356,
"grad_norm": 1.8119760172996933,
"learning_rate": 1.8500152578577969e-06,
"loss": 0.22690942287445068,
"step": 2425,
"token_acc": 0.9115115653577192
},
{
"epoch": 0.01853850936652464,
"grad_norm": 1.3406730695016253,
"learning_rate": 1.8538297223069882e-06,
"loss": 0.18345394134521484,
"step": 2430,
"token_acc": 0.9293214862681745
},
{
"epoch": 0.01857665444752572,
"grad_norm": 1.3631166798123526,
"learning_rate": 1.8576441867561796e-06,
"loss": 0.1886366844177246,
"step": 2435,
"token_acc": 0.9293960574884159
},
{
"epoch": 0.0186147995285268,
"grad_norm": 1.1991865751441018,
"learning_rate": 1.8614586512053709e-06,
"loss": 0.20665998458862306,
"step": 2440,
"token_acc": 0.9221501390176089
},
{
"epoch": 0.018652944609527877,
"grad_norm": 1.4335840990483475,
"learning_rate": 1.8652731156545622e-06,
"loss": 0.20270237922668458,
"step": 2445,
"token_acc": 0.9229033197437391
},
{
"epoch": 0.018691089690528958,
"grad_norm": 2.885663189592157,
"learning_rate": 1.8690875801037536e-06,
"loss": 0.23508081436157227,
"step": 2450,
"token_acc": 0.9112227805695142
},
{
"epoch": 0.01872923477153004,
"grad_norm": 1.7218620924617174,
"learning_rate": 1.872902044552945e-06,
"loss": 0.22273545265197753,
"step": 2455,
"token_acc": 0.9135168628839515
},
{
"epoch": 0.018767379852531116,
"grad_norm": 1.4362676119651934,
"learning_rate": 1.8767165090021365e-06,
"loss": 0.17815079689025878,
"step": 2460,
"token_acc": 0.9292741498179564
},
{
"epoch": 0.018805524933532197,
"grad_norm": 1.6355840560514923,
"learning_rate": 1.8805309734513274e-06,
"loss": 0.1880096197128296,
"step": 2465,
"token_acc": 0.9285714285714286
},
{
"epoch": 0.018843670014533277,
"grad_norm": 1.3027374560425038,
"learning_rate": 1.884345437900519e-06,
"loss": 0.18501565456390381,
"step": 2470,
"token_acc": 0.9285664795953718
},
{
"epoch": 0.018881815095534354,
"grad_norm": 2.048830636586346,
"learning_rate": 1.8881599023497103e-06,
"loss": 0.20168228149414064,
"step": 2475,
"token_acc": 0.921498926164556
},
{
"epoch": 0.018919960176535435,
"grad_norm": 1.5195396806427874,
"learning_rate": 1.8919743667989016e-06,
"loss": 0.20652539730072023,
"step": 2480,
"token_acc": 0.919762776905634
},
{
"epoch": 0.018958105257536516,
"grad_norm": 1.3453390383371566,
"learning_rate": 1.895788831248093e-06,
"loss": 0.19405163526535035,
"step": 2485,
"token_acc": 0.924301545184954
},
{
"epoch": 0.018996250338537593,
"grad_norm": 1.3030026483550121,
"learning_rate": 1.8996032956972843e-06,
"loss": 0.22250971794128419,
"step": 2490,
"token_acc": 0.9139876880984952
},
{
"epoch": 0.019034395419538674,
"grad_norm": 1.2869428954176683,
"learning_rate": 1.9034177601464756e-06,
"loss": 0.20713326930999756,
"step": 2495,
"token_acc": 0.9180003048315806
},
{
"epoch": 0.019072540500539754,
"grad_norm": 2.202482237967539,
"learning_rate": 1.907232224595667e-06,
"loss": 0.20494112968444825,
"step": 2500,
"token_acc": 0.9188058035714286
},
{
"epoch": 0.01911068558154083,
"grad_norm": 1.817881570038458,
"learning_rate": 1.9110466890448583e-06,
"loss": 0.20258455276489257,
"step": 2505,
"token_acc": 0.9216130940642872
},
{
"epoch": 0.019148830662541912,
"grad_norm": 1.2145700658553251,
"learning_rate": 1.9148611534940494e-06,
"loss": 0.2241365909576416,
"step": 2510,
"token_acc": 0.9165954803137377
},
{
"epoch": 0.019186975743542993,
"grad_norm": 1.9878463603684455,
"learning_rate": 1.918675617943241e-06,
"loss": 0.22702600955963134,
"step": 2515,
"token_acc": 0.9109577512203333
},
{
"epoch": 0.01922512082454407,
"grad_norm": 1.76923988096184,
"learning_rate": 1.922490082392432e-06,
"loss": 0.19853074550628663,
"step": 2520,
"token_acc": 0.921753460777851
},
{
"epoch": 0.01926326590554515,
"grad_norm": 2.167604085590848,
"learning_rate": 1.9263045468416236e-06,
"loss": 0.20722584724426268,
"step": 2525,
"token_acc": 0.9188640973630832
},
{
"epoch": 0.01930141098654623,
"grad_norm": 2.3051422126078163,
"learning_rate": 1.9301190112908148e-06,
"loss": 0.20600132942199706,
"step": 2530,
"token_acc": 0.9217038539553752
},
{
"epoch": 0.01933955606754731,
"grad_norm": 1.6977591185650496,
"learning_rate": 1.9339334757400063e-06,
"loss": 0.20889570713043212,
"step": 2535,
"token_acc": 0.916250279662913
},
{
"epoch": 0.01937770114854839,
"grad_norm": 1.6183398443764982,
"learning_rate": 1.937747940189198e-06,
"loss": 0.1942192792892456,
"step": 2540,
"token_acc": 0.9275437684174034
},
{
"epoch": 0.019415846229549467,
"grad_norm": 7.237693079661838,
"learning_rate": 1.941562404638389e-06,
"loss": 0.19048376083374025,
"step": 2545,
"token_acc": 0.9276581704198631
},
{
"epoch": 0.019453991310550547,
"grad_norm": 1.4962984776153763,
"learning_rate": 1.94537686908758e-06,
"loss": 0.22876057624816895,
"step": 2550,
"token_acc": 0.9109613852061146
},
{
"epoch": 0.019492136391551628,
"grad_norm": 1.9179380244038937,
"learning_rate": 1.9491913335367717e-06,
"loss": 0.1788722515106201,
"step": 2555,
"token_acc": 0.9300826296195406
},
{
"epoch": 0.019530281472552705,
"grad_norm": 1.9934246872566406,
"learning_rate": 1.953005797985963e-06,
"loss": 0.22490510940551758,
"step": 2560,
"token_acc": 0.9131477042227549
},
{
"epoch": 0.019568426553553786,
"grad_norm": 1.2180102355318678,
"learning_rate": 1.9568202624351544e-06,
"loss": 0.20247552394866944,
"step": 2565,
"token_acc": 0.920934858279463
},
{
"epoch": 0.019606571634554867,
"grad_norm": 1.332013870563317,
"learning_rate": 1.9606347268843455e-06,
"loss": 0.19542100429534912,
"step": 2570,
"token_acc": 0.9209403254972875
},
{
"epoch": 0.019644716715555944,
"grad_norm": 2.127333241097001,
"learning_rate": 1.964449191333537e-06,
"loss": 0.2115682601928711,
"step": 2575,
"token_acc": 0.9171516603902773
},
{
"epoch": 0.019682861796557025,
"grad_norm": 7.468367347055794,
"learning_rate": 1.9682636557827286e-06,
"loss": 0.21477277278900148,
"step": 2580,
"token_acc": 0.9163501308348105
},
{
"epoch": 0.019721006877558105,
"grad_norm": 1.900045519498587,
"learning_rate": 1.9720781202319193e-06,
"loss": 0.21193952560424806,
"step": 2585,
"token_acc": 0.9168292506684009
},
{
"epoch": 0.019759151958559183,
"grad_norm": 1.3226908762110294,
"learning_rate": 1.975892584681111e-06,
"loss": 0.20381035804748535,
"step": 2590,
"token_acc": 0.9197420279469725
},
{
"epoch": 0.019797297039560263,
"grad_norm": 2.1349391596394627,
"learning_rate": 1.9797070491303024e-06,
"loss": 0.22477612495422364,
"step": 2595,
"token_acc": 0.9165057915057915
},
{
"epoch": 0.019835442120561344,
"grad_norm": 2.306680141228622,
"learning_rate": 1.9835215135794935e-06,
"loss": 0.19850780963897705,
"step": 2600,
"token_acc": 0.920951976200595
},
{
"epoch": 0.01987358720156242,
"grad_norm": 2.6125506140044834,
"learning_rate": 1.987335978028685e-06,
"loss": 0.2011402130126953,
"step": 2605,
"token_acc": 0.9217556005486252
},
{
"epoch": 0.019911732282563502,
"grad_norm": 1.2246492634435528,
"learning_rate": 1.991150442477876e-06,
"loss": 0.16237325668334962,
"step": 2610,
"token_acc": 0.9352309965944869
},
{
"epoch": 0.019949877363564582,
"grad_norm": 1.8777652004646164,
"learning_rate": 1.9949649069270677e-06,
"loss": 0.21197285652160644,
"step": 2615,
"token_acc": 0.9173806203094906
},
{
"epoch": 0.01998802244456566,
"grad_norm": 3.715721198194247,
"learning_rate": 1.998779371376259e-06,
"loss": 0.18457635641098022,
"step": 2620,
"token_acc": 0.9286711118002791
},
{
"epoch": 0.02002616752556674,
"grad_norm": 1.592587942286185,
"learning_rate": 2.00259383582545e-06,
"loss": 0.19520969390869142,
"step": 2625,
"token_acc": 0.9199297012302284
},
{
"epoch": 0.02006431260656782,
"grad_norm": 1.5696944509697894,
"learning_rate": 2.0064083002746415e-06,
"loss": 0.21010656356811525,
"step": 2630,
"token_acc": 0.9182804027885361
},
{
"epoch": 0.0201024576875689,
"grad_norm": 1.2781727976375845,
"learning_rate": 2.0102227647238327e-06,
"loss": 0.21386294364929198,
"step": 2635,
"token_acc": 0.9152439844065062
},
{
"epoch": 0.02014060276856998,
"grad_norm": 1.3258099655072049,
"learning_rate": 2.0140372291730242e-06,
"loss": 0.20661239624023436,
"step": 2640,
"token_acc": 0.918448930641615
},
{
"epoch": 0.02017874784957106,
"grad_norm": 1.7403961980036085,
"learning_rate": 2.0178516936222158e-06,
"loss": 0.19582462310791016,
"step": 2645,
"token_acc": 0.9218826491363032
},
{
"epoch": 0.020216892930572137,
"grad_norm": 1.7233440008613232,
"learning_rate": 2.021666158071407e-06,
"loss": 0.20986742973327638,
"step": 2650,
"token_acc": 0.9167579408543264
},
{
"epoch": 0.020255038011573218,
"grad_norm": 3.5577944737211977,
"learning_rate": 2.0254806225205984e-06,
"loss": 0.19719557762145995,
"step": 2655,
"token_acc": 0.9237371260421775
},
{
"epoch": 0.0202931830925743,
"grad_norm": 1.158552999787391,
"learning_rate": 2.0292950869697896e-06,
"loss": 0.20484368801116942,
"step": 2660,
"token_acc": 0.9190930787589499
},
{
"epoch": 0.020331328173575376,
"grad_norm": 1.9125347639980081,
"learning_rate": 2.0331095514189807e-06,
"loss": 0.2020045042037964,
"step": 2665,
"token_acc": 0.9193178561192319
},
{
"epoch": 0.020369473254576456,
"grad_norm": 1.7699797077242856,
"learning_rate": 2.0369240158681723e-06,
"loss": 0.23643569946289061,
"step": 2670,
"token_acc": 0.9077784960137901
},
{
"epoch": 0.020407618335577537,
"grad_norm": 2.0466798934801127,
"learning_rate": 2.0407384803173634e-06,
"loss": 0.21667494773864746,
"step": 2675,
"token_acc": 0.9170589190066548
},
{
"epoch": 0.020445763416578614,
"grad_norm": 1.2664469868102024,
"learning_rate": 2.044552944766555e-06,
"loss": 0.21387147903442383,
"step": 2680,
"token_acc": 0.9174318456829668
},
{
"epoch": 0.020483908497579695,
"grad_norm": 1.7701009312674256,
"learning_rate": 2.0483674092157465e-06,
"loss": 0.2026449680328369,
"step": 2685,
"token_acc": 0.9215110833593506
},
{
"epoch": 0.020522053578580775,
"grad_norm": 2.8552688036512675,
"learning_rate": 2.0521818736649376e-06,
"loss": 0.23055758476257324,
"step": 2690,
"token_acc": 0.9108095387511767
},
{
"epoch": 0.020560198659581853,
"grad_norm": 1.0337694095268017,
"learning_rate": 2.055996338114129e-06,
"loss": 0.21216511726379395,
"step": 2695,
"token_acc": 0.9183030303030303
},
{
"epoch": 0.020598343740582933,
"grad_norm": 1.2653385155254577,
"learning_rate": 2.0598108025633203e-06,
"loss": 0.20372607707977294,
"step": 2700,
"token_acc": 0.9236231774571277
},
{
"epoch": 0.020636488821584014,
"grad_norm": 1.3160865096013206,
"learning_rate": 2.0636252670125114e-06,
"loss": 0.19869842529296874,
"step": 2705,
"token_acc": 0.9218445550985062
},
{
"epoch": 0.02067463390258509,
"grad_norm": 1.0164745238804,
"learning_rate": 2.067439731461703e-06,
"loss": 0.2025317907333374,
"step": 2710,
"token_acc": 0.9241326725123904
},
{
"epoch": 0.020712778983586172,
"grad_norm": 1.0355263589391157,
"learning_rate": 2.071254195910894e-06,
"loss": 0.2157978057861328,
"step": 2715,
"token_acc": 0.9175509033553197
},
{
"epoch": 0.020750924064587253,
"grad_norm": 1.1447404787247333,
"learning_rate": 2.0750686603600856e-06,
"loss": 0.2149195671081543,
"step": 2720,
"token_acc": 0.9178747411269014
},
{
"epoch": 0.02078906914558833,
"grad_norm": 1.5203438117368853,
"learning_rate": 2.0788831248092768e-06,
"loss": 0.18731241226196288,
"step": 2725,
"token_acc": 0.9227862883835956
},
{
"epoch": 0.02082721422658941,
"grad_norm": 1.4403453667638735,
"learning_rate": 2.0826975892584683e-06,
"loss": 0.17171252965927125,
"step": 2730,
"token_acc": 0.9289990421455939
},
{
"epoch": 0.02086535930759049,
"grad_norm": 1.4359740074667224,
"learning_rate": 2.08651205370766e-06,
"loss": 0.21033456325531005,
"step": 2735,
"token_acc": 0.9199894142554693
},
{
"epoch": 0.02090350438859157,
"grad_norm": 1.0051354828686356,
"learning_rate": 2.090326518156851e-06,
"loss": 0.161898934841156,
"step": 2740,
"token_acc": 0.9368178829717291
},
{
"epoch": 0.02094164946959265,
"grad_norm": 2.2046914820981813,
"learning_rate": 2.094140982606042e-06,
"loss": 0.1903095245361328,
"step": 2745,
"token_acc": 0.9243278404163053
},
{
"epoch": 0.02097979455059373,
"grad_norm": 1.7566394361994613,
"learning_rate": 2.0979554470552337e-06,
"loss": 0.1944281578063965,
"step": 2750,
"token_acc": 0.92602587506945
},
{
"epoch": 0.021017939631594807,
"grad_norm": 1.8093662410761455,
"learning_rate": 2.101769911504425e-06,
"loss": 0.21700360774993896,
"step": 2755,
"token_acc": 0.9197230693150911
},
{
"epoch": 0.021056084712595888,
"grad_norm": 1.7618955134872507,
"learning_rate": 2.1055843759536163e-06,
"loss": 0.23112616539001465,
"step": 2760,
"token_acc": 0.9134187082405345
},
{
"epoch": 0.021094229793596965,
"grad_norm": 1.224869239508112,
"learning_rate": 2.1093988404028075e-06,
"loss": 0.20494866371154785,
"step": 2765,
"token_acc": 0.9221903937112784
},
{
"epoch": 0.021132374874598046,
"grad_norm": 1.266718773309995,
"learning_rate": 2.113213304851999e-06,
"loss": 0.21830267906188966,
"step": 2770,
"token_acc": 0.9143821172501054
},
{
"epoch": 0.021170519955599126,
"grad_norm": 2.13213432225318,
"learning_rate": 2.11702776930119e-06,
"loss": 0.19345738887786865,
"step": 2775,
"token_acc": 0.9253987730061349
},
{
"epoch": 0.021208665036600204,
"grad_norm": 1.0344893468856853,
"learning_rate": 2.1208422337503817e-06,
"loss": 0.18992245197296143,
"step": 2780,
"token_acc": 0.9282378580323786
},
{
"epoch": 0.021246810117601284,
"grad_norm": 1.2323028830401854,
"learning_rate": 2.124656698199573e-06,
"loss": 0.18465198278427125,
"step": 2785,
"token_acc": 0.9250308938624193
},
{
"epoch": 0.021284955198602365,
"grad_norm": 1.389054744990813,
"learning_rate": 2.1284711626487644e-06,
"loss": 0.21194868087768554,
"step": 2790,
"token_acc": 0.913845754012085
},
{
"epoch": 0.021323100279603442,
"grad_norm": 1.4099272429460736,
"learning_rate": 2.1322856270979555e-06,
"loss": 0.19589532613754274,
"step": 2795,
"token_acc": 0.9238692688971499
},
{
"epoch": 0.021361245360604523,
"grad_norm": 1.2162462268915564,
"learning_rate": 2.136100091547147e-06,
"loss": 0.21031720638275148,
"step": 2800,
"token_acc": 0.916545904015003
},
{
"epoch": 0.021399390441605604,
"grad_norm": 5.896646635810368,
"learning_rate": 2.139914555996338e-06,
"loss": 0.21370596885681153,
"step": 2805,
"token_acc": 0.9173048720873391
},
{
"epoch": 0.02143753552260668,
"grad_norm": 1.2032756025108589,
"learning_rate": 2.1437290204455297e-06,
"loss": 0.2118002414703369,
"step": 2810,
"token_acc": 0.9175477092891815
},
{
"epoch": 0.02147568060360776,
"grad_norm": 2.214841001020114,
"learning_rate": 2.147543484894721e-06,
"loss": 0.20006299018859863,
"step": 2815,
"token_acc": 0.924581777962118
},
{
"epoch": 0.021513825684608842,
"grad_norm": 1.0448783580490337,
"learning_rate": 2.1513579493439124e-06,
"loss": 0.21090760231018066,
"step": 2820,
"token_acc": 0.9187909686817188
},
{
"epoch": 0.02155197076560992,
"grad_norm": 1.6632623908618909,
"learning_rate": 2.1551724137931035e-06,
"loss": 0.19327199459075928,
"step": 2825,
"token_acc": 0.9247885038822575
},
{
"epoch": 0.021590115846611,
"grad_norm": 1.572764511331945,
"learning_rate": 2.1589868782422947e-06,
"loss": 0.20100698471069336,
"step": 2830,
"token_acc": 0.9204061346208463
},
{
"epoch": 0.02162826092761208,
"grad_norm": 2.063521437303773,
"learning_rate": 2.1628013426914862e-06,
"loss": 0.2081470251083374,
"step": 2835,
"token_acc": 0.9202662229617304
},
{
"epoch": 0.021666406008613158,
"grad_norm": 1.015133448143875,
"learning_rate": 2.1666158071406778e-06,
"loss": 0.2131657600402832,
"step": 2840,
"token_acc": 0.9186417106101872
},
{
"epoch": 0.02170455108961424,
"grad_norm": 2.325376074366735,
"learning_rate": 2.170430271589869e-06,
"loss": 0.21315178871154786,
"step": 2845,
"token_acc": 0.9162956366874443
},
{
"epoch": 0.02174269617061532,
"grad_norm": 1.637925489889917,
"learning_rate": 2.1742447360390604e-06,
"loss": 0.2165147304534912,
"step": 2850,
"token_acc": 0.9122819714844033
},
{
"epoch": 0.021780841251616397,
"grad_norm": 1.1201898107049064,
"learning_rate": 2.1780592004882516e-06,
"loss": 0.2119300365447998,
"step": 2855,
"token_acc": 0.915927471954083
},
{
"epoch": 0.021818986332617477,
"grad_norm": 1.7019134053879024,
"learning_rate": 2.181873664937443e-06,
"loss": 0.21481680870056152,
"step": 2860,
"token_acc": 0.9155999437332958
},
{
"epoch": 0.021857131413618558,
"grad_norm": 1.7388751836707037,
"learning_rate": 2.1856881293866342e-06,
"loss": 0.1959786295890808,
"step": 2865,
"token_acc": 0.9204308313699091
},
{
"epoch": 0.021895276494619635,
"grad_norm": 1.3109775878916654,
"learning_rate": 2.1895025938358254e-06,
"loss": 0.20571684837341309,
"step": 2870,
"token_acc": 0.9216351074718526
},
{
"epoch": 0.021933421575620716,
"grad_norm": 1.3621079705691053,
"learning_rate": 2.193317058285017e-06,
"loss": 0.1924302577972412,
"step": 2875,
"token_acc": 0.9243261921216309
},
{
"epoch": 0.021971566656621797,
"grad_norm": 1.5183335484576328,
"learning_rate": 2.1971315227342085e-06,
"loss": 0.20640120506286622,
"step": 2880,
"token_acc": 0.9213394408466277
},
{
"epoch": 0.022009711737622874,
"grad_norm": 1.4540571468385162,
"learning_rate": 2.2009459871833996e-06,
"loss": 0.18939340114593506,
"step": 2885,
"token_acc": 0.9273614310882233
},
{
"epoch": 0.022047856818623954,
"grad_norm": 3.513247029595685,
"learning_rate": 2.204760451632591e-06,
"loss": 0.1810924530029297,
"step": 2890,
"token_acc": 0.9278501503121869
},
{
"epoch": 0.022086001899625035,
"grad_norm": 3.4868763798376365,
"learning_rate": 2.2085749160817823e-06,
"loss": 0.20305850505828857,
"step": 2895,
"token_acc": 0.9209842690142878
},
{
"epoch": 0.022124146980626112,
"grad_norm": 1.4774157899005798,
"learning_rate": 2.212389380530974e-06,
"loss": 0.20870618820190429,
"step": 2900,
"token_acc": 0.9168318184924485
},
{
"epoch": 0.022162292061627193,
"grad_norm": 1.3545222032499986,
"learning_rate": 2.216203844980165e-06,
"loss": 0.19565773010253906,
"step": 2905,
"token_acc": 0.9203842515100793
},
{
"epoch": 0.022200437142628274,
"grad_norm": 1.2194715747741562,
"learning_rate": 2.220018309429356e-06,
"loss": 0.2221846103668213,
"step": 2910,
"token_acc": 0.9159565271088057
},
{
"epoch": 0.02223858222362935,
"grad_norm": 2.408027510773736,
"learning_rate": 2.2238327738785476e-06,
"loss": 0.2199941635131836,
"step": 2915,
"token_acc": 0.9174886488179114
},
{
"epoch": 0.02227672730463043,
"grad_norm": 1.473524898006836,
"learning_rate": 2.2276472383277388e-06,
"loss": 0.20053207874298096,
"step": 2920,
"token_acc": 0.9217468692390821
},
{
"epoch": 0.022314872385631512,
"grad_norm": 2.3335414538548296,
"learning_rate": 2.2314617027769303e-06,
"loss": 0.16574041843414306,
"step": 2925,
"token_acc": 0.9340620839464648
},
{
"epoch": 0.02235301746663259,
"grad_norm": 2.8391238040884152,
"learning_rate": 2.235276167226122e-06,
"loss": 0.2059793472290039,
"step": 2930,
"token_acc": 0.9182191135920371
},
{
"epoch": 0.02239116254763367,
"grad_norm": 3.4476729762701646,
"learning_rate": 2.239090631675313e-06,
"loss": 0.19382208585739136,
"step": 2935,
"token_acc": 0.9197308715381004
},
{
"epoch": 0.02242930762863475,
"grad_norm": 1.745871323045488,
"learning_rate": 2.2429050961245045e-06,
"loss": 0.19769505262374878,
"step": 2940,
"token_acc": 0.9222211657316726
},
{
"epoch": 0.022467452709635828,
"grad_norm": 1.5026693850345831,
"learning_rate": 2.2467195605736957e-06,
"loss": 0.2234032392501831,
"step": 2945,
"token_acc": 0.9132177207811134
},
{
"epoch": 0.02250559779063691,
"grad_norm": 2.154091505878964,
"learning_rate": 2.250534025022887e-06,
"loss": 0.19003331661224365,
"step": 2950,
"token_acc": 0.9294983893235159
},
{
"epoch": 0.02254374287163799,
"grad_norm": 6.086425675730291,
"learning_rate": 2.2543484894720783e-06,
"loss": 0.2481135129928589,
"step": 2955,
"token_acc": 0.9025681950087058
},
{
"epoch": 0.022581887952639067,
"grad_norm": 1.7831563423132075,
"learning_rate": 2.2581629539212695e-06,
"loss": 0.19258670806884765,
"step": 2960,
"token_acc": 0.9227014483476034
},
{
"epoch": 0.022620033033640147,
"grad_norm": 2.3307267894467265,
"learning_rate": 2.261977418370461e-06,
"loss": 0.20832488536834717,
"step": 2965,
"token_acc": 0.9174521949905736
},
{
"epoch": 0.022658178114641228,
"grad_norm": 2.2591268415245045,
"learning_rate": 2.265791882819652e-06,
"loss": 0.1919557571411133,
"step": 2970,
"token_acc": 0.9269755895738518
},
{
"epoch": 0.022696323195642305,
"grad_norm": 5.614166345053465,
"learning_rate": 2.2696063472688437e-06,
"loss": 0.2213520050048828,
"step": 2975,
"token_acc": 0.9145066889632107
},
{
"epoch": 0.022734468276643386,
"grad_norm": 3.306989118009514,
"learning_rate": 2.2734208117180352e-06,
"loss": 0.16601232290267945,
"step": 2980,
"token_acc": 0.9331041183944988
},
{
"epoch": 0.022772613357644463,
"grad_norm": 5.665509606810251,
"learning_rate": 2.2772352761672264e-06,
"loss": 0.21386363506317138,
"step": 2985,
"token_acc": 0.9149357748018584
},
{
"epoch": 0.022810758438645544,
"grad_norm": 1.2613879879384444,
"learning_rate": 2.2810497406164175e-06,
"loss": 0.20539891719818115,
"step": 2990,
"token_acc": 0.9198976844372644
},
{
"epoch": 0.022848903519646625,
"grad_norm": 1.942429322955949,
"learning_rate": 2.284864205065609e-06,
"loss": 0.19969228506088257,
"step": 2995,
"token_acc": 0.9207992296581609
},
{
"epoch": 0.022887048600647702,
"grad_norm": 2.3751063848925504,
"learning_rate": 2.2886786695148e-06,
"loss": 0.21538619995117186,
"step": 3000,
"token_acc": 0.9176108546649882
},
{
"epoch": 0.022925193681648783,
"grad_norm": 2.2539640488018953,
"learning_rate": 2.2924931339639917e-06,
"loss": 0.20669665336608886,
"step": 3005,
"token_acc": 0.9196124441753084
},
{
"epoch": 0.022963338762649863,
"grad_norm": 2.3342461671014743,
"learning_rate": 2.296307598413183e-06,
"loss": 0.21201112270355224,
"step": 3010,
"token_acc": 0.9176204936541307
},
{
"epoch": 0.02300148384365094,
"grad_norm": 1.1594973426070225,
"learning_rate": 2.3001220628623744e-06,
"loss": 0.20576462745666504,
"step": 3015,
"token_acc": 0.917910447761194
},
{
"epoch": 0.02303962892465202,
"grad_norm": 2.0814712571459677,
"learning_rate": 2.303936527311566e-06,
"loss": 0.20728278160095215,
"step": 3020,
"token_acc": 0.9153217637922885
},
{
"epoch": 0.023077774005653102,
"grad_norm": 3.8353558139676855,
"learning_rate": 2.3077509917607567e-06,
"loss": 0.1788616418838501,
"step": 3025,
"token_acc": 0.9276512069433144
},
{
"epoch": 0.02311591908665418,
"grad_norm": 1.9031319685385915,
"learning_rate": 2.311565456209948e-06,
"loss": 0.20528368949890136,
"step": 3030,
"token_acc": 0.9175701530612245
},
{
"epoch": 0.02315406416765526,
"grad_norm": 1.635938805565002,
"learning_rate": 2.3153799206591398e-06,
"loss": 0.21890780925750733,
"step": 3035,
"token_acc": 0.9092424646572419
},
{
"epoch": 0.02319220924865634,
"grad_norm": 1.617065272870264,
"learning_rate": 2.319194385108331e-06,
"loss": 0.1962704062461853,
"step": 3040,
"token_acc": 0.9193251307743314
},
{
"epoch": 0.023230354329657418,
"grad_norm": 1.7699180790824782,
"learning_rate": 2.3230088495575224e-06,
"loss": 0.21197433471679689,
"step": 3045,
"token_acc": 0.9177777777777778
},
{
"epoch": 0.0232684994106585,
"grad_norm": 1.3123572586017935,
"learning_rate": 2.3268233140067136e-06,
"loss": 0.19481242895126344,
"step": 3050,
"token_acc": 0.9221951568992692
},
{
"epoch": 0.02330664449165958,
"grad_norm": 1.4549497586963682,
"learning_rate": 2.330637778455905e-06,
"loss": 0.19850339889526367,
"step": 3055,
"token_acc": 0.923897353648757
},
{
"epoch": 0.023344789572660656,
"grad_norm": 1.8898439288581246,
"learning_rate": 2.3344522429050962e-06,
"loss": 0.21541030406951905,
"step": 3060,
"token_acc": 0.9192298661657666
},
{
"epoch": 0.023382934653661737,
"grad_norm": 1.4209395493866182,
"learning_rate": 2.3382667073542874e-06,
"loss": 0.2228672981262207,
"step": 3065,
"token_acc": 0.9140913893291073
},
{
"epoch": 0.023421079734662818,
"grad_norm": 1.8235166473958186,
"learning_rate": 2.342081171803479e-06,
"loss": 0.20130834579467774,
"step": 3070,
"token_acc": 0.9229288856304986
},
{
"epoch": 0.023459224815663895,
"grad_norm": 1.1443484612627595,
"learning_rate": 2.34589563625267e-06,
"loss": 0.1719369888305664,
"step": 3075,
"token_acc": 0.9326820044122281
},
{
"epoch": 0.023497369896664975,
"grad_norm": 1.4985911396997922,
"learning_rate": 2.3497101007018616e-06,
"loss": 0.21691513061523438,
"step": 3080,
"token_acc": 0.9165984538426557
},
{
"epoch": 0.023535514977666056,
"grad_norm": 1.355031532209808,
"learning_rate": 2.353524565151053e-06,
"loss": 0.21523776054382324,
"step": 3085,
"token_acc": 0.9149363697253852
},
{
"epoch": 0.023573660058667133,
"grad_norm": 1.4195898122543265,
"learning_rate": 2.3573390296002443e-06,
"loss": 0.1943108320236206,
"step": 3090,
"token_acc": 0.9240119256937543
},
{
"epoch": 0.023611805139668214,
"grad_norm": 1.4255428652736284,
"learning_rate": 2.361153494049436e-06,
"loss": 0.19203507900238037,
"step": 3095,
"token_acc": 0.9247487095897854
},
{
"epoch": 0.023649950220669295,
"grad_norm": 1.3597613464161376,
"learning_rate": 2.364967958498627e-06,
"loss": 0.21402199268341066,
"step": 3100,
"token_acc": 0.9176813707619395
},
{
"epoch": 0.023688095301670372,
"grad_norm": 3.4733567078442906,
"learning_rate": 2.368782422947818e-06,
"loss": 0.2028500556945801,
"step": 3105,
"token_acc": 0.9203402938901779
},
{
"epoch": 0.023726240382671453,
"grad_norm": 1.6462836492318036,
"learning_rate": 2.3725968873970096e-06,
"loss": 0.20602803230285643,
"step": 3110,
"token_acc": 0.919665417792106
},
{
"epoch": 0.023764385463672533,
"grad_norm": 2.243924768194205,
"learning_rate": 2.3764113518462008e-06,
"loss": 0.18997081518173217,
"step": 3115,
"token_acc": 0.9257795017156497
},
{
"epoch": 0.02380253054467361,
"grad_norm": 1.3143427208284315,
"learning_rate": 2.3802258162953923e-06,
"loss": 0.20612406730651855,
"step": 3120,
"token_acc": 0.9191098784997359
},
{
"epoch": 0.02384067562567469,
"grad_norm": 2.610116758813615,
"learning_rate": 2.384040280744584e-06,
"loss": 0.20351636409759521,
"step": 3125,
"token_acc": 0.9208366962604251
},
{
"epoch": 0.023878820706675772,
"grad_norm": 1.8220359727412347,
"learning_rate": 2.387854745193775e-06,
"loss": 0.19077625274658203,
"step": 3130,
"token_acc": 0.9252701289647961
},
{
"epoch": 0.02391696578767685,
"grad_norm": 2.1560327293094668,
"learning_rate": 2.3916692096429665e-06,
"loss": 0.20846571922302246,
"step": 3135,
"token_acc": 0.9206925228698873
},
{
"epoch": 0.02395511086867793,
"grad_norm": 1.695867570321,
"learning_rate": 2.3954836740921577e-06,
"loss": 0.19898052215576173,
"step": 3140,
"token_acc": 0.9240560763739571
},
{
"epoch": 0.02399325594967901,
"grad_norm": 1.3466421625417941,
"learning_rate": 2.3992981385413488e-06,
"loss": 0.19682128429412843,
"step": 3145,
"token_acc": 0.9224083438685209
},
{
"epoch": 0.024031401030680088,
"grad_norm": 1.6462002502795419,
"learning_rate": 2.4031126029905403e-06,
"loss": 0.1966496467590332,
"step": 3150,
"token_acc": 0.9246347684934684
},
{
"epoch": 0.02406954611168117,
"grad_norm": 1.5214096430344561,
"learning_rate": 2.4069270674397315e-06,
"loss": 0.1902092218399048,
"step": 3155,
"token_acc": 0.9270326615705351
},
{
"epoch": 0.02410769119268225,
"grad_norm": 3.27704727333946,
"learning_rate": 2.410741531888923e-06,
"loss": 0.18937910795211793,
"step": 3160,
"token_acc": 0.9251128022089029
},
{
"epoch": 0.024145836273683326,
"grad_norm": 3.94438683770782,
"learning_rate": 2.414555996338114e-06,
"loss": 0.2085925579071045,
"step": 3165,
"token_acc": 0.9192110619779603
},
{
"epoch": 0.024183981354684407,
"grad_norm": 1.1639758214771085,
"learning_rate": 2.4183704607873057e-06,
"loss": 0.20401809215545655,
"step": 3170,
"token_acc": 0.921358872425009
},
{
"epoch": 0.024222126435685488,
"grad_norm": 1.396187513528409,
"learning_rate": 2.4221849252364972e-06,
"loss": 0.181234872341156,
"step": 3175,
"token_acc": 0.9256796258403975
},
{
"epoch": 0.024260271516686565,
"grad_norm": 1.6484137736550526,
"learning_rate": 2.4259993896856884e-06,
"loss": 0.1904531717300415,
"step": 3180,
"token_acc": 0.9242170490705075
},
{
"epoch": 0.024298416597687646,
"grad_norm": 1.6070168465829955,
"learning_rate": 2.4298138541348795e-06,
"loss": 0.21354050636291505,
"step": 3185,
"token_acc": 0.9180774138664398
},
{
"epoch": 0.024336561678688726,
"grad_norm": 1.2994161075874269,
"learning_rate": 2.433628318584071e-06,
"loss": 0.22085328102111818,
"step": 3190,
"token_acc": 0.9128524945770065
},
{
"epoch": 0.024374706759689804,
"grad_norm": 1.8277865014342922,
"learning_rate": 2.437442783033262e-06,
"loss": 0.19820194244384765,
"step": 3195,
"token_acc": 0.9210979106923392
},
{
"epoch": 0.024412851840690884,
"grad_norm": 1.197738457095394,
"learning_rate": 2.4412572474824537e-06,
"loss": 0.1977736234664917,
"step": 3200,
"token_acc": 0.9222513985614796
},
{
"epoch": 0.02445099692169196,
"grad_norm": 1.5871368173652445,
"learning_rate": 2.445071711931645e-06,
"loss": 0.20693984031677246,
"step": 3205,
"token_acc": 0.9180889972569338
},
{
"epoch": 0.024489142002693042,
"grad_norm": 2.3236783153908633,
"learning_rate": 2.4488861763808364e-06,
"loss": 0.19624040126800538,
"step": 3210,
"token_acc": 0.9228678929765887
},
{
"epoch": 0.024527287083694123,
"grad_norm": 1.096677899266777,
"learning_rate": 2.452700640830028e-06,
"loss": 0.2062000036239624,
"step": 3215,
"token_acc": 0.9165614145879382
},
{
"epoch": 0.0245654321646952,
"grad_norm": 2.4450043077879027,
"learning_rate": 2.456515105279219e-06,
"loss": 0.17640507221221924,
"step": 3220,
"token_acc": 0.9317507418397626
},
{
"epoch": 0.02460357724569628,
"grad_norm": 2.2625165763316093,
"learning_rate": 2.46032956972841e-06,
"loss": 0.2207273244857788,
"step": 3225,
"token_acc": 0.9162966399363774
},
{
"epoch": 0.02464172232669736,
"grad_norm": 1.5748549677779555,
"learning_rate": 2.4641440341776017e-06,
"loss": 0.21585614681243898,
"step": 3230,
"token_acc": 0.9186153717207275
},
{
"epoch": 0.02467986740769844,
"grad_norm": 1.80505402570258,
"learning_rate": 2.467958498626793e-06,
"loss": 0.21426246166229249,
"step": 3235,
"token_acc": 0.9174377224199288
},
{
"epoch": 0.02471801248869952,
"grad_norm": 1.8419688794266353,
"learning_rate": 2.4717729630759844e-06,
"loss": 0.20132458209991455,
"step": 3240,
"token_acc": 0.9175627240143369
},
{
"epoch": 0.0247561575697006,
"grad_norm": 1.734880588981232,
"learning_rate": 2.4755874275251756e-06,
"loss": 0.20023577213287352,
"step": 3245,
"token_acc": 0.92313328042005
},
{
"epoch": 0.024794302650701677,
"grad_norm": 1.1892617955743958,
"learning_rate": 2.479401891974367e-06,
"loss": 0.20981893539428711,
"step": 3250,
"token_acc": 0.9183657942787584
},
{
"epoch": 0.024832447731702758,
"grad_norm": 1.4891682706604996,
"learning_rate": 2.4832163564235582e-06,
"loss": 0.18011443614959716,
"step": 3255,
"token_acc": 0.9264728039777008
},
{
"epoch": 0.02487059281270384,
"grad_norm": 2.620534540249001,
"learning_rate": 2.4870308208727498e-06,
"loss": 0.19824955463409424,
"step": 3260,
"token_acc": 0.9220367048433835
},
{
"epoch": 0.024908737893704916,
"grad_norm": 1.302379431668048,
"learning_rate": 2.490845285321941e-06,
"loss": 0.18277753591537477,
"step": 3265,
"token_acc": 0.9270991838428552
},
{
"epoch": 0.024946882974705997,
"grad_norm": 1.979624742539282,
"learning_rate": 2.494659749771132e-06,
"loss": 0.20292062759399415,
"step": 3270,
"token_acc": 0.9199508146326468
},
{
"epoch": 0.024985028055707077,
"grad_norm": 2.7821871581456294,
"learning_rate": 2.4984742142203236e-06,
"loss": 0.19786319732666016,
"step": 3275,
"token_acc": 0.9237000753579503
},
{
"epoch": 0.025023173136708154,
"grad_norm": 2.613811755308233,
"learning_rate": 2.502288678669515e-06,
"loss": 0.2162404775619507,
"step": 3280,
"token_acc": 0.9180237751666324
},
{
"epoch": 0.025061318217709235,
"grad_norm": 1.9425003637482738,
"learning_rate": 2.5061031431187067e-06,
"loss": 0.19339216947555543,
"step": 3285,
"token_acc": 0.9244024050447279
},
{
"epoch": 0.025099463298710316,
"grad_norm": 1.4282072187087982,
"learning_rate": 2.509917607567898e-06,
"loss": 0.2097069263458252,
"step": 3290,
"token_acc": 0.9177074422583404
},
{
"epoch": 0.025137608379711393,
"grad_norm": 4.466852558002352,
"learning_rate": 2.513732072017089e-06,
"loss": 0.21624383926391602,
"step": 3295,
"token_acc": 0.9177210277109369
},
{
"epoch": 0.025175753460712474,
"grad_norm": 1.2956082977304848,
"learning_rate": 2.5175465364662805e-06,
"loss": 0.20223746299743653,
"step": 3300,
"token_acc": 0.9227829287281478
},
{
"epoch": 0.025213898541713554,
"grad_norm": 1.6461607405015957,
"learning_rate": 2.5213610009154716e-06,
"loss": 0.1988580584526062,
"step": 3305,
"token_acc": 0.9235935524652339
},
{
"epoch": 0.02525204362271463,
"grad_norm": 1.4458333406851063,
"learning_rate": 2.525175465364663e-06,
"loss": 0.18063056468963623,
"step": 3310,
"token_acc": 0.9278569062603511
},
{
"epoch": 0.025290188703715712,
"grad_norm": 1.3097623495348556,
"learning_rate": 2.5289899298138543e-06,
"loss": 0.20000226497650148,
"step": 3315,
"token_acc": 0.9213914614863339
},
{
"epoch": 0.025328333784716793,
"grad_norm": 0.9392387797184764,
"learning_rate": 2.532804394263046e-06,
"loss": 0.19979712963104249,
"step": 3320,
"token_acc": 0.9209222298692361
},
{
"epoch": 0.02536647886571787,
"grad_norm": 1.5938085586796704,
"learning_rate": 2.536618858712237e-06,
"loss": 0.20666780471801757,
"step": 3325,
"token_acc": 0.9198485580795875
},
{
"epoch": 0.02540462394671895,
"grad_norm": 1.78426978823132,
"learning_rate": 2.540433323161428e-06,
"loss": 0.20396008491516113,
"step": 3330,
"token_acc": 0.92052622094071
},
{
"epoch": 0.02544276902772003,
"grad_norm": 1.889068202724405,
"learning_rate": 2.5442477876106196e-06,
"loss": 0.1926923155784607,
"step": 3335,
"token_acc": 0.924373508353222
},
{
"epoch": 0.02548091410872111,
"grad_norm": 1.4183778245264154,
"learning_rate": 2.5480622520598108e-06,
"loss": 0.1744776487350464,
"step": 3340,
"token_acc": 0.9291473421334284
},
{
"epoch": 0.02551905918972219,
"grad_norm": 1.230090398402074,
"learning_rate": 2.5518767165090023e-06,
"loss": 0.22086596488952637,
"step": 3345,
"token_acc": 0.9121761246224765
},
{
"epoch": 0.02555720427072327,
"grad_norm": 1.141019361828116,
"learning_rate": 2.5556911809581935e-06,
"loss": 0.19518723487854003,
"step": 3350,
"token_acc": 0.9193296640105151
},
{
"epoch": 0.025595349351724347,
"grad_norm": 1.4903963956970512,
"learning_rate": 2.5595056454073854e-06,
"loss": 0.21103744506835936,
"step": 3355,
"token_acc": 0.9164084403128228
},
{
"epoch": 0.025633494432725428,
"grad_norm": 1.7960488417356926,
"learning_rate": 2.563320109856576e-06,
"loss": 0.19511945247650148,
"step": 3360,
"token_acc": 0.9239013107170393
},
{
"epoch": 0.02567163951372651,
"grad_norm": 1.5568121800630565,
"learning_rate": 2.567134574305768e-06,
"loss": 0.20403335094451905,
"step": 3365,
"token_acc": 0.9206216195627333
},
{
"epoch": 0.025709784594727586,
"grad_norm": 2.101844090844681,
"learning_rate": 2.5709490387549592e-06,
"loss": 0.2135997533798218,
"step": 3370,
"token_acc": 0.916172807695611
},
{
"epoch": 0.025747929675728667,
"grad_norm": 2.0526990050519607,
"learning_rate": 2.57476350320415e-06,
"loss": 0.20049848556518554,
"step": 3375,
"token_acc": 0.9205975421296637
},
{
"epoch": 0.025786074756729747,
"grad_norm": 1.2489143353812984,
"learning_rate": 2.578577967653342e-06,
"loss": 0.1879721164703369,
"step": 3380,
"token_acc": 0.9274835189815868
},
{
"epoch": 0.025824219837730825,
"grad_norm": 1.1078867212375774,
"learning_rate": 2.582392432102533e-06,
"loss": 0.18349864482879638,
"step": 3385,
"token_acc": 0.9273663139834144
},
{
"epoch": 0.025862364918731905,
"grad_norm": 1.0985860792253743,
"learning_rate": 2.5862068965517246e-06,
"loss": 0.18642749786376953,
"step": 3390,
"token_acc": 0.9265118500388037
},
{
"epoch": 0.025900509999732986,
"grad_norm": 1.3459947905225353,
"learning_rate": 2.5900213610009157e-06,
"loss": 0.2053903579711914,
"step": 3395,
"token_acc": 0.9212346597248048
},
{
"epoch": 0.025938655080734063,
"grad_norm": 1.1921273755345934,
"learning_rate": 2.5938358254501073e-06,
"loss": 0.1947965979576111,
"step": 3400,
"token_acc": 0.9229361826697893
},
{
"epoch": 0.025976800161735144,
"grad_norm": 2.2914526407679974,
"learning_rate": 2.5976502898992984e-06,
"loss": 0.21972281932830812,
"step": 3405,
"token_acc": 0.9162527593818984
},
{
"epoch": 0.026014945242736225,
"grad_norm": 1.8068309515672767,
"learning_rate": 2.6014647543484895e-06,
"loss": 0.23329036235809325,
"step": 3410,
"token_acc": 0.9087927533968452
},
{
"epoch": 0.026053090323737302,
"grad_norm": 1.8504503763219065,
"learning_rate": 2.605279218797681e-06,
"loss": 0.19596837759017943,
"step": 3415,
"token_acc": 0.92435394177298
},
{
"epoch": 0.026091235404738383,
"grad_norm": 1.8232734385810558,
"learning_rate": 2.609093683246872e-06,
"loss": 0.19747570753097535,
"step": 3420,
"token_acc": 0.9244764847236526
},
{
"epoch": 0.026129380485739463,
"grad_norm": 1.3737563654260403,
"learning_rate": 2.6129081476960637e-06,
"loss": 0.2016169548034668,
"step": 3425,
"token_acc": 0.9190569154527707
},
{
"epoch": 0.02616752556674054,
"grad_norm": 1.1824459258465985,
"learning_rate": 2.616722612145255e-06,
"loss": 0.19958848953247071,
"step": 3430,
"token_acc": 0.9205338186229906
},
{
"epoch": 0.02620567064774162,
"grad_norm": 1.2964613761223518,
"learning_rate": 2.6205370765944464e-06,
"loss": 0.1926640272140503,
"step": 3435,
"token_acc": 0.9220590746115759
},
{
"epoch": 0.0262438157287427,
"grad_norm": 1.4236505920817293,
"learning_rate": 2.6243515410436375e-06,
"loss": 0.19189447164535522,
"step": 3440,
"token_acc": 0.9263172724711186
},
{
"epoch": 0.02628196080974378,
"grad_norm": 2.068754695101069,
"learning_rate": 2.628166005492829e-06,
"loss": 0.19884355068206788,
"step": 3445,
"token_acc": 0.9203148057058534
},
{
"epoch": 0.02632010589074486,
"grad_norm": 1.7624065023039683,
"learning_rate": 2.6319804699420202e-06,
"loss": 0.1670260787010193,
"step": 3450,
"token_acc": 0.936337360112651
},
{
"epoch": 0.026358250971745937,
"grad_norm": 1.7736179776757142,
"learning_rate": 2.6357949343912113e-06,
"loss": 0.207686448097229,
"step": 3455,
"token_acc": 0.9205982115325316
},
{
"epoch": 0.026396396052747018,
"grad_norm": 1.3139041300543912,
"learning_rate": 2.6396093988404033e-06,
"loss": 0.21785390377044678,
"step": 3460,
"token_acc": 0.9146677378629634
},
{
"epoch": 0.0264345411337481,
"grad_norm": 1.4333028167641677,
"learning_rate": 2.643423863289594e-06,
"loss": 0.20164916515350342,
"step": 3465,
"token_acc": 0.9204484783769354
},
{
"epoch": 0.026472686214749176,
"grad_norm": 1.3406003328725824,
"learning_rate": 2.647238327738786e-06,
"loss": 0.18156652450561522,
"step": 3470,
"token_acc": 0.927774471883368
},
{
"epoch": 0.026510831295750256,
"grad_norm": 3.2105844484858896,
"learning_rate": 2.651052792187977e-06,
"loss": 0.20250210762023926,
"step": 3475,
"token_acc": 0.9198271365770062
},
{
"epoch": 0.026548976376751337,
"grad_norm": 1.3886197479602163,
"learning_rate": 2.6548672566371687e-06,
"loss": 0.1934921622276306,
"step": 3480,
"token_acc": 0.923725271663416
},
{
"epoch": 0.026587121457752414,
"grad_norm": 1.5414973728752523,
"learning_rate": 2.65868172108636e-06,
"loss": 0.18831499814987182,
"step": 3485,
"token_acc": 0.9270644576753818
},
{
"epoch": 0.026625266538753495,
"grad_norm": 2.5912066719395694,
"learning_rate": 2.662496185535551e-06,
"loss": 0.21828749179840087,
"step": 3490,
"token_acc": 0.9120089316865536
},
{
"epoch": 0.026663411619754575,
"grad_norm": 3.1360895953541386,
"learning_rate": 2.6663106499847425e-06,
"loss": 0.2132286548614502,
"step": 3495,
"token_acc": 0.9180106702643037
},
{
"epoch": 0.026701556700755653,
"grad_norm": 1.0979789660251023,
"learning_rate": 2.6701251144339336e-06,
"loss": 0.18332486152648925,
"step": 3500,
"token_acc": 0.9251136503255928
},
{
"epoch": 0.026739701781756733,
"grad_norm": 2.1455022936300785,
"learning_rate": 2.673939578883125e-06,
"loss": 0.20165183544158935,
"step": 3505,
"token_acc": 0.923957309366662
},
{
"epoch": 0.026777846862757814,
"grad_norm": 2.588923329543811,
"learning_rate": 2.6777540433323163e-06,
"loss": 0.19635795354843139,
"step": 3510,
"token_acc": 0.9267547544489422
},
{
"epoch": 0.02681599194375889,
"grad_norm": 1.4551640150977048,
"learning_rate": 2.681568507781508e-06,
"loss": 0.20422911643981934,
"step": 3515,
"token_acc": 0.9184940008274721
},
{
"epoch": 0.026854137024759972,
"grad_norm": 1.775168257408535,
"learning_rate": 2.685382972230699e-06,
"loss": 0.205169677734375,
"step": 3520,
"token_acc": 0.9201095072175212
},
{
"epoch": 0.026892282105761053,
"grad_norm": 1.74914388979897,
"learning_rate": 2.6891974366798905e-06,
"loss": 0.21436622142791747,
"step": 3525,
"token_acc": 0.9189360857483129
},
{
"epoch": 0.02693042718676213,
"grad_norm": 2.958239337038096,
"learning_rate": 2.6930119011290816e-06,
"loss": 0.22306289672851562,
"step": 3530,
"token_acc": 0.9199970775188134
},
{
"epoch": 0.02696857226776321,
"grad_norm": 1.160297695323484,
"learning_rate": 2.6968263655782728e-06,
"loss": 0.20923199653625488,
"step": 3535,
"token_acc": 0.9172434547095998
},
{
"epoch": 0.02700671734876429,
"grad_norm": 1.0827037684759806,
"learning_rate": 2.7006408300274643e-06,
"loss": 0.16745710372924805,
"step": 3540,
"token_acc": 0.9324505853003262
},
{
"epoch": 0.02704486242976537,
"grad_norm": 1.5145795323870532,
"learning_rate": 2.7044552944766554e-06,
"loss": 0.19411576986312867,
"step": 3545,
"token_acc": 0.9214267252520031
},
{
"epoch": 0.02708300751076645,
"grad_norm": 2.001414013876113,
"learning_rate": 2.708269758925847e-06,
"loss": 0.20656776428222656,
"step": 3550,
"token_acc": 0.9177293659227772
},
{
"epoch": 0.02712115259176753,
"grad_norm": 1.7419024962232494,
"learning_rate": 2.712084223375038e-06,
"loss": 0.20097751617431642,
"step": 3555,
"token_acc": 0.9209418999838161
},
{
"epoch": 0.027159297672768607,
"grad_norm": 1.4577041816150211,
"learning_rate": 2.71589868782423e-06,
"loss": 0.18280582427978515,
"step": 3560,
"token_acc": 0.9285591913654274
},
{
"epoch": 0.027197442753769688,
"grad_norm": 1.4853218322134096,
"learning_rate": 2.7197131522734212e-06,
"loss": 0.18355511426925658,
"step": 3565,
"token_acc": 0.9286125135808782
},
{
"epoch": 0.02723558783477077,
"grad_norm": 1.2481361015982042,
"learning_rate": 2.723527616722612e-06,
"loss": 0.20470716953277587,
"step": 3570,
"token_acc": 0.9189206975978941
},
{
"epoch": 0.027273732915771846,
"grad_norm": 1.6311460259680775,
"learning_rate": 2.727342081171804e-06,
"loss": 0.20143959522247315,
"step": 3575,
"token_acc": 0.9201599657216311
},
{
"epoch": 0.027311877996772926,
"grad_norm": 1.3545433816329115,
"learning_rate": 2.731156545620995e-06,
"loss": 0.18310431241989136,
"step": 3580,
"token_acc": 0.9275694750771946
},
{
"epoch": 0.027350023077774007,
"grad_norm": 6.862113513604494,
"learning_rate": 2.7349710100701866e-06,
"loss": 0.21209301948547363,
"step": 3585,
"token_acc": 0.9138872823083349
},
{
"epoch": 0.027388168158775084,
"grad_norm": 1.0572018447429878,
"learning_rate": 2.7387854745193777e-06,
"loss": 0.1815540075302124,
"step": 3590,
"token_acc": 0.9257654329757761
},
{
"epoch": 0.027426313239776165,
"grad_norm": 0.7962772529483024,
"learning_rate": 2.7425999389685693e-06,
"loss": 0.19189114570617677,
"step": 3595,
"token_acc": 0.9243278880668083
},
{
"epoch": 0.027464458320777246,
"grad_norm": 1.0633957558676406,
"learning_rate": 2.7464144034177604e-06,
"loss": 0.1664803981781006,
"step": 3600,
"token_acc": 0.9321279364427094
},
{
"epoch": 0.027502603401778323,
"grad_norm": 1.889115399911927,
"learning_rate": 2.750228867866952e-06,
"loss": 0.20531563758850097,
"step": 3605,
"token_acc": 0.9209059541714589
},
{
"epoch": 0.027540748482779404,
"grad_norm": 2.322933860358111,
"learning_rate": 2.754043332316143e-06,
"loss": 0.20830049514770507,
"step": 3610,
"token_acc": 0.9170935899037861
},
{
"epoch": 0.027578893563780484,
"grad_norm": 1.6918445597699396,
"learning_rate": 2.757857796765334e-06,
"loss": 0.20824193954467773,
"step": 3615,
"token_acc": 0.916390838077585
},
{
"epoch": 0.02761703864478156,
"grad_norm": 1.4459218316667777,
"learning_rate": 2.7616722612145257e-06,
"loss": 0.19788497686386108,
"step": 3620,
"token_acc": 0.9232068698528895
},
{
"epoch": 0.027655183725782642,
"grad_norm": 0.9164309158448356,
"learning_rate": 2.765486725663717e-06,
"loss": 0.21459462642669677,
"step": 3625,
"token_acc": 0.9149860378529321
},
{
"epoch": 0.027693328806783723,
"grad_norm": 1.1240369012011204,
"learning_rate": 2.7693011901129084e-06,
"loss": 0.16905201673507692,
"step": 3630,
"token_acc": 0.9317073170731708
},
{
"epoch": 0.0277314738877848,
"grad_norm": 1.5916211991686071,
"learning_rate": 2.7731156545620995e-06,
"loss": 0.21215219497680665,
"step": 3635,
"token_acc": 0.9190266703326918
},
{
"epoch": 0.02776961896878588,
"grad_norm": 1.4125466019773558,
"learning_rate": 2.776930119011291e-06,
"loss": 0.17141737937927246,
"step": 3640,
"token_acc": 0.931858632087465
},
{
"epoch": 0.02780776404978696,
"grad_norm": 1.420878513598927,
"learning_rate": 2.7807445834604822e-06,
"loss": 0.19016761779785157,
"step": 3645,
"token_acc": 0.9247868231178485
},
{
"epoch": 0.02784590913078804,
"grad_norm": 1.9564894593842872,
"learning_rate": 2.7845590479096733e-06,
"loss": 0.1922837018966675,
"step": 3650,
"token_acc": 0.9231418918918919
},
{
"epoch": 0.02788405421178912,
"grad_norm": 1.4032647726476815,
"learning_rate": 2.7883735123588653e-06,
"loss": 0.1849084258079529,
"step": 3655,
"token_acc": 0.9272778127403678
},
{
"epoch": 0.027922199292790197,
"grad_norm": 1.4023537174732466,
"learning_rate": 2.792187976808056e-06,
"loss": 0.17962974309921265,
"step": 3660,
"token_acc": 0.9287585504293407
},
{
"epoch": 0.027960344373791277,
"grad_norm": 1.620132137742645,
"learning_rate": 2.796002441257248e-06,
"loss": 0.18202762603759765,
"step": 3665,
"token_acc": 0.9219573772926413
},
{
"epoch": 0.027998489454792358,
"grad_norm": 1.713948253902535,
"learning_rate": 2.799816905706439e-06,
"loss": 0.1887749671936035,
"step": 3670,
"token_acc": 0.9236799197290857
},
{
"epoch": 0.028036634535793435,
"grad_norm": 1.1755096183945846,
"learning_rate": 2.8036313701556307e-06,
"loss": 0.19311631917953492,
"step": 3675,
"token_acc": 0.9269267235541108
},
{
"epoch": 0.028074779616794516,
"grad_norm": 1.2351916888680017,
"learning_rate": 2.807445834604822e-06,
"loss": 0.20339250564575195,
"step": 3680,
"token_acc": 0.9169700807345259
},
{
"epoch": 0.028112924697795597,
"grad_norm": 1.6348688797486821,
"learning_rate": 2.8112602990540133e-06,
"loss": 0.2240605354309082,
"step": 3685,
"token_acc": 0.914444087403599
},
{
"epoch": 0.028151069778796674,
"grad_norm": 1.3220125589867708,
"learning_rate": 2.8150747635032045e-06,
"loss": 0.18837164640426635,
"step": 3690,
"token_acc": 0.9276987318280235
},
{
"epoch": 0.028189214859797754,
"grad_norm": 1.3031108805525653,
"learning_rate": 2.8188892279523956e-06,
"loss": 0.20531167984008789,
"step": 3695,
"token_acc": 0.9203928279145897
},
{
"epoch": 0.028227359940798835,
"grad_norm": 2.3036385678829894,
"learning_rate": 2.822703692401587e-06,
"loss": 0.1809326171875,
"step": 3700,
"token_acc": 0.9282380791342577
},
{
"epoch": 0.028265505021799912,
"grad_norm": 0.9921800492429008,
"learning_rate": 2.8265181568507783e-06,
"loss": 0.1938246726989746,
"step": 3705,
"token_acc": 0.9251956652618905
},
{
"epoch": 0.028303650102800993,
"grad_norm": 0.9669556560959542,
"learning_rate": 2.83033262129997e-06,
"loss": 0.2065743923187256,
"step": 3710,
"token_acc": 0.9183449285975833
},
{
"epoch": 0.028341795183802074,
"grad_norm": 2.708408755844107,
"learning_rate": 2.834147085749161e-06,
"loss": 0.2093135356903076,
"step": 3715,
"token_acc": 0.9176694745286803
},
{
"epoch": 0.02837994026480315,
"grad_norm": 2.8691192491752053,
"learning_rate": 2.8379615501983525e-06,
"loss": 0.20307013988494874,
"step": 3720,
"token_acc": 0.9224725763415357
},
{
"epoch": 0.02841808534580423,
"grad_norm": 2.5760131026606174,
"learning_rate": 2.8417760146475436e-06,
"loss": 0.20344066619873047,
"step": 3725,
"token_acc": 0.9184155131721631
},
{
"epoch": 0.028456230426805312,
"grad_norm": 1.2801170424352877,
"learning_rate": 2.8455904790967348e-06,
"loss": 0.20083198547363282,
"step": 3730,
"token_acc": 0.9237481650627957
},
{
"epoch": 0.02849437550780639,
"grad_norm": 1.0916836118928501,
"learning_rate": 2.8494049435459263e-06,
"loss": 0.19578756093978883,
"step": 3735,
"token_acc": 0.9209138840070299
},
{
"epoch": 0.02853252058880747,
"grad_norm": 2.0550046310921664,
"learning_rate": 2.8532194079951174e-06,
"loss": 0.24391551017761232,
"step": 3740,
"token_acc": 0.9034128931519072
},
{
"epoch": 0.02857066566980855,
"grad_norm": 3.3510381408825904,
"learning_rate": 2.857033872444309e-06,
"loss": 0.21118056774139404,
"step": 3745,
"token_acc": 0.9214160669130519
},
{
"epoch": 0.028608810750809628,
"grad_norm": 3.792775353793664,
"learning_rate": 2.8608483368935e-06,
"loss": 0.20475456714630128,
"step": 3750,
"token_acc": 0.9231483631297888
},
{
"epoch": 0.02864695583181071,
"grad_norm": 2.2185488625263416,
"learning_rate": 2.864662801342692e-06,
"loss": 0.2053576946258545,
"step": 3755,
"token_acc": 0.9238636363636363
},
{
"epoch": 0.02868510091281179,
"grad_norm": 1.891491051406715,
"learning_rate": 2.8684772657918832e-06,
"loss": 0.19211243391036986,
"step": 3760,
"token_acc": 0.9240767642786146
},
{
"epoch": 0.028723245993812867,
"grad_norm": 1.0633991863613896,
"learning_rate": 2.8722917302410748e-06,
"loss": 0.17564024925231933,
"step": 3765,
"token_acc": 0.9308428310116786
},
{
"epoch": 0.028761391074813947,
"grad_norm": 1.9446934459687018,
"learning_rate": 2.876106194690266e-06,
"loss": 0.22697651386260986,
"step": 3770,
"token_acc": 0.9146782841823057
},
{
"epoch": 0.028799536155815028,
"grad_norm": 1.738474624904149,
"learning_rate": 2.879920659139457e-06,
"loss": 0.1840140700340271,
"step": 3775,
"token_acc": 0.927884991014923
},
{
"epoch": 0.028837681236816105,
"grad_norm": 1.1397327141029683,
"learning_rate": 2.8837351235886486e-06,
"loss": 0.1996319532394409,
"step": 3780,
"token_acc": 0.9207894206887293
},
{
"epoch": 0.028875826317817186,
"grad_norm": 2.0010810592728316,
"learning_rate": 2.8875495880378397e-06,
"loss": 0.21771843433380128,
"step": 3785,
"token_acc": 0.9183735627807812
},
{
"epoch": 0.028913971398818267,
"grad_norm": 0.8589476632325155,
"learning_rate": 2.8913640524870312e-06,
"loss": 0.20254058837890626,
"step": 3790,
"token_acc": 0.9199303742384682
},
{
"epoch": 0.028952116479819344,
"grad_norm": 1.1903116308482833,
"learning_rate": 2.8951785169362224e-06,
"loss": 0.18127316236495972,
"step": 3795,
"token_acc": 0.9276018099547512
},
{
"epoch": 0.028990261560820425,
"grad_norm": 1.3252150677765582,
"learning_rate": 2.898992981385414e-06,
"loss": 0.219500732421875,
"step": 3800,
"token_acc": 0.9147205849436694
},
{
"epoch": 0.029028406641821505,
"grad_norm": 1.429053710025562,
"learning_rate": 2.902807445834605e-06,
"loss": 0.19799134731292725,
"step": 3805,
"token_acc": 0.9229177444386963
},
{
"epoch": 0.029066551722822583,
"grad_norm": 1.6683615340424383,
"learning_rate": 2.906621910283796e-06,
"loss": 0.22626845836639403,
"step": 3810,
"token_acc": 0.911969650046454
},
{
"epoch": 0.029104696803823663,
"grad_norm": 1.8097731739691143,
"learning_rate": 2.9104363747329877e-06,
"loss": 0.18414243459701538,
"step": 3815,
"token_acc": 0.9260662177328844
},
{
"epoch": 0.029142841884824744,
"grad_norm": 1.1750023450553644,
"learning_rate": 2.914250839182179e-06,
"loss": 0.2139209508895874,
"step": 3820,
"token_acc": 0.9180434499804865
},
{
"epoch": 0.02918098696582582,
"grad_norm": 1.3265417401375221,
"learning_rate": 2.9180653036313704e-06,
"loss": 0.19648045301437378,
"step": 3825,
"token_acc": 0.9244593359731953
},
{
"epoch": 0.029219132046826902,
"grad_norm": 4.089263242623879,
"learning_rate": 2.9218797680805615e-06,
"loss": 0.2120136260986328,
"step": 3830,
"token_acc": 0.9149560117302052
},
{
"epoch": 0.029257277127827983,
"grad_norm": 2.9951773248291667,
"learning_rate": 2.925694232529753e-06,
"loss": 0.1790635585784912,
"step": 3835,
"token_acc": 0.9295267489711934
},
{
"epoch": 0.02929542220882906,
"grad_norm": 1.1985757081172763,
"learning_rate": 2.929508696978944e-06,
"loss": 0.18016217947006224,
"step": 3840,
"token_acc": 0.9291431460106159
},
{
"epoch": 0.02933356728983014,
"grad_norm": 3.786594549881491,
"learning_rate": 2.933323161428136e-06,
"loss": 0.2000204563140869,
"step": 3845,
"token_acc": 0.9177666095984979
},
{
"epoch": 0.02937171237083122,
"grad_norm": 1.4358906709929617,
"learning_rate": 2.937137625877327e-06,
"loss": 0.21723394393920897,
"step": 3850,
"token_acc": 0.9156148289275224
},
{
"epoch": 0.0294098574518323,
"grad_norm": 1.6149865226829878,
"learning_rate": 2.940952090326518e-06,
"loss": 0.18792085647583007,
"step": 3855,
"token_acc": 0.9265557163531114
},
{
"epoch": 0.02944800253283338,
"grad_norm": 6.687140858620651,
"learning_rate": 2.94476655477571e-06,
"loss": 0.1787867784500122,
"step": 3860,
"token_acc": 0.9314399036758977
},
{
"epoch": 0.02948614761383446,
"grad_norm": 2.330461552680994,
"learning_rate": 2.948581019224901e-06,
"loss": 0.20301666259765624,
"step": 3865,
"token_acc": 0.9209180146550532
},
{
"epoch": 0.029524292694835537,
"grad_norm": 1.5496342823927245,
"learning_rate": 2.9523954836740927e-06,
"loss": 0.1925230383872986,
"step": 3870,
"token_acc": 0.9241680305510093
},
{
"epoch": 0.029562437775836618,
"grad_norm": 1.113569629202286,
"learning_rate": 2.956209948123284e-06,
"loss": 0.1838328719139099,
"step": 3875,
"token_acc": 0.9333010447657926
},
{
"epoch": 0.029600582856837695,
"grad_norm": 1.1293587320000253,
"learning_rate": 2.9600244125724753e-06,
"loss": 0.19594515562057496,
"step": 3880,
"token_acc": 0.9206130268199234
},
{
"epoch": 0.029638727937838776,
"grad_norm": 1.5085073109583709,
"learning_rate": 2.9638388770216665e-06,
"loss": 0.2004699468612671,
"step": 3885,
"token_acc": 0.9216602342744938
},
{
"epoch": 0.029676873018839856,
"grad_norm": 0.9613468934507051,
"learning_rate": 2.9676533414708576e-06,
"loss": 0.1836915969848633,
"step": 3890,
"token_acc": 0.9260555983474547
},
{
"epoch": 0.029715018099840933,
"grad_norm": 0.9512768251729952,
"learning_rate": 2.971467805920049e-06,
"loss": 0.1740575313568115,
"step": 3895,
"token_acc": 0.9287118701196562
},
{
"epoch": 0.029753163180842014,
"grad_norm": 1.073568534348093,
"learning_rate": 2.9752822703692403e-06,
"loss": 0.17895724773406982,
"step": 3900,
"token_acc": 0.928657637119485
},
{
"epoch": 0.029791308261843095,
"grad_norm": 1.3038297504697949,
"learning_rate": 2.979096734818432e-06,
"loss": 0.20097880363464354,
"step": 3905,
"token_acc": 0.9232682060390763
},
{
"epoch": 0.029829453342844172,
"grad_norm": 3.0321806578855575,
"learning_rate": 2.982911199267623e-06,
"loss": 0.1921984910964966,
"step": 3910,
"token_acc": 0.9280540408561895
},
{
"epoch": 0.029867598423845253,
"grad_norm": 0.944480409021657,
"learning_rate": 2.9867256637168145e-06,
"loss": 0.16717551946640014,
"step": 3915,
"token_acc": 0.9331492802208637
},
{
"epoch": 0.029905743504846333,
"grad_norm": 1.318727008868968,
"learning_rate": 2.9905401281660056e-06,
"loss": 0.1812400221824646,
"step": 3920,
"token_acc": 0.9287547169811321
},
{
"epoch": 0.02994388858584741,
"grad_norm": 1.4483467160016403,
"learning_rate": 2.994354592615197e-06,
"loss": 0.18201427459716796,
"step": 3925,
"token_acc": 0.926509384186327
},
{
"epoch": 0.02998203366684849,
"grad_norm": 1.720297984348613,
"learning_rate": 2.9981690570643883e-06,
"loss": 0.19895753860473633,
"step": 3930,
"token_acc": 0.9236267266302602
},
{
"epoch": 0.030020178747849572,
"grad_norm": 0.9996651095140233,
"learning_rate": 3.0019835215135794e-06,
"loss": 0.18112491369247435,
"step": 3935,
"token_acc": 0.9296546896873542
},
{
"epoch": 0.03005832382885065,
"grad_norm": 1.283101685119387,
"learning_rate": 3.005797985962771e-06,
"loss": 0.1879702925682068,
"step": 3940,
"token_acc": 0.9251506760058641
},
{
"epoch": 0.03009646890985173,
"grad_norm": 1.4474884168184794,
"learning_rate": 3.009612450411962e-06,
"loss": 0.204351806640625,
"step": 3945,
"token_acc": 0.919713461717369
},
{
"epoch": 0.03013461399085281,
"grad_norm": 1.6961839850384683,
"learning_rate": 3.013426914861154e-06,
"loss": 0.21911368370056153,
"step": 3950,
"token_acc": 0.9175461741424802
},
{
"epoch": 0.030172759071853888,
"grad_norm": 1.2100343112925807,
"learning_rate": 3.017241379310345e-06,
"loss": 0.18599618673324586,
"step": 3955,
"token_acc": 0.9262127078114325
},
{
"epoch": 0.03021090415285497,
"grad_norm": 1.3555996447444714,
"learning_rate": 3.0210558437595368e-06,
"loss": 0.19445273876190186,
"step": 3960,
"token_acc": 0.9245685712543448
},
{
"epoch": 0.03024904923385605,
"grad_norm": 5.168158234427423,
"learning_rate": 3.024870308208728e-06,
"loss": 0.18503272533416748,
"step": 3965,
"token_acc": 0.9273064728172609
},
{
"epoch": 0.030287194314857126,
"grad_norm": 1.7180517175963503,
"learning_rate": 3.028684772657919e-06,
"loss": 0.1992701292037964,
"step": 3970,
"token_acc": 0.9214704134476611
},
{
"epoch": 0.030325339395858207,
"grad_norm": 1.5831398403012196,
"learning_rate": 3.0324992371071106e-06,
"loss": 0.22213096618652345,
"step": 3975,
"token_acc": 0.9156822522458796
},
{
"epoch": 0.030363484476859288,
"grad_norm": 1.529404501460422,
"learning_rate": 3.0363137015563017e-06,
"loss": 0.21425409317016603,
"step": 3980,
"token_acc": 0.9196020985214962
},
{
"epoch": 0.030401629557860365,
"grad_norm": 1.7255614148231966,
"learning_rate": 3.0401281660054932e-06,
"loss": 0.1851879835128784,
"step": 3985,
"token_acc": 0.9263255240443896
},
{
"epoch": 0.030439774638861446,
"grad_norm": 1.0099747703647042,
"learning_rate": 3.0439426304546844e-06,
"loss": 0.16479753255844115,
"step": 3990,
"token_acc": 0.9368120991369071
},
{
"epoch": 0.030477919719862526,
"grad_norm": 1.8077632303300455,
"learning_rate": 3.047757094903876e-06,
"loss": 0.191705322265625,
"step": 3995,
"token_acc": 0.9255334181150204
},
{
"epoch": 0.030516064800863604,
"grad_norm": 1.7740221324081489,
"learning_rate": 3.051571559353067e-06,
"loss": 0.19576795101165773,
"step": 4000,
"token_acc": 0.9241032885958715
},
{
"epoch": 0.030554209881864684,
"grad_norm": 1.1053414146409208,
"learning_rate": 3.0553860238022586e-06,
"loss": 0.19408992528915406,
"step": 4005,
"token_acc": 0.9230924505450141
},
{
"epoch": 0.030592354962865765,
"grad_norm": 1.6971402391897856,
"learning_rate": 3.0592004882514497e-06,
"loss": 0.19552085399627686,
"step": 4010,
"token_acc": 0.9229929851909587
},
{
"epoch": 0.030630500043866842,
"grad_norm": 7.675129785952881,
"learning_rate": 3.063014952700641e-06,
"loss": 0.19060148000717164,
"step": 4015,
"token_acc": 0.9252086044553318
},
{
"epoch": 0.030668645124867923,
"grad_norm": 1.52935230884688,
"learning_rate": 3.0668294171498324e-06,
"loss": 0.21632587909698486,
"step": 4020,
"token_acc": 0.9138138328165538
},
{
"epoch": 0.030706790205869004,
"grad_norm": 1.3806489826709554,
"learning_rate": 3.0706438815990235e-06,
"loss": 0.19226115942001343,
"step": 4025,
"token_acc": 0.9239113346958026
},
{
"epoch": 0.03074493528687008,
"grad_norm": 1.3555410716452954,
"learning_rate": 3.074458346048215e-06,
"loss": 0.19386298656463624,
"step": 4030,
"token_acc": 0.9222626315105857
},
{
"epoch": 0.03078308036787116,
"grad_norm": 1.828040146371655,
"learning_rate": 3.078272810497406e-06,
"loss": 0.2037498712539673,
"step": 4035,
"token_acc": 0.9204927211646137
},
{
"epoch": 0.030821225448872242,
"grad_norm": 1.3175080584510894,
"learning_rate": 3.082087274946598e-06,
"loss": 0.20440731048583985,
"step": 4040,
"token_acc": 0.9201301048066498
},
{
"epoch": 0.03085937052987332,
"grad_norm": 1.1777901232555745,
"learning_rate": 3.085901739395789e-06,
"loss": 0.20805349349975585,
"step": 4045,
"token_acc": 0.9202988023212743
},
{
"epoch": 0.0308975156108744,
"grad_norm": 1.237596817863281,
"learning_rate": 3.08971620384498e-06,
"loss": 0.21010746955871581,
"step": 4050,
"token_acc": 0.9170735611942687
},
{
"epoch": 0.03093566069187548,
"grad_norm": 1.1860394506152796,
"learning_rate": 3.093530668294172e-06,
"loss": 0.18669127225875853,
"step": 4055,
"token_acc": 0.9262059485128717
},
{
"epoch": 0.030973805772876558,
"grad_norm": 1.3889878598720948,
"learning_rate": 3.097345132743363e-06,
"loss": 0.17644023895263672,
"step": 4060,
"token_acc": 0.9319327181098239
},
{
"epoch": 0.03101195085387764,
"grad_norm": 1.5567893298709274,
"learning_rate": 3.1011595971925547e-06,
"loss": 0.19679009914398193,
"step": 4065,
"token_acc": 0.923408071748879
},
{
"epoch": 0.03105009593487872,
"grad_norm": 1.0713063581206868,
"learning_rate": 3.1049740616417458e-06,
"loss": 0.1980377435684204,
"step": 4070,
"token_acc": 0.9227026275545669
},
{
"epoch": 0.031088241015879797,
"grad_norm": 0.8997825449921926,
"learning_rate": 3.1087885260909373e-06,
"loss": 0.21196327209472657,
"step": 4075,
"token_acc": 0.9173516854648569
},
{
"epoch": 0.031126386096880877,
"grad_norm": 1.399487799217613,
"learning_rate": 3.1126029905401285e-06,
"loss": 0.18792858123779296,
"step": 4080,
"token_acc": 0.9250527797325827
},
{
"epoch": 0.031164531177881958,
"grad_norm": 1.424376567974957,
"learning_rate": 3.11641745498932e-06,
"loss": 0.20289034843444825,
"step": 4085,
"token_acc": 0.9218360655737705
},
{
"epoch": 0.031202676258883035,
"grad_norm": 2.2151217627106705,
"learning_rate": 3.120231919438511e-06,
"loss": 0.2270430564880371,
"step": 4090,
"token_acc": 0.9116272677295217
},
{
"epoch": 0.031240821339884116,
"grad_norm": 2.135710489213465,
"learning_rate": 3.1240463838877023e-06,
"loss": 0.15556539297103883,
"step": 4095,
"token_acc": 0.93789288408348
},
{
"epoch": 0.03127896642088519,
"grad_norm": 2.4562238568130246,
"learning_rate": 3.127860848336894e-06,
"loss": 0.18782906532287597,
"step": 4100,
"token_acc": 0.9251013491061341
},
{
"epoch": 0.03131711150188628,
"grad_norm": 1.241750902898804,
"learning_rate": 3.131675312786085e-06,
"loss": 0.20374877452850343,
"step": 4105,
"token_acc": 0.9193150944588362
},
{
"epoch": 0.031355256582887354,
"grad_norm": 1.4124623077434573,
"learning_rate": 3.1354897772352765e-06,
"loss": 0.14972851276397706,
"step": 4110,
"token_acc": 0.9433177327093084
},
{
"epoch": 0.03139340166388843,
"grad_norm": 1.365258153193531,
"learning_rate": 3.1393042416844676e-06,
"loss": 0.19449151754379274,
"step": 4115,
"token_acc": 0.9252552648372687
},
{
"epoch": 0.031431546744889516,
"grad_norm": 1.2297770812385165,
"learning_rate": 3.143118706133659e-06,
"loss": 0.19626073837280272,
"step": 4120,
"token_acc": 0.9240330425384081
},
{
"epoch": 0.03146969182589059,
"grad_norm": 0.9688977073020094,
"learning_rate": 3.1469331705828503e-06,
"loss": 0.18966922760009766,
"step": 4125,
"token_acc": 0.924347061778001
},
{
"epoch": 0.03150783690689167,
"grad_norm": 1.233316461449362,
"learning_rate": 3.1507476350320414e-06,
"loss": 0.16788265705108643,
"step": 4130,
"token_acc": 0.934992490881785
},
{
"epoch": 0.031545981987892754,
"grad_norm": 1.2746413239950252,
"learning_rate": 3.154562099481233e-06,
"loss": 0.17925671339035035,
"step": 4135,
"token_acc": 0.929021746830063
},
{
"epoch": 0.03158412706889383,
"grad_norm": 1.3950047985818248,
"learning_rate": 3.158376563930424e-06,
"loss": 0.22052040100097656,
"step": 4140,
"token_acc": 0.9110111480539801
},
{
"epoch": 0.03162227214989491,
"grad_norm": 1.5514933575147345,
"learning_rate": 3.162191028379616e-06,
"loss": 0.21991984844207763,
"step": 4145,
"token_acc": 0.9153456340956341
},
{
"epoch": 0.03166041723089599,
"grad_norm": 1.3882376611341833,
"learning_rate": 3.1660054928288068e-06,
"loss": 0.18932245969772338,
"step": 4150,
"token_acc": 0.9287331048308867
},
{
"epoch": 0.03169856231189707,
"grad_norm": 1.229650208018774,
"learning_rate": 3.1698199572779987e-06,
"loss": 0.18681371212005615,
"step": 4155,
"token_acc": 0.9264337590783872
},
{
"epoch": 0.03173670739289815,
"grad_norm": 1.3370060505988737,
"learning_rate": 3.17363442172719e-06,
"loss": 0.19819024801254273,
"step": 4160,
"token_acc": 0.9221357063403782
},
{
"epoch": 0.03177485247389923,
"grad_norm": 1.907061793648975,
"learning_rate": 3.1774488861763814e-06,
"loss": 0.18505465984344482,
"step": 4165,
"token_acc": 0.9267986614612381
},
{
"epoch": 0.03181299755490031,
"grad_norm": 1.231176856586619,
"learning_rate": 3.1812633506255726e-06,
"loss": 0.19385309219360353,
"step": 4170,
"token_acc": 0.9215822784810127
},
{
"epoch": 0.031851142635901386,
"grad_norm": 1.378763020765013,
"learning_rate": 3.1850778150747637e-06,
"loss": 0.19081203937530516,
"step": 4175,
"token_acc": 0.9240742241309456
},
{
"epoch": 0.03188928771690247,
"grad_norm": 3.3504358388227797,
"learning_rate": 3.1888922795239552e-06,
"loss": 0.19464006423950195,
"step": 4180,
"token_acc": 0.9258165246012322
},
{
"epoch": 0.03192743279790355,
"grad_norm": 1.247515523105034,
"learning_rate": 3.1927067439731464e-06,
"loss": 0.17227525711059571,
"step": 4185,
"token_acc": 0.9312586022327572
},
{
"epoch": 0.031965577878904625,
"grad_norm": 6.095942024898531,
"learning_rate": 3.196521208422338e-06,
"loss": 0.2009768009185791,
"step": 4190,
"token_acc": 0.9168563192231831
},
{
"epoch": 0.03200372295990571,
"grad_norm": 1.5155119429442323,
"learning_rate": 3.200335672871529e-06,
"loss": 0.18182271718978882,
"step": 4195,
"token_acc": 0.930840400925212
},
{
"epoch": 0.032041868040906786,
"grad_norm": 1.4827775531997158,
"learning_rate": 3.2041501373207206e-06,
"loss": 0.235754656791687,
"step": 4200,
"token_acc": 0.9086977089950665
},
{
"epoch": 0.03208001312190786,
"grad_norm": 1.301149022280892,
"learning_rate": 3.2079646017699117e-06,
"loss": 0.1856813907623291,
"step": 4205,
"token_acc": 0.9264301773669747
},
{
"epoch": 0.03211815820290894,
"grad_norm": 1.5208849535594273,
"learning_rate": 3.211779066219103e-06,
"loss": 0.21679787635803222,
"step": 4210,
"token_acc": 0.9135308246597278
},
{
"epoch": 0.032156303283910025,
"grad_norm": 1.9604492069483286,
"learning_rate": 3.2155935306682944e-06,
"loss": 0.19855906963348388,
"step": 4215,
"token_acc": 0.9217003985309057
},
{
"epoch": 0.0321944483649111,
"grad_norm": 1.3406838456600245,
"learning_rate": 3.2194079951174855e-06,
"loss": 0.17857465744018555,
"step": 4220,
"token_acc": 0.9279414646893578
},
{
"epoch": 0.03223259344591218,
"grad_norm": 1.1323856377548835,
"learning_rate": 3.223222459566677e-06,
"loss": 0.19370107650756835,
"step": 4225,
"token_acc": 0.9249237029501526
},
{
"epoch": 0.03227073852691326,
"grad_norm": 1.404063556239358,
"learning_rate": 3.227036924015868e-06,
"loss": 0.1866621732711792,
"step": 4230,
"token_acc": 0.9295297805642633
},
{
"epoch": 0.03230888360791434,
"grad_norm": 1.893220440556427,
"learning_rate": 3.23085138846506e-06,
"loss": 0.1865388870239258,
"step": 4235,
"token_acc": 0.9260938562684199
},
{
"epoch": 0.03234702868891542,
"grad_norm": 1.5395084257597322,
"learning_rate": 3.234665852914251e-06,
"loss": 0.20243444442749023,
"step": 4240,
"token_acc": 0.9196350707721205
},
{
"epoch": 0.0323851737699165,
"grad_norm": 1.5559146134257837,
"learning_rate": 3.238480317363443e-06,
"loss": 0.21359536647796631,
"step": 4245,
"token_acc": 0.9184399907142304
},
{
"epoch": 0.03242331885091758,
"grad_norm": 1.6064268325635596,
"learning_rate": 3.242294781812634e-06,
"loss": 0.14891519546508789,
"step": 4250,
"token_acc": 0.937641894959891
},
{
"epoch": 0.032461463931918656,
"grad_norm": 1.5787382687684244,
"learning_rate": 3.246109246261825e-06,
"loss": 0.15965185165405274,
"step": 4255,
"token_acc": 0.9405324707638716
},
{
"epoch": 0.03249960901291974,
"grad_norm": 1.3795333808011254,
"learning_rate": 3.2499237107110166e-06,
"loss": 0.1852767825126648,
"step": 4260,
"token_acc": 0.9294036148372063
},
{
"epoch": 0.03253775409392082,
"grad_norm": 1.4895657996899223,
"learning_rate": 3.2537381751602078e-06,
"loss": 0.192819344997406,
"step": 4265,
"token_acc": 0.9245401666404653
},
{
"epoch": 0.032575899174921895,
"grad_norm": 1.2328522769477055,
"learning_rate": 3.2575526396093993e-06,
"loss": 0.18221015930175782,
"step": 4270,
"token_acc": 0.9267710754924752
},
{
"epoch": 0.03261404425592298,
"grad_norm": 1.209682864509687,
"learning_rate": 3.2613671040585905e-06,
"loss": 0.2271733522415161,
"step": 4275,
"token_acc": 0.9130706691682301
},
{
"epoch": 0.032652189336924056,
"grad_norm": 1.2913186177046156,
"learning_rate": 3.265181568507782e-06,
"loss": 0.2268014907836914,
"step": 4280,
"token_acc": 0.9100142551674982
},
{
"epoch": 0.03269033441792513,
"grad_norm": 1.5394196674138498,
"learning_rate": 3.268996032956973e-06,
"loss": 0.21484310626983644,
"step": 4285,
"token_acc": 0.9175599181018909
},
{
"epoch": 0.03272847949892622,
"grad_norm": 1.0167276423058904,
"learning_rate": 3.2728104974061643e-06,
"loss": 0.19501055479049684,
"step": 4290,
"token_acc": 0.9230512546716497
},
{
"epoch": 0.032766624579927295,
"grad_norm": 1.9778914839599584,
"learning_rate": 3.276624961855356e-06,
"loss": 0.20547893047332763,
"step": 4295,
"token_acc": 0.9216037368625924
},
{
"epoch": 0.03280476966092837,
"grad_norm": 1.2914478897753081,
"learning_rate": 3.280439426304547e-06,
"loss": 0.20135602951049805,
"step": 4300,
"token_acc": 0.9194859487497097
},
{
"epoch": 0.032842914741929456,
"grad_norm": 1.964194073243358,
"learning_rate": 3.2842538907537385e-06,
"loss": 0.19254244565963746,
"step": 4305,
"token_acc": 0.9256360078277887
},
{
"epoch": 0.03288105982293053,
"grad_norm": 1.067852120624141,
"learning_rate": 3.2880683552029296e-06,
"loss": 0.16231229305267333,
"step": 4310,
"token_acc": 0.9347159457714654
},
{
"epoch": 0.03291920490393161,
"grad_norm": 3.115935552311204,
"learning_rate": 3.291882819652121e-06,
"loss": 0.1989269733428955,
"step": 4315,
"token_acc": 0.9199965178027335
},
{
"epoch": 0.032957349984932695,
"grad_norm": 1.3261319959179994,
"learning_rate": 3.2956972841013123e-06,
"loss": 0.203765344619751,
"step": 4320,
"token_acc": 0.9218821729150727
},
{
"epoch": 0.03299549506593377,
"grad_norm": 1.3166279930548022,
"learning_rate": 3.299511748550504e-06,
"loss": 0.19871768951416016,
"step": 4325,
"token_acc": 0.9229404309252218
},
{
"epoch": 0.03303364014693485,
"grad_norm": 1.3379477521261212,
"learning_rate": 3.303326212999695e-06,
"loss": 0.19905929565429686,
"step": 4330,
"token_acc": 0.9207920792079208
},
{
"epoch": 0.03307178522793593,
"grad_norm": 1.7007595374435627,
"learning_rate": 3.307140677448886e-06,
"loss": 0.22140722274780272,
"step": 4335,
"token_acc": 0.9125604274446611
},
{
"epoch": 0.03310993030893701,
"grad_norm": 1.490401750451036,
"learning_rate": 3.310955141898078e-06,
"loss": 0.20570905208587648,
"step": 4340,
"token_acc": 0.9196816208393632
},
{
"epoch": 0.03314807538993809,
"grad_norm": 1.6845288207188736,
"learning_rate": 3.3147696063472688e-06,
"loss": 0.20891788005828857,
"step": 4345,
"token_acc": 0.9194862710363153
},
{
"epoch": 0.03318622047093917,
"grad_norm": 1.453340122491242,
"learning_rate": 3.3185840707964607e-06,
"loss": 0.16959222555160522,
"step": 4350,
"token_acc": 0.9353018474143928
},
{
"epoch": 0.03322436555194025,
"grad_norm": 1.3988136413660832,
"learning_rate": 3.322398535245652e-06,
"loss": 0.21306605339050294,
"step": 4355,
"token_acc": 0.9172180214240706
},
{
"epoch": 0.033262510632941326,
"grad_norm": 1.1651218783256203,
"learning_rate": 3.3262129996948434e-06,
"loss": 0.20567469596862792,
"step": 4360,
"token_acc": 0.9187889077532541
},
{
"epoch": 0.03330065571394241,
"grad_norm": 1.1182236086031767,
"learning_rate": 3.3300274641440345e-06,
"loss": 0.20992355346679686,
"step": 4365,
"token_acc": 0.9190948942753997
},
{
"epoch": 0.03333880079494349,
"grad_norm": 1.1419586361477712,
"learning_rate": 3.3338419285932257e-06,
"loss": 0.2016165018081665,
"step": 4370,
"token_acc": 0.9203829841922632
},
{
"epoch": 0.033376945875944565,
"grad_norm": 1.2287414874914733,
"learning_rate": 3.3376563930424172e-06,
"loss": 0.18167028427124024,
"step": 4375,
"token_acc": 0.924721885987665
},
{
"epoch": 0.03341509095694565,
"grad_norm": 1.5130461084813085,
"learning_rate": 3.3414708574916083e-06,
"loss": 0.1922617793083191,
"step": 4380,
"token_acc": 0.9219984705582462
},
{
"epoch": 0.033453236037946726,
"grad_norm": 3.298724048809877,
"learning_rate": 3.3452853219408e-06,
"loss": 0.20182492733001708,
"step": 4385,
"token_acc": 0.9179416809605488
},
{
"epoch": 0.033491381118947804,
"grad_norm": 1.3836622590023513,
"learning_rate": 3.349099786389991e-06,
"loss": 0.1804134726524353,
"step": 4390,
"token_acc": 0.9295729508853458
},
{
"epoch": 0.03352952619994889,
"grad_norm": 1.1096692149699199,
"learning_rate": 3.3529142508391826e-06,
"loss": 0.19002759456634521,
"step": 4395,
"token_acc": 0.925955356497869
},
{
"epoch": 0.033567671280949965,
"grad_norm": 1.0956634553013571,
"learning_rate": 3.3567287152883737e-06,
"loss": 0.19112770557403563,
"step": 4400,
"token_acc": 0.925659146507203
},
{
"epoch": 0.03360581636195104,
"grad_norm": 1.883821846083063,
"learning_rate": 3.3605431797375653e-06,
"loss": 0.20746521949768065,
"step": 4405,
"token_acc": 0.9208971902937421
},
{
"epoch": 0.033643961442952126,
"grad_norm": 2.3966086942468414,
"learning_rate": 3.3643576441867564e-06,
"loss": 0.1930229902267456,
"step": 4410,
"token_acc": 0.9248863101771994
},
{
"epoch": 0.033682106523953204,
"grad_norm": 1.9764005425535567,
"learning_rate": 3.3681721086359475e-06,
"loss": 0.2252532958984375,
"step": 4415,
"token_acc": 0.9115563780231
},
{
"epoch": 0.03372025160495428,
"grad_norm": 1.305072579245311,
"learning_rate": 3.371986573085139e-06,
"loss": 0.2133772611618042,
"step": 4420,
"token_acc": 0.9161680717976611
},
{
"epoch": 0.033758396685955365,
"grad_norm": 0.8926723787469454,
"learning_rate": 3.37580103753433e-06,
"loss": 0.2033097267150879,
"step": 4425,
"token_acc": 0.92083940542116
},
{
"epoch": 0.03379654176695644,
"grad_norm": 1.4008706311509471,
"learning_rate": 3.379615501983522e-06,
"loss": 0.21097609996795655,
"step": 4430,
"token_acc": 0.9180301312165884
},
{
"epoch": 0.03383468684795752,
"grad_norm": 4.295872333834146,
"learning_rate": 3.383429966432713e-06,
"loss": 0.21316566467285156,
"step": 4435,
"token_acc": 0.9184354673057172
},
{
"epoch": 0.033872831928958604,
"grad_norm": 1.2232520943527658,
"learning_rate": 3.387244430881905e-06,
"loss": 0.18360961675643922,
"step": 4440,
"token_acc": 0.929012923768496
},
{
"epoch": 0.03391097700995968,
"grad_norm": 1.8738336262848434,
"learning_rate": 3.391058895331096e-06,
"loss": 0.19892446994781493,
"step": 4445,
"token_acc": 0.9220344144967261
},
{
"epoch": 0.03394912209096076,
"grad_norm": 1.1677837083340437,
"learning_rate": 3.3948733597802867e-06,
"loss": 0.211161470413208,
"step": 4450,
"token_acc": 0.9177242572829536
},
{
"epoch": 0.03398726717196184,
"grad_norm": 2.5660853936636387,
"learning_rate": 3.3986878242294786e-06,
"loss": 0.18049848079681396,
"step": 4455,
"token_acc": 0.9283002588438309
},
{
"epoch": 0.03402541225296292,
"grad_norm": 1.3454857444378412,
"learning_rate": 3.4025022886786698e-06,
"loss": 0.18342503309249877,
"step": 4460,
"token_acc": 0.9298885900883596
},
{
"epoch": 0.034063557333964,
"grad_norm": 1.0265617861666363,
"learning_rate": 3.4063167531278613e-06,
"loss": 0.1845909833908081,
"step": 4465,
"token_acc": 0.9273385893884509
},
{
"epoch": 0.03410170241496508,
"grad_norm": 1.2769753645467474,
"learning_rate": 3.4101312175770524e-06,
"loss": 0.18961447477340698,
"step": 4470,
"token_acc": 0.925330696006923
},
{
"epoch": 0.03413984749596616,
"grad_norm": 1.197029649136069,
"learning_rate": 3.413945682026244e-06,
"loss": 0.1997068405151367,
"step": 4475,
"token_acc": 0.9214889157085688
},
{
"epoch": 0.034177992576967235,
"grad_norm": 0.9105631720861997,
"learning_rate": 3.417760146475435e-06,
"loss": 0.20405330657958984,
"step": 4480,
"token_acc": 0.9211218836565097
},
{
"epoch": 0.03421613765796832,
"grad_norm": 1.0326170607910674,
"learning_rate": 3.4215746109246267e-06,
"loss": 0.2117034912109375,
"step": 4485,
"token_acc": 0.9175376575468798
},
{
"epoch": 0.0342542827389694,
"grad_norm": 2.133917541486272,
"learning_rate": 3.425389075373818e-06,
"loss": 0.20000073909759522,
"step": 4490,
"token_acc": 0.9216012788916272
},
{
"epoch": 0.034292427819970474,
"grad_norm": 0.9803957220883355,
"learning_rate": 3.429203539823009e-06,
"loss": 0.20116126537322998,
"step": 4495,
"token_acc": 0.9222745625841184
},
{
"epoch": 0.03433057290097156,
"grad_norm": 1.3788173542900473,
"learning_rate": 3.4330180042722005e-06,
"loss": 0.20180737972259521,
"step": 4500,
"token_acc": 0.921762038965813
},
{
"epoch": 0.034368717981972635,
"grad_norm": 1.4877574944005867,
"learning_rate": 3.4368324687213916e-06,
"loss": 0.167411208152771,
"step": 4505,
"token_acc": 0.9326003021655196
},
{
"epoch": 0.03440686306297371,
"grad_norm": 1.0153344676125826,
"learning_rate": 3.440646933170583e-06,
"loss": 0.2086881160736084,
"step": 4510,
"token_acc": 0.9179098394065379
},
{
"epoch": 0.0344450081439748,
"grad_norm": 1.062372351307854,
"learning_rate": 3.4444613976197743e-06,
"loss": 0.17328981161117554,
"step": 4515,
"token_acc": 0.9288730694980695
},
{
"epoch": 0.034483153224975874,
"grad_norm": 1.6466776021823493,
"learning_rate": 3.448275862068966e-06,
"loss": 0.18375781774520875,
"step": 4520,
"token_acc": 0.9334014151287475
},
{
"epoch": 0.03452129830597695,
"grad_norm": 13.58832608656192,
"learning_rate": 3.452090326518157e-06,
"loss": 0.20321755409240722,
"step": 4525,
"token_acc": 0.9197571989044341
},
{
"epoch": 0.034559443386978035,
"grad_norm": 0.9907633925648729,
"learning_rate": 3.455904790967348e-06,
"loss": 0.16616424322128295,
"step": 4530,
"token_acc": 0.9342551854179761
},
{
"epoch": 0.03459758846797911,
"grad_norm": 1.5319091662195916,
"learning_rate": 3.45971925541654e-06,
"loss": 0.19910019636154175,
"step": 4535,
"token_acc": 0.9225499151768237
},
{
"epoch": 0.03463573354898019,
"grad_norm": 1.0840269899748796,
"learning_rate": 3.4635337198657308e-06,
"loss": 0.1993529677391052,
"step": 4540,
"token_acc": 0.920117762512267
},
{
"epoch": 0.034673878629981274,
"grad_norm": 0.783136484221215,
"learning_rate": 3.4673481843149227e-06,
"loss": 0.20680639743804932,
"step": 4545,
"token_acc": 0.916354556803995
},
{
"epoch": 0.03471202371098235,
"grad_norm": 0.9996457251855007,
"learning_rate": 3.471162648764114e-06,
"loss": 0.18114088773727416,
"step": 4550,
"token_acc": 0.9310844844692056
},
{
"epoch": 0.03475016879198343,
"grad_norm": 1.1171570162245272,
"learning_rate": 3.4749771132133054e-06,
"loss": 0.20022878646850586,
"step": 4555,
"token_acc": 0.9233644859813084
},
{
"epoch": 0.03478831387298451,
"grad_norm": 1.7912516285866034,
"learning_rate": 3.4787915776624965e-06,
"loss": 0.2127835750579834,
"step": 4560,
"token_acc": 0.9194244604316547
},
{
"epoch": 0.03482645895398559,
"grad_norm": 2.524797243253283,
"learning_rate": 3.482606042111688e-06,
"loss": 0.17914170026779175,
"step": 4565,
"token_acc": 0.9318651066758431
},
{
"epoch": 0.03486460403498667,
"grad_norm": 0.874483882870503,
"learning_rate": 3.4864205065608792e-06,
"loss": 0.18493621349334716,
"step": 4570,
"token_acc": 0.9272653861683106
},
{
"epoch": 0.03490274911598775,
"grad_norm": 0.8288145751330266,
"learning_rate": 3.4902349710100703e-06,
"loss": 0.17861564159393312,
"step": 4575,
"token_acc": 0.9307804334789685
},
{
"epoch": 0.03494089419698883,
"grad_norm": 1.4567538583956594,
"learning_rate": 3.494049435459262e-06,
"loss": 0.1920659899711609,
"step": 4580,
"token_acc": 0.9228076874124984
},
{
"epoch": 0.034979039277989905,
"grad_norm": 2.303644502890191,
"learning_rate": 3.497863899908453e-06,
"loss": 0.21063036918640138,
"step": 4585,
"token_acc": 0.914778167890201
},
{
"epoch": 0.03501718435899099,
"grad_norm": 1.5757141326980766,
"learning_rate": 3.5016783643576446e-06,
"loss": 0.21858203411102295,
"step": 4590,
"token_acc": 0.9161388949552304
},
{
"epoch": 0.03505532943999207,
"grad_norm": 1.605844786815046,
"learning_rate": 3.5054928288068357e-06,
"loss": 0.19421610832214356,
"step": 4595,
"token_acc": 0.9229766147961401
},
{
"epoch": 0.035093474520993144,
"grad_norm": 1.1959483398538873,
"learning_rate": 3.5093072932560272e-06,
"loss": 0.23491010665893555,
"step": 4600,
"token_acc": 0.9084976694155611
},
{
"epoch": 0.03513161960199423,
"grad_norm": 1.2204233228422685,
"learning_rate": 3.5131217577052184e-06,
"loss": 0.16732492446899414,
"step": 4605,
"token_acc": 0.9352314714192567
},
{
"epoch": 0.035169764682995305,
"grad_norm": 1.0486376328895832,
"learning_rate": 3.5169362221544095e-06,
"loss": 0.18333892822265624,
"step": 4610,
"token_acc": 0.9249541159676432
},
{
"epoch": 0.03520790976399638,
"grad_norm": 1.4386067614200613,
"learning_rate": 3.520750686603601e-06,
"loss": 0.20781965255737306,
"step": 4615,
"token_acc": 0.9179694799127998
},
{
"epoch": 0.03524605484499747,
"grad_norm": 1.7227179028237356,
"learning_rate": 3.524565151052792e-06,
"loss": 0.20078675746917723,
"step": 4620,
"token_acc": 0.9206438631790744
},
{
"epoch": 0.035284199925998544,
"grad_norm": 1.0681920586361395,
"learning_rate": 3.5283796155019837e-06,
"loss": 0.17395143508911132,
"step": 4625,
"token_acc": 0.9317450863609291
},
{
"epoch": 0.03532234500699962,
"grad_norm": 1.1571162640225507,
"learning_rate": 3.532194079951175e-06,
"loss": 0.18423476219177246,
"step": 4630,
"token_acc": 0.9282626317211564
},
{
"epoch": 0.035360490088000705,
"grad_norm": 1.579647197005868,
"learning_rate": 3.536008544400367e-06,
"loss": 0.18670377731323243,
"step": 4635,
"token_acc": 0.9276224631184286
},
{
"epoch": 0.03539863516900178,
"grad_norm": 1.3245342006826428,
"learning_rate": 3.539823008849558e-06,
"loss": 0.1820981502532959,
"step": 4640,
"token_acc": 0.9261430246189918
},
{
"epoch": 0.03543678025000286,
"grad_norm": 1.6841284741245828,
"learning_rate": 3.5436374732987495e-06,
"loss": 0.21025495529174804,
"step": 4645,
"token_acc": 0.9165961375702085
},
{
"epoch": 0.03547492533100394,
"grad_norm": 1.9311786430845685,
"learning_rate": 3.5474519377479406e-06,
"loss": 0.19530566930770873,
"step": 4650,
"token_acc": 0.9222356313169408
},
{
"epoch": 0.03551307041200502,
"grad_norm": 1.135227987674949,
"learning_rate": 3.5512664021971318e-06,
"loss": 0.1924229860305786,
"step": 4655,
"token_acc": 0.9220117570215546
},
{
"epoch": 0.0355512154930061,
"grad_norm": 1.1084152410582266,
"learning_rate": 3.5550808666463233e-06,
"loss": 0.18804261684417725,
"step": 4660,
"token_acc": 0.9246747139050008
},
{
"epoch": 0.035589360574007176,
"grad_norm": 1.5971821612427188,
"learning_rate": 3.5588953310955144e-06,
"loss": 0.19889142513275146,
"step": 4665,
"token_acc": 0.9225494150867285
},
{
"epoch": 0.03562750565500826,
"grad_norm": 0.7555037043168267,
"learning_rate": 3.562709795544706e-06,
"loss": 0.1885276198387146,
"step": 4670,
"token_acc": 0.9278821074595722
},
{
"epoch": 0.03566565073600934,
"grad_norm": 1.2619697595724308,
"learning_rate": 3.566524259993897e-06,
"loss": 0.2120218276977539,
"step": 4675,
"token_acc": 0.9156688333903524
},
{
"epoch": 0.035703795817010414,
"grad_norm": 1.3361577352288174,
"learning_rate": 3.5703387244430887e-06,
"loss": 0.20398688316345215,
"step": 4680,
"token_acc": 0.9225897841846513
},
{
"epoch": 0.0357419408980115,
"grad_norm": 1.3862786036410368,
"learning_rate": 3.57415318889228e-06,
"loss": 0.18577802181243896,
"step": 4685,
"token_acc": 0.9271190146490781
},
{
"epoch": 0.035780085979012576,
"grad_norm": 1.19723430476584,
"learning_rate": 3.577967653341471e-06,
"loss": 0.2018489122390747,
"step": 4690,
"token_acc": 0.9190664092015783
},
{
"epoch": 0.03581823106001365,
"grad_norm": 1.275837221866477,
"learning_rate": 3.5817821177906625e-06,
"loss": 0.18619383573532106,
"step": 4695,
"token_acc": 0.9308565531475749
},
{
"epoch": 0.03585637614101474,
"grad_norm": 1.1433879546033165,
"learning_rate": 3.5855965822398536e-06,
"loss": 0.20321109294891357,
"step": 4700,
"token_acc": 0.920124021851469
},
{
"epoch": 0.035894521222015814,
"grad_norm": 1.2237784549589745,
"learning_rate": 3.589411046689045e-06,
"loss": 0.1738419532775879,
"step": 4705,
"token_acc": 0.930321896939992
},
{
"epoch": 0.03593266630301689,
"grad_norm": 1.2283255648419102,
"learning_rate": 3.5932255111382363e-06,
"loss": 0.2010645866394043,
"step": 4710,
"token_acc": 0.921566885770119
},
{
"epoch": 0.035970811384017976,
"grad_norm": 1.6016903630531252,
"learning_rate": 3.597039975587428e-06,
"loss": 0.20425634384155272,
"step": 4715,
"token_acc": 0.9221378591633169
},
{
"epoch": 0.03600895646501905,
"grad_norm": 0.9872845449849743,
"learning_rate": 3.600854440036619e-06,
"loss": 0.17373206615447997,
"step": 4720,
"token_acc": 0.928167148362235
},
{
"epoch": 0.03604710154602013,
"grad_norm": 1.6493104359751685,
"learning_rate": 3.604668904485811e-06,
"loss": 0.21091775894165038,
"step": 4725,
"token_acc": 0.919090294217112
},
{
"epoch": 0.036085246627021214,
"grad_norm": 1.1700371939270466,
"learning_rate": 3.608483368935002e-06,
"loss": 0.20188605785369873,
"step": 4730,
"token_acc": 0.9211062208500296
},
{
"epoch": 0.03612339170802229,
"grad_norm": 1.488953606114595,
"learning_rate": 3.6122978333841928e-06,
"loss": 0.19887063503265381,
"step": 4735,
"token_acc": 0.9201634877384196
},
{
"epoch": 0.03616153678902337,
"grad_norm": 1.99614020397488,
"learning_rate": 3.6161122978333847e-06,
"loss": 0.1961009383201599,
"step": 4740,
"token_acc": 0.9211907618476305
},
{
"epoch": 0.03619968187002445,
"grad_norm": 1.3097272993164855,
"learning_rate": 3.619926762282576e-06,
"loss": 0.20877888202667236,
"step": 4745,
"token_acc": 0.9193976335604159
},
{
"epoch": 0.03623782695102553,
"grad_norm": 1.148256364633608,
"learning_rate": 3.6237412267317674e-06,
"loss": 0.201180362701416,
"step": 4750,
"token_acc": 0.9194849519081602
},
{
"epoch": 0.03627597203202661,
"grad_norm": 1.6413022924415255,
"learning_rate": 3.6275556911809585e-06,
"loss": 0.16963256597518922,
"step": 4755,
"token_acc": 0.9342214056040422
},
{
"epoch": 0.03631411711302769,
"grad_norm": 1.0360245119506337,
"learning_rate": 3.63137015563015e-06,
"loss": 0.19635586738586425,
"step": 4760,
"token_acc": 0.9222570065593322
},
{
"epoch": 0.03635226219402877,
"grad_norm": 1.4220485680690975,
"learning_rate": 3.635184620079341e-06,
"loss": 0.18384374380111695,
"step": 4765,
"token_acc": 0.9263365360926337
},
{
"epoch": 0.036390407275029846,
"grad_norm": 1.2887444636711287,
"learning_rate": 3.6389990845285323e-06,
"loss": 0.1849268674850464,
"step": 4770,
"token_acc": 0.930059035749426
},
{
"epoch": 0.03642855235603093,
"grad_norm": 0.8265619160701162,
"learning_rate": 3.642813548977724e-06,
"loss": 0.1880925178527832,
"step": 4775,
"token_acc": 0.9256587874026995
},
{
"epoch": 0.03646669743703201,
"grad_norm": 2.2907162408230746,
"learning_rate": 3.646628013426915e-06,
"loss": 0.18768199682235717,
"step": 4780,
"token_acc": 0.9248212330090833
},
{
"epoch": 0.036504842518033084,
"grad_norm": 5.281483863492951,
"learning_rate": 3.6504424778761066e-06,
"loss": 0.191192889213562,
"step": 4785,
"token_acc": 0.926709557604298
},
{
"epoch": 0.03654298759903417,
"grad_norm": 1.409792095316201,
"learning_rate": 3.6542569423252977e-06,
"loss": 0.18201837539672852,
"step": 4790,
"token_acc": 0.9261794787198945
},
{
"epoch": 0.036581132680035246,
"grad_norm": 10.820767117445637,
"learning_rate": 3.6580714067744892e-06,
"loss": 0.20197932720184325,
"step": 4795,
"token_acc": 0.9194280526554698
},
{
"epoch": 0.03661927776103632,
"grad_norm": 1.20165134000025,
"learning_rate": 3.6618858712236804e-06,
"loss": 0.17642291784286498,
"step": 4800,
"token_acc": 0.9253318956803019
},
{
"epoch": 0.03665742284203741,
"grad_norm": 1.4130533861235786,
"learning_rate": 3.665700335672872e-06,
"loss": 0.1893955111503601,
"step": 4805,
"token_acc": 0.9278597050617776
},
{
"epoch": 0.036695567923038484,
"grad_norm": 1.5284395023270398,
"learning_rate": 3.669514800122063e-06,
"loss": 0.1943805694580078,
"step": 4810,
"token_acc": 0.9211874425325102
},
{
"epoch": 0.03673371300403956,
"grad_norm": 4.01595886775926,
"learning_rate": 3.673329264571254e-06,
"loss": 0.19269106388092042,
"step": 4815,
"token_acc": 0.9235645584377733
},
{
"epoch": 0.036771858085040646,
"grad_norm": 1.706328088251388,
"learning_rate": 3.6771437290204457e-06,
"loss": 0.1907578706741333,
"step": 4820,
"token_acc": 0.9254122142806596
},
{
"epoch": 0.03681000316604172,
"grad_norm": 1.4132479875279278,
"learning_rate": 3.680958193469637e-06,
"loss": 0.18836854696273803,
"step": 4825,
"token_acc": 0.927105283696246
},
{
"epoch": 0.0368481482470428,
"grad_norm": 1.267894149108726,
"learning_rate": 3.684772657918829e-06,
"loss": 0.16857354640960692,
"step": 4830,
"token_acc": 0.9333333333333333
},
{
"epoch": 0.036886293328043884,
"grad_norm": 1.1281032243811113,
"learning_rate": 3.68858712236802e-06,
"loss": 0.22880544662475585,
"step": 4835,
"token_acc": 0.9115356932539422
},
{
"epoch": 0.03692443840904496,
"grad_norm": 2.082958304571172,
"learning_rate": 3.6924015868172115e-06,
"loss": 0.20400071144104004,
"step": 4840,
"token_acc": 0.9228743162017965
},
{
"epoch": 0.03696258349004604,
"grad_norm": 1.5385614498642348,
"learning_rate": 3.6962160512664026e-06,
"loss": 0.20122828483581542,
"step": 4845,
"token_acc": 0.9186180977161029
},
{
"epoch": 0.03700072857104712,
"grad_norm": 1.7976115690885215,
"learning_rate": 3.7000305157155938e-06,
"loss": 0.19796953201293946,
"step": 4850,
"token_acc": 0.925753442884705
},
{
"epoch": 0.0370388736520482,
"grad_norm": 1.186096156506103,
"learning_rate": 3.7038449801647853e-06,
"loss": 0.21562557220458983,
"step": 4855,
"token_acc": 0.9181612660135644
},
{
"epoch": 0.03707701873304928,
"grad_norm": 1.4392356566829416,
"learning_rate": 3.7076594446139764e-06,
"loss": 0.19940098524093627,
"step": 4860,
"token_acc": 0.9186329730552596
},
{
"epoch": 0.03711516381405036,
"grad_norm": 2.4124781239253807,
"learning_rate": 3.711473909063168e-06,
"loss": 0.2155149459838867,
"step": 4865,
"token_acc": 0.9143595124707402
},
{
"epoch": 0.03715330889505144,
"grad_norm": 0.8149812146962572,
"learning_rate": 3.715288373512359e-06,
"loss": 0.2081230640411377,
"step": 4870,
"token_acc": 0.91744145403705
},
{
"epoch": 0.037191453976052516,
"grad_norm": 1.1079012929462566,
"learning_rate": 3.7191028379615507e-06,
"loss": 0.21206936836242676,
"step": 4875,
"token_acc": 0.9173213206614295
},
{
"epoch": 0.0372295990570536,
"grad_norm": 1.1318139049509908,
"learning_rate": 3.7229173024107418e-06,
"loss": 0.21933393478393554,
"step": 4880,
"token_acc": 0.9154421411702965
},
{
"epoch": 0.03726774413805468,
"grad_norm": 1.3653931157313959,
"learning_rate": 3.7267317668599333e-06,
"loss": 0.2180100202560425,
"step": 4885,
"token_acc": 0.9178968974500671
},
{
"epoch": 0.037305889219055755,
"grad_norm": 1.5591956043552042,
"learning_rate": 3.7305462313091245e-06,
"loss": 0.20957231521606445,
"step": 4890,
"token_acc": 0.9155541360587671
},
{
"epoch": 0.03734403430005684,
"grad_norm": 0.9281506343547913,
"learning_rate": 3.7343606957583156e-06,
"loss": 0.2087615966796875,
"step": 4895,
"token_acc": 0.918942731277533
},
{
"epoch": 0.037382179381057916,
"grad_norm": 1.387321637762542,
"learning_rate": 3.738175160207507e-06,
"loss": 0.22257075309753419,
"step": 4900,
"token_acc": 0.9117294822120465
},
{
"epoch": 0.03742032446205899,
"grad_norm": 1.9332313868766162,
"learning_rate": 3.7419896246566983e-06,
"loss": 0.21306405067443848,
"step": 4905,
"token_acc": 0.9168977914740627
},
{
"epoch": 0.03745846954306008,
"grad_norm": 1.2341793934550365,
"learning_rate": 3.74580408910589e-06,
"loss": 0.1533681869506836,
"step": 4910,
"token_acc": 0.9394854586129754
},
{
"epoch": 0.037496614624061154,
"grad_norm": 0.8690615822200775,
"learning_rate": 3.749618553555081e-06,
"loss": 0.1840430736541748,
"step": 4915,
"token_acc": 0.9282443746729461
},
{
"epoch": 0.03753475970506223,
"grad_norm": 1.1017459600215964,
"learning_rate": 3.753433018004273e-06,
"loss": 0.23162190914154052,
"step": 4920,
"token_acc": 0.9124168514412417
},
{
"epoch": 0.037572904786063316,
"grad_norm": 5.287422229518523,
"learning_rate": 3.7572474824534636e-06,
"loss": 0.1864027738571167,
"step": 4925,
"token_acc": 0.9278933256033046
},
{
"epoch": 0.03761104986706439,
"grad_norm": 0.9739475351475809,
"learning_rate": 3.7610619469026547e-06,
"loss": 0.19523341655731202,
"step": 4930,
"token_acc": 0.923131946737097
},
{
"epoch": 0.03764919494806547,
"grad_norm": 1.2123404999407168,
"learning_rate": 3.7648764113518467e-06,
"loss": 0.184981632232666,
"step": 4935,
"token_acc": 0.928767960868236
},
{
"epoch": 0.037687340029066554,
"grad_norm": 1.555138617430549,
"learning_rate": 3.768690875801038e-06,
"loss": 0.17744894027709962,
"step": 4940,
"token_acc": 0.9293309703145769
},
{
"epoch": 0.03772548511006763,
"grad_norm": 1.4276540757055345,
"learning_rate": 3.7725053402502294e-06,
"loss": 0.19839925765991212,
"step": 4945,
"token_acc": 0.922281993204983
},
{
"epoch": 0.03776363019106871,
"grad_norm": 1.7056188770862295,
"learning_rate": 3.7763198046994205e-06,
"loss": 0.17661449909210206,
"step": 4950,
"token_acc": 0.9315087621308386
},
{
"epoch": 0.03780177527206979,
"grad_norm": 1.1428599578981349,
"learning_rate": 3.780134269148612e-06,
"loss": 0.21469569206237793,
"step": 4955,
"token_acc": 0.9146308963311717
},
{
"epoch": 0.03783992035307087,
"grad_norm": 1.2450759631658221,
"learning_rate": 3.783948733597803e-06,
"loss": 0.2074908971786499,
"step": 4960,
"token_acc": 0.917936694021102
},
{
"epoch": 0.03787806543407195,
"grad_norm": 1.1493411235282904,
"learning_rate": 3.7877631980469947e-06,
"loss": 0.19364709854125978,
"step": 4965,
"token_acc": 0.924304476768818
},
{
"epoch": 0.03791621051507303,
"grad_norm": 1.185630586162837,
"learning_rate": 3.791577662496186e-06,
"loss": 0.16621183156967162,
"step": 4970,
"token_acc": 0.9322709163346613
},
{
"epoch": 0.03795435559607411,
"grad_norm": 1.571149747616013,
"learning_rate": 3.795392126945377e-06,
"loss": 0.17076672315597535,
"step": 4975,
"token_acc": 0.9333277842517064
},
{
"epoch": 0.037992500677075186,
"grad_norm": 1.232263398202259,
"learning_rate": 3.7992065913945686e-06,
"loss": 0.21470823287963867,
"step": 4980,
"token_acc": 0.9152279471270569
},
{
"epoch": 0.03803064575807627,
"grad_norm": 1.2694799248600706,
"learning_rate": 3.8030210558437597e-06,
"loss": 0.19774235486984254,
"step": 4985,
"token_acc": 0.9214192574291401
},
{
"epoch": 0.03806879083907735,
"grad_norm": 0.6853213604117129,
"learning_rate": 3.8068355202929512e-06,
"loss": 0.18826234340667725,
"step": 4990,
"token_acc": 0.9254212454212454
},
{
"epoch": 0.038106935920078425,
"grad_norm": 1.4763069192096154,
"learning_rate": 3.8106499847421424e-06,
"loss": 0.19704945087432862,
"step": 4995,
"token_acc": 0.9233115468409586
},
{
"epoch": 0.03814508100107951,
"grad_norm": 3.8217789772842066,
"learning_rate": 3.814464449191334e-06,
"loss": 0.21327171325683594,
"step": 5000,
"token_acc": 0.9167379304983285
}
],
"logging_steps": 5,
"max_steps": 262158,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2096988551708672.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}