memrag-mem / trainer_state.json
hyunseoki's picture
checkpoint-846 (inference)
97ace16 verified
Raw
History Blame Contribute Delete
221 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 846,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0035555555555555557,
"grad_norm": 20.983922958374023,
"learning_rate": 2.5000000000000004e-07,
"loss": 1.1340906620025635,
"loss_sft": 1.1340906620025635,
"step": 1,
"token_acc": 0.6925138263298933
},
{
"epoch": 0.0071111111111111115,
"grad_norm": 21.411361694335938,
"learning_rate": 5.000000000000001e-07,
"loss": 1.1090593338012695,
"loss_sft": 1.1090593338012695,
"step": 2,
"token_acc": 0.6990699238009861
},
{
"epoch": 0.010666666666666666,
"grad_norm": 21.33144187927246,
"learning_rate": 7.5e-07,
"loss": 1.1518434286117554,
"loss_sft": 1.1518434286117554,
"step": 3,
"token_acc": 0.6946262270324691
},
{
"epoch": 0.014222222222222223,
"grad_norm": 20.271839141845703,
"learning_rate": 1.0000000000000002e-06,
"loss": 1.071044683456421,
"loss_sft": 1.071044683456421,
"step": 4,
"token_acc": 0.7028114604571306
},
{
"epoch": 0.017777777777777778,
"grad_norm": 16.812294006347656,
"learning_rate": 1.25e-06,
"loss": 1.082520604133606,
"loss_sft": 1.082520604133606,
"step": 5,
"token_acc": 0.6990645676477298
},
{
"epoch": 0.021333333333333333,
"grad_norm": 11.476592063903809,
"learning_rate": 1.5e-06,
"loss": 1.0654475688934326,
"loss_sft": 1.0654475688934326,
"step": 6,
"token_acc": 0.7033498466619486
},
{
"epoch": 0.024888888888888887,
"grad_norm": 10.128439903259277,
"learning_rate": 1.75e-06,
"loss": 1.0348063707351685,
"loss_sft": 1.0348063707351685,
"step": 7,
"token_acc": 0.71674241493674
},
{
"epoch": 0.028444444444444446,
"grad_norm": 8.450559616088867,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.9253971576690674,
"loss_sft": 0.9253971576690674,
"step": 8,
"token_acc": 0.7334515704559758
},
{
"epoch": 0.032,
"grad_norm": 7.736713409423828,
"learning_rate": 2.25e-06,
"loss": 0.9541059732437134,
"loss_sft": 0.9541059732437134,
"step": 9,
"token_acc": 0.7284861257463997
},
{
"epoch": 0.035555555555555556,
"grad_norm": 6.614664077758789,
"learning_rate": 2.5e-06,
"loss": 0.8742601871490479,
"loss_sft": 0.8742601871490479,
"step": 10,
"token_acc": 0.7462323208903315
},
{
"epoch": 0.03911111111111111,
"grad_norm": 5.6005988121032715,
"learning_rate": 2.7500000000000004e-06,
"loss": 0.8437846899032593,
"loss_sft": 0.8437846899032593,
"step": 11,
"token_acc": 0.7580404096318849
},
{
"epoch": 0.042666666666666665,
"grad_norm": 4.773887634277344,
"learning_rate": 3e-06,
"loss": 0.821181058883667,
"loss_sft": 0.821181058883667,
"step": 12,
"token_acc": 0.759621451104101
},
{
"epoch": 0.04622222222222222,
"grad_norm": 4.401590347290039,
"learning_rate": 3.2500000000000002e-06,
"loss": 0.8202284574508667,
"loss_sft": 0.8202284574508667,
"step": 13,
"token_acc": 0.7584763499372122
},
{
"epoch": 0.049777777777777775,
"grad_norm": 4.867381572723389,
"learning_rate": 3.5e-06,
"loss": 0.7727222442626953,
"loss_sft": 0.7727222442626953,
"step": 14,
"token_acc": 0.7709236224461079
},
{
"epoch": 0.05333333333333334,
"grad_norm": 3.7348971366882324,
"learning_rate": 3.7500000000000005e-06,
"loss": 0.7707201838493347,
"loss_sft": 0.7707201838493347,
"step": 15,
"token_acc": 0.7755102040816326
},
{
"epoch": 0.05688888888888889,
"grad_norm": 3.442126512527466,
"learning_rate": 4.000000000000001e-06,
"loss": 0.7654933929443359,
"loss_sft": 0.7654933929443359,
"step": 16,
"token_acc": 0.7764313507504169
},
{
"epoch": 0.060444444444444446,
"grad_norm": 3.7610056400299072,
"learning_rate": 4.25e-06,
"loss": 0.704330563545227,
"loss_sft": 0.704330563545227,
"step": 17,
"token_acc": 0.7872063512722067
},
{
"epoch": 0.064,
"grad_norm": 2.9932496547698975,
"learning_rate": 4.5e-06,
"loss": 0.7594562768936157,
"loss_sft": 0.7594562768936157,
"step": 18,
"token_acc": 0.7800751879699248
},
{
"epoch": 0.06755555555555555,
"grad_norm": 3.2831130027770996,
"learning_rate": 4.75e-06,
"loss": 0.7128193378448486,
"loss_sft": 0.7128193378448486,
"step": 19,
"token_acc": 0.784909299102031
},
{
"epoch": 0.07111111111111111,
"grad_norm": 3.4387524127960205,
"learning_rate": 5e-06,
"loss": 0.7347320318222046,
"loss_sft": 0.7347320318222046,
"step": 20,
"token_acc": 0.7835679640364368
},
{
"epoch": 0.07466666666666667,
"grad_norm": 3.813136577606201,
"learning_rate": 5.2500000000000006e-06,
"loss": 0.7200600504875183,
"loss_sft": 0.7200600504875183,
"step": 21,
"token_acc": 0.7873118044948724
},
{
"epoch": 0.07822222222222222,
"grad_norm": 3.248347043991089,
"learning_rate": 5.500000000000001e-06,
"loss": 0.6967421770095825,
"loss_sft": 0.6967421770095825,
"step": 22,
"token_acc": 0.7923794297895017
},
{
"epoch": 0.08177777777777778,
"grad_norm": 2.9261245727539062,
"learning_rate": 5.75e-06,
"loss": 0.7348852753639221,
"loss_sft": 0.7348852753639221,
"step": 23,
"token_acc": 0.7791546106428482
},
{
"epoch": 0.08533333333333333,
"grad_norm": 2.7344608306884766,
"learning_rate": 6e-06,
"loss": 0.6594898700714111,
"loss_sft": 0.6594898700714111,
"step": 24,
"token_acc": 0.8044193408230401
},
{
"epoch": 0.08888888888888889,
"grad_norm": 2.738558769226074,
"learning_rate": 6.25e-06,
"loss": 0.7026379108428955,
"loss_sft": 0.7026379108428955,
"step": 25,
"token_acc": 0.790961079340403
},
{
"epoch": 0.09244444444444444,
"grad_norm": 2.6157360076904297,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.70719313621521,
"loss_sft": 0.70719313621521,
"step": 26,
"token_acc": 0.7903861765046968
},
{
"epoch": 0.096,
"grad_norm": 2.609368324279785,
"learning_rate": 6.750000000000001e-06,
"loss": 0.6963019371032715,
"loss_sft": 0.6963019371032715,
"step": 27,
"token_acc": 0.7886773767605634
},
{
"epoch": 0.09955555555555555,
"grad_norm": 2.7674100399017334,
"learning_rate": 7e-06,
"loss": 0.7104215025901794,
"loss_sft": 0.7104215025901794,
"step": 28,
"token_acc": 0.7928346326090562
},
{
"epoch": 0.10311111111111111,
"grad_norm": 2.560457229614258,
"learning_rate": 7.25e-06,
"loss": 0.7051395177841187,
"loss_sft": 0.7051395177841187,
"step": 29,
"token_acc": 0.7864459780030192
},
{
"epoch": 0.10666666666666667,
"grad_norm": 2.5793039798736572,
"learning_rate": 7.500000000000001e-06,
"loss": 0.6692949533462524,
"loss_sft": 0.6692949533462524,
"step": 30,
"token_acc": 0.7924174387540499
},
{
"epoch": 0.11022222222222222,
"grad_norm": 2.9297451972961426,
"learning_rate": 7.75e-06,
"loss": 0.7029169797897339,
"loss_sft": 0.7029169797897339,
"step": 31,
"token_acc": 0.7913652138109523
},
{
"epoch": 0.11377777777777778,
"grad_norm": 2.656677007675171,
"learning_rate": 8.000000000000001e-06,
"loss": 0.6783644556999207,
"loss_sft": 0.6783644556999207,
"step": 32,
"token_acc": 0.7904994635832638
},
{
"epoch": 0.11733333333333333,
"grad_norm": 2.7285196781158447,
"learning_rate": 8.25e-06,
"loss": 0.7081842422485352,
"loss_sft": 0.7081842422485352,
"step": 33,
"token_acc": 0.7891266558772403
},
{
"epoch": 0.12088888888888889,
"grad_norm": 2.551206111907959,
"learning_rate": 8.5e-06,
"loss": 0.6748272180557251,
"loss_sft": 0.6748272180557251,
"step": 34,
"token_acc": 0.7955828836742377
},
{
"epoch": 0.12444444444444444,
"grad_norm": 2.531101703643799,
"learning_rate": 8.750000000000001e-06,
"loss": 0.6638057231903076,
"loss_sft": 0.6638057231903076,
"step": 35,
"token_acc": 0.798465829846583
},
{
"epoch": 0.128,
"grad_norm": 2.491770029067993,
"learning_rate": 9e-06,
"loss": 0.6896926164627075,
"loss_sft": 0.6896926164627075,
"step": 36,
"token_acc": 0.7869425186964354
},
{
"epoch": 0.13155555555555556,
"grad_norm": 2.6662817001342773,
"learning_rate": 9.250000000000001e-06,
"loss": 0.6853880286216736,
"loss_sft": 0.6853880286216736,
"step": 37,
"token_acc": 0.7954451590064311
},
{
"epoch": 0.1351111111111111,
"grad_norm": 2.650941848754883,
"learning_rate": 9.5e-06,
"loss": 0.6828696727752686,
"loss_sft": 0.6828696727752686,
"step": 38,
"token_acc": 0.7936775921290097
},
{
"epoch": 0.13866666666666666,
"grad_norm": 2.5845916271209717,
"learning_rate": 9.75e-06,
"loss": 0.6833871603012085,
"loss_sft": 0.6833871603012085,
"step": 39,
"token_acc": 0.7933317977657491
},
{
"epoch": 0.14222222222222222,
"grad_norm": 2.6829824447631836,
"learning_rate": 1e-05,
"loss": 0.673632025718689,
"loss_sft": 0.673632025718689,
"step": 40,
"token_acc": 0.7936316001612254
},
{
"epoch": 0.14577777777777778,
"grad_norm": 2.528928756713867,
"learning_rate": 9.99996201876164e-06,
"loss": 0.6603695154190063,
"loss_sft": 0.6603695154190063,
"step": 41,
"token_acc": 0.797522793738173
},
{
"epoch": 0.14933333333333335,
"grad_norm": 2.5398335456848145,
"learning_rate": 9.999848075623584e-06,
"loss": 0.687311053276062,
"loss_sft": 0.687311053276062,
"step": 42,
"token_acc": 0.7938973647711511
},
{
"epoch": 0.15288888888888888,
"grad_norm": 2.4450557231903076,
"learning_rate": 9.999658172316915e-06,
"loss": 0.6976714134216309,
"loss_sft": 0.6976714134216309,
"step": 43,
"token_acc": 0.7889001806684733
},
{
"epoch": 0.15644444444444444,
"grad_norm": 2.45168399810791,
"learning_rate": 9.999392311726738e-06,
"loss": 0.7229422330856323,
"loss_sft": 0.7229422330856323,
"step": 44,
"token_acc": 0.7890711535418534
},
{
"epoch": 0.16,
"grad_norm": 2.445906639099121,
"learning_rate": 9.99905049789214e-06,
"loss": 0.6838053464889526,
"loss_sft": 0.6838053464889526,
"step": 45,
"token_acc": 0.7923917893797412
},
{
"epoch": 0.16355555555555557,
"grad_norm": 2.466531991958618,
"learning_rate": 9.998632736006124e-06,
"loss": 0.686069130897522,
"loss_sft": 0.686069130897522,
"step": 46,
"token_acc": 0.7914860773021434
},
{
"epoch": 0.1671111111111111,
"grad_norm": 2.718491554260254,
"learning_rate": 9.998139032415534e-06,
"loss": 0.6970874071121216,
"loss_sft": 0.6970874071121216,
"step": 47,
"token_acc": 0.7882946642870331
},
{
"epoch": 0.17066666666666666,
"grad_norm": 2.4150822162628174,
"learning_rate": 9.997569394620965e-06,
"loss": 0.6699665784835815,
"loss_sft": 0.6699665784835815,
"step": 48,
"token_acc": 0.794776521270867
},
{
"epoch": 0.17422222222222222,
"grad_norm": 2.298213481903076,
"learning_rate": 9.996923831276632e-06,
"loss": 0.6545356512069702,
"loss_sft": 0.6545356512069702,
"step": 49,
"token_acc": 0.8027815539056782
},
{
"epoch": 0.17777777777777778,
"grad_norm": 2.6017403602600098,
"learning_rate": 9.996202352190256e-06,
"loss": 0.6707948446273804,
"loss_sft": 0.6707948446273804,
"step": 50,
"token_acc": 0.7993915069911963
},
{
"epoch": 0.18133333333333335,
"grad_norm": 2.4087908267974854,
"learning_rate": 9.995404968322902e-06,
"loss": 0.6896227598190308,
"loss_sft": 0.6896227598190308,
"step": 51,
"token_acc": 0.7926623855142647
},
{
"epoch": 0.18488888888888888,
"grad_norm": 2.3799080848693848,
"learning_rate": 9.994531691788822e-06,
"loss": 0.6673187017440796,
"loss_sft": 0.6673187017440796,
"step": 52,
"token_acc": 0.7992457721996347
},
{
"epoch": 0.18844444444444444,
"grad_norm": 2.300178050994873,
"learning_rate": 9.993582535855265e-06,
"loss": 0.6866666674613953,
"loss_sft": 0.6866666674613953,
"step": 53,
"token_acc": 0.7960427866762734
},
{
"epoch": 0.192,
"grad_norm": 2.3645646572113037,
"learning_rate": 9.992557514942278e-06,
"loss": 0.6767413020133972,
"loss_sft": 0.6767413020133972,
"step": 54,
"token_acc": 0.7919062238347754
},
{
"epoch": 0.19555555555555557,
"grad_norm": 2.661588430404663,
"learning_rate": 9.991456644622489e-06,
"loss": 0.6490683555603027,
"loss_sft": 0.6490683555603027,
"step": 55,
"token_acc": 0.8024441249509868
},
{
"epoch": 0.1991111111111111,
"grad_norm": 2.4531829357147217,
"learning_rate": 9.990279941620861e-06,
"loss": 0.6744770407676697,
"loss_sft": 0.6744770407676697,
"step": 56,
"token_acc": 0.7957093821510297
},
{
"epoch": 0.20266666666666666,
"grad_norm": 2.4367525577545166,
"learning_rate": 9.989027423814454e-06,
"loss": 0.6794407367706299,
"loss_sft": 0.6794407367706299,
"step": 57,
"token_acc": 0.7992387661194115
},
{
"epoch": 0.20622222222222222,
"grad_norm": 2.7172770500183105,
"learning_rate": 9.987699110232134e-06,
"loss": 0.6924586296081543,
"loss_sft": 0.6924586296081543,
"step": 58,
"token_acc": 0.7921842209744201
},
{
"epoch": 0.20977777777777779,
"grad_norm": 2.506361961364746,
"learning_rate": 9.986295021054302e-06,
"loss": 0.6553081274032593,
"loss_sft": 0.6553081274032593,
"step": 59,
"token_acc": 0.7987840372550288
},
{
"epoch": 0.21333333333333335,
"grad_norm": 2.5046894550323486,
"learning_rate": 9.984815177612574e-06,
"loss": 0.7028884291648865,
"loss_sft": 0.7028884291648865,
"step": 60,
"token_acc": 0.7910873656730011
},
{
"epoch": 0.21688888888888888,
"grad_norm": 2.5686514377593994,
"learning_rate": 9.983259602389469e-06,
"loss": 0.6981114745140076,
"loss_sft": 0.6981114745140076,
"step": 61,
"token_acc": 0.7903784396448587
},
{
"epoch": 0.22044444444444444,
"grad_norm": 2.3537237644195557,
"learning_rate": 9.98162831901805e-06,
"loss": 0.6965206861495972,
"loss_sft": 0.6965206861495972,
"step": 62,
"token_acc": 0.7908831908831909
},
{
"epoch": 0.224,
"grad_norm": 2.3147056102752686,
"learning_rate": 9.979921352281585e-06,
"loss": 0.7216426134109497,
"loss_sft": 0.7216426134109497,
"step": 63,
"token_acc": 0.784184048386185
},
{
"epoch": 0.22755555555555557,
"grad_norm": 2.256810188293457,
"learning_rate": 9.97813872811316e-06,
"loss": 0.6454267501831055,
"loss_sft": 0.6454267501831055,
"step": 64,
"token_acc": 0.8023249377248823
},
{
"epoch": 0.2311111111111111,
"grad_norm": 2.3185648918151855,
"learning_rate": 9.976280473595284e-06,
"loss": 0.6606327295303345,
"loss_sft": 0.6606327295303345,
"step": 65,
"token_acc": 0.7970721256843609
},
{
"epoch": 0.23466666666666666,
"grad_norm": 2.4192910194396973,
"learning_rate": 9.974346616959476e-06,
"loss": 0.7121932506561279,
"loss_sft": 0.7121932506561279,
"step": 66,
"token_acc": 0.7860363032388183
},
{
"epoch": 0.23822222222222222,
"grad_norm": 2.4552114009857178,
"learning_rate": 9.972337187585848e-06,
"loss": 0.6743242740631104,
"loss_sft": 0.6743242740631104,
"step": 67,
"token_acc": 0.7963865446333056
},
{
"epoch": 0.24177777777777779,
"grad_norm": 2.484628677368164,
"learning_rate": 9.970252216002647e-06,
"loss": 0.6334826946258545,
"loss_sft": 0.6334826946258545,
"step": 68,
"token_acc": 0.8025587828492393
},
{
"epoch": 0.24533333333333332,
"grad_norm": 2.2711801528930664,
"learning_rate": 9.96809173388579e-06,
"loss": 0.6617833971977234,
"loss_sft": 0.6617833971977234,
"step": 69,
"token_acc": 0.7961126310989513
},
{
"epoch": 0.24888888888888888,
"grad_norm": 2.3135502338409424,
"learning_rate": 9.965855774058395e-06,
"loss": 0.6995804309844971,
"loss_sft": 0.6995804309844971,
"step": 70,
"token_acc": 0.7893893252123738
},
{
"epoch": 0.25244444444444447,
"grad_norm": 2.2933452129364014,
"learning_rate": 9.96354437049027e-06,
"loss": 0.7320924997329712,
"loss_sft": 0.7320924997329712,
"step": 71,
"token_acc": 0.78335769471706
},
{
"epoch": 0.256,
"grad_norm": 2.2438836097717285,
"learning_rate": 9.961157558297404e-06,
"loss": 0.6782565712928772,
"loss_sft": 0.6782565712928772,
"step": 72,
"token_acc": 0.7954691173327634
},
{
"epoch": 0.25955555555555554,
"grad_norm": 2.092151641845703,
"learning_rate": 9.958695373741428e-06,
"loss": 0.6640706062316895,
"loss_sft": 0.6640706062316895,
"step": 73,
"token_acc": 0.7954522280004078
},
{
"epoch": 0.26311111111111113,
"grad_norm": 2.23357892036438,
"learning_rate": 9.956157854229072e-06,
"loss": 0.6789810657501221,
"loss_sft": 0.6789810657501221,
"step": 74,
"token_acc": 0.7947180192572214
},
{
"epoch": 0.26666666666666666,
"grad_norm": 2.246324300765991,
"learning_rate": 9.953545038311587e-06,
"loss": 0.6890798807144165,
"loss_sft": 0.6890798807144165,
"step": 75,
"token_acc": 0.7892845820564303
},
{
"epoch": 0.2702222222222222,
"grad_norm": 2.1419386863708496,
"learning_rate": 9.950856965684167e-06,
"loss": 0.671385645866394,
"loss_sft": 0.671385645866394,
"step": 76,
"token_acc": 0.7958000656239747
},
{
"epoch": 0.2737777777777778,
"grad_norm": 2.2745158672332764,
"learning_rate": 9.948093677185345e-06,
"loss": 0.6130911111831665,
"loss_sft": 0.6130911111831665,
"step": 77,
"token_acc": 0.8117005046356062
},
{
"epoch": 0.2773333333333333,
"grad_norm": 2.233564853668213,
"learning_rate": 9.945255214796366e-06,
"loss": 0.6761378645896912,
"loss_sft": 0.6761378645896912,
"step": 78,
"token_acc": 0.795233892321271
},
{
"epoch": 0.2808888888888889,
"grad_norm": 2.2238430976867676,
"learning_rate": 9.942341621640558e-06,
"loss": 0.6803784370422363,
"loss_sft": 0.6803784370422363,
"step": 79,
"token_acc": 0.7964225336750832
},
{
"epoch": 0.28444444444444444,
"grad_norm": 2.1541366577148438,
"learning_rate": 9.939352941982671e-06,
"loss": 0.6466889381408691,
"loss_sft": 0.6466889381408691,
"step": 80,
"token_acc": 0.80746859004188
},
{
"epoch": 0.288,
"grad_norm": 2.2607672214508057,
"learning_rate": 9.936289221228207e-06,
"loss": 0.7033430337905884,
"loss_sft": 0.7033430337905884,
"step": 81,
"token_acc": 0.7853936942428105
},
{
"epoch": 0.29155555555555557,
"grad_norm": 2.2796926498413086,
"learning_rate": 9.93315050592273e-06,
"loss": 0.6681029796600342,
"loss_sft": 0.6681029796600342,
"step": 82,
"token_acc": 0.7938898217864687
},
{
"epoch": 0.2951111111111111,
"grad_norm": 2.2661917209625244,
"learning_rate": 9.929936843751158e-06,
"loss": 0.727078914642334,
"loss_sft": 0.727078914642334,
"step": 83,
"token_acc": 0.7853594080338266
},
{
"epoch": 0.2986666666666667,
"grad_norm": 2.127492904663086,
"learning_rate": 9.926648283537037e-06,
"loss": 0.6307873725891113,
"loss_sft": 0.6307873725891113,
"step": 84,
"token_acc": 0.8066015533066604
},
{
"epoch": 0.3022222222222222,
"grad_norm": 2.2362189292907715,
"learning_rate": 9.923284875241802e-06,
"loss": 0.6705547571182251,
"loss_sft": 0.6705547571182251,
"step": 85,
"token_acc": 0.7986414093996796
},
{
"epoch": 0.30577777777777776,
"grad_norm": 2.2825276851654053,
"learning_rate": 9.91984666996402e-06,
"loss": 0.7046873569488525,
"loss_sft": 0.7046873569488525,
"step": 86,
"token_acc": 0.7860974395841398
},
{
"epoch": 0.30933333333333335,
"grad_norm": 2.295374631881714,
"learning_rate": 9.916333719938608e-06,
"loss": 0.6466997861862183,
"loss_sft": 0.6466997861862183,
"step": 87,
"token_acc": 0.7971345396426737
},
{
"epoch": 0.3128888888888889,
"grad_norm": 2.1010236740112305,
"learning_rate": 9.912746078536044e-06,
"loss": 0.681746780872345,
"loss_sft": 0.681746780872345,
"step": 88,
"token_acc": 0.7895611167675901
},
{
"epoch": 0.3164444444444444,
"grad_norm": 2.3137731552124023,
"learning_rate": 9.90908380026155e-06,
"loss": 0.6900076866149902,
"loss_sft": 0.6900076866149902,
"step": 89,
"token_acc": 0.7924449233716475
},
{
"epoch": 0.32,
"grad_norm": 2.110109806060791,
"learning_rate": 9.905346940754274e-06,
"loss": 0.6393294930458069,
"loss_sft": 0.6393294930458069,
"step": 90,
"token_acc": 0.8044213973799127
},
{
"epoch": 0.32355555555555554,
"grad_norm": 2.319997549057007,
"learning_rate": 9.90153555678644e-06,
"loss": 0.666654109954834,
"loss_sft": 0.666654109954834,
"step": 91,
"token_acc": 0.7977658405567086
},
{
"epoch": 0.32711111111111113,
"grad_norm": 2.233067512512207,
"learning_rate": 9.897649706262474e-06,
"loss": 0.7031636834144592,
"loss_sft": 0.7031636834144592,
"step": 92,
"token_acc": 0.7854753203973441
},
{
"epoch": 0.33066666666666666,
"grad_norm": 2.1728761196136475,
"learning_rate": 9.893689448218146e-06,
"loss": 0.6725103855133057,
"loss_sft": 0.6725103855133057,
"step": 93,
"token_acc": 0.79289843104872
},
{
"epoch": 0.3342222222222222,
"grad_norm": 2.073380470275879,
"learning_rate": 9.889654842819658e-06,
"loss": 0.6568620204925537,
"loss_sft": 0.6568620204925537,
"step": 94,
"token_acc": 0.7996971906016935
},
{
"epoch": 0.3377777777777778,
"grad_norm": 2.0992860794067383,
"learning_rate": 9.885545951362733e-06,
"loss": 0.6902315616607666,
"loss_sft": 0.6902315616607666,
"step": 95,
"token_acc": 0.792477825135353
},
{
"epoch": 0.3413333333333333,
"grad_norm": 2.199206829071045,
"learning_rate": 9.881362836271686e-06,
"loss": 0.7037885189056396,
"loss_sft": 0.7037885189056396,
"step": 96,
"token_acc": 0.79005189356299
},
{
"epoch": 0.3448888888888889,
"grad_norm": 2.041038990020752,
"learning_rate": 9.877105561098473e-06,
"loss": 0.6494865417480469,
"loss_sft": 0.6494865417480469,
"step": 97,
"token_acc": 0.7996281061587874
},
{
"epoch": 0.34844444444444445,
"grad_norm": 2.305077075958252,
"learning_rate": 9.872774190521727e-06,
"loss": 0.6209917664527893,
"loss_sft": 0.6209917664527893,
"step": 98,
"token_acc": 0.8064258098920144
},
{
"epoch": 0.352,
"grad_norm": 2.1822192668914795,
"learning_rate": 9.868368790345777e-06,
"loss": 0.6678800582885742,
"loss_sft": 0.6678800582885742,
"step": 99,
"token_acc": 0.7979810191621834
},
{
"epoch": 0.35555555555555557,
"grad_norm": 2.130117893218994,
"learning_rate": 9.863889427499641e-06,
"loss": 0.6377717852592468,
"loss_sft": 0.6377717852592468,
"step": 100,
"token_acc": 0.8031002248254644
},
{
"epoch": 0.3591111111111111,
"grad_norm": 2.2308602333068848,
"learning_rate": 9.859336170036022e-06,
"loss": 0.6418084502220154,
"loss_sft": 0.6418084502220154,
"step": 101,
"token_acc": 0.8024010013654984
},
{
"epoch": 0.3626666666666667,
"grad_norm": 2.3035035133361816,
"learning_rate": 9.854709087130261e-06,
"loss": 0.6098614931106567,
"loss_sft": 0.6098614931106567,
"step": 102,
"token_acc": 0.8089721137006601
},
{
"epoch": 0.3662222222222222,
"grad_norm": 2.4704132080078125,
"learning_rate": 9.850008249079295e-06,
"loss": 0.645072340965271,
"loss_sft": 0.645072340965271,
"step": 103,
"token_acc": 0.8059923071732236
},
{
"epoch": 0.36977777777777776,
"grad_norm": 2.1937694549560547,
"learning_rate": 9.84523372730058e-06,
"loss": 0.6707781553268433,
"loss_sft": 0.6707781553268433,
"step": 104,
"token_acc": 0.7937119797626935
},
{
"epoch": 0.37333333333333335,
"grad_norm": 2.3002634048461914,
"learning_rate": 9.840385594331022e-06,
"loss": 0.7387837171554565,
"loss_sft": 0.7387837171554565,
"step": 105,
"token_acc": 0.7824325095599566
},
{
"epoch": 0.3768888888888889,
"grad_norm": 2.1478371620178223,
"learning_rate": 9.835463923825854e-06,
"loss": 0.652603030204773,
"loss_sft": 0.652603030204773,
"step": 106,
"token_acc": 0.7982686400446802
},
{
"epoch": 0.3804444444444444,
"grad_norm": 2.1360690593719482,
"learning_rate": 9.830468790557536e-06,
"loss": 0.68009352684021,
"loss_sft": 0.68009352684021,
"step": 107,
"token_acc": 0.7938009074509376
},
{
"epoch": 0.384,
"grad_norm": 2.09120774269104,
"learning_rate": 9.825400270414602e-06,
"loss": 0.636631965637207,
"loss_sft": 0.636631965637207,
"step": 108,
"token_acc": 0.8020736900573968
},
{
"epoch": 0.38755555555555554,
"grad_norm": 2.0506725311279297,
"learning_rate": 9.820258440400525e-06,
"loss": 0.697675347328186,
"loss_sft": 0.697675347328186,
"step": 109,
"token_acc": 0.7911266469481043
},
{
"epoch": 0.39111111111111113,
"grad_norm": 2.057025909423828,
"learning_rate": 9.81504337863253e-06,
"loss": 0.6447837352752686,
"loss_sft": 0.6447837352752686,
"step": 110,
"token_acc": 0.8059669107675617
},
{
"epoch": 0.39466666666666667,
"grad_norm": 2.1300220489501953,
"learning_rate": 9.809755164340423e-06,
"loss": 0.6317578554153442,
"loss_sft": 0.6317578554153442,
"step": 111,
"token_acc": 0.804151728923326
},
{
"epoch": 0.3982222222222222,
"grad_norm": 2.0897693634033203,
"learning_rate": 9.804393877865373e-06,
"loss": 0.6446391940116882,
"loss_sft": 0.6446391940116882,
"step": 112,
"token_acc": 0.8046760041419151
},
{
"epoch": 0.4017777777777778,
"grad_norm": 2.40112566947937,
"learning_rate": 9.798959600658697e-06,
"loss": 0.6558912992477417,
"loss_sft": 0.6558912992477417,
"step": 113,
"token_acc": 0.8000542225837061
},
{
"epoch": 0.4053333333333333,
"grad_norm": 2.1476519107818604,
"learning_rate": 9.79345241528063e-06,
"loss": 0.5815258026123047,
"loss_sft": 0.5815258026123047,
"step": 114,
"token_acc": 0.8215657128700607
},
{
"epoch": 0.4088888888888889,
"grad_norm": 2.2261154651641846,
"learning_rate": 9.787872405399059e-06,
"loss": 0.6672112941741943,
"loss_sft": 0.6672112941741943,
"step": 115,
"token_acc": 0.7952070908792471
},
{
"epoch": 0.41244444444444445,
"grad_norm": 2.1991844177246094,
"learning_rate": 9.782219655788257e-06,
"loss": 0.6576797962188721,
"loss_sft": 0.6576797962188721,
"step": 116,
"token_acc": 0.8013130615065653
},
{
"epoch": 0.416,
"grad_norm": 2.104767322540283,
"learning_rate": 9.776494252327597e-06,
"loss": 0.6630063056945801,
"loss_sft": 0.6630063056945801,
"step": 117,
"token_acc": 0.794594885020417
},
{
"epoch": 0.41955555555555557,
"grad_norm": 2.061361312866211,
"learning_rate": 9.770696282000245e-06,
"loss": 0.6720774173736572,
"loss_sft": 0.6720774173736572,
"step": 118,
"token_acc": 0.7923928532281048
},
{
"epoch": 0.4231111111111111,
"grad_norm": 2.1530919075012207,
"learning_rate": 9.764825832891837e-06,
"loss": 0.7166485786437988,
"loss_sft": 0.7166485786437988,
"step": 119,
"token_acc": 0.7821944809461235
},
{
"epoch": 0.4266666666666667,
"grad_norm": 2.1180672645568848,
"learning_rate": 9.758882994189145e-06,
"loss": 0.6884403824806213,
"loss_sft": 0.6884403824806213,
"step": 120,
"token_acc": 0.7955883161890093
},
{
"epoch": 0.43022222222222223,
"grad_norm": 2.1795127391815186,
"learning_rate": 9.752867856178719e-06,
"loss": 0.6919851303100586,
"loss_sft": 0.6919851303100586,
"step": 121,
"token_acc": 0.7879579696447434
},
{
"epoch": 0.43377777777777776,
"grad_norm": 2.1765785217285156,
"learning_rate": 9.746780510245512e-06,
"loss": 0.7012777328491211,
"loss_sft": 0.7012777328491211,
"step": 122,
"token_acc": 0.7918989646148595
},
{
"epoch": 0.43733333333333335,
"grad_norm": 2.2151670455932617,
"learning_rate": 9.740621048871501e-06,
"loss": 0.6641297936439514,
"loss_sft": 0.6641297936439514,
"step": 123,
"token_acc": 0.792056477070539
},
{
"epoch": 0.4408888888888889,
"grad_norm": 1.9977933168411255,
"learning_rate": 9.734389565634277e-06,
"loss": 0.6629228591918945,
"loss_sft": 0.6629228591918945,
"step": 124,
"token_acc": 0.7995346432264736
},
{
"epoch": 0.4444444444444444,
"grad_norm": 2.1483852863311768,
"learning_rate": 9.728086155205614e-06,
"loss": 0.6678029298782349,
"loss_sft": 0.6678029298782349,
"step": 125,
"token_acc": 0.7960780111444492
},
{
"epoch": 0.448,
"grad_norm": 2.1776058673858643,
"learning_rate": 9.721710913350048e-06,
"loss": 0.6489057540893555,
"loss_sft": 0.6489057540893555,
"step": 126,
"token_acc": 0.8038780165686157
},
{
"epoch": 0.45155555555555554,
"grad_norm": 2.2860937118530273,
"learning_rate": 9.715263936923413e-06,
"loss": 0.6795281171798706,
"loss_sft": 0.6795281171798706,
"step": 127,
"token_acc": 0.7961333333333334
},
{
"epoch": 0.45511111111111113,
"grad_norm": 2.0711753368377686,
"learning_rate": 9.708745323871369e-06,
"loss": 0.6966684460639954,
"loss_sft": 0.6966684460639954,
"step": 128,
"token_acc": 0.7885511651469098
},
{
"epoch": 0.45866666666666667,
"grad_norm": 2.217532157897949,
"learning_rate": 9.702155173227911e-06,
"loss": 0.633077085018158,
"loss_sft": 0.633077085018158,
"step": 129,
"token_acc": 0.8083073198700389
},
{
"epoch": 0.4622222222222222,
"grad_norm": 2.129143714904785,
"learning_rate": 9.695493585113873e-06,
"loss": 0.6834061741828918,
"loss_sft": 0.6834061741828918,
"step": 130,
"token_acc": 0.7937014040387089
},
{
"epoch": 0.4657777777777778,
"grad_norm": 2.1016604900360107,
"learning_rate": 9.688760660735403e-06,
"loss": 0.6638543605804443,
"loss_sft": 0.6638543605804443,
"step": 131,
"token_acc": 0.8011290593035605
},
{
"epoch": 0.4693333333333333,
"grad_norm": 2.0630922317504883,
"learning_rate": 9.681956502382423e-06,
"loss": 0.6465827226638794,
"loss_sft": 0.6465827226638794,
"step": 132,
"token_acc": 0.7966269841269841
},
{
"epoch": 0.4728888888888889,
"grad_norm": 2.152106285095215,
"learning_rate": 9.675081213427076e-06,
"loss": 0.6958965063095093,
"loss_sft": 0.6958965063095093,
"step": 133,
"token_acc": 0.7919225866264921
},
{
"epoch": 0.47644444444444445,
"grad_norm": 2.1455864906311035,
"learning_rate": 9.668134898322157e-06,
"loss": 0.6584903001785278,
"loss_sft": 0.6584903001785278,
"step": 134,
"token_acc": 0.7958133348753831
},
{
"epoch": 0.48,
"grad_norm": 2.005760669708252,
"learning_rate": 9.661117662599527e-06,
"loss": 0.6611692309379578,
"loss_sft": 0.6611692309379578,
"step": 135,
"token_acc": 0.7998969868658254
},
{
"epoch": 0.48355555555555557,
"grad_norm": 2.193725824356079,
"learning_rate": 9.654029612868507e-06,
"loss": 0.6380483508110046,
"loss_sft": 0.6380483508110046,
"step": 136,
"token_acc": 0.804847169470067
},
{
"epoch": 0.4871111111111111,
"grad_norm": 2.190033435821533,
"learning_rate": 9.646870856814259e-06,
"loss": 0.6512812972068787,
"loss_sft": 0.6512812972068787,
"step": 137,
"token_acc": 0.8042041326743286
},
{
"epoch": 0.49066666666666664,
"grad_norm": 2.0834288597106934,
"learning_rate": 9.639641503196152e-06,
"loss": 0.6663111448287964,
"loss_sft": 0.6663111448287964,
"step": 138,
"token_acc": 0.7974926253687316
},
{
"epoch": 0.49422222222222223,
"grad_norm": 2.1515750885009766,
"learning_rate": 9.632341661846107e-06,
"loss": 0.6522279977798462,
"loss_sft": 0.6522279977798462,
"step": 139,
"token_acc": 0.7999557668915183
},
{
"epoch": 0.49777777777777776,
"grad_norm": 2.1535730361938477,
"learning_rate": 9.62497144366693e-06,
"loss": 0.7094186544418335,
"loss_sft": 0.7094186544418335,
"step": 140,
"token_acc": 0.788243006993007
},
{
"epoch": 0.5013333333333333,
"grad_norm": 2.11590313911438,
"learning_rate": 9.617530960630624e-06,
"loss": 0.6448466777801514,
"loss_sft": 0.6448466777801514,
"step": 141,
"token_acc": 0.8037015945330296
},
{
"epoch": 0.5048888888888889,
"grad_norm": 2.0922439098358154,
"learning_rate": 9.610020325776694e-06,
"loss": 0.6817845106124878,
"loss_sft": 0.6817845106124878,
"step": 142,
"token_acc": 0.7948980141004252
},
{
"epoch": 0.5084444444444445,
"grad_norm": 2.195615768432617,
"learning_rate": 9.602439653210426e-06,
"loss": 0.6335958242416382,
"loss_sft": 0.6335958242416382,
"step": 143,
"token_acc": 0.8032353118046786
},
{
"epoch": 0.512,
"grad_norm": 1.9804186820983887,
"learning_rate": 9.594789058101154e-06,
"loss": 0.6651124954223633,
"loss_sft": 0.6651124954223633,
"step": 144,
"token_acc": 0.7988301650302904
},
{
"epoch": 0.5155555555555555,
"grad_norm": 2.0500376224517822,
"learning_rate": 9.587068656680506e-06,
"loss": 0.6702920198440552,
"loss_sft": 0.6702920198440552,
"step": 145,
"token_acc": 0.7985091109884042
},
{
"epoch": 0.5191111111111111,
"grad_norm": 1.9637221097946167,
"learning_rate": 9.579278566240646e-06,
"loss": 0.663544774055481,
"loss_sft": 0.663544774055481,
"step": 146,
"token_acc": 0.7968317827508172
},
{
"epoch": 0.5226666666666666,
"grad_norm": 2.0351431369781494,
"learning_rate": 9.571418905132486e-06,
"loss": 0.6361479759216309,
"loss_sft": 0.6361479759216309,
"step": 147,
"token_acc": 0.8042215537807281
},
{
"epoch": 0.5262222222222223,
"grad_norm": 2.1730854511260986,
"learning_rate": 9.56348979276389e-06,
"loss": 0.6433050632476807,
"loss_sft": 0.6433050632476807,
"step": 148,
"token_acc": 0.8028481597928611
},
{
"epoch": 0.5297777777777778,
"grad_norm": 2.0182387828826904,
"learning_rate": 9.555491349597862e-06,
"loss": 0.6097121238708496,
"loss_sft": 0.6097121238708496,
"step": 149,
"token_acc": 0.8099349497338852
},
{
"epoch": 0.5333333333333333,
"grad_norm": 2.0734705924987793,
"learning_rate": 9.547423697150714e-06,
"loss": 0.6620005369186401,
"loss_sft": 0.6620005369186401,
"step": 150,
"token_acc": 0.7960768751062985
},
{
"epoch": 0.5368888888888889,
"grad_norm": 2.11378812789917,
"learning_rate": 9.539286957990215e-06,
"loss": 0.6536435484886169,
"loss_sft": 0.6536435484886169,
"step": 151,
"token_acc": 0.7974818230182656
},
{
"epoch": 0.5404444444444444,
"grad_norm": 2.0858607292175293,
"learning_rate": 9.53108125573374e-06,
"loss": 0.6139059066772461,
"loss_sft": 0.6139059066772461,
"step": 152,
"token_acc": 0.8129766581927432
},
{
"epoch": 0.544,
"grad_norm": 2.166447401046753,
"learning_rate": 9.52280671504638e-06,
"loss": 0.7133034467697144,
"loss_sft": 0.7133034467697144,
"step": 153,
"token_acc": 0.7866457187745484
},
{
"epoch": 0.5475555555555556,
"grad_norm": 2.184514045715332,
"learning_rate": 9.514463461639055e-06,
"loss": 0.6937499046325684,
"loss_sft": 0.6937499046325684,
"step": 154,
"token_acc": 0.787934682363153
},
{
"epoch": 0.5511111111111111,
"grad_norm": 2.1372385025024414,
"learning_rate": 9.506051622266608e-06,
"loss": 0.6579703092575073,
"loss_sft": 0.6579703092575073,
"step": 155,
"token_acc": 0.8003354734224073
},
{
"epoch": 0.5546666666666666,
"grad_norm": 2.1216418743133545,
"learning_rate": 9.497571324725865e-06,
"loss": 0.6675810813903809,
"loss_sft": 0.6675810813903809,
"step": 156,
"token_acc": 0.8022306542485915
},
{
"epoch": 0.5582222222222222,
"grad_norm": 2.00091290473938,
"learning_rate": 9.48902269785371e-06,
"loss": 0.697437047958374,
"loss_sft": 0.697437047958374,
"step": 157,
"token_acc": 0.7910054333207811
},
{
"epoch": 0.5617777777777778,
"grad_norm": 1.9830737113952637,
"learning_rate": 9.480405871525114e-06,
"loss": 0.6428194046020508,
"loss_sft": 0.6428194046020508,
"step": 158,
"token_acc": 0.8031120990790728
},
{
"epoch": 0.5653333333333334,
"grad_norm": 2.1295888423919678,
"learning_rate": 9.471720976651173e-06,
"loss": 0.6710842251777649,
"loss_sft": 0.6710842251777649,
"step": 159,
"token_acc": 0.8010291595197255
},
{
"epoch": 0.5688888888888889,
"grad_norm": 1.983245611190796,
"learning_rate": 9.462968145177112e-06,
"loss": 0.6440168619155884,
"loss_sft": 0.6440168619155884,
"step": 160,
"token_acc": 0.8022438134092007
},
{
"epoch": 0.5724444444444444,
"grad_norm": 2.134579658508301,
"learning_rate": 9.45414751008028e-06,
"loss": 0.6128056645393372,
"loss_sft": 0.6128056645393372,
"step": 161,
"token_acc": 0.8123558381647029
},
{
"epoch": 0.576,
"grad_norm": 2.35969614982605,
"learning_rate": 9.445259205368138e-06,
"loss": 0.659021258354187,
"loss_sft": 0.659021258354187,
"step": 162,
"token_acc": 0.8013130408833185
},
{
"epoch": 0.5795555555555556,
"grad_norm": 2.1321892738342285,
"learning_rate": 9.436303366076213e-06,
"loss": 0.6367583274841309,
"loss_sft": 0.6367583274841309,
"step": 163,
"token_acc": 0.8067535327583043
},
{
"epoch": 0.5831111111111111,
"grad_norm": 2.199120044708252,
"learning_rate": 9.427280128266049e-06,
"loss": 0.6567786931991577,
"loss_sft": 0.6567786931991577,
"step": 164,
"token_acc": 0.7986356151493766
},
{
"epoch": 0.5866666666666667,
"grad_norm": 2.1931631565093994,
"learning_rate": 9.418189629023149e-06,
"loss": 0.6438684463500977,
"loss_sft": 0.6438684463500977,
"step": 165,
"token_acc": 0.798000858527013
},
{
"epoch": 0.5902222222222222,
"grad_norm": 2.1690075397491455,
"learning_rate": 9.409032006454877e-06,
"loss": 0.6535179615020752,
"loss_sft": 0.6535179615020752,
"step": 166,
"token_acc": 0.8021361656423415
},
{
"epoch": 0.5937777777777777,
"grad_norm": 2.1426050662994385,
"learning_rate": 9.399807399688371e-06,
"loss": 0.6235388517379761,
"loss_sft": 0.6235388517379761,
"step": 167,
"token_acc": 0.8151931967353462
},
{
"epoch": 0.5973333333333334,
"grad_norm": 2.1936612129211426,
"learning_rate": 9.39051594886843e-06,
"loss": 0.6342451572418213,
"loss_sft": 0.6342451572418213,
"step": 168,
"token_acc": 0.8044199549709681
},
{
"epoch": 0.6008888888888889,
"grad_norm": 2.2286295890808105,
"learning_rate": 9.381157795155374e-06,
"loss": 0.7048568725585938,
"loss_sft": 0.7048568725585938,
"step": 169,
"token_acc": 0.7904340124003543
},
{
"epoch": 0.6044444444444445,
"grad_norm": 2.1595396995544434,
"learning_rate": 9.371733080722911e-06,
"loss": 0.6420679092407227,
"loss_sft": 0.6420679092407227,
"step": 170,
"token_acc": 0.8060135215276955
},
{
"epoch": 0.608,
"grad_norm": 2.0006065368652344,
"learning_rate": 9.36224194875597e-06,
"loss": 0.6651041507720947,
"loss_sft": 0.6651041507720947,
"step": 171,
"token_acc": 0.7975910993161172
},
{
"epoch": 0.6115555555555555,
"grad_norm": 2.109405040740967,
"learning_rate": 9.352684543448532e-06,
"loss": 0.6676123738288879,
"loss_sft": 0.6676123738288879,
"step": 172,
"token_acc": 0.7941523093327347
},
{
"epoch": 0.6151111111111112,
"grad_norm": 2.0356075763702393,
"learning_rate": 9.343061010001428e-06,
"loss": 0.6631208658218384,
"loss_sft": 0.6631208658218384,
"step": 173,
"token_acc": 0.7933867217773185
},
{
"epoch": 0.6186666666666667,
"grad_norm": 2.2451894283294678,
"learning_rate": 9.33337149462015e-06,
"loss": 0.5808688998222351,
"loss_sft": 0.5808688998222351,
"step": 174,
"token_acc": 0.8190159670544036
},
{
"epoch": 0.6222222222222222,
"grad_norm": 2.1735124588012695,
"learning_rate": 9.323616144512612e-06,
"loss": 0.650094747543335,
"loss_sft": 0.650094747543335,
"step": 175,
"token_acc": 0.8004563004563005
},
{
"epoch": 0.6257777777777778,
"grad_norm": 2.082838296890259,
"learning_rate": 9.313795107886925e-06,
"loss": 0.6318097114562988,
"loss_sft": 0.6318097114562988,
"step": 176,
"token_acc": 0.8082224909310762
},
{
"epoch": 0.6293333333333333,
"grad_norm": 1.9577778577804565,
"learning_rate": 9.303908533949146e-06,
"loss": 0.6116292476654053,
"loss_sft": 0.6116292476654053,
"step": 177,
"token_acc": 0.8136783325040085
},
{
"epoch": 0.6328888888888888,
"grad_norm": 1.940348505973816,
"learning_rate": 9.293956572900999e-06,
"loss": 0.6606093645095825,
"loss_sft": 0.6606093645095825,
"step": 178,
"token_acc": 0.7973079138951448
},
{
"epoch": 0.6364444444444445,
"grad_norm": 1.9303661584854126,
"learning_rate": 9.283939375937609e-06,
"loss": 0.6310110688209534,
"loss_sft": 0.6310110688209534,
"step": 179,
"token_acc": 0.8062589060616026
},
{
"epoch": 0.64,
"grad_norm": 2.0357861518859863,
"learning_rate": 9.273857095245192e-06,
"loss": 0.6442798972129822,
"loss_sft": 0.6442798972129822,
"step": 180,
"token_acc": 0.8027325676337707
},
{
"epoch": 0.6435555555555555,
"grad_norm": 1.9845647811889648,
"learning_rate": 9.263709883998753e-06,
"loss": 0.6144827008247375,
"loss_sft": 0.6144827008247375,
"step": 181,
"token_acc": 0.8132859979541489
},
{
"epoch": 0.6471111111111111,
"grad_norm": 2.0927419662475586,
"learning_rate": 9.253497896359749e-06,
"loss": 0.6944558620452881,
"loss_sft": 0.6944558620452881,
"step": 182,
"token_acc": 0.7908486299547752
},
{
"epoch": 0.6506666666666666,
"grad_norm": 2.098994731903076,
"learning_rate": 9.243221287473755e-06,
"loss": 0.6277338266372681,
"loss_sft": 0.6277338266372681,
"step": 183,
"token_acc": 0.802647412755716
},
{
"epoch": 0.6542222222222223,
"grad_norm": 1.9601129293441772,
"learning_rate": 9.232880213468106e-06,
"loss": 0.6347959041595459,
"loss_sft": 0.6347959041595459,
"step": 184,
"token_acc": 0.8064264139152172
},
{
"epoch": 0.6577777777777778,
"grad_norm": 2.09283447265625,
"learning_rate": 9.222474831449519e-06,
"loss": 0.669002890586853,
"loss_sft": 0.669002890586853,
"step": 185,
"token_acc": 0.7967453036260376
},
{
"epoch": 0.6613333333333333,
"grad_norm": 2.1455302238464355,
"learning_rate": 9.212005299501712e-06,
"loss": 0.6421587467193604,
"loss_sft": 0.6421587467193604,
"step": 186,
"token_acc": 0.8034950507148967
},
{
"epoch": 0.6648888888888889,
"grad_norm": 2.1961469650268555,
"learning_rate": 9.201471776682999e-06,
"loss": 0.6340146660804749,
"loss_sft": 0.6340146660804749,
"step": 187,
"token_acc": 0.8066413662239089
},
{
"epoch": 0.6684444444444444,
"grad_norm": 2.169645309448242,
"learning_rate": 9.19087442302388e-06,
"loss": 0.6266807317733765,
"loss_sft": 0.6266807317733765,
"step": 188,
"token_acc": 0.8067930936880838
},
{
"epoch": 0.672,
"grad_norm": 1.990903615951538,
"learning_rate": 9.180213399524599e-06,
"loss": 0.6344912648200989,
"loss_sft": 0.6344912648200989,
"step": 189,
"token_acc": 0.8035234899328859
},
{
"epoch": 0.6755555555555556,
"grad_norm": 2.0581541061401367,
"learning_rate": 9.169488868152704e-06,
"loss": 0.6445058584213257,
"loss_sft": 0.6445058584213257,
"step": 190,
"token_acc": 0.7947607825943991
},
{
"epoch": 0.6791111111111111,
"grad_norm": 2.0418686866760254,
"learning_rate": 9.15870099184059e-06,
"loss": 0.6409372687339783,
"loss_sft": 0.6409372687339783,
"step": 191,
"token_acc": 0.8020778614369929
},
{
"epoch": 0.6826666666666666,
"grad_norm": 1.9903651475906372,
"learning_rate": 9.147849934483019e-06,
"loss": 0.6447004079818726,
"loss_sft": 0.6447004079818726,
"step": 192,
"token_acc": 0.8021308113035551
},
{
"epoch": 0.6862222222222222,
"grad_norm": 2.2193522453308105,
"learning_rate": 9.136935860934628e-06,
"loss": 0.6359076499938965,
"loss_sft": 0.6359076499938965,
"step": 193,
"token_acc": 0.8066659096070861
},
{
"epoch": 0.6897777777777778,
"grad_norm": 2.0499746799468994,
"learning_rate": 9.125958937007427e-06,
"loss": 0.6424558162689209,
"loss_sft": 0.6424558162689209,
"step": 194,
"token_acc": 0.8034006376195537
},
{
"epoch": 0.6933333333333334,
"grad_norm": 1.9909120798110962,
"learning_rate": 9.114919329468283e-06,
"loss": 0.6117143630981445,
"loss_sft": 0.6117143630981445,
"step": 195,
"token_acc": 0.8090642770980527
},
{
"epoch": 0.6968888888888889,
"grad_norm": 2.161841630935669,
"learning_rate": 9.103817206036383e-06,
"loss": 0.6553783416748047,
"loss_sft": 0.6553783416748047,
"step": 196,
"token_acc": 0.800908540278619
},
{
"epoch": 0.7004444444444444,
"grad_norm": 2.1325244903564453,
"learning_rate": 9.092652735380683e-06,
"loss": 0.615043044090271,
"loss_sft": 0.615043044090271,
"step": 197,
"token_acc": 0.8108505296675035
},
{
"epoch": 0.704,
"grad_norm": 2.2787342071533203,
"learning_rate": 9.081426087117356e-06,
"loss": 0.6704057455062866,
"loss_sft": 0.6704057455062866,
"step": 198,
"token_acc": 0.7961885537876989
},
{
"epoch": 0.7075555555555556,
"grad_norm": 2.039710283279419,
"learning_rate": 9.0701374318072e-06,
"loss": 0.6609984636306763,
"loss_sft": 0.6609984636306763,
"step": 199,
"token_acc": 0.7982543365373992
},
{
"epoch": 0.7111111111111111,
"grad_norm": 2.134953260421753,
"learning_rate": 9.058786940953057e-06,
"loss": 0.6441085338592529,
"loss_sft": 0.6441085338592529,
"step": 200,
"token_acc": 0.8000582920431362
},
{
"epoch": 0.7146666666666667,
"grad_norm": 2.0990445613861084,
"learning_rate": 9.04737478699721e-06,
"loss": 0.6577693819999695,
"loss_sft": 0.6577693819999695,
"step": 201,
"token_acc": 0.8014115216892357
},
{
"epoch": 0.7182222222222222,
"grad_norm": 2.0252952575683594,
"learning_rate": 9.035901143318753e-06,
"loss": 0.6100600361824036,
"loss_sft": 0.6100600361824036,
"step": 202,
"token_acc": 0.810319554382879
},
{
"epoch": 0.7217777777777777,
"grad_norm": 2.1375560760498047,
"learning_rate": 9.024366184230964e-06,
"loss": 0.6831796765327454,
"loss_sft": 0.6831796765327454,
"step": 203,
"token_acc": 0.7924018621759871
},
{
"epoch": 0.7253333333333334,
"grad_norm": 2.042707681655884,
"learning_rate": 9.012770084978657e-06,
"loss": 0.6535006761550903,
"loss_sft": 0.6535006761550903,
"step": 204,
"token_acc": 0.8003682419237851
},
{
"epoch": 0.7288888888888889,
"grad_norm": 1.9784481525421143,
"learning_rate": 9.001113021735512e-06,
"loss": 0.5847540497779846,
"loss_sft": 0.5847540497779846,
"step": 205,
"token_acc": 0.8174256965043913
},
{
"epoch": 0.7324444444444445,
"grad_norm": 2.0748870372772217,
"learning_rate": 8.989395171601413e-06,
"loss": 0.6814305782318115,
"loss_sft": 0.6814305782318115,
"step": 206,
"token_acc": 0.7949538024164889
},
{
"epoch": 0.736,
"grad_norm": 1.9863920211791992,
"learning_rate": 8.97761671259974e-06,
"loss": 0.6189029216766357,
"loss_sft": 0.6189029216766357,
"step": 207,
"token_acc": 0.8083671520649319
},
{
"epoch": 0.7395555555555555,
"grad_norm": 1.9961540699005127,
"learning_rate": 8.965777823674679e-06,
"loss": 0.640105128288269,
"loss_sft": 0.640105128288269,
"step": 208,
"token_acc": 0.8055312157721797
},
{
"epoch": 0.7431111111111111,
"grad_norm": 2.0450522899627686,
"learning_rate": 8.953878684688492e-06,
"loss": 0.6534634232521057,
"loss_sft": 0.6534634232521057,
"step": 209,
"token_acc": 0.7987610619469027
},
{
"epoch": 0.7466666666666667,
"grad_norm": 2.0796287059783936,
"learning_rate": 8.941919476418797e-06,
"loss": 0.6454188823699951,
"loss_sft": 0.6454188823699951,
"step": 210,
"token_acc": 0.8024406332453826
},
{
"epoch": 0.7502222222222222,
"grad_norm": 2.0922389030456543,
"learning_rate": 8.929900380555805e-06,
"loss": 0.710419237613678,
"loss_sft": 0.710419237613678,
"step": 211,
"token_acc": 0.7875075206475961
},
{
"epoch": 0.7537777777777778,
"grad_norm": 2.1682066917419434,
"learning_rate": 8.917821579699578e-06,
"loss": 0.6921146512031555,
"loss_sft": 0.6921146512031555,
"step": 212,
"token_acc": 0.793841642228739
},
{
"epoch": 0.7573333333333333,
"grad_norm": 1.9257030487060547,
"learning_rate": 8.905683257357238e-06,
"loss": 0.6323363184928894,
"loss_sft": 0.6323363184928894,
"step": 213,
"token_acc": 0.8046088600480302
},
{
"epoch": 0.7608888888888888,
"grad_norm": 1.9910633563995361,
"learning_rate": 8.893485597940195e-06,
"loss": 0.6313146948814392,
"loss_sft": 0.6313146948814392,
"step": 214,
"token_acc": 0.8061831153388823
},
{
"epoch": 0.7644444444444445,
"grad_norm": 2.0893025398254395,
"learning_rate": 8.881228786761329e-06,
"loss": 0.6637563109397888,
"loss_sft": 0.6637563109397888,
"step": 215,
"token_acc": 0.7985502542464568
},
{
"epoch": 0.768,
"grad_norm": 1.9198391437530518,
"learning_rate": 8.868913010032188e-06,
"loss": 0.6042695045471191,
"loss_sft": 0.6042695045471191,
"step": 216,
"token_acc": 0.814913524241565
},
{
"epoch": 0.7715555555555556,
"grad_norm": 2.0637130737304688,
"learning_rate": 8.856538454860155e-06,
"loss": 0.6680890321731567,
"loss_sft": 0.6680890321731567,
"step": 217,
"token_acc": 0.798783165025403
},
{
"epoch": 0.7751111111111111,
"grad_norm": 2.116689920425415,
"learning_rate": 8.8441053092456e-06,
"loss": 0.6844155788421631,
"loss_sft": 0.6844155788421631,
"step": 218,
"token_acc": 0.792506370164466
},
{
"epoch": 0.7786666666666666,
"grad_norm": 2.026621103286743,
"learning_rate": 8.831613762079026e-06,
"loss": 0.6676284670829773,
"loss_sft": 0.6676284670829773,
"step": 219,
"token_acc": 0.8025002422715379
},
{
"epoch": 0.7822222222222223,
"grad_norm": 1.9405319690704346,
"learning_rate": 8.819064003138211e-06,
"loss": 0.6332445740699768,
"loss_sft": 0.6332445740699768,
"step": 220,
"token_acc": 0.805747794389561
},
{
"epoch": 0.7857777777777778,
"grad_norm": 2.1476385593414307,
"learning_rate": 8.806456223085308e-06,
"loss": 0.6292698383331299,
"loss_sft": 0.6292698383331299,
"step": 221,
"token_acc": 0.8039263356226828
},
{
"epoch": 0.7893333333333333,
"grad_norm": 2.0222296714782715,
"learning_rate": 8.793790613463956e-06,
"loss": 0.6251882910728455,
"loss_sft": 0.6251882910728455,
"step": 222,
"token_acc": 0.8049246683157185
},
{
"epoch": 0.7928888888888889,
"grad_norm": 2.0721542835235596,
"learning_rate": 8.781067366696368e-06,
"loss": 0.6587234735488892,
"loss_sft": 0.6587234735488892,
"step": 223,
"token_acc": 0.7976418482164233
},
{
"epoch": 0.7964444444444444,
"grad_norm": 2.000278949737549,
"learning_rate": 8.768286676080415e-06,
"loss": 0.5966362953186035,
"loss_sft": 0.5966362953186035,
"step": 224,
"token_acc": 0.8129665302191187
},
{
"epoch": 0.8,
"grad_norm": 1.9993627071380615,
"learning_rate": 8.75544873578668e-06,
"loss": 0.6348760724067688,
"loss_sft": 0.6348760724067688,
"step": 225,
"token_acc": 0.8057276838050578
},
{
"epoch": 0.8035555555555556,
"grad_norm": 2.0028460025787354,
"learning_rate": 8.742553740855507e-06,
"loss": 0.6157115697860718,
"loss_sft": 0.6157115697860718,
"step": 226,
"token_acc": 0.8112164756530339
},
{
"epoch": 0.8071111111111111,
"grad_norm": 2.0482850074768066,
"learning_rate": 8.729601887194048e-06,
"loss": 0.6731700897216797,
"loss_sft": 0.6731700897216797,
"step": 227,
"token_acc": 0.7923266052947567
},
{
"epoch": 0.8106666666666666,
"grad_norm": 2.022529125213623,
"learning_rate": 8.71659337157328e-06,
"loss": 0.6152519583702087,
"loss_sft": 0.6152519583702087,
"step": 228,
"token_acc": 0.8080425783560024
},
{
"epoch": 0.8142222222222222,
"grad_norm": 2.013263463973999,
"learning_rate": 8.70352839162502e-06,
"loss": 0.630630373954773,
"loss_sft": 0.630630373954773,
"step": 229,
"token_acc": 0.8056728310638045
},
{
"epoch": 0.8177777777777778,
"grad_norm": 2.1523144245147705,
"learning_rate": 8.690407145838909e-06,
"loss": 0.680114209651947,
"loss_sft": 0.680114209651947,
"step": 230,
"token_acc": 0.7924459778463774
},
{
"epoch": 0.8213333333333334,
"grad_norm": 1.9573442935943604,
"learning_rate": 8.677229833559413e-06,
"loss": 0.6561951041221619,
"loss_sft": 0.6561951041221619,
"step": 231,
"token_acc": 0.8001333128236682
},
{
"epoch": 0.8248888888888889,
"grad_norm": 1.9068033695220947,
"learning_rate": 8.663996654982791e-06,
"loss": 0.6676413416862488,
"loss_sft": 0.6676413416862488,
"step": 232,
"token_acc": 0.7988162581185837
},
{
"epoch": 0.8284444444444444,
"grad_norm": 1.8816215991973877,
"learning_rate": 8.650707811154045e-06,
"loss": 0.5788356065750122,
"loss_sft": 0.5788356065750122,
"step": 233,
"token_acc": 0.8179523946085461
},
{
"epoch": 0.832,
"grad_norm": 1.9978104829788208,
"learning_rate": 8.637363503963873e-06,
"loss": 0.6836059093475342,
"loss_sft": 0.6836059093475342,
"step": 234,
"token_acc": 0.7910746123507483
},
{
"epoch": 0.8355555555555556,
"grad_norm": 1.9295011758804321,
"learning_rate": 8.6239639361456e-06,
"loss": 0.6758168935775757,
"loss_sft": 0.6758168935775757,
"step": 235,
"token_acc": 0.7973735562470334
},
{
"epoch": 0.8391111111111111,
"grad_norm": 2.05021071434021,
"learning_rate": 8.610509311272099e-06,
"loss": 0.6622954607009888,
"loss_sft": 0.6622954607009888,
"step": 236,
"token_acc": 0.7970913008349044
},
{
"epoch": 0.8426666666666667,
"grad_norm": 2.044762134552002,
"learning_rate": 8.596999833752694e-06,
"loss": 0.6156852841377258,
"loss_sft": 0.6156852841377258,
"step": 237,
"token_acc": 0.8131106988249845
},
{
"epoch": 0.8462222222222222,
"grad_norm": 1.9513144493103027,
"learning_rate": 8.583435708830058e-06,
"loss": 0.6836113929748535,
"loss_sft": 0.6836113929748535,
"step": 238,
"token_acc": 0.792070707070707
},
{
"epoch": 0.8497777777777777,
"grad_norm": 2.0153276920318604,
"learning_rate": 8.569817142577099e-06,
"loss": 0.6637976765632629,
"loss_sft": 0.6637976765632629,
"step": 239,
"token_acc": 0.7986599297012302
},
{
"epoch": 0.8533333333333334,
"grad_norm": 1.8882949352264404,
"learning_rate": 8.556144341893819e-06,
"loss": 0.5988000631332397,
"loss_sft": 0.5988000631332397,
"step": 240,
"token_acc": 0.8124131294771731
},
{
"epoch": 0.8568888888888889,
"grad_norm": 1.911758303642273,
"learning_rate": 8.54241751450418e-06,
"loss": 0.6740436553955078,
"loss_sft": 0.6740436553955078,
"step": 241,
"token_acc": 0.7965611761774234
},
{
"epoch": 0.8604444444444445,
"grad_norm": 1.9943009614944458,
"learning_rate": 8.528636868952944e-06,
"loss": 0.655754804611206,
"loss_sft": 0.655754804611206,
"step": 242,
"token_acc": 0.8008454106280193
},
{
"epoch": 0.864,
"grad_norm": 2.186279058456421,
"learning_rate": 8.514802614602503e-06,
"loss": 0.6526083946228027,
"loss_sft": 0.6526083946228027,
"step": 243,
"token_acc": 0.7990930708400155
},
{
"epoch": 0.8675555555555555,
"grad_norm": 1.9700061082839966,
"learning_rate": 8.5009149616297e-06,
"loss": 0.6805433630943298,
"loss_sft": 0.6805433630943298,
"step": 244,
"token_acc": 0.7921971922951355
},
{
"epoch": 0.8711111111111111,
"grad_norm": 2.0039520263671875,
"learning_rate": 8.486974121022642e-06,
"loss": 0.63426673412323,
"loss_sft": 0.63426673412323,
"step": 245,
"token_acc": 0.8073060941828255
},
{
"epoch": 0.8746666666666667,
"grad_norm": 2.001971960067749,
"learning_rate": 8.472980304577483e-06,
"loss": 0.641181230545044,
"loss_sft": 0.641181230545044,
"step": 246,
"token_acc": 0.8014227891573654
},
{
"epoch": 0.8782222222222222,
"grad_norm": 1.9824213981628418,
"learning_rate": 8.458933724895216e-06,
"loss": 0.6483989953994751,
"loss_sft": 0.6483989953994751,
"step": 247,
"token_acc": 0.8027881672900374
},
{
"epoch": 0.8817777777777778,
"grad_norm": 1.9733328819274902,
"learning_rate": 8.444834595378434e-06,
"loss": 0.6219915151596069,
"loss_sft": 0.6219915151596069,
"step": 248,
"token_acc": 0.8064610011641443
},
{
"epoch": 0.8853333333333333,
"grad_norm": 1.9841550588607788,
"learning_rate": 8.4306831302281e-06,
"loss": 0.6062391996383667,
"loss_sft": 0.6062391996383667,
"step": 249,
"token_acc": 0.8100934153565732
},
{
"epoch": 0.8888888888888888,
"grad_norm": 2.093348979949951,
"learning_rate": 8.416479544440279e-06,
"loss": 0.6224026679992676,
"loss_sft": 0.6224026679992676,
"step": 250,
"token_acc": 0.8074651547365934
},
{
"epoch": 0.8924444444444445,
"grad_norm": 1.9712640047073364,
"learning_rate": 8.402224053802884e-06,
"loss": 0.6228599548339844,
"loss_sft": 0.6228599548339844,
"step": 251,
"token_acc": 0.806829268292683
},
{
"epoch": 0.896,
"grad_norm": 1.9803690910339355,
"learning_rate": 8.38791687489239e-06,
"loss": 0.6189337968826294,
"loss_sft": 0.6189337968826294,
"step": 252,
"token_acc": 0.8092676640331709
},
{
"epoch": 0.8995555555555556,
"grad_norm": 2.0156350135803223,
"learning_rate": 8.373558225070546e-06,
"loss": 0.6873805522918701,
"loss_sft": 0.6873805522918701,
"step": 253,
"token_acc": 0.7929888507718696
},
{
"epoch": 0.9031111111111111,
"grad_norm": 1.9648245573043823,
"learning_rate": 8.359148322481073e-06,
"loss": 0.6371824145317078,
"loss_sft": 0.6371824145317078,
"step": 254,
"token_acc": 0.8062453188915135
},
{
"epoch": 0.9066666666666666,
"grad_norm": 1.9671170711517334,
"learning_rate": 8.344687386046348e-06,
"loss": 0.6354299783706665,
"loss_sft": 0.6354299783706665,
"step": 255,
"token_acc": 0.8036082746944606
},
{
"epoch": 0.9102222222222223,
"grad_norm": 2.2807400226593018,
"learning_rate": 8.330175635464082e-06,
"loss": 0.6177355051040649,
"loss_sft": 0.6177355051040649,
"step": 256,
"token_acc": 0.814567821301392
},
{
"epoch": 0.9137777777777778,
"grad_norm": 1.985668659210205,
"learning_rate": 8.315613291203977e-06,
"loss": 0.6435043811798096,
"loss_sft": 0.6435043811798096,
"step": 257,
"token_acc": 0.8027281063732233
},
{
"epoch": 0.9173333333333333,
"grad_norm": 1.9402862787246704,
"learning_rate": 8.30100057450438e-06,
"loss": 0.6300252676010132,
"loss_sft": 0.6300252676010132,
"step": 258,
"token_acc": 0.805009633911368
},
{
"epoch": 0.9208888888888889,
"grad_norm": 1.9983580112457275,
"learning_rate": 8.286337707368922e-06,
"loss": 0.6859633922576904,
"loss_sft": 0.6859633922576904,
"step": 259,
"token_acc": 0.79475234270415
},
{
"epoch": 0.9244444444444444,
"grad_norm": 1.9765851497650146,
"learning_rate": 8.271624912563143e-06,
"loss": 0.5987871885299683,
"loss_sft": 0.5987871885299683,
"step": 260,
"token_acc": 0.8132356473696853
},
{
"epoch": 0.928,
"grad_norm": 1.9230178594589233,
"learning_rate": 8.256862413611113e-06,
"loss": 0.6346046924591064,
"loss_sft": 0.6346046924591064,
"step": 261,
"token_acc": 0.8040879290397224
},
{
"epoch": 0.9315555555555556,
"grad_norm": 1.9823640584945679,
"learning_rate": 8.242050434792022e-06,
"loss": 0.6048730611801147,
"loss_sft": 0.6048730611801147,
"step": 262,
"token_acc": 0.8079008997904598
},
{
"epoch": 0.9351111111111111,
"grad_norm": 1.9911458492279053,
"learning_rate": 8.227189201136796e-06,
"loss": 0.6229594945907593,
"loss_sft": 0.6229594945907593,
"step": 263,
"token_acc": 0.8070793394156369
},
{
"epoch": 0.9386666666666666,
"grad_norm": 2.022102117538452,
"learning_rate": 8.212278938424654e-06,
"loss": 0.6361520290374756,
"loss_sft": 0.6361520290374756,
"step": 264,
"token_acc": 0.8019678805700068
},
{
"epoch": 0.9422222222222222,
"grad_norm": 1.8416802883148193,
"learning_rate": 8.197319873179694e-06,
"loss": 0.6217108964920044,
"loss_sft": 0.6217108964920044,
"step": 265,
"token_acc": 0.8058184177797004
},
{
"epoch": 0.9457777777777778,
"grad_norm": 1.9203720092773438,
"learning_rate": 8.182312232667446e-06,
"loss": 0.6299636363983154,
"loss_sft": 0.6299636363983154,
"step": 266,
"token_acc": 0.8090100111234705
},
{
"epoch": 0.9493333333333334,
"grad_norm": 1.9958794116973877,
"learning_rate": 8.167256244891416e-06,
"loss": 0.6250180602073669,
"loss_sft": 0.6250180602073669,
"step": 267,
"token_acc": 0.806598744224618
},
{
"epoch": 0.9528888888888889,
"grad_norm": 2.0192413330078125,
"learning_rate": 8.152152138589633e-06,
"loss": 0.6231001019477844,
"loss_sft": 0.6231001019477844,
"step": 268,
"token_acc": 0.8057962858750704
},
{
"epoch": 0.9564444444444444,
"grad_norm": 2.0096864700317383,
"learning_rate": 8.137000143231156e-06,
"loss": 0.6231207847595215,
"loss_sft": 0.6231207847595215,
"step": 269,
"token_acc": 0.8096854995825216
},
{
"epoch": 0.96,
"grad_norm": 2.020512104034424,
"learning_rate": 8.121800489012608e-06,
"loss": 0.5984187126159668,
"loss_sft": 0.5984187126159668,
"step": 270,
"token_acc": 0.8176981541802388
},
{
"epoch": 0.9635555555555556,
"grad_norm": 1.9126023054122925,
"learning_rate": 8.106553406854664e-06,
"loss": 0.6417933702468872,
"loss_sft": 0.6417933702468872,
"step": 271,
"token_acc": 0.8031815512731544
},
{
"epoch": 0.9671111111111111,
"grad_norm": 2.1562108993530273,
"learning_rate": 8.091259128398548e-06,
"loss": 0.6182734370231628,
"loss_sft": 0.6182734370231628,
"step": 272,
"token_acc": 0.8112527716186253
},
{
"epoch": 0.9706666666666667,
"grad_norm": 2.0692408084869385,
"learning_rate": 8.075917886002514e-06,
"loss": 0.6280180215835571,
"loss_sft": 0.6280180215835571,
"step": 273,
"token_acc": 0.8055349923125107
},
{
"epoch": 0.9742222222222222,
"grad_norm": 2.059676170349121,
"learning_rate": 8.060529912738316e-06,
"loss": 0.6234342455863953,
"loss_sft": 0.6234342455863953,
"step": 274,
"token_acc": 0.8058777531604326
},
{
"epoch": 0.9777777777777777,
"grad_norm": 2.004762649536133,
"learning_rate": 8.045095442387666e-06,
"loss": 0.6629883050918579,
"loss_sft": 0.6629883050918579,
"step": 275,
"token_acc": 0.7998265394622723
},
{
"epoch": 0.9813333333333333,
"grad_norm": 1.9370667934417725,
"learning_rate": 8.029614709438684e-06,
"loss": 0.6546408534049988,
"loss_sft": 0.6546408534049988,
"step": 276,
"token_acc": 0.8045176720701568
},
{
"epoch": 0.9848888888888889,
"grad_norm": 1.982785940170288,
"learning_rate": 8.014087949082333e-06,
"loss": 0.6044105291366577,
"loss_sft": 0.6044105291366577,
"step": 277,
"token_acc": 0.8108814589665654
},
{
"epoch": 0.9884444444444445,
"grad_norm": 1.9405336380004883,
"learning_rate": 7.998515397208846e-06,
"loss": 0.633181095123291,
"loss_sft": 0.633181095123291,
"step": 278,
"token_acc": 0.8021648332254838
},
{
"epoch": 0.992,
"grad_norm": 2.1792025566101074,
"learning_rate": 7.982897290404146e-06,
"loss": 0.5948256254196167,
"loss_sft": 0.5948256254196167,
"step": 279,
"token_acc": 0.8133802816901409
},
{
"epoch": 0.9955555555555555,
"grad_norm": 1.9870167970657349,
"learning_rate": 7.967233865946249e-06,
"loss": 0.6368255019187927,
"loss_sft": 0.6368255019187927,
"step": 280,
"token_acc": 0.7998420844848007
},
{
"epoch": 0.9991111111111111,
"grad_norm": 1.9441301822662354,
"learning_rate": 7.951525361801655e-06,
"loss": 0.5970394611358643,
"loss_sft": 0.5970394611358643,
"step": 281,
"token_acc": 0.8161379388065588
},
{
"epoch": 1.0,
"grad_norm": 3.6429951190948486,
"learning_rate": 7.935772016621744e-06,
"loss": 0.6557987332344055,
"loss_sft": 0.6557987332344055,
"step": 282,
"token_acc": 0.808817384036774
},
{
"epoch": 1.0035555555555555,
"grad_norm": 1.8796638250350952,
"learning_rate": 7.919974069739136e-06,
"loss": 0.45198220014572144,
"loss_sft": 0.45198220014572144,
"step": 283,
"token_acc": 0.8566094100074683
},
{
"epoch": 1.007111111111111,
"grad_norm": 1.9316012859344482,
"learning_rate": 7.904131761164068e-06,
"loss": 0.45736163854599,
"loss_sft": 0.45736163854599,
"step": 284,
"token_acc": 0.8534147695724598
},
{
"epoch": 1.0106666666666666,
"grad_norm": 1.8426058292388916,
"learning_rate": 7.888245331580737e-06,
"loss": 0.4319244623184204,
"loss_sft": 0.4319244623184204,
"step": 285,
"token_acc": 0.8551612723525152
},
{
"epoch": 1.0142222222222221,
"grad_norm": 1.8539133071899414,
"learning_rate": 7.872315022343654e-06,
"loss": 0.4552215337753296,
"loss_sft": 0.4552215337753296,
"step": 286,
"token_acc": 0.8531102384611007
},
{
"epoch": 1.0177777777777777,
"grad_norm": 1.9627256393432617,
"learning_rate": 7.856341075473963e-06,
"loss": 0.4469631612300873,
"loss_sft": 0.4469631612300873,
"step": 287,
"token_acc": 0.8565242832065535
},
{
"epoch": 1.0213333333333334,
"grad_norm": 2.003913164138794,
"learning_rate": 7.84032373365578e-06,
"loss": 0.446913480758667,
"loss_sft": 0.446913480758667,
"step": 288,
"token_acc": 0.8570271583607797
},
{
"epoch": 1.024888888888889,
"grad_norm": 2.0955398082733154,
"learning_rate": 7.824263240232497e-06,
"loss": 0.42114806175231934,
"loss_sft": 0.42114806175231934,
"step": 289,
"token_acc": 0.8642726719415703
},
{
"epoch": 1.0284444444444445,
"grad_norm": 2.32963490486145,
"learning_rate": 7.808159839203085e-06,
"loss": 0.43851760029792786,
"loss_sft": 0.43851760029792786,
"step": 290,
"token_acc": 0.8581108472809406
},
{
"epoch": 1.032,
"grad_norm": 2.4058187007904053,
"learning_rate": 7.792013775218385e-06,
"loss": 0.4426143765449524,
"loss_sft": 0.4426143765449524,
"step": 291,
"token_acc": 0.8589058758519065
},
{
"epoch": 1.0355555555555556,
"grad_norm": 2.1952006816864014,
"learning_rate": 7.775825293577407e-06,
"loss": 0.4106180667877197,
"loss_sft": 0.4106180667877197,
"step": 292,
"token_acc": 0.866868042736477
},
{
"epoch": 1.039111111111111,
"grad_norm": 2.399146556854248,
"learning_rate": 7.759594640223576e-06,
"loss": 0.4342702031135559,
"loss_sft": 0.4342702031135559,
"step": 293,
"token_acc": 0.8569778188539742
},
{
"epoch": 1.0426666666666666,
"grad_norm": 2.259556770324707,
"learning_rate": 7.743322061741024e-06,
"loss": 0.4169110655784607,
"loss_sft": 0.4169110655784607,
"step": 294,
"token_acc": 0.8623610375860243
},
{
"epoch": 1.0462222222222222,
"grad_norm": 2.184926986694336,
"learning_rate": 7.727007805350815e-06,
"loss": 0.4324934482574463,
"loss_sft": 0.4324934482574463,
"step": 295,
"token_acc": 0.8581070136601695
},
{
"epoch": 1.0497777777777777,
"grad_norm": 2.299448251724243,
"learning_rate": 7.71065211890722e-06,
"loss": 0.4577699303627014,
"loss_sft": 0.4577699303627014,
"step": 296,
"token_acc": 0.8515471512770137
},
{
"epoch": 1.0533333333333332,
"grad_norm": 1.9239484071731567,
"learning_rate": 7.694255250893927e-06,
"loss": 0.41096746921539307,
"loss_sft": 0.41096746921539307,
"step": 297,
"token_acc": 0.8665798384996093
},
{
"epoch": 1.056888888888889,
"grad_norm": 2.0346615314483643,
"learning_rate": 7.677817450420279e-06,
"loss": 0.41789641976356506,
"loss_sft": 0.41789641976356506,
"step": 298,
"token_acc": 0.864607190412783
},
{
"epoch": 1.0604444444444445,
"grad_norm": 2.03113055229187,
"learning_rate": 7.661338967217483e-06,
"loss": 0.4331536889076233,
"loss_sft": 0.4331536889076233,
"step": 299,
"token_acc": 0.8597536522486393
},
{
"epoch": 1.064,
"grad_norm": 2.095858097076416,
"learning_rate": 7.644820051634813e-06,
"loss": 0.3948267698287964,
"loss_sft": 0.3948267698287964,
"step": 300,
"token_acc": 0.8748241912798875
},
{
"epoch": 1.0675555555555556,
"grad_norm": 2.001725673675537,
"learning_rate": 7.628260954635822e-06,
"loss": 0.4145166575908661,
"loss_sft": 0.4145166575908661,
"step": 301,
"token_acc": 0.8624406590337894
},
{
"epoch": 1.0711111111111111,
"grad_norm": 2.0921318531036377,
"learning_rate": 7.6116619277945135e-06,
"loss": 0.4647698998451233,
"loss_sft": 0.4647698998451233,
"step": 302,
"token_acc": 0.8554048984404309
},
{
"epoch": 1.0746666666666667,
"grad_norm": 2.232443332672119,
"learning_rate": 7.595023223291525e-06,
"loss": 0.4536513388156891,
"loss_sft": 0.4536513388156891,
"step": 303,
"token_acc": 0.8545784224841342
},
{
"epoch": 1.0782222222222222,
"grad_norm": 2.04292893409729,
"learning_rate": 7.578345093910298e-06,
"loss": 0.4145389795303345,
"loss_sft": 0.4145389795303345,
"step": 304,
"token_acc": 0.8660278943190838
},
{
"epoch": 1.0817777777777777,
"grad_norm": 2.215210437774658,
"learning_rate": 7.561627793033233e-06,
"loss": 0.41918593645095825,
"loss_sft": 0.41918593645095825,
"step": 305,
"token_acc": 0.8641311147253847
},
{
"epoch": 1.0853333333333333,
"grad_norm": 2.0715889930725098,
"learning_rate": 7.54487157463785e-06,
"loss": 0.4150959849357605,
"loss_sft": 0.4150959849357605,
"step": 306,
"token_acc": 0.8655424141248373
},
{
"epoch": 1.0888888888888888,
"grad_norm": 2.0739564895629883,
"learning_rate": 7.528076693292914e-06,
"loss": 0.47890666127204895,
"loss_sft": 0.47890666127204895,
"step": 307,
"token_acc": 0.8511529340531813
},
{
"epoch": 1.0924444444444443,
"grad_norm": 2.01745867729187,
"learning_rate": 7.511243404154586e-06,
"loss": 0.4373610317707062,
"loss_sft": 0.4373610317707062,
"step": 308,
"token_acc": 0.8587376790284812
},
{
"epoch": 1.096,
"grad_norm": 2.0189902782440186,
"learning_rate": 7.49437196296253e-06,
"loss": 0.44800865650177,
"loss_sft": 0.44800865650177,
"step": 309,
"token_acc": 0.8522212148685403
},
{
"epoch": 1.0995555555555556,
"grad_norm": 1.9877958297729492,
"learning_rate": 7.4774626260360384e-06,
"loss": 0.41285446286201477,
"loss_sft": 0.41285446286201477,
"step": 310,
"token_acc": 0.8651653525441034
},
{
"epoch": 1.1031111111111112,
"grad_norm": 1.9735313653945923,
"learning_rate": 7.460515650270134e-06,
"loss": 0.39050233364105225,
"loss_sft": 0.39050233364105225,
"step": 311,
"token_acc": 0.8713266416737211
},
{
"epoch": 1.1066666666666667,
"grad_norm": 2.1215789318084717,
"learning_rate": 7.443531293131667e-06,
"loss": 0.43831193447113037,
"loss_sft": 0.43831193447113037,
"step": 312,
"token_acc": 0.8566848154869934
},
{
"epoch": 1.1102222222222222,
"grad_norm": 2.0347213745117188,
"learning_rate": 7.4265098126554065e-06,
"loss": 0.4095308780670166,
"loss_sft": 0.4095308780670166,
"step": 313,
"token_acc": 0.871372502378687
},
{
"epoch": 1.1137777777777778,
"grad_norm": 2.122067451477051,
"learning_rate": 7.409451467440111e-06,
"loss": 0.3899034559726715,
"loss_sft": 0.3899034559726715,
"step": 314,
"token_acc": 0.8731713442581779
},
{
"epoch": 1.1173333333333333,
"grad_norm": 2.079582929611206,
"learning_rate": 7.392356516644614e-06,
"loss": 0.42709362506866455,
"loss_sft": 0.42709362506866455,
"step": 315,
"token_acc": 0.865532414197712
},
{
"epoch": 1.1208888888888888,
"grad_norm": 2.2067623138427734,
"learning_rate": 7.375225219983876e-06,
"loss": 0.43742042779922485,
"loss_sft": 0.43742042779922485,
"step": 316,
"token_acc": 0.8600528102602791
},
{
"epoch": 1.1244444444444444,
"grad_norm": 2.0631027221679688,
"learning_rate": 7.358057837725039e-06,
"loss": 0.4304013252258301,
"loss_sft": 0.4304013252258301,
"step": 317,
"token_acc": 0.8630193977845961
},
{
"epoch": 1.1280000000000001,
"grad_norm": 2.1489007472991943,
"learning_rate": 7.340854630683481e-06,
"loss": 0.4300694167613983,
"loss_sft": 0.4300694167613983,
"step": 318,
"token_acc": 0.8608665630933886
},
{
"epoch": 1.1315555555555556,
"grad_norm": 2.2351152896881104,
"learning_rate": 7.323615860218844e-06,
"loss": 0.4287225008010864,
"loss_sft": 0.4287225008010864,
"step": 319,
"token_acc": 0.8593641362745718
},
{
"epoch": 1.1351111111111112,
"grad_norm": 2.0943422317504883,
"learning_rate": 7.306341788231067e-06,
"loss": 0.3972741365432739,
"loss_sft": 0.3972741365432739,
"step": 320,
"token_acc": 0.8697066477407145
},
{
"epoch": 1.1386666666666667,
"grad_norm": 2.039888381958008,
"learning_rate": 7.289032677156408e-06,
"loss": 0.441580593585968,
"loss_sft": 0.441580593585968,
"step": 321,
"token_acc": 0.8587555216229481
},
{
"epoch": 1.1422222222222222,
"grad_norm": 1.9970409870147705,
"learning_rate": 7.271688789963458e-06,
"loss": 0.40510284900665283,
"loss_sft": 0.40510284900665283,
"step": 322,
"token_acc": 0.8659458853845723
},
{
"epoch": 1.1457777777777778,
"grad_norm": 1.9442920684814453,
"learning_rate": 7.254310390149144e-06,
"loss": 0.4107077121734619,
"loss_sft": 0.4107077121734619,
"step": 323,
"token_acc": 0.865978818648324
},
{
"epoch": 1.1493333333333333,
"grad_norm": 1.9443494081497192,
"learning_rate": 7.236897741734721e-06,
"loss": 0.40924397110939026,
"loss_sft": 0.40924397110939026,
"step": 324,
"token_acc": 0.8665360791995341
},
{
"epoch": 1.1528888888888889,
"grad_norm": 1.9774799346923828,
"learning_rate": 7.219451109261768e-06,
"loss": 0.40714192390441895,
"loss_sft": 0.40714192390441895,
"step": 325,
"token_acc": 0.8686944476509293
},
{
"epoch": 1.1564444444444444,
"grad_norm": 2.033506393432617,
"learning_rate": 7.201970757788172e-06,
"loss": 0.39068448543548584,
"loss_sft": 0.39068448543548584,
"step": 326,
"token_acc": 0.8731719012227284
},
{
"epoch": 1.16,
"grad_norm": 2.06431245803833,
"learning_rate": 7.184456952884089e-06,
"loss": 0.44274166226387024,
"loss_sft": 0.44274166226387024,
"step": 327,
"token_acc": 0.8565521803848062
},
{
"epoch": 1.1635555555555555,
"grad_norm": 1.9392691850662231,
"learning_rate": 7.166909960627918e-06,
"loss": 0.428322434425354,
"loss_sft": 0.428322434425354,
"step": 328,
"token_acc": 0.8638397198988523
},
{
"epoch": 1.1671111111111112,
"grad_norm": 2.162794589996338,
"learning_rate": 7.149330047602258e-06,
"loss": 0.3970050811767578,
"loss_sft": 0.3970050811767578,
"step": 329,
"token_acc": 0.8698859409868585
},
{
"epoch": 1.1706666666666667,
"grad_norm": 2.0334274768829346,
"learning_rate": 7.131717480889854e-06,
"loss": 0.40429121255874634,
"loss_sft": 0.40429121255874634,
"step": 330,
"token_acc": 0.8662298225986894
},
{
"epoch": 1.1742222222222223,
"grad_norm": 2.1451988220214844,
"learning_rate": 7.114072528069549e-06,
"loss": 0.4276900887489319,
"loss_sft": 0.4276900887489319,
"step": 331,
"token_acc": 0.8622798159991024
},
{
"epoch": 1.1777777777777778,
"grad_norm": 2.331615447998047,
"learning_rate": 7.0963954572122e-06,
"loss": 0.387137234210968,
"loss_sft": 0.387137234210968,
"step": 332,
"token_acc": 0.8720646124609747
},
{
"epoch": 1.1813333333333333,
"grad_norm": 2.172874689102173,
"learning_rate": 7.078686536876627e-06,
"loss": 0.42369771003723145,
"loss_sft": 0.42369771003723145,
"step": 333,
"token_acc": 0.8624027445935184
},
{
"epoch": 1.1848888888888889,
"grad_norm": 2.044041395187378,
"learning_rate": 7.06094603610552e-06,
"loss": 0.4190555512905121,
"loss_sft": 0.4190555512905121,
"step": 334,
"token_acc": 0.86600178292846
},
{
"epoch": 1.1884444444444444,
"grad_norm": 2.065030574798584,
"learning_rate": 7.043174224421353e-06,
"loss": 0.4139631986618042,
"loss_sft": 0.4139631986618042,
"step": 335,
"token_acc": 0.8676819511118469
},
{
"epoch": 1.192,
"grad_norm": 2.0200564861297607,
"learning_rate": 7.025371371822294e-06,
"loss": 0.39490020275115967,
"loss_sft": 0.39490020275115967,
"step": 336,
"token_acc": 0.8737214574003428
},
{
"epoch": 1.1955555555555555,
"grad_norm": 2.0914556980133057,
"learning_rate": 7.007537748778097e-06,
"loss": 0.3916926383972168,
"loss_sft": 0.3916926383972168,
"step": 337,
"token_acc": 0.8692341576993267
},
{
"epoch": 1.199111111111111,
"grad_norm": 2.2042653560638428,
"learning_rate": 6.989673626225997e-06,
"loss": 0.4564780592918396,
"loss_sft": 0.4564780592918396,
"step": 338,
"token_acc": 0.8545752557508619
},
{
"epoch": 1.2026666666666666,
"grad_norm": 2.047900438308716,
"learning_rate": 6.971779275566593e-06,
"loss": 0.4163472056388855,
"loss_sft": 0.4163472056388855,
"step": 339,
"token_acc": 0.8650571495218101
},
{
"epoch": 1.2062222222222223,
"grad_norm": 2.059612512588501,
"learning_rate": 6.953854968659726e-06,
"loss": 0.4669194221496582,
"loss_sft": 0.4669194221496582,
"step": 340,
"token_acc": 0.8516977796266445
},
{
"epoch": 1.2097777777777778,
"grad_norm": 2.2702395915985107,
"learning_rate": 6.935900977820341e-06,
"loss": 0.4309927821159363,
"loss_sft": 0.4309927821159363,
"step": 341,
"token_acc": 0.8571656729131175
},
{
"epoch": 1.2133333333333334,
"grad_norm": 1.9873970746994019,
"learning_rate": 6.917917575814364e-06,
"loss": 0.3964521288871765,
"loss_sft": 0.3964521288871765,
"step": 342,
"token_acc": 0.869855329199882
},
{
"epoch": 1.216888888888889,
"grad_norm": 2.0788300037384033,
"learning_rate": 6.899905035854544e-06,
"loss": 0.41934633255004883,
"loss_sft": 0.41934633255004883,
"step": 343,
"token_acc": 0.8623936654004601
},
{
"epoch": 1.2204444444444444,
"grad_norm": 2.1086623668670654,
"learning_rate": 6.881863631596313e-06,
"loss": 0.44192826747894287,
"loss_sft": 0.44192826747894287,
"step": 344,
"token_acc": 0.8593427982596749
},
{
"epoch": 1.224,
"grad_norm": 2.113551139831543,
"learning_rate": 6.863793637133618e-06,
"loss": 0.4093806743621826,
"loss_sft": 0.4093806743621826,
"step": 345,
"token_acc": 0.868071818891491
},
{
"epoch": 1.2275555555555555,
"grad_norm": 2.028347969055176,
"learning_rate": 6.845695326994768e-06,
"loss": 0.3931807279586792,
"loss_sft": 0.3931807279586792,
"step": 346,
"token_acc": 0.8703492353343986
},
{
"epoch": 1.231111111111111,
"grad_norm": 2.1860127449035645,
"learning_rate": 6.827568976138255e-06,
"loss": 0.4532551169395447,
"loss_sft": 0.4532551169395447,
"step": 347,
"token_acc": 0.8518605195413059
},
{
"epoch": 1.2346666666666666,
"grad_norm": 2.230043411254883,
"learning_rate": 6.809414859948579e-06,
"loss": 0.43102413415908813,
"loss_sft": 0.43102413415908813,
"step": 348,
"token_acc": 0.859845870715436
},
{
"epoch": 1.2382222222222223,
"grad_norm": 2.13657808303833,
"learning_rate": 6.791233254232066e-06,
"loss": 0.40593427419662476,
"loss_sft": 0.40593427419662476,
"step": 349,
"token_acc": 0.8668020547126986
},
{
"epoch": 1.2417777777777779,
"grad_norm": 2.088085651397705,
"learning_rate": 6.773024435212678e-06,
"loss": 0.41960766911506653,
"loss_sft": 0.41960766911506653,
"step": 350,
"token_acc": 0.8668713674745038
},
{
"epoch": 1.2453333333333334,
"grad_norm": 2.2578799724578857,
"learning_rate": 6.7547886795278136e-06,
"loss": 0.4201706349849701,
"loss_sft": 0.4201706349849701,
"step": 351,
"token_acc": 0.8626221772834513
},
{
"epoch": 1.248888888888889,
"grad_norm": 2.108464241027832,
"learning_rate": 6.736526264224101e-06,
"loss": 0.4046739935874939,
"loss_sft": 0.4046739935874939,
"step": 352,
"token_acc": 0.8661211699164345
},
{
"epoch": 1.2524444444444445,
"grad_norm": 2.146855115890503,
"learning_rate": 6.718237466753206e-06,
"loss": 0.4080267548561096,
"loss_sft": 0.4080267548561096,
"step": 353,
"token_acc": 0.8646153846153846
},
{
"epoch": 1.256,
"grad_norm": 2.099658489227295,
"learning_rate": 6.6999225649675955e-06,
"loss": 0.40895289182662964,
"loss_sft": 0.40895289182662964,
"step": 354,
"token_acc": 0.8670667934395055
},
{
"epoch": 1.2595555555555555,
"grad_norm": 2.0408408641815186,
"learning_rate": 6.681581837116331e-06,
"loss": 0.3808214068412781,
"loss_sft": 0.3808214068412781,
"step": 355,
"token_acc": 0.8734803094642546
},
{
"epoch": 1.263111111111111,
"grad_norm": 2.115304708480835,
"learning_rate": 6.6632155618408335e-06,
"loss": 0.39477837085723877,
"loss_sft": 0.39477837085723877,
"step": 356,
"token_acc": 0.8717051842598376
},
{
"epoch": 1.2666666666666666,
"grad_norm": 2.059264659881592,
"learning_rate": 6.644824018170655e-06,
"loss": 0.3914453089237213,
"loss_sft": 0.3914453089237213,
"step": 357,
"token_acc": 0.8695928019379398
},
{
"epoch": 1.2702222222222221,
"grad_norm": 1.9908522367477417,
"learning_rate": 6.6264074855192385e-06,
"loss": 0.4108723998069763,
"loss_sft": 0.4108723998069763,
"step": 358,
"token_acc": 0.8642107036558657
},
{
"epoch": 1.2737777777777777,
"grad_norm": 2.1818549633026123,
"learning_rate": 6.607966243679669e-06,
"loss": 0.3807740807533264,
"loss_sft": 0.3807740807533264,
"step": 359,
"token_acc": 0.8740432306656053
},
{
"epoch": 1.2773333333333334,
"grad_norm": 2.1877641677856445,
"learning_rate": 6.589500572820428e-06,
"loss": 0.4238789975643158,
"loss_sft": 0.4238789975643158,
"step": 360,
"token_acc": 0.8610257040954372
},
{
"epoch": 1.280888888888889,
"grad_norm": 2.2202279567718506,
"learning_rate": 6.571010753481135e-06,
"loss": 0.4421176612377167,
"loss_sft": 0.4421176612377167,
"step": 361,
"token_acc": 0.8553278910093529
},
{
"epoch": 1.2844444444444445,
"grad_norm": 2.1197445392608643,
"learning_rate": 6.552497066568282e-06,
"loss": 0.4149945378303528,
"loss_sft": 0.4149945378303528,
"step": 362,
"token_acc": 0.8652888303658327
},
{
"epoch": 1.288,
"grad_norm": 2.1825129985809326,
"learning_rate": 6.533959793350974e-06,
"loss": 0.3928068280220032,
"loss_sft": 0.3928068280220032,
"step": 363,
"token_acc": 0.8691647346324958
},
{
"epoch": 1.2915555555555556,
"grad_norm": 2.181267499923706,
"learning_rate": 6.5153992154566445e-06,
"loss": 0.4269263744354248,
"loss_sft": 0.4269263744354248,
"step": 364,
"token_acc": 0.8601386278195489
},
{
"epoch": 1.295111111111111,
"grad_norm": 2.0992515087127686,
"learning_rate": 6.496815614866792e-06,
"loss": 0.4099985957145691,
"loss_sft": 0.4099985957145691,
"step": 365,
"token_acc": 0.8613764529865346
},
{
"epoch": 1.2986666666666666,
"grad_norm": 2.2896366119384766,
"learning_rate": 6.478209273912675e-06,
"loss": 0.41681671142578125,
"loss_sft": 0.41681671142578125,
"step": 366,
"token_acc": 0.8647914982095414
},
{
"epoch": 1.3022222222222222,
"grad_norm": 2.1159205436706543,
"learning_rate": 6.4595804752710475e-06,
"loss": 0.45071905851364136,
"loss_sft": 0.45071905851364136,
"step": 367,
"token_acc": 0.8551527358173884
},
{
"epoch": 1.3057777777777777,
"grad_norm": 1.992004156112671,
"learning_rate": 6.440929501959844e-06,
"loss": 0.3944794535636902,
"loss_sft": 0.3944794535636902,
"step": 368,
"token_acc": 0.8682188456872147
},
{
"epoch": 1.3093333333333335,
"grad_norm": 2.001436710357666,
"learning_rate": 6.42225663733389e-06,
"loss": 0.41189777851104736,
"loss_sft": 0.41189777851104736,
"step": 369,
"token_acc": 0.867511582086373
},
{
"epoch": 1.3128888888888888,
"grad_norm": 2.31205415725708,
"learning_rate": 6.403562165080594e-06,
"loss": 0.4462592899799347,
"loss_sft": 0.4462592899799347,
"step": 370,
"token_acc": 0.8568150527765029
},
{
"epoch": 1.3164444444444445,
"grad_norm": 1.98836350440979,
"learning_rate": 6.3848463692156396e-06,
"loss": 0.42124196887016296,
"loss_sft": 0.42124196887016296,
"step": 371,
"token_acc": 0.8648098553829673
},
{
"epoch": 1.32,
"grad_norm": 1.9472687244415283,
"learning_rate": 6.366109534078667e-06,
"loss": 0.4333266615867615,
"loss_sft": 0.4333266615867615,
"step": 372,
"token_acc": 0.8565029428616073
},
{
"epoch": 1.3235555555555556,
"grad_norm": 2.1749327182769775,
"learning_rate": 6.347351944328958e-06,
"loss": 0.4078105390071869,
"loss_sft": 0.4078105390071869,
"step": 373,
"token_acc": 0.8684759916492694
},
{
"epoch": 1.3271111111111111,
"grad_norm": 2.082580804824829,
"learning_rate": 6.328573884941107e-06,
"loss": 0.38568437099456787,
"loss_sft": 0.38568437099456787,
"step": 374,
"token_acc": 0.8727032469166877
},
{
"epoch": 1.3306666666666667,
"grad_norm": 2.1083035469055176,
"learning_rate": 6.309775641200695e-06,
"loss": 0.4161454439163208,
"loss_sft": 0.4161454439163208,
"step": 375,
"token_acc": 0.8634233165232869
},
{
"epoch": 1.3342222222222222,
"grad_norm": 2.0515077114105225,
"learning_rate": 6.290957498699949e-06,
"loss": 0.40313780307769775,
"loss_sft": 0.40313780307769775,
"step": 376,
"token_acc": 0.8661456394988215
},
{
"epoch": 1.3377777777777777,
"grad_norm": 2.109354019165039,
"learning_rate": 6.272119743333416e-06,
"loss": 0.42586779594421387,
"loss_sft": 0.42586779594421387,
"step": 377,
"token_acc": 0.8588957055214724
},
{
"epoch": 1.3413333333333333,
"grad_norm": 2.1871895790100098,
"learning_rate": 6.2532626612936035e-06,
"loss": 0.4273819625377655,
"loss_sft": 0.4273819625377655,
"step": 378,
"token_acc": 0.8611840534733827
},
{
"epoch": 1.3448888888888888,
"grad_norm": 2.244476795196533,
"learning_rate": 6.234386539066643e-06,
"loss": 0.4375607371330261,
"loss_sft": 0.4375607371330261,
"step": 379,
"token_acc": 0.8610177350179669
},
{
"epoch": 1.3484444444444446,
"grad_norm": 2.0864622592926025,
"learning_rate": 6.215491663427936e-06,
"loss": 0.4264943301677704,
"loss_sft": 0.4264943301677704,
"step": 380,
"token_acc": 0.8594910162670748
},
{
"epoch": 1.3519999999999999,
"grad_norm": 2.0926661491394043,
"learning_rate": 6.1965783214377895e-06,
"loss": 0.43840712308883667,
"loss_sft": 0.43840712308883667,
"step": 381,
"token_acc": 0.8592106751102567
},
{
"epoch": 1.3555555555555556,
"grad_norm": 1.9400382041931152,
"learning_rate": 6.177646800437066e-06,
"loss": 0.4148487448692322,
"loss_sft": 0.4148487448692322,
"step": 382,
"token_acc": 0.8605477195511171
},
{
"epoch": 1.3591111111111112,
"grad_norm": 2.0739121437072754,
"learning_rate": 6.158697388042807e-06,
"loss": 0.4656825661659241,
"loss_sft": 0.4656825661659241,
"step": 383,
"token_acc": 0.8545405258877745
},
{
"epoch": 1.3626666666666667,
"grad_norm": 2.072519063949585,
"learning_rate": 6.139730372143877e-06,
"loss": 0.4354017674922943,
"loss_sft": 0.4354017674922943,
"step": 384,
"token_acc": 0.8596860961677112
},
{
"epoch": 1.3662222222222222,
"grad_norm": 2.0000264644622803,
"learning_rate": 6.120746040896572e-06,
"loss": 0.40153205394744873,
"loss_sft": 0.40153205394744873,
"step": 385,
"token_acc": 0.8690829279867125
},
{
"epoch": 1.3697777777777778,
"grad_norm": 1.9740139245986938,
"learning_rate": 6.10174468272026e-06,
"loss": 0.4312746822834015,
"loss_sft": 0.4312746822834015,
"step": 386,
"token_acc": 0.8571791355579258
},
{
"epoch": 1.3733333333333333,
"grad_norm": 2.038445472717285,
"learning_rate": 6.082726586292982e-06,
"loss": 0.421906054019928,
"loss_sft": 0.421906054019928,
"step": 387,
"token_acc": 0.8600680749905452
},
{
"epoch": 1.3768888888888888,
"grad_norm": 2.019009828567505,
"learning_rate": 6.063692040547083e-06,
"loss": 0.4096180200576782,
"loss_sft": 0.4096180200576782,
"step": 388,
"token_acc": 0.8684788895315211
},
{
"epoch": 1.3804444444444444,
"grad_norm": 2.2725658416748047,
"learning_rate": 6.044641334664812e-06,
"loss": 0.41742199659347534,
"loss_sft": 0.41742199659347534,
"step": 389,
"token_acc": 0.8634841420973144
},
{
"epoch": 1.384,
"grad_norm": 2.089355945587158,
"learning_rate": 6.025574758073925e-06,
"loss": 0.42201587557792664,
"loss_sft": 0.42201587557792664,
"step": 390,
"token_acc": 0.8627428307123034
},
{
"epoch": 1.3875555555555557,
"grad_norm": 2.1601922512054443,
"learning_rate": 6.006492600443301e-06,
"loss": 0.3951718211174011,
"loss_sft": 0.3951718211174011,
"step": 391,
"token_acc": 0.8695599393019727
},
{
"epoch": 1.3911111111111112,
"grad_norm": 2.048130989074707,
"learning_rate": 5.98739515167853e-06,
"loss": 0.40731945633888245,
"loss_sft": 0.40731945633888245,
"step": 392,
"token_acc": 0.8659113408865912
},
{
"epoch": 1.3946666666666667,
"grad_norm": 2.0659661293029785,
"learning_rate": 5.968282701917514e-06,
"loss": 0.39317312836647034,
"loss_sft": 0.39317312836647034,
"step": 393,
"token_acc": 0.8704396893681211
},
{
"epoch": 1.3982222222222223,
"grad_norm": 2.111722230911255,
"learning_rate": 5.949155541526057e-06,
"loss": 0.4161897301673889,
"loss_sft": 0.4161897301673889,
"step": 394,
"token_acc": 0.8668742921857304
},
{
"epoch": 1.4017777777777778,
"grad_norm": 2.1751391887664795,
"learning_rate": 5.930013961093454e-06,
"loss": 0.4339696764945984,
"loss_sft": 0.4339696764945984,
"step": 395,
"token_acc": 0.8604475443185121
},
{
"epoch": 1.4053333333333333,
"grad_norm": 2.0277295112609863,
"learning_rate": 5.910858251428077e-06,
"loss": 0.42175161838531494,
"loss_sft": 0.42175161838531494,
"step": 396,
"token_acc": 0.8647277646263104
},
{
"epoch": 1.4088888888888889,
"grad_norm": 2.056373119354248,
"learning_rate": 5.891688703552953e-06,
"loss": 0.4363563656806946,
"loss_sft": 0.4363563656806946,
"step": 397,
"token_acc": 0.8593149974052932
},
{
"epoch": 1.4124444444444444,
"grad_norm": 2.125103712081909,
"learning_rate": 5.872505608701354e-06,
"loss": 0.4381544589996338,
"loss_sft": 0.4381544589996338,
"step": 398,
"token_acc": 0.8529967752696542
},
{
"epoch": 1.416,
"grad_norm": 2.0783627033233643,
"learning_rate": 5.853309258312356e-06,
"loss": 0.41318410634994507,
"loss_sft": 0.41318410634994507,
"step": 399,
"token_acc": 0.8663534039995628
},
{
"epoch": 1.4195555555555557,
"grad_norm": 2.3527917861938477,
"learning_rate": 5.834099944026422e-06,
"loss": 0.4116997718811035,
"loss_sft": 0.4116997718811035,
"step": 400,
"token_acc": 0.8664490861618799
},
{
"epoch": 1.423111111111111,
"grad_norm": 2.21236515045166,
"learning_rate": 5.814877957680973e-06,
"loss": 0.4189806580543518,
"loss_sft": 0.4189806580543518,
"step": 401,
"token_acc": 0.8665842711246992
},
{
"epoch": 1.4266666666666667,
"grad_norm": 2.0903818607330322,
"learning_rate": 5.795643591305945e-06,
"loss": 0.3932293951511383,
"loss_sft": 0.3932293951511383,
"step": 402,
"token_acc": 0.8708672605060386
},
{
"epoch": 1.4302222222222223,
"grad_norm": 2.1492693424224854,
"learning_rate": 5.776397137119362e-06,
"loss": 0.42269301414489746,
"loss_sft": 0.42269301414489746,
"step": 403,
"token_acc": 0.858822830670547
},
{
"epoch": 1.4337777777777778,
"grad_norm": 2.1094796657562256,
"learning_rate": 5.757138887522884e-06,
"loss": 0.42915230989456177,
"loss_sft": 0.42915230989456177,
"step": 404,
"token_acc": 0.8588242048455621
},
{
"epoch": 1.4373333333333334,
"grad_norm": 2.1783599853515625,
"learning_rate": 5.7378691350973835e-06,
"loss": 0.43106937408447266,
"loss_sft": 0.43106937408447266,
"step": 405,
"token_acc": 0.85428145087236
},
{
"epoch": 1.4408888888888889,
"grad_norm": 2.249063491821289,
"learning_rate": 5.7185881725984835e-06,
"loss": 0.4146789312362671,
"loss_sft": 0.4146789312362671,
"step": 406,
"token_acc": 0.8634489167683104
},
{
"epoch": 1.4444444444444444,
"grad_norm": 2.0479493141174316,
"learning_rate": 5.699296292952117e-06,
"loss": 0.40824148058891296,
"loss_sft": 0.40824148058891296,
"step": 407,
"token_acc": 0.8637382339757956
},
{
"epoch": 1.448,
"grad_norm": 1.9829336404800415,
"learning_rate": 5.679993789250075e-06,
"loss": 0.4057304561138153,
"loss_sft": 0.4057304561138153,
"step": 408,
"token_acc": 0.868186638388123
},
{
"epoch": 1.4515555555555555,
"grad_norm": 2.214102268218994,
"learning_rate": 5.660680954745554e-06,
"loss": 0.43437254428863525,
"loss_sft": 0.43437254428863525,
"step": 409,
"token_acc": 0.8594101622645979
},
{
"epoch": 1.455111111111111,
"grad_norm": 2.0235214233398438,
"learning_rate": 5.641358082848705e-06,
"loss": 0.4238203167915344,
"loss_sft": 0.4238203167915344,
"step": 410,
"token_acc": 0.8627439822454902
},
{
"epoch": 1.4586666666666668,
"grad_norm": 2.224508285522461,
"learning_rate": 5.622025467122167e-06,
"loss": 0.4427334666252136,
"loss_sft": 0.4427334666252136,
"step": 411,
"token_acc": 0.8546247274992215
},
{
"epoch": 1.462222222222222,
"grad_norm": 2.0751185417175293,
"learning_rate": 5.6026834012766155e-06,
"loss": 0.4205954670906067,
"loss_sft": 0.4205954670906067,
"step": 412,
"token_acc": 0.8647512263489839
},
{
"epoch": 1.4657777777777778,
"grad_norm": 2.059248685836792,
"learning_rate": 5.5833321791662975e-06,
"loss": 0.4332770109176636,
"loss_sft": 0.4332770109176636,
"step": 413,
"token_acc": 0.8589489081600669
},
{
"epoch": 1.4693333333333334,
"grad_norm": 2.159562349319458,
"learning_rate": 5.563972094784561e-06,
"loss": 0.4090723991394043,
"loss_sft": 0.4090723991394043,
"step": 414,
"token_acc": 0.8648569589890599
},
{
"epoch": 1.472888888888889,
"grad_norm": 2.1778712272644043,
"learning_rate": 5.544603442259401e-06,
"loss": 0.4404616355895996,
"loss_sft": 0.4404616355895996,
"step": 415,
"token_acc": 0.857877777149641
},
{
"epoch": 1.4764444444444444,
"grad_norm": 2.149751901626587,
"learning_rate": 5.525226515848979e-06,
"loss": 0.42241090536117554,
"loss_sft": 0.42241090536117554,
"step": 416,
"token_acc": 0.864554824433704
},
{
"epoch": 1.48,
"grad_norm": 2.0396487712860107,
"learning_rate": 5.505841609937162e-06,
"loss": 0.4617084860801697,
"loss_sft": 0.4617084860801697,
"step": 417,
"token_acc": 0.8518164942985945
},
{
"epoch": 1.4835555555555555,
"grad_norm": 2.021590232849121,
"learning_rate": 5.4864490190290386e-06,
"loss": 0.3814869523048401,
"loss_sft": 0.3814869523048401,
"step": 418,
"token_acc": 0.8718812817784084
},
{
"epoch": 1.487111111111111,
"grad_norm": 1.9816710948944092,
"learning_rate": 5.46704903774646e-06,
"loss": 0.3968183100223541,
"loss_sft": 0.3968183100223541,
"step": 419,
"token_acc": 0.872141317491194
},
{
"epoch": 1.4906666666666666,
"grad_norm": 1.9432770013809204,
"learning_rate": 5.447641960823549e-06,
"loss": 0.42878788709640503,
"loss_sft": 0.42878788709640503,
"step": 420,
"token_acc": 0.861723961005644
},
{
"epoch": 1.4942222222222221,
"grad_norm": 2.1680777072906494,
"learning_rate": 5.42822808310223e-06,
"loss": 0.4524781107902527,
"loss_sft": 0.4524781107902527,
"step": 421,
"token_acc": 0.853664061615132
},
{
"epoch": 1.4977777777777779,
"grad_norm": 1.9948804378509521,
"learning_rate": 5.4088076995277564e-06,
"loss": 0.3889770805835724,
"loss_sft": 0.3889770805835724,
"step": 422,
"token_acc": 0.8710159362549801
},
{
"epoch": 1.5013333333333332,
"grad_norm": 1.9953668117523193,
"learning_rate": 5.389381105144208e-06,
"loss": 0.38785499334335327,
"loss_sft": 0.38785499334335327,
"step": 423,
"token_acc": 0.8721294660664425
},
{
"epoch": 1.504888888888889,
"grad_norm": 2.0792336463928223,
"learning_rate": 5.369948595090033e-06,
"loss": 0.39852064847946167,
"loss_sft": 0.39852064847946167,
"step": 424,
"token_acc": 0.8671501857420838
},
{
"epoch": 1.5084444444444445,
"grad_norm": 2.2840240001678467,
"learning_rate": 5.350510464593548e-06,
"loss": 0.41780519485473633,
"loss_sft": 0.41780519485473633,
"step": 425,
"token_acc": 0.8665506237307804
},
{
"epoch": 1.512,
"grad_norm": 2.2083730697631836,
"learning_rate": 5.331067008968462e-06,
"loss": 0.4437114894390106,
"loss_sft": 0.4437114894390106,
"step": 426,
"token_acc": 0.8560632484594815
},
{
"epoch": 1.5155555555555555,
"grad_norm": 1.9596586227416992,
"learning_rate": 5.311618523609386e-06,
"loss": 0.40260934829711914,
"loss_sft": 0.40260934829711914,
"step": 427,
"token_acc": 0.8683822407567513
},
{
"epoch": 1.519111111111111,
"grad_norm": 2.002098560333252,
"learning_rate": 5.292165303987336e-06,
"loss": 0.4126218557357788,
"loss_sft": 0.4126218557357788,
"step": 428,
"token_acc": 0.8640786912839935
},
{
"epoch": 1.5226666666666666,
"grad_norm": 2.0283186435699463,
"learning_rate": 5.27270764564527e-06,
"loss": 0.4139862060546875,
"loss_sft": 0.4139862060546875,
"step": 429,
"token_acc": 0.8626638001638002
},
{
"epoch": 1.5262222222222221,
"grad_norm": 2.088085174560547,
"learning_rate": 5.253245844193564e-06,
"loss": 0.42426472902297974,
"loss_sft": 0.42426472902297974,
"step": 430,
"token_acc": 0.8606607954226735
},
{
"epoch": 1.529777777777778,
"grad_norm": 2.189093589782715,
"learning_rate": 5.23378019530555e-06,
"loss": 0.38969558477401733,
"loss_sft": 0.38969558477401733,
"step": 431,
"token_acc": 0.8720338983050847
},
{
"epoch": 1.5333333333333332,
"grad_norm": 2.1219334602355957,
"learning_rate": 5.214310994713008e-06,
"loss": 0.41368842124938965,
"loss_sft": 0.41368842124938965,
"step": 432,
"token_acc": 0.8649772209567198
},
{
"epoch": 1.536888888888889,
"grad_norm": 2.108358860015869,
"learning_rate": 5.194838538201676e-06,
"loss": 0.42121070623397827,
"loss_sft": 0.42121070623397827,
"step": 433,
"token_acc": 0.8624245748765771
},
{
"epoch": 1.5404444444444443,
"grad_norm": 2.103247880935669,
"learning_rate": 5.175363121606759e-06,
"loss": 0.40116065740585327,
"loss_sft": 0.40116065740585327,
"step": 434,
"token_acc": 0.8683508272549368
},
{
"epoch": 1.544,
"grad_norm": 2.1058623790740967,
"learning_rate": 5.155885040808432e-06,
"loss": 0.43327242136001587,
"loss_sft": 0.43327242136001587,
"step": 435,
"token_acc": 0.8573345259391771
},
{
"epoch": 1.5475555555555556,
"grad_norm": 2.0265440940856934,
"learning_rate": 5.1364045917273505e-06,
"loss": 0.42016279697418213,
"loss_sft": 0.42016279697418213,
"step": 436,
"token_acc": 0.8640215716486903
},
{
"epoch": 1.551111111111111,
"grad_norm": 2.579517126083374,
"learning_rate": 5.116922070320144e-06,
"loss": 0.40481728315353394,
"loss_sft": 0.40481728315353394,
"step": 437,
"token_acc": 0.8704780179468368
},
{
"epoch": 1.5546666666666666,
"grad_norm": 2.0824403762817383,
"learning_rate": 5.097437772574927e-06,
"loss": 0.42053020000457764,
"loss_sft": 0.42053020000457764,
"step": 438,
"token_acc": 0.8621852395714363
},
{
"epoch": 1.5582222222222222,
"grad_norm": 2.0338289737701416,
"learning_rate": 5.077951994506805e-06,
"loss": 0.4205910563468933,
"loss_sft": 0.4205910563468933,
"step": 439,
"token_acc": 0.8643185127113898
},
{
"epoch": 1.561777777777778,
"grad_norm": 2.067195177078247,
"learning_rate": 5.058465032153368e-06,
"loss": 0.3816699981689453,
"loss_sft": 0.3816699981689453,
"step": 440,
"token_acc": 0.8755683912990045
},
{
"epoch": 1.5653333333333332,
"grad_norm": 1.9996615648269653,
"learning_rate": 5.038977181570204e-06,
"loss": 0.4421253800392151,
"loss_sft": 0.4421253800392151,
"step": 441,
"token_acc": 0.8619064692926292
},
{
"epoch": 1.568888888888889,
"grad_norm": 2.1808149814605713,
"learning_rate": 5.0194887388263895e-06,
"loss": 0.4402797818183899,
"loss_sft": 0.4402797818183899,
"step": 442,
"token_acc": 0.8590721591230152
},
{
"epoch": 1.5724444444444443,
"grad_norm": 2.2986810207366943,
"learning_rate": 5e-06,
"loss": 0.4166395664215088,
"loss_sft": 0.4166395664215088,
"step": 443,
"token_acc": 0.868903862832744
},
{
"epoch": 1.576,
"grad_norm": 2.078742027282715,
"learning_rate": 4.980511261173613e-06,
"loss": 0.42019397020339966,
"loss_sft": 0.42019397020339966,
"step": 444,
"token_acc": 0.8630479218714513
},
{
"epoch": 1.5795555555555556,
"grad_norm": 2.030142068862915,
"learning_rate": 4.961022818429798e-06,
"loss": 0.42731574177742004,
"loss_sft": 0.42731574177742004,
"step": 445,
"token_acc": 0.8596580905321747
},
{
"epoch": 1.5831111111111111,
"grad_norm": 1.9992742538452148,
"learning_rate": 4.9415349678466335e-06,
"loss": 0.412159264087677,
"loss_sft": 0.412159264087677,
"step": 446,
"token_acc": 0.8650519031141869
},
{
"epoch": 1.5866666666666667,
"grad_norm": 2.1405060291290283,
"learning_rate": 4.922048005493196e-06,
"loss": 0.3980518579483032,
"loss_sft": 0.3980518579483032,
"step": 447,
"token_acc": 0.8683663833805477
},
{
"epoch": 1.5902222222222222,
"grad_norm": 2.1107187271118164,
"learning_rate": 4.902562227425075e-06,
"loss": 0.375371515750885,
"loss_sft": 0.375371515750885,
"step": 448,
"token_acc": 0.8768915415686984
},
{
"epoch": 1.5937777777777777,
"grad_norm": 2.133552312850952,
"learning_rate": 4.883077929679859e-06,
"loss": 0.44094976782798767,
"loss_sft": 0.44094976782798767,
"step": 449,
"token_acc": 0.8569249381936773
},
{
"epoch": 1.5973333333333333,
"grad_norm": 2.1417441368103027,
"learning_rate": 4.86359540827265e-06,
"loss": 0.3865364193916321,
"loss_sft": 0.3865364193916321,
"step": 450,
"token_acc": 0.8736435534302005
},
{
"epoch": 1.600888888888889,
"grad_norm": 2.0163345336914062,
"learning_rate": 4.844114959191569e-06,
"loss": 0.4052085876464844,
"loss_sft": 0.4052085876464844,
"step": 451,
"token_acc": 0.870444897734164
},
{
"epoch": 1.6044444444444443,
"grad_norm": 1.976217269897461,
"learning_rate": 4.824636878393243e-06,
"loss": 0.40031105279922485,
"loss_sft": 0.40031105279922485,
"step": 452,
"token_acc": 0.8713493924536346
},
{
"epoch": 1.608,
"grad_norm": 2.247943639755249,
"learning_rate": 4.805161461798326e-06,
"loss": 0.4370366334915161,
"loss_sft": 0.4370366334915161,
"step": 453,
"token_acc": 0.8583845173252984
},
{
"epoch": 1.6115555555555554,
"grad_norm": 2.0089313983917236,
"learning_rate": 4.785689005286995e-06,
"loss": 0.43705374002456665,
"loss_sft": 0.43705374002456665,
"step": 454,
"token_acc": 0.8586118251928021
},
{
"epoch": 1.6151111111111112,
"grad_norm": 2.1519558429718018,
"learning_rate": 4.766219804694451e-06,
"loss": 0.413113534450531,
"loss_sft": 0.413113534450531,
"step": 455,
"token_acc": 0.8672015719665958
},
{
"epoch": 1.6186666666666667,
"grad_norm": 2.148524284362793,
"learning_rate": 4.746754155806437e-06,
"loss": 0.41832467913627625,
"loss_sft": 0.41832467913627625,
"step": 456,
"token_acc": 0.8632517462661139
},
{
"epoch": 1.6222222222222222,
"grad_norm": 2.0033304691314697,
"learning_rate": 4.727292354354731e-06,
"loss": 0.4066150188446045,
"loss_sft": 0.4066150188446045,
"step": 457,
"token_acc": 0.8704140615783886
},
{
"epoch": 1.6257777777777778,
"grad_norm": 2.1065657138824463,
"learning_rate": 4.7078346960126645e-06,
"loss": 0.42339444160461426,
"loss_sft": 0.42339444160461426,
"step": 458,
"token_acc": 0.8636465449658416
},
{
"epoch": 1.6293333333333333,
"grad_norm": 2.136706829071045,
"learning_rate": 4.688381476390617e-06,
"loss": 0.4289117455482483,
"loss_sft": 0.4289117455482483,
"step": 459,
"token_acc": 0.8628943240573985
},
{
"epoch": 1.6328888888888888,
"grad_norm": 1.9764639139175415,
"learning_rate": 4.668932991031538e-06,
"loss": 0.40687328577041626,
"loss_sft": 0.40687328577041626,
"step": 460,
"token_acc": 0.8657798584151633
},
{
"epoch": 1.6364444444444444,
"grad_norm": 2.036417007446289,
"learning_rate": 4.649489535406454e-06,
"loss": 0.39480650424957275,
"loss_sft": 0.39480650424957275,
"step": 461,
"token_acc": 0.8714285714285714
},
{
"epoch": 1.6400000000000001,
"grad_norm": 2.0405428409576416,
"learning_rate": 4.6300514049099694e-06,
"loss": 0.398843914270401,
"loss_sft": 0.398843914270401,
"step": 462,
"token_acc": 0.8684040617479604
},
{
"epoch": 1.6435555555555554,
"grad_norm": 2.133903980255127,
"learning_rate": 4.6106188948557935e-06,
"loss": 0.42368829250335693,
"loss_sft": 0.42368829250335693,
"step": 463,
"token_acc": 0.8638733820011042
},
{
"epoch": 1.6471111111111112,
"grad_norm": 2.061514377593994,
"learning_rate": 4.591192300472247e-06,
"loss": 0.428686261177063,
"loss_sft": 0.428686261177063,
"step": 464,
"token_acc": 0.8636491583628891
},
{
"epoch": 1.6506666666666665,
"grad_norm": 2.080209732055664,
"learning_rate": 4.5717719168977696e-06,
"loss": 0.4320542812347412,
"loss_sft": 0.4320542812347412,
"step": 465,
"token_acc": 0.8582339168246181
},
{
"epoch": 1.6542222222222223,
"grad_norm": 2.143778085708618,
"learning_rate": 4.552358039176453e-06,
"loss": 0.3890891671180725,
"loss_sft": 0.3890891671180725,
"step": 466,
"token_acc": 0.87114662214537
},
{
"epoch": 1.6577777777777778,
"grad_norm": 2.1197397708892822,
"learning_rate": 4.532950962253543e-06,
"loss": 0.38661035895347595,
"loss_sft": 0.38661035895347595,
"step": 467,
"token_acc": 0.8737529899072399
},
{
"epoch": 1.6613333333333333,
"grad_norm": 1.9991018772125244,
"learning_rate": 4.513550980970962e-06,
"loss": 0.411617249250412,
"loss_sft": 0.411617249250412,
"step": 468,
"token_acc": 0.8652980877390326
},
{
"epoch": 1.6648888888888889,
"grad_norm": 2.0429420471191406,
"learning_rate": 4.49415839006284e-06,
"loss": 0.4422208070755005,
"loss_sft": 0.4422208070755005,
"step": 469,
"token_acc": 0.8600126876718123
},
{
"epoch": 1.6684444444444444,
"grad_norm": 2.142669200897217,
"learning_rate": 4.474773484151021e-06,
"loss": 0.4309496283531189,
"loss_sft": 0.4309496283531189,
"step": 470,
"token_acc": 0.8614294764729598
},
{
"epoch": 1.6720000000000002,
"grad_norm": 2.1032681465148926,
"learning_rate": 4.4553965577406006e-06,
"loss": 0.4220542907714844,
"loss_sft": 0.4220542907714844,
"step": 471,
"token_acc": 0.8621151271753681
},
{
"epoch": 1.6755555555555555,
"grad_norm": 2.0916078090667725,
"learning_rate": 4.4360279052154406e-06,
"loss": 0.4344140887260437,
"loss_sft": 0.4344140887260437,
"step": 472,
"token_acc": 0.8587638569336958
},
{
"epoch": 1.6791111111111112,
"grad_norm": 2.0320165157318115,
"learning_rate": 4.416667820833704e-06,
"loss": 0.3744451403617859,
"loss_sft": 0.3744451403617859,
"step": 473,
"token_acc": 0.8786209935014869
},
{
"epoch": 1.6826666666666665,
"grad_norm": 1.946739673614502,
"learning_rate": 4.397316598723385e-06,
"loss": 0.4068688750267029,
"loss_sft": 0.4068688750267029,
"step": 474,
"token_acc": 0.8643133513542164
},
{
"epoch": 1.6862222222222223,
"grad_norm": 2.187589168548584,
"learning_rate": 4.377974532877834e-06,
"loss": 0.431757390499115,
"loss_sft": 0.431757390499115,
"step": 475,
"token_acc": 0.8634345586929497
},
{
"epoch": 1.6897777777777778,
"grad_norm": 2.0243496894836426,
"learning_rate": 4.358641917151297e-06,
"loss": 0.42587560415267944,
"loss_sft": 0.42587560415267944,
"step": 476,
"token_acc": 0.8635711629389576
},
{
"epoch": 1.6933333333333334,
"grad_norm": 2.0677926540374756,
"learning_rate": 4.339319045254448e-06,
"loss": 0.4221827983856201,
"loss_sft": 0.4221827983856201,
"step": 477,
"token_acc": 0.863139508677034
},
{
"epoch": 1.696888888888889,
"grad_norm": 1.953474521636963,
"learning_rate": 4.320006210749928e-06,
"loss": 0.41056370735168457,
"loss_sft": 0.41056370735168457,
"step": 478,
"token_acc": 0.8640124902419984
},
{
"epoch": 1.7004444444444444,
"grad_norm": 2.083036422729492,
"learning_rate": 4.3007037070478855e-06,
"loss": 0.41453826427459717,
"loss_sft": 0.41453826427459717,
"step": 479,
"token_acc": 0.8618951028687337
},
{
"epoch": 1.704,
"grad_norm": 2.150409460067749,
"learning_rate": 4.281411827401517e-06,
"loss": 0.3892195224761963,
"loss_sft": 0.3892195224761963,
"step": 480,
"token_acc": 0.8719496532237349
},
{
"epoch": 1.7075555555555555,
"grad_norm": 2.11389422416687,
"learning_rate": 4.262130864902617e-06,
"loss": 0.40103036165237427,
"loss_sft": 0.40103036165237427,
"step": 481,
"token_acc": 0.8687618821292775
},
{
"epoch": 1.7111111111111112,
"grad_norm": 2.0252602100372314,
"learning_rate": 4.2428611124771184e-06,
"loss": 0.4143048822879791,
"loss_sft": 0.4143048822879791,
"step": 482,
"token_acc": 0.8651578947368421
},
{
"epoch": 1.7146666666666666,
"grad_norm": 1.9856598377227783,
"learning_rate": 4.22360286288064e-06,
"loss": 0.4202113151550293,
"loss_sft": 0.4202113151550293,
"step": 483,
"token_acc": 0.8633540372670807
},
{
"epoch": 1.7182222222222223,
"grad_norm": 2.0671544075012207,
"learning_rate": 4.2043564086940565e-06,
"loss": 0.3882021903991699,
"loss_sft": 0.3882021903991699,
"step": 484,
"token_acc": 0.8722292408407029
},
{
"epoch": 1.7217777777777776,
"grad_norm": 2.126646041870117,
"learning_rate": 4.185122042319027e-06,
"loss": 0.466674268245697,
"loss_sft": 0.466674268245697,
"step": 485,
"token_acc": 0.8490584352552902
},
{
"epoch": 1.7253333333333334,
"grad_norm": 2.009033203125,
"learning_rate": 4.165900055973579e-06,
"loss": 0.36831024289131165,
"loss_sft": 0.36831024289131165,
"step": 486,
"token_acc": 0.8761332665887981
},
{
"epoch": 1.728888888888889,
"grad_norm": 2.0723743438720703,
"learning_rate": 4.1466907416876466e-06,
"loss": 0.4306026101112366,
"loss_sft": 0.4306026101112366,
"step": 487,
"token_acc": 0.8595511708586296
},
{
"epoch": 1.7324444444444445,
"grad_norm": 2.0858848094940186,
"learning_rate": 4.127494391298647e-06,
"loss": 0.38521188497543335,
"loss_sft": 0.38521188497543335,
"step": 488,
"token_acc": 0.8730092454518342
},
{
"epoch": 1.736,
"grad_norm": 2.135643482208252,
"learning_rate": 4.108311296447048e-06,
"loss": 0.4089963436126709,
"loss_sft": 0.4089963436126709,
"step": 489,
"token_acc": 0.8660353389536898
},
{
"epoch": 1.7395555555555555,
"grad_norm": 2.136782169342041,
"learning_rate": 4.089141748571926e-06,
"loss": 0.44046175479888916,
"loss_sft": 0.44046175479888916,
"step": 490,
"token_acc": 0.8559193408499567
},
{
"epoch": 1.743111111111111,
"grad_norm": 2.021350145339966,
"learning_rate": 4.069986038906547e-06,
"loss": 0.36756423115730286,
"loss_sft": 0.36756423115730286,
"step": 491,
"token_acc": 0.8812554489973845
},
{
"epoch": 1.7466666666666666,
"grad_norm": 2.10075306892395,
"learning_rate": 4.050844458473945e-06,
"loss": 0.4100453555583954,
"loss_sft": 0.4100453555583954,
"step": 492,
"token_acc": 0.863740022805017
},
{
"epoch": 1.7502222222222223,
"grad_norm": 2.0675134658813477,
"learning_rate": 4.031717298082487e-06,
"loss": 0.41361191868782043,
"loss_sft": 0.41361191868782043,
"step": 493,
"token_acc": 0.8653119942456393
},
{
"epoch": 1.7537777777777777,
"grad_norm": 2.1129140853881836,
"learning_rate": 4.012604848321471e-06,
"loss": 0.42755448818206787,
"loss_sft": 0.42755448818206787,
"step": 494,
"token_acc": 0.8613538895825417
},
{
"epoch": 1.7573333333333334,
"grad_norm": 2.1215624809265137,
"learning_rate": 3.993507399556699e-06,
"loss": 0.4140801429748535,
"loss_sft": 0.4140801429748535,
"step": 495,
"token_acc": 0.8645008381987398
},
{
"epoch": 1.7608888888888887,
"grad_norm": 2.0439512729644775,
"learning_rate": 3.974425241926076e-06,
"loss": 0.3860655426979065,
"loss_sft": 0.3860655426979065,
"step": 496,
"token_acc": 0.8685189359783588
},
{
"epoch": 1.7644444444444445,
"grad_norm": 2.078835964202881,
"learning_rate": 3.95535866533519e-06,
"loss": 0.4328363239765167,
"loss_sft": 0.4328363239765167,
"step": 497,
"token_acc": 0.8562522541089186
},
{
"epoch": 1.768,
"grad_norm": 2.1487245559692383,
"learning_rate": 3.936307959452917e-06,
"loss": 0.4036521911621094,
"loss_sft": 0.4036521911621094,
"step": 498,
"token_acc": 0.8710146613916686
},
{
"epoch": 1.7715555555555556,
"grad_norm": 2.125499963760376,
"learning_rate": 3.91727341370702e-06,
"loss": 0.4277384281158447,
"loss_sft": 0.4277384281158447,
"step": 499,
"token_acc": 0.8604188481675392
},
{
"epoch": 1.775111111111111,
"grad_norm": 2.0366432666778564,
"learning_rate": 3.898255317279744e-06,
"loss": 0.4030233919620514,
"loss_sft": 0.4030233919620514,
"step": 500,
"token_acc": 0.8657188841201717
},
{
"epoch": 1.7786666666666666,
"grad_norm": 2.219578266143799,
"learning_rate": 3.879253959103429e-06,
"loss": 0.4163818359375,
"loss_sft": 0.4163818359375,
"step": 501,
"token_acc": 0.862064854554125
},
{
"epoch": 1.7822222222222224,
"grad_norm": 2.0823135375976562,
"learning_rate": 3.860269627856126e-06,
"loss": 0.4303836226463318,
"loss_sft": 0.4303836226463318,
"step": 502,
"token_acc": 0.8584603699886074
},
{
"epoch": 1.7857777777777777,
"grad_norm": 2.1446592807769775,
"learning_rate": 3.841302611957193e-06,
"loss": 0.41120094060897827,
"loss_sft": 0.41120094060897827,
"step": 503,
"token_acc": 0.8640704095720203
},
{
"epoch": 1.7893333333333334,
"grad_norm": 2.0287466049194336,
"learning_rate": 3.822353199562936e-06,
"loss": 0.38759124279022217,
"loss_sft": 0.38759124279022217,
"step": 504,
"token_acc": 0.8734872905987423
},
{
"epoch": 1.7928888888888888,
"grad_norm": 2.269944906234741,
"learning_rate": 3.803421678562213e-06,
"loss": 0.4101651906967163,
"loss_sft": 0.4101651906967163,
"step": 505,
"token_acc": 0.8653293482009426
},
{
"epoch": 1.7964444444444445,
"grad_norm": 2.147164821624756,
"learning_rate": 3.784508336572066e-06,
"loss": 0.4444081783294678,
"loss_sft": 0.4444081783294678,
"step": 506,
"token_acc": 0.8558062740781508
},
{
"epoch": 1.8,
"grad_norm": 2.113614797592163,
"learning_rate": 3.7656134609333576e-06,
"loss": 0.4107510447502136,
"loss_sft": 0.4107510447502136,
"step": 507,
"token_acc": 0.8662553769526828
},
{
"epoch": 1.8035555555555556,
"grad_norm": 2.201674699783325,
"learning_rate": 3.7467373387063973e-06,
"loss": 0.4345109462738037,
"loss_sft": 0.4345109462738037,
"step": 508,
"token_acc": 0.8577196577076052
},
{
"epoch": 1.8071111111111111,
"grad_norm": 2.115177869796753,
"learning_rate": 3.727880256666586e-06,
"loss": 0.42980408668518066,
"loss_sft": 0.42980408668518066,
"step": 509,
"token_acc": 0.8630675122060721
},
{
"epoch": 1.8106666666666666,
"grad_norm": 2.050177574157715,
"learning_rate": 3.709042501300052e-06,
"loss": 0.41575467586517334,
"loss_sft": 0.41575467586517334,
"step": 510,
"token_acc": 0.8641130367590242
},
{
"epoch": 1.8142222222222222,
"grad_norm": 2.065338611602783,
"learning_rate": 3.6902243587993068e-06,
"loss": 0.4081893563270569,
"loss_sft": 0.4081893563270569,
"step": 511,
"token_acc": 0.8658730594108289
},
{
"epoch": 1.8177777777777777,
"grad_norm": 2.1498680114746094,
"learning_rate": 3.6714261150588947e-06,
"loss": 0.4343623220920563,
"loss_sft": 0.4343623220920563,
"step": 512,
"token_acc": 0.8613245931283906
},
{
"epoch": 1.8213333333333335,
"grad_norm": 2.0592949390411377,
"learning_rate": 3.652648055671043e-06,
"loss": 0.4178444743156433,
"loss_sft": 0.4178444743156433,
"step": 513,
"token_acc": 0.863631234484315
},
{
"epoch": 1.8248888888888888,
"grad_norm": 2.1280148029327393,
"learning_rate": 3.6338904659213335e-06,
"loss": 0.41738808155059814,
"loss_sft": 0.41738808155059814,
"step": 514,
"token_acc": 0.8658208151295709
},
{
"epoch": 1.8284444444444445,
"grad_norm": 2.0445332527160645,
"learning_rate": 3.6151536307843625e-06,
"loss": 0.42178139090538025,
"loss_sft": 0.42178139090538025,
"step": 515,
"token_acc": 0.8629355622070367
},
{
"epoch": 1.8319999999999999,
"grad_norm": 1.9766736030578613,
"learning_rate": 3.5964378349194075e-06,
"loss": 0.3943728804588318,
"loss_sft": 0.3943728804588318,
"step": 516,
"token_acc": 0.8735380507343125
},
{
"epoch": 1.8355555555555556,
"grad_norm": 2.126513957977295,
"learning_rate": 3.577743362666112e-06,
"loss": 0.4394909143447876,
"loss_sft": 0.4394909143447876,
"step": 517,
"token_acc": 0.8551081895371131
},
{
"epoch": 1.8391111111111111,
"grad_norm": 2.24725341796875,
"learning_rate": 3.5590704980401564e-06,
"loss": 0.40318551659584045,
"loss_sft": 0.40318551659584045,
"step": 518,
"token_acc": 0.8716417910447761
},
{
"epoch": 1.8426666666666667,
"grad_norm": 2.1285533905029297,
"learning_rate": 3.540419524728954e-06,
"loss": 0.412613183259964,
"loss_sft": 0.412613183259964,
"step": 519,
"token_acc": 0.8662673533363189
},
{
"epoch": 1.8462222222222222,
"grad_norm": 2.0844173431396484,
"learning_rate": 3.5217907260873265e-06,
"loss": 0.3816235661506653,
"loss_sft": 0.3816235661506653,
"step": 520,
"token_acc": 0.8734657442535149
},
{
"epoch": 1.8497777777777777,
"grad_norm": 2.0110743045806885,
"learning_rate": 3.5031843851332105e-06,
"loss": 0.3994944095611572,
"loss_sft": 0.3994944095611572,
"step": 521,
"token_acc": 0.8749266627553469
},
{
"epoch": 1.8533333333333335,
"grad_norm": 2.195115804672241,
"learning_rate": 3.4846007845433568e-06,
"loss": 0.41000989079475403,
"loss_sft": 0.41000989079475403,
"step": 522,
"token_acc": 0.8674979606106514
},
{
"epoch": 1.8568888888888888,
"grad_norm": 2.0471348762512207,
"learning_rate": 3.4660402066490274e-06,
"loss": 0.3831998109817505,
"loss_sft": 0.3831998109817505,
"step": 523,
"token_acc": 0.8717228464419475
},
{
"epoch": 1.8604444444444446,
"grad_norm": 2.127256393432617,
"learning_rate": 3.4475029334317195e-06,
"loss": 0.41519564390182495,
"loss_sft": 0.41519564390182495,
"step": 524,
"token_acc": 0.8658607126330402
},
{
"epoch": 1.8639999999999999,
"grad_norm": 2.1483802795410156,
"learning_rate": 3.428989246518867e-06,
"loss": 0.39058467745780945,
"loss_sft": 0.39058467745780945,
"step": 525,
"token_acc": 0.8687976539589443
},
{
"epoch": 1.8675555555555556,
"grad_norm": 1.9400416612625122,
"learning_rate": 3.410499427179572e-06,
"loss": 0.3935142159461975,
"loss_sft": 0.3935142159461975,
"step": 526,
"token_acc": 0.8684092158528662
},
{
"epoch": 1.871111111111111,
"grad_norm": 2.1435794830322266,
"learning_rate": 3.392033756320333e-06,
"loss": 0.41104888916015625,
"loss_sft": 0.41104888916015625,
"step": 527,
"token_acc": 0.865040650406504
},
{
"epoch": 1.8746666666666667,
"grad_norm": 2.0702149868011475,
"learning_rate": 3.3735925144807623e-06,
"loss": 0.4158642888069153,
"loss_sft": 0.4158642888069153,
"step": 528,
"token_acc": 0.8647888846840028
},
{
"epoch": 1.8782222222222222,
"grad_norm": 2.2176244258880615,
"learning_rate": 3.355175981829346e-06,
"loss": 0.4156597852706909,
"loss_sft": 0.4156597852706909,
"step": 529,
"token_acc": 0.8642835504392609
},
{
"epoch": 1.8817777777777778,
"grad_norm": 2.1262834072113037,
"learning_rate": 3.33678443815917e-06,
"loss": 0.41688382625579834,
"loss_sft": 0.41688382625579834,
"step": 530,
"token_acc": 0.8633536861148802
},
{
"epoch": 1.8853333333333333,
"grad_norm": 1.9204524755477905,
"learning_rate": 3.318418162883671e-06,
"loss": 0.3793541491031647,
"loss_sft": 0.3793541491031647,
"step": 531,
"token_acc": 0.8763848994665573
},
{
"epoch": 1.8888888888888888,
"grad_norm": 2.105459451675415,
"learning_rate": 3.300077435032406e-06,
"loss": 0.4083092212677002,
"loss_sft": 0.4083092212677002,
"step": 532,
"token_acc": 0.868655435132525
},
{
"epoch": 1.8924444444444446,
"grad_norm": 2.1021809577941895,
"learning_rate": 3.281762533246794e-06,
"loss": 0.4291813373565674,
"loss_sft": 0.4291813373565674,
"step": 533,
"token_acc": 0.8602228412256268
},
{
"epoch": 1.896,
"grad_norm": 2.3498737812042236,
"learning_rate": 3.2634737357758994e-06,
"loss": 0.4291939437389374,
"loss_sft": 0.4291939437389374,
"step": 534,
"token_acc": 0.8617167919799499
},
{
"epoch": 1.8995555555555557,
"grad_norm": 2.1913115978240967,
"learning_rate": 3.245211320472189e-06,
"loss": 0.3991416096687317,
"loss_sft": 0.3991416096687317,
"step": 535,
"token_acc": 0.8679076427420799
},
{
"epoch": 1.903111111111111,
"grad_norm": 2.0638091564178467,
"learning_rate": 3.226975564787322e-06,
"loss": 0.408561646938324,
"loss_sft": 0.408561646938324,
"step": 536,
"token_acc": 0.8670682496052401
},
{
"epoch": 1.9066666666666667,
"grad_norm": 2.1876537799835205,
"learning_rate": 3.208766745767935e-06,
"loss": 0.40613359212875366,
"loss_sft": 0.40613359212875366,
"step": 537,
"token_acc": 0.8674216354877057
},
{
"epoch": 1.9102222222222223,
"grad_norm": 2.1358630657196045,
"learning_rate": 3.1905851400514232e-06,
"loss": 0.40880149602890015,
"loss_sft": 0.40880149602890015,
"step": 538,
"token_acc": 0.8671139287253031
},
{
"epoch": 1.9137777777777778,
"grad_norm": 2.0311546325683594,
"learning_rate": 3.1724310238617475e-06,
"loss": 0.3915891647338867,
"loss_sft": 0.3915891647338867,
"step": 539,
"token_acc": 0.8688214019174465
},
{
"epoch": 1.9173333333333333,
"grad_norm": 2.1547634601593018,
"learning_rate": 3.154304673005235e-06,
"loss": 0.39122992753982544,
"loss_sft": 0.39122992753982544,
"step": 540,
"token_acc": 0.8704589206980066
},
{
"epoch": 1.9208888888888889,
"grad_norm": 2.2282280921936035,
"learning_rate": 3.1362063628663836e-06,
"loss": 0.44803735613822937,
"loss_sft": 0.44803735613822937,
"step": 541,
"token_acc": 0.8549925313110421
},
{
"epoch": 1.9244444444444444,
"grad_norm": 1.9914199113845825,
"learning_rate": 3.118136368403689e-06,
"loss": 0.40213990211486816,
"loss_sft": 0.40213990211486816,
"step": 542,
"token_acc": 0.8693961213013087
},
{
"epoch": 1.928,
"grad_norm": 2.1156744956970215,
"learning_rate": 3.100094964145458e-06,
"loss": 0.3796846270561218,
"loss_sft": 0.3796846270561218,
"step": 543,
"token_acc": 0.8778777536456718
},
{
"epoch": 1.9315555555555557,
"grad_norm": 2.1139566898345947,
"learning_rate": 3.0820824241856377e-06,
"loss": 0.38775634765625,
"loss_sft": 0.38775634765625,
"step": 544,
"token_acc": 0.8717824480649264
},
{
"epoch": 1.935111111111111,
"grad_norm": 2.048475503921509,
"learning_rate": 3.064099022179661e-06,
"loss": 0.39649468660354614,
"loss_sft": 0.39649468660354614,
"step": 545,
"token_acc": 0.8703424389698899
},
{
"epoch": 1.9386666666666668,
"grad_norm": 1.9958182573318481,
"learning_rate": 3.046145031340275e-06,
"loss": 0.41872507333755493,
"loss_sft": 0.41872507333755493,
"step": 546,
"token_acc": 0.8660214235223751
},
{
"epoch": 1.942222222222222,
"grad_norm": 2.1013219356536865,
"learning_rate": 3.0282207244334084e-06,
"loss": 0.4178723096847534,
"loss_sft": 0.4178723096847534,
"step": 547,
"token_acc": 0.8607965321051205
},
{
"epoch": 1.9457777777777778,
"grad_norm": 2.1734778881073,
"learning_rate": 3.010326373774004e-06,
"loss": 0.3943750560283661,
"loss_sft": 0.3943750560283661,
"step": 548,
"token_acc": 0.868645945432677
},
{
"epoch": 1.9493333333333334,
"grad_norm": 2.0213682651519775,
"learning_rate": 2.9924622512219037e-06,
"loss": 0.3996032476425171,
"loss_sft": 0.3996032476425171,
"step": 549,
"token_acc": 0.8714108771534738
},
{
"epoch": 1.952888888888889,
"grad_norm": 2.0122392177581787,
"learning_rate": 2.9746286281777075e-06,
"loss": 0.39854031801223755,
"loss_sft": 0.39854031801223755,
"step": 550,
"token_acc": 0.8720117891059928
},
{
"epoch": 1.9564444444444444,
"grad_norm": 2.0792717933654785,
"learning_rate": 2.9568257755786474e-06,
"loss": 0.4097796678543091,
"loss_sft": 0.4097796678543091,
"step": 551,
"token_acc": 0.866356866916746
},
{
"epoch": 1.96,
"grad_norm": 2.078012228012085,
"learning_rate": 2.939053963894481e-06,
"loss": 0.4321756958961487,
"loss_sft": 0.4321756958961487,
"step": 552,
"token_acc": 0.8608705185834886
},
{
"epoch": 1.9635555555555557,
"grad_norm": 2.068402051925659,
"learning_rate": 2.921313463123375e-06,
"loss": 0.3985518515110016,
"loss_sft": 0.3985518515110016,
"step": 553,
"token_acc": 0.8717814892136395
},
{
"epoch": 1.967111111111111,
"grad_norm": 2.1032450199127197,
"learning_rate": 2.9036045427878e-06,
"loss": 0.3798477351665497,
"loss_sft": 0.3798477351665497,
"step": 554,
"token_acc": 0.8731815306767868
},
{
"epoch": 1.9706666666666668,
"grad_norm": 2.1064600944519043,
"learning_rate": 2.8859274719304543e-06,
"loss": 0.4290257692337036,
"loss_sft": 0.4290257692337036,
"step": 555,
"token_acc": 0.8605027932960894
},
{
"epoch": 1.974222222222222,
"grad_norm": 2.1887898445129395,
"learning_rate": 2.8682825191101447e-06,
"loss": 0.38799673318862915,
"loss_sft": 0.38799673318862915,
"step": 556,
"token_acc": 0.8706955388719985
},
{
"epoch": 1.9777777777777779,
"grad_norm": 2.055995225906372,
"learning_rate": 2.850669952397743e-06,
"loss": 0.4083443284034729,
"loss_sft": 0.4083443284034729,
"step": 557,
"token_acc": 0.8700440528634361
},
{
"epoch": 1.9813333333333332,
"grad_norm": 2.176994562149048,
"learning_rate": 2.8330900393720846e-06,
"loss": 0.44645291566848755,
"loss_sft": 0.44645291566848755,
"step": 558,
"token_acc": 0.8595201030651135
},
{
"epoch": 1.984888888888889,
"grad_norm": 1.9817695617675781,
"learning_rate": 2.8155430471159118e-06,
"loss": 0.3868786692619324,
"loss_sft": 0.3868786692619324,
"step": 559,
"token_acc": 0.8751240489579888
},
{
"epoch": 1.9884444444444445,
"grad_norm": 2.2523109912872314,
"learning_rate": 2.7980292422118282e-06,
"loss": 0.4071633815765381,
"loss_sft": 0.4071633815765381,
"step": 560,
"token_acc": 0.8678319533885311
},
{
"epoch": 1.992,
"grad_norm": 2.1368203163146973,
"learning_rate": 2.7805488907382316e-06,
"loss": 0.43625980615615845,
"loss_sft": 0.43625980615615845,
"step": 561,
"token_acc": 0.8613791987922575
},
{
"epoch": 1.9955555555555555,
"grad_norm": 2.0597100257873535,
"learning_rate": 2.7631022582652808e-06,
"loss": 0.37715715169906616,
"loss_sft": 0.37715715169906616,
"step": 562,
"token_acc": 0.8763434777689784
},
{
"epoch": 1.999111111111111,
"grad_norm": 2.1500675678253174,
"learning_rate": 2.745689609850859e-06,
"loss": 0.377418577671051,
"loss_sft": 0.377418577671051,
"step": 563,
"token_acc": 0.8753953610964681
},
{
"epoch": 2.0,
"grad_norm": 4.0047926902771,
"learning_rate": 2.728311210036542e-06,
"loss": 0.3969249427318573,
"loss_sft": 0.3969249427318573,
"step": 564,
"token_acc": 0.8686021505376345
},
{
"epoch": 2.0035555555555558,
"grad_norm": 1.9045144319534302,
"learning_rate": 2.7109673228435925e-06,
"loss": 0.2640722393989563,
"loss_sft": 0.2640722393989563,
"step": 565,
"token_acc": 0.9178060328775869
},
{
"epoch": 2.007111111111111,
"grad_norm": 2.002485513687134,
"learning_rate": 2.6936582117689347e-06,
"loss": 0.28322356939315796,
"loss_sft": 0.28322356939315796,
"step": 566,
"token_acc": 0.912829311489927
},
{
"epoch": 2.010666666666667,
"grad_norm": 1.993098497390747,
"learning_rate": 2.6763841397811576e-06,
"loss": 0.24907240271568298,
"loss_sft": 0.24907240271568298,
"step": 567,
"token_acc": 0.9214471564861851
},
{
"epoch": 2.014222222222222,
"grad_norm": 2.0112924575805664,
"learning_rate": 2.6591453693165204e-06,
"loss": 0.25736403465270996,
"loss_sft": 0.25736403465270996,
"step": 568,
"token_acc": 0.9153152124257652
},
{
"epoch": 2.017777777777778,
"grad_norm": 1.9399166107177734,
"learning_rate": 2.641942162274962e-06,
"loss": 0.22330206632614136,
"loss_sft": 0.22330206632614136,
"step": 569,
"token_acc": 0.9280085323220951
},
{
"epoch": 2.021333333333333,
"grad_norm": 2.019472360610962,
"learning_rate": 2.624774780016126e-06,
"loss": 0.22442716360092163,
"loss_sft": 0.22442716360092163,
"step": 570,
"token_acc": 0.92497479390309
},
{
"epoch": 2.024888888888889,
"grad_norm": 2.200451135635376,
"learning_rate": 2.607643483355387e-06,
"loss": 0.21958491206169128,
"loss_sft": 0.21958491206169128,
"step": 571,
"token_acc": 0.9281897328129691
},
{
"epoch": 2.0284444444444443,
"grad_norm": 2.454540729522705,
"learning_rate": 2.59054853255989e-06,
"loss": 0.2276432067155838,
"loss_sft": 0.2276432067155838,
"step": 572,
"token_acc": 0.9281120394301857
},
{
"epoch": 2.032,
"grad_norm": 2.893754243850708,
"learning_rate": 2.573490187344596e-06,
"loss": 0.2316417396068573,
"loss_sft": 0.2316417396068573,
"step": 573,
"token_acc": 0.9252141568981064
},
{
"epoch": 2.0355555555555553,
"grad_norm": 2.72332763671875,
"learning_rate": 2.5564687068683335e-06,
"loss": 0.23335057497024536,
"loss_sft": 0.23335057497024536,
"step": 574,
"token_acc": 0.9263855421686747
},
{
"epoch": 2.039111111111111,
"grad_norm": 3.0100057125091553,
"learning_rate": 2.539484349729868e-06,
"loss": 0.2272835671901703,
"loss_sft": 0.2272835671901703,
"step": 575,
"token_acc": 0.9250593055855079
},
{
"epoch": 2.042666666666667,
"grad_norm": 2.727752447128296,
"learning_rate": 2.5225373739639637e-06,
"loss": 0.24300718307495117,
"loss_sft": 0.24300718307495117,
"step": 576,
"token_acc": 0.9186033329554473
},
{
"epoch": 2.046222222222222,
"grad_norm": 2.5885136127471924,
"learning_rate": 2.5056280370374725e-06,
"loss": 0.2558279037475586,
"loss_sft": 0.2558279037475586,
"step": 577,
"token_acc": 0.9146695508624125
},
{
"epoch": 2.049777777777778,
"grad_norm": 2.411102294921875,
"learning_rate": 2.488756595845417e-06,
"loss": 0.2502776086330414,
"loss_sft": 0.2502776086330414,
"step": 578,
"token_acc": 0.9200146858281758
},
{
"epoch": 2.0533333333333332,
"grad_norm": 2.4136903285980225,
"learning_rate": 2.471923306707086e-06,
"loss": 0.21444688737392426,
"loss_sft": 0.21444688737392426,
"step": 579,
"token_acc": 0.9298335796440275
},
{
"epoch": 2.056888888888889,
"grad_norm": 2.271580219268799,
"learning_rate": 2.455128425362153e-06,
"loss": 0.23721033334732056,
"loss_sft": 0.23721033334732056,
"step": 580,
"token_acc": 0.9240779768177029
},
{
"epoch": 2.0604444444444443,
"grad_norm": 2.2007665634155273,
"learning_rate": 2.4383722069667683e-06,
"loss": 0.2211160808801651,
"loss_sft": 0.2211160808801651,
"step": 581,
"token_acc": 0.9300287895856804
},
{
"epoch": 2.064,
"grad_norm": 2.361142158508301,
"learning_rate": 2.4216549060897026e-06,
"loss": 0.240919828414917,
"loss_sft": 0.240919828414917,
"step": 582,
"token_acc": 0.9231724627395316
},
{
"epoch": 2.0675555555555554,
"grad_norm": 2.27839994430542,
"learning_rate": 2.404976776708477e-06,
"loss": 0.21856752038002014,
"loss_sft": 0.21856752038002014,
"step": 583,
"token_acc": 0.929429604572495
},
{
"epoch": 2.071111111111111,
"grad_norm": 2.153813362121582,
"learning_rate": 2.3883380722054865e-06,
"loss": 0.23440015316009521,
"loss_sft": 0.23440015316009521,
"step": 584,
"token_acc": 0.9247323880513366
},
{
"epoch": 2.074666666666667,
"grad_norm": 2.155805826187134,
"learning_rate": 2.371739045364178e-06,
"loss": 0.2542189955711365,
"loss_sft": 0.2542189955711365,
"step": 585,
"token_acc": 0.917460839084237
},
{
"epoch": 2.078222222222222,
"grad_norm": 2.1631388664245605,
"learning_rate": 2.3551799483651894e-06,
"loss": 0.2051049768924713,
"loss_sft": 0.2051049768924713,
"step": 586,
"token_acc": 0.9327362685538924
},
{
"epoch": 2.081777777777778,
"grad_norm": 2.194667100906372,
"learning_rate": 2.3386610327825194e-06,
"loss": 0.23108305037021637,
"loss_sft": 0.23108305037021637,
"step": 587,
"token_acc": 0.9259168404268368
},
{
"epoch": 2.0853333333333333,
"grad_norm": 2.2058894634246826,
"learning_rate": 2.3221825495797213e-06,
"loss": 0.22139829397201538,
"loss_sft": 0.22139829397201538,
"step": 588,
"token_acc": 0.9298180004986287
},
{
"epoch": 2.088888888888889,
"grad_norm": 2.3936402797698975,
"learning_rate": 2.3057447491060725e-06,
"loss": 0.2293173223733902,
"loss_sft": 0.2293173223733902,
"step": 589,
"token_acc": 0.9277040206380167
},
{
"epoch": 2.0924444444444443,
"grad_norm": 2.2239348888397217,
"learning_rate": 2.28934788109278e-06,
"loss": 0.2269078493118286,
"loss_sft": 0.2269078493118286,
"step": 590,
"token_acc": 0.9260611310481376
},
{
"epoch": 2.096,
"grad_norm": 2.281940460205078,
"learning_rate": 2.2729921946491855e-06,
"loss": 0.22220298647880554,
"loss_sft": 0.22220298647880554,
"step": 591,
"token_acc": 0.9265550383848153
},
{
"epoch": 2.0995555555555554,
"grad_norm": 2.245628833770752,
"learning_rate": 2.2566779382589788e-06,
"loss": 0.2203902304172516,
"loss_sft": 0.2203902304172516,
"step": 592,
"token_acc": 0.9247014115092291
},
{
"epoch": 2.103111111111111,
"grad_norm": 2.3485538959503174,
"learning_rate": 2.240405359776424e-06,
"loss": 0.22701899707317352,
"loss_sft": 0.22701899707317352,
"step": 593,
"token_acc": 0.925434003857812
},
{
"epoch": 2.1066666666666665,
"grad_norm": 2.4488658905029297,
"learning_rate": 2.2241747064225942e-06,
"loss": 0.23555168509483337,
"loss_sft": 0.23555168509483337,
"step": 594,
"token_acc": 0.9227388799809965
},
{
"epoch": 2.110222222222222,
"grad_norm": 2.4494524002075195,
"learning_rate": 2.2079862247816148e-06,
"loss": 0.22785550355911255,
"loss_sft": 0.22785550355911255,
"step": 595,
"token_acc": 0.9270127058287277
},
{
"epoch": 2.113777777777778,
"grad_norm": 2.1819405555725098,
"learning_rate": 2.191840160796918e-06,
"loss": 0.22543713450431824,
"loss_sft": 0.22543713450431824,
"step": 596,
"token_acc": 0.926160680426892
},
{
"epoch": 2.1173333333333333,
"grad_norm": 2.2094991207122803,
"learning_rate": 2.175736759767505e-06,
"loss": 0.22682014107704163,
"loss_sft": 0.22682014107704163,
"step": 597,
"token_acc": 0.9274349602236078
},
{
"epoch": 2.120888888888889,
"grad_norm": 2.484931468963623,
"learning_rate": 2.159676266344222e-06,
"loss": 0.23292362689971924,
"loss_sft": 0.23292362689971924,
"step": 598,
"token_acc": 0.925633470830878
},
{
"epoch": 2.1244444444444444,
"grad_norm": 2.3404879570007324,
"learning_rate": 2.1436589245260375e-06,
"loss": 0.23116694390773773,
"loss_sft": 0.23116694390773773,
"step": 599,
"token_acc": 0.9248289016043981
},
{
"epoch": 2.128,
"grad_norm": 2.2433395385742188,
"learning_rate": 2.127684977656349e-06,
"loss": 0.23524919152259827,
"loss_sft": 0.23524919152259827,
"step": 600,
"token_acc": 0.9223147939653231
},
{
"epoch": 2.1315555555555554,
"grad_norm": 2.2880399227142334,
"learning_rate": 2.111754668419264e-06,
"loss": 0.23501253128051758,
"loss_sft": 0.23501253128051758,
"step": 601,
"token_acc": 0.925569078205995
},
{
"epoch": 2.135111111111111,
"grad_norm": 2.199207067489624,
"learning_rate": 2.095868238835932e-06,
"loss": 0.2567347288131714,
"loss_sft": 0.2567347288131714,
"step": 602,
"token_acc": 0.9164026457673281
},
{
"epoch": 2.1386666666666665,
"grad_norm": 2.0758249759674072,
"learning_rate": 2.0800259302608656e-06,
"loss": 0.23210355639457703,
"loss_sft": 0.23210355639457703,
"step": 603,
"token_acc": 0.9254466098254162
},
{
"epoch": 2.1422222222222222,
"grad_norm": 2.1163580417633057,
"learning_rate": 2.064227983378256e-06,
"loss": 0.22319991886615753,
"loss_sft": 0.22319991886615753,
"step": 604,
"token_acc": 0.9281840497802135
},
{
"epoch": 2.145777777777778,
"grad_norm": 2.1162099838256836,
"learning_rate": 2.048474638198347e-06,
"loss": 0.21498337388038635,
"loss_sft": 0.21498337388038635,
"step": 605,
"token_acc": 0.9294320137693631
},
{
"epoch": 2.1493333333333333,
"grad_norm": 2.196406602859497,
"learning_rate": 2.0327661340537536e-06,
"loss": 0.22724901139736176,
"loss_sft": 0.22724901139736176,
"step": 606,
"token_acc": 0.9259239507226281
},
{
"epoch": 2.152888888888889,
"grad_norm": 2.0686843395233154,
"learning_rate": 2.0171027095958543e-06,
"loss": 0.23336181044578552,
"loss_sft": 0.23336181044578552,
"step": 607,
"token_acc": 0.9254922551850879
},
{
"epoch": 2.1564444444444444,
"grad_norm": 2.2436225414276123,
"learning_rate": 2.001484602791157e-06,
"loss": 0.24389603734016418,
"loss_sft": 0.24389603734016418,
"step": 608,
"token_acc": 0.9204148411069825
},
{
"epoch": 2.16,
"grad_norm": 2.36028790473938,
"learning_rate": 1.9859120509176706e-06,
"loss": 0.2247694432735443,
"loss_sft": 0.2247694432735443,
"step": 609,
"token_acc": 0.9258828739067863
},
{
"epoch": 2.1635555555555555,
"grad_norm": 2.2761528491973877,
"learning_rate": 1.970385290561317e-06,
"loss": 0.24648919701576233,
"loss_sft": 0.24648919701576233,
"step": 610,
"token_acc": 0.919445343692673
},
{
"epoch": 2.167111111111111,
"grad_norm": 2.1917624473571777,
"learning_rate": 1.954904557612337e-06,
"loss": 0.22748097777366638,
"loss_sft": 0.22748097777366638,
"step": 611,
"token_acc": 0.9236482796286183
},
{
"epoch": 2.1706666666666665,
"grad_norm": 2.3685874938964844,
"learning_rate": 1.9394700872616856e-06,
"loss": 0.2315749078989029,
"loss_sft": 0.2315749078989029,
"step": 612,
"token_acc": 0.9223126125566664
},
{
"epoch": 2.1742222222222223,
"grad_norm": 2.4561285972595215,
"learning_rate": 1.924082113997488e-06,
"loss": 0.21668648719787598,
"loss_sft": 0.21668648719787598,
"step": 613,
"token_acc": 0.926342237503306
},
{
"epoch": 2.1777777777777776,
"grad_norm": 2.407071590423584,
"learning_rate": 1.9087408716014562e-06,
"loss": 0.24422219395637512,
"loss_sft": 0.24422219395637512,
"step": 614,
"token_acc": 0.9214019885574627
},
{
"epoch": 2.1813333333333333,
"grad_norm": 2.2460415363311768,
"learning_rate": 1.8934465931453378e-06,
"loss": 0.22939246892929077,
"loss_sft": 0.22939246892929077,
"step": 615,
"token_acc": 0.9265570803800331
},
{
"epoch": 2.1848888888888887,
"grad_norm": 2.3530876636505127,
"learning_rate": 1.8781995109873929e-06,
"loss": 0.2310415357351303,
"loss_sft": 0.2310415357351303,
"step": 616,
"token_acc": 0.924309423993568
},
{
"epoch": 2.1884444444444444,
"grad_norm": 2.2692759037017822,
"learning_rate": 1.8629998567688445e-06,
"loss": 0.21300077438354492,
"loss_sft": 0.21300077438354492,
"step": 617,
"token_acc": 0.9311182184316512
},
{
"epoch": 2.192,
"grad_norm": 2.3331189155578613,
"learning_rate": 1.8478478614103684e-06,
"loss": 0.2366311401128769,
"loss_sft": 0.2366311401128769,
"step": 618,
"token_acc": 0.9262598882557946
},
{
"epoch": 2.1955555555555555,
"grad_norm": 2.183361768722534,
"learning_rate": 1.8327437551085842e-06,
"loss": 0.22136721014976501,
"loss_sft": 0.22136721014976501,
"step": 619,
"token_acc": 0.9283729371065615
},
{
"epoch": 2.1991111111111112,
"grad_norm": 2.2662577629089355,
"learning_rate": 1.8176877673325555e-06,
"loss": 0.2098524272441864,
"loss_sft": 0.2098524272441864,
"step": 620,
"token_acc": 0.9300883498875048
},
{
"epoch": 2.2026666666666666,
"grad_norm": 2.428753614425659,
"learning_rate": 1.802680126820307e-06,
"loss": 0.2367994487285614,
"loss_sft": 0.2367994487285614,
"step": 621,
"token_acc": 0.9235941684022613
},
{
"epoch": 2.2062222222222223,
"grad_norm": 2.2775909900665283,
"learning_rate": 1.7877210615753477e-06,
"loss": 0.207049161195755,
"loss_sft": 0.207049161195755,
"step": 622,
"token_acc": 0.9330305782621859
},
{
"epoch": 2.2097777777777776,
"grad_norm": 2.7947118282318115,
"learning_rate": 1.772810798863206e-06,
"loss": 0.21759605407714844,
"loss_sft": 0.21759605407714844,
"step": 623,
"token_acc": 0.9265487819957957
},
{
"epoch": 2.2133333333333334,
"grad_norm": 2.3573431968688965,
"learning_rate": 1.7579495652079786e-06,
"loss": 0.22782224416732788,
"loss_sft": 0.22782224416732788,
"step": 624,
"token_acc": 0.9269208211143695
},
{
"epoch": 2.2168888888888887,
"grad_norm": 2.163207530975342,
"learning_rate": 1.74313758638889e-06,
"loss": 0.22399967908859253,
"loss_sft": 0.22399967908859253,
"step": 625,
"token_acc": 0.9274890011406225
},
{
"epoch": 2.2204444444444444,
"grad_norm": 2.6319117546081543,
"learning_rate": 1.7283750874368577e-06,
"loss": 0.20773431658744812,
"loss_sft": 0.20773431658744812,
"step": 626,
"token_acc": 0.932035078669074
},
{
"epoch": 2.224,
"grad_norm": 2.4332780838012695,
"learning_rate": 1.713662292631078e-06,
"loss": 0.22071672976016998,
"loss_sft": 0.22071672976016998,
"step": 627,
"token_acc": 0.9254120213281629
},
{
"epoch": 2.2275555555555555,
"grad_norm": 2.3868393898010254,
"learning_rate": 1.6989994254956222e-06,
"loss": 0.24235326051712036,
"loss_sft": 0.24235326051712036,
"step": 628,
"token_acc": 0.9214900046490004
},
{
"epoch": 2.2311111111111113,
"grad_norm": 2.245732307434082,
"learning_rate": 1.6843867087960252e-06,
"loss": 0.22326138615608215,
"loss_sft": 0.22326138615608215,
"step": 629,
"token_acc": 0.9268034340506188
},
{
"epoch": 2.2346666666666666,
"grad_norm": 2.1434824466705322,
"learning_rate": 1.669824364535918e-06,
"loss": 0.20635294914245605,
"loss_sft": 0.20635294914245605,
"step": 630,
"token_acc": 0.9335148302049004
},
{
"epoch": 2.2382222222222223,
"grad_norm": 2.241582155227661,
"learning_rate": 1.6553126139536534e-06,
"loss": 0.22539973258972168,
"loss_sft": 0.22539973258972168,
"step": 631,
"token_acc": 0.9268402154398564
},
{
"epoch": 2.2417777777777776,
"grad_norm": 2.173891305923462,
"learning_rate": 1.640851677518927e-06,
"loss": 0.22469571232795715,
"loss_sft": 0.22469571232795715,
"step": 632,
"token_acc": 0.9270023480302635
},
{
"epoch": 2.2453333333333334,
"grad_norm": 2.215585947036743,
"learning_rate": 1.6264417749294543e-06,
"loss": 0.220083549618721,
"loss_sft": 0.220083549618721,
"step": 633,
"token_acc": 0.9289827255278311
},
{
"epoch": 2.2488888888888887,
"grad_norm": 2.354578733444214,
"learning_rate": 1.6120831251076118e-06,
"loss": 0.23738837242126465,
"loss_sft": 0.23738837242126465,
"step": 634,
"token_acc": 0.9206971921762036
},
{
"epoch": 2.2524444444444445,
"grad_norm": 2.184593439102173,
"learning_rate": 1.597775946197117e-06,
"loss": 0.19693312048912048,
"loss_sft": 0.19693312048912048,
"step": 635,
"token_acc": 0.9342680365555998
},
{
"epoch": 2.2560000000000002,
"grad_norm": 2.274911403656006,
"learning_rate": 1.5835204555597227e-06,
"loss": 0.22760219871997833,
"loss_sft": 0.22760219871997833,
"step": 636,
"token_acc": 0.9245377869630789
},
{
"epoch": 2.2595555555555555,
"grad_norm": 2.3674378395080566,
"learning_rate": 1.5693168697719024e-06,
"loss": 0.26029878854751587,
"loss_sft": 0.26029878854751587,
"step": 637,
"token_acc": 0.9171414348138724
},
{
"epoch": 2.2631111111111113,
"grad_norm": 2.100651502609253,
"learning_rate": 1.555165404621567e-06,
"loss": 0.22421355545520782,
"loss_sft": 0.22421355545520782,
"step": 638,
"token_acc": 0.9254526091586794
},
{
"epoch": 2.2666666666666666,
"grad_norm": 2.4591217041015625,
"learning_rate": 1.5410662751047855e-06,
"loss": 0.21270984411239624,
"loss_sft": 0.21270984411239624,
"step": 639,
"token_acc": 0.92981938813122
},
{
"epoch": 2.2702222222222224,
"grad_norm": 2.35151743888855,
"learning_rate": 1.5270196954225175e-06,
"loss": 0.2397022247314453,
"loss_sft": 0.2397022247314453,
"step": 640,
"token_acc": 0.9193046762190381
},
{
"epoch": 2.2737777777777777,
"grad_norm": 2.3232421875,
"learning_rate": 1.5130258789773583e-06,
"loss": 0.23034314811229706,
"loss_sft": 0.23034314811229706,
"step": 641,
"token_acc": 0.9261403508771929
},
{
"epoch": 2.2773333333333334,
"grad_norm": 2.2469310760498047,
"learning_rate": 1.4990850383703005e-06,
"loss": 0.23017071187496185,
"loss_sft": 0.23017071187496185,
"step": 642,
"token_acc": 0.9259094448770719
},
{
"epoch": 2.2808888888888887,
"grad_norm": 2.287365198135376,
"learning_rate": 1.4851973853974987e-06,
"loss": 0.20179936289787292,
"loss_sft": 0.20179936289787292,
"step": 643,
"token_acc": 0.9300446663163426
},
{
"epoch": 2.2844444444444445,
"grad_norm": 2.4203362464904785,
"learning_rate": 1.4713631310470571e-06,
"loss": 0.23404662311077118,
"loss_sft": 0.23404662311077118,
"step": 644,
"token_acc": 0.9235795272564218
},
{
"epoch": 2.288,
"grad_norm": 2.208920478820801,
"learning_rate": 1.457582485495821e-06,
"loss": 0.2153758406639099,
"loss_sft": 0.2153758406639099,
"step": 645,
"token_acc": 0.9302123026673925
},
{
"epoch": 2.2915555555555556,
"grad_norm": 2.303738832473755,
"learning_rate": 1.4438556581061819e-06,
"loss": 0.2312268316745758,
"loss_sft": 0.2312268316745758,
"step": 646,
"token_acc": 0.9265285151567049
},
{
"epoch": 2.295111111111111,
"grad_norm": 2.2923800945281982,
"learning_rate": 1.4301828574229026e-06,
"loss": 0.21543818712234497,
"loss_sft": 0.21543818712234497,
"step": 647,
"token_acc": 0.9304871690899612
},
{
"epoch": 2.2986666666666666,
"grad_norm": 2.255633592605591,
"learning_rate": 1.4165642911699435e-06,
"loss": 0.2189178615808487,
"loss_sft": 0.2189178615808487,
"step": 648,
"token_acc": 0.9276665304454434
},
{
"epoch": 2.3022222222222224,
"grad_norm": 2.287534475326538,
"learning_rate": 1.4030001662473086e-06,
"loss": 0.23445531725883484,
"loss_sft": 0.23445531725883484,
"step": 649,
"token_acc": 0.9227652969958436
},
{
"epoch": 2.3057777777777777,
"grad_norm": 2.2217423915863037,
"learning_rate": 1.389490688727903e-06,
"loss": 0.22768665850162506,
"loss_sft": 0.22768665850162506,
"step": 650,
"token_acc": 0.9245046686404008
},
{
"epoch": 2.3093333333333335,
"grad_norm": 2.1874356269836426,
"learning_rate": 1.3760360638544012e-06,
"loss": 0.22763150930404663,
"loss_sft": 0.22763150930404663,
"step": 651,
"token_acc": 0.9265893559743218
},
{
"epoch": 2.3128888888888888,
"grad_norm": 2.3325886726379395,
"learning_rate": 1.3626364960361282e-06,
"loss": 0.21652105450630188,
"loss_sft": 0.21652105450630188,
"step": 652,
"token_acc": 0.9292892156862745
},
{
"epoch": 2.3164444444444445,
"grad_norm": 2.342155933380127,
"learning_rate": 1.3492921888459566e-06,
"loss": 0.21013683080673218,
"loss_sft": 0.21013683080673218,
"step": 653,
"token_acc": 0.9317246434008006
},
{
"epoch": 2.32,
"grad_norm": 2.149019479751587,
"learning_rate": 1.3360033450172106e-06,
"loss": 0.22255712747573853,
"loss_sft": 0.22255712747573853,
"step": 654,
"token_acc": 0.9261175496688742
},
{
"epoch": 2.3235555555555556,
"grad_norm": 2.420893430709839,
"learning_rate": 1.3227701664405868e-06,
"loss": 0.21703636646270752,
"loss_sft": 0.21703636646270752,
"step": 655,
"token_acc": 0.9300050492299924
},
{
"epoch": 2.327111111111111,
"grad_norm": 2.3207695484161377,
"learning_rate": 1.3095928541610936e-06,
"loss": 0.20958241820335388,
"loss_sft": 0.20958241820335388,
"step": 656,
"token_acc": 0.9306846445869879
},
{
"epoch": 2.3306666666666667,
"grad_norm": 2.0305697917938232,
"learning_rate": 1.2964716083749829e-06,
"loss": 0.20475006103515625,
"loss_sft": 0.20475006103515625,
"step": 657,
"token_acc": 0.9350241039614337
},
{
"epoch": 2.3342222222222224,
"grad_norm": 2.2730162143707275,
"learning_rate": 1.2834066284267189e-06,
"loss": 0.24420136213302612,
"loss_sft": 0.24420136213302612,
"step": 658,
"token_acc": 0.9210352786020441
},
{
"epoch": 2.3377777777777777,
"grad_norm": 2.300133228302002,
"learning_rate": 1.2703981128059534e-06,
"loss": 0.23302239179611206,
"loss_sft": 0.23302239179611206,
"step": 659,
"token_acc": 0.924833023103033
},
{
"epoch": 2.3413333333333335,
"grad_norm": 2.323042154312134,
"learning_rate": 1.257446259144494e-06,
"loss": 0.21388910710811615,
"loss_sft": 0.21388910710811615,
"step": 660,
"token_acc": 0.9307247446782331
},
{
"epoch": 2.344888888888889,
"grad_norm": 2.2700846195220947,
"learning_rate": 1.2445512642133218e-06,
"loss": 0.2315957099199295,
"loss_sft": 0.2315957099199295,
"step": 661,
"token_acc": 0.9251601787541054
},
{
"epoch": 2.3484444444444446,
"grad_norm": 2.3249659538269043,
"learning_rate": 1.2317133239195866e-06,
"loss": 0.21437618136405945,
"loss_sft": 0.21437618136405945,
"step": 662,
"token_acc": 0.929119712559203
},
{
"epoch": 2.352,
"grad_norm": 2.288879632949829,
"learning_rate": 1.2189326333036323e-06,
"loss": 0.21847400069236755,
"loss_sft": 0.21847400069236755,
"step": 663,
"token_acc": 0.9282767196066241
},
{
"epoch": 2.3555555555555556,
"grad_norm": 2.321206569671631,
"learning_rate": 1.2062093865360458e-06,
"loss": 0.21765762567520142,
"loss_sft": 0.21765762567520142,
"step": 664,
"token_acc": 0.9302363311881456
},
{
"epoch": 2.359111111111111,
"grad_norm": 2.234557867050171,
"learning_rate": 1.193543776914693e-06,
"loss": 0.213822603225708,
"loss_sft": 0.213822603225708,
"step": 665,
"token_acc": 0.9318563458100861
},
{
"epoch": 2.3626666666666667,
"grad_norm": 2.3301467895507812,
"learning_rate": 1.1809359968617894e-06,
"loss": 0.23474878072738647,
"loss_sft": 0.23474878072738647,
"step": 666,
"token_acc": 0.9235434956105347
},
{
"epoch": 2.3662222222222224,
"grad_norm": 2.311751365661621,
"learning_rate": 1.1683862379209747e-06,
"loss": 0.2049456536769867,
"loss_sft": 0.2049456536769867,
"step": 667,
"token_acc": 0.934585635359116
},
{
"epoch": 2.3697777777777778,
"grad_norm": 2.3276891708374023,
"learning_rate": 1.1558946907544034e-06,
"loss": 0.23988944292068481,
"loss_sft": 0.23988944292068481,
"step": 668,
"token_acc": 0.9234197948552709
},
{
"epoch": 2.3733333333333335,
"grad_norm": 2.351001024246216,
"learning_rate": 1.1434615451398467e-06,
"loss": 0.2275446057319641,
"loss_sft": 0.2275446057319641,
"step": 669,
"token_acc": 0.9258532998345487
},
{
"epoch": 2.376888888888889,
"grad_norm": 2.213923454284668,
"learning_rate": 1.1310869899678122e-06,
"loss": 0.21022680401802063,
"loss_sft": 0.21022680401802063,
"step": 670,
"token_acc": 0.9326544781300625
},
{
"epoch": 2.3804444444444446,
"grad_norm": 2.255131483078003,
"learning_rate": 1.1187712132386723e-06,
"loss": 0.2283816635608673,
"loss_sft": 0.2283816635608673,
"step": 671,
"token_acc": 0.9264109085006222
},
{
"epoch": 2.384,
"grad_norm": 2.1072750091552734,
"learning_rate": 1.1065144020598067e-06,
"loss": 0.22100964188575745,
"loss_sft": 0.22100964188575745,
"step": 672,
"token_acc": 0.9259102231172823
},
{
"epoch": 2.3875555555555557,
"grad_norm": 2.1687779426574707,
"learning_rate": 1.0943167426427625e-06,
"loss": 0.21291211247444153,
"loss_sft": 0.21291211247444153,
"step": 673,
"token_acc": 0.9302440633245382
},
{
"epoch": 2.391111111111111,
"grad_norm": 2.214552164077759,
"learning_rate": 1.0821784203004238e-06,
"loss": 0.20286612212657928,
"loss_sft": 0.20286612212657928,
"step": 674,
"token_acc": 0.933087191138881
},
{
"epoch": 2.3946666666666667,
"grad_norm": 2.2621734142303467,
"learning_rate": 1.0700996194441944e-06,
"loss": 0.21771803498268127,
"loss_sft": 0.21771803498268127,
"step": 675,
"token_acc": 0.928151414920985
},
{
"epoch": 2.398222222222222,
"grad_norm": 2.2273988723754883,
"learning_rate": 1.0580805235812042e-06,
"loss": 0.2276049554347992,
"loss_sft": 0.2276049554347992,
"step": 676,
"token_acc": 0.9253663264724329
},
{
"epoch": 2.401777777777778,
"grad_norm": 2.3285980224609375,
"learning_rate": 1.046121315311508e-06,
"loss": 0.2154673933982849,
"loss_sft": 0.2154673933982849,
"step": 677,
"token_acc": 0.9301025163094129
},
{
"epoch": 2.405333333333333,
"grad_norm": 2.3074052333831787,
"learning_rate": 1.0342221763253207e-06,
"loss": 0.23719647526741028,
"loss_sft": 0.23719647526741028,
"step": 678,
"token_acc": 0.9255713076597546
},
{
"epoch": 2.408888888888889,
"grad_norm": 2.2709662914276123,
"learning_rate": 1.0223832874002603e-06,
"loss": 0.22504989802837372,
"loss_sft": 0.22504989802837372,
"step": 679,
"token_acc": 0.92686154908505
},
{
"epoch": 2.4124444444444446,
"grad_norm": 2.275752305984497,
"learning_rate": 1.010604828398587e-06,
"loss": 0.2162834256887436,
"loss_sft": 0.2162834256887436,
"step": 680,
"token_acc": 0.9294919956076981
},
{
"epoch": 2.416,
"grad_norm": 2.2758357524871826,
"learning_rate": 9.988869782644872e-07,
"loss": 0.21630850434303284,
"loss_sft": 0.21630850434303284,
"step": 681,
"token_acc": 0.9274036475240512
},
{
"epoch": 2.4195555555555557,
"grad_norm": 2.3172664642333984,
"learning_rate": 9.872299150213455e-07,
"loss": 0.24233470857143402,
"loss_sft": 0.24233470857143402,
"step": 682,
"token_acc": 0.9204110762203825
},
{
"epoch": 2.423111111111111,
"grad_norm": 2.223998546600342,
"learning_rate": 9.75633815769036e-07,
"loss": 0.2369760125875473,
"loss_sft": 0.2369760125875473,
"step": 683,
"token_acc": 0.923796581821983
},
{
"epoch": 2.4266666666666667,
"grad_norm": 2.217749834060669,
"learning_rate": 9.640988566812475e-07,
"loss": 0.24139933288097382,
"loss_sft": 0.24139933288097382,
"step": 684,
"token_acc": 0.9205305154194522
},
{
"epoch": 2.430222222222222,
"grad_norm": 2.2150449752807617,
"learning_rate": 9.526252130027919e-07,
"loss": 0.21766090393066406,
"loss_sft": 0.21766090393066406,
"step": 685,
"token_acc": 0.9294198264047511
},
{
"epoch": 2.433777777777778,
"grad_norm": 2.3251469135284424,
"learning_rate": 9.412130590469438e-07,
"loss": 0.22623226046562195,
"loss_sft": 0.22623226046562195,
"step": 686,
"token_acc": 0.9280179198450176
},
{
"epoch": 2.437333333333333,
"grad_norm": 2.674274444580078,
"learning_rate": 9.298625681928031e-07,
"loss": 0.23587551712989807,
"loss_sft": 0.23587551712989807,
"step": 687,
"token_acc": 0.92281676530982
},
{
"epoch": 2.440888888888889,
"grad_norm": 2.2052040100097656,
"learning_rate": 9.185739128826454e-07,
"loss": 0.20603349804878235,
"loss_sft": 0.20603349804878235,
"step": 688,
"token_acc": 0.933254716981132
},
{
"epoch": 2.4444444444444446,
"grad_norm": 2.258056402206421,
"learning_rate": 9.073472646193171e-07,
"loss": 0.2301127314567566,
"loss_sft": 0.2301127314567566,
"step": 689,
"token_acc": 0.9237278573846327
},
{
"epoch": 2.448,
"grad_norm": 2.1317646503448486,
"learning_rate": 8.961827939636198e-07,
"loss": 0.22052863240242004,
"loss_sft": 0.22052863240242004,
"step": 690,
"token_acc": 0.9286256643887624
},
{
"epoch": 2.4515555555555557,
"grad_norm": 2.2725841999053955,
"learning_rate": 8.850806705317183e-07,
"loss": 0.21507984399795532,
"loss_sft": 0.21507984399795532,
"step": 691,
"token_acc": 0.9299817292391113
},
{
"epoch": 2.455111111111111,
"grad_norm": 2.3653690814971924,
"learning_rate": 8.740410629925744e-07,
"loss": 0.2642444968223572,
"loss_sft": 0.2642444968223572,
"step": 692,
"token_acc": 0.9142736930682654
},
{
"epoch": 2.458666666666667,
"grad_norm": 2.2154908180236816,
"learning_rate": 8.630641390653743e-07,
"loss": 0.23430384695529938,
"loss_sft": 0.23430384695529938,
"step": 693,
"token_acc": 0.9252841542233118
},
{
"epoch": 2.462222222222222,
"grad_norm": 2.238935708999634,
"learning_rate": 8.521500655169823e-07,
"loss": 0.21517524123191833,
"loss_sft": 0.21517524123191833,
"step": 694,
"token_acc": 0.9300866788321168
},
{
"epoch": 2.465777777777778,
"grad_norm": 2.3367300033569336,
"learning_rate": 8.412990081594102e-07,
"loss": 0.2334917038679123,
"loss_sft": 0.2334917038679123,
"step": 695,
"token_acc": 0.922547862238875
},
{
"epoch": 2.469333333333333,
"grad_norm": 2.195610761642456,
"learning_rate": 8.30511131847297e-07,
"loss": 0.21901236474514008,
"loss_sft": 0.21901236474514008,
"step": 696,
"token_acc": 0.9289079985864059
},
{
"epoch": 2.472888888888889,
"grad_norm": 2.2124106884002686,
"learning_rate": 8.197866004754029e-07,
"loss": 0.22410106658935547,
"loss_sft": 0.22410106658935547,
"step": 697,
"token_acc": 0.927457264957265
},
{
"epoch": 2.4764444444444447,
"grad_norm": 2.3393123149871826,
"learning_rate": 8.091255769761213e-07,
"loss": 0.22128678858280182,
"loss_sft": 0.22128678858280182,
"step": 698,
"token_acc": 0.9276012132401424
},
{
"epoch": 2.48,
"grad_norm": 2.167865037918091,
"learning_rate": 7.985282233170011e-07,
"loss": 0.20742401480674744,
"loss_sft": 0.20742401480674744,
"step": 699,
"token_acc": 0.932900196549465
},
{
"epoch": 2.4835555555555557,
"grad_norm": 2.155754566192627,
"learning_rate": 7.879947004982896e-07,
"loss": 0.2224690020084381,
"loss_sft": 0.2224690020084381,
"step": 700,
"token_acc": 0.9277827201842064
},
{
"epoch": 2.487111111111111,
"grad_norm": 2.258657932281494,
"learning_rate": 7.775251685504826e-07,
"loss": 0.20096063613891602,
"loss_sft": 0.20096063613891602,
"step": 701,
"token_acc": 0.9330745772109786
},
{
"epoch": 2.490666666666667,
"grad_norm": 2.176412582397461,
"learning_rate": 7.671197865318952e-07,
"loss": 0.22116267681121826,
"loss_sft": 0.22116267681121826,
"step": 702,
"token_acc": 0.9291611624834875
},
{
"epoch": 2.494222222222222,
"grad_norm": 2.298837661743164,
"learning_rate": 7.567787125262449e-07,
"loss": 0.22052553296089172,
"loss_sft": 0.22052553296089172,
"step": 703,
"token_acc": 0.9281712567081442
},
{
"epoch": 2.497777777777778,
"grad_norm": 2.5039660930633545,
"learning_rate": 7.465021036402531e-07,
"loss": 0.20655038952827454,
"loss_sft": 0.20655038952827454,
"step": 704,
"token_acc": 0.9312034621014447
},
{
"epoch": 2.501333333333333,
"grad_norm": 2.252969264984131,
"learning_rate": 7.362901160012492e-07,
"loss": 0.22675952315330505,
"loss_sft": 0.22675952315330505,
"step": 705,
"token_acc": 0.9252372332839669
},
{
"epoch": 2.504888888888889,
"grad_norm": 2.312624216079712,
"learning_rate": 7.261429047548085e-07,
"loss": 0.2267449051141739,
"loss_sft": 0.2267449051141739,
"step": 706,
"token_acc": 0.9274555575507273
},
{
"epoch": 2.5084444444444447,
"grad_norm": 2.360414981842041,
"learning_rate": 7.16060624062393e-07,
"loss": 0.21422916650772095,
"loss_sft": 0.21422916650772095,
"step": 707,
"token_acc": 0.9297883324728963
},
{
"epoch": 2.512,
"grad_norm": 2.181637763977051,
"learning_rate": 7.060434270990013e-07,
"loss": 0.2177937626838684,
"loss_sft": 0.2177937626838684,
"step": 708,
"token_acc": 0.9296344225169848
},
{
"epoch": 2.5155555555555553,
"grad_norm": 2.5070977210998535,
"learning_rate": 6.96091466050855e-07,
"loss": 0.1984608769416809,
"loss_sft": 0.1984608769416809,
"step": 709,
"token_acc": 0.9352184734513275
},
{
"epoch": 2.519111111111111,
"grad_norm": 2.2668845653533936,
"learning_rate": 6.86204892113076e-07,
"loss": 0.19761916995048523,
"loss_sft": 0.19761916995048523,
"step": 710,
"token_acc": 0.935682517960999
},
{
"epoch": 2.522666666666667,
"grad_norm": 2.4548606872558594,
"learning_rate": 6.763838554873892e-07,
"loss": 0.22925253212451935,
"loss_sft": 0.22925253212451935,
"step": 711,
"token_acc": 0.9273095858673617
},
{
"epoch": 2.526222222222222,
"grad_norm": 2.2812607288360596,
"learning_rate": 6.66628505379851e-07,
"loss": 0.2384592890739441,
"loss_sft": 0.2384592890739441,
"step": 712,
"token_acc": 0.9221725106648632
},
{
"epoch": 2.529777777777778,
"grad_norm": 2.1901040077209473,
"learning_rate": 6.569389899985723e-07,
"loss": 0.2233884185552597,
"loss_sft": 0.2233884185552597,
"step": 713,
"token_acc": 0.926458419528341
},
{
"epoch": 2.533333333333333,
"grad_norm": 2.2454679012298584,
"learning_rate": 6.473154565514695e-07,
"loss": 0.21911823749542236,
"loss_sft": 0.21911823749542236,
"step": 714,
"token_acc": 0.9268028914565428
},
{
"epoch": 2.536888888888889,
"grad_norm": 2.268723964691162,
"learning_rate": 6.377580512440301e-07,
"loss": 0.22316551208496094,
"loss_sft": 0.22316551208496094,
"step": 715,
"token_acc": 0.9258381120423467
},
{
"epoch": 2.5404444444444443,
"grad_norm": 2.134340524673462,
"learning_rate": 6.282669192770896e-07,
"loss": 0.21278002858161926,
"loss_sft": 0.21278002858161926,
"step": 716,
"token_acc": 0.9323845299190519
},
{
"epoch": 2.544,
"grad_norm": 2.3874874114990234,
"learning_rate": 6.188422048446263e-07,
"loss": 0.2318885326385498,
"loss_sft": 0.2318885326385498,
"step": 717,
"token_acc": 0.9230263157894737
},
{
"epoch": 2.5475555555555554,
"grad_norm": 2.400909423828125,
"learning_rate": 6.094840511315703e-07,
"loss": 0.20906594395637512,
"loss_sft": 0.20906594395637512,
"step": 718,
"token_acc": 0.9310097252137589
},
{
"epoch": 2.551111111111111,
"grad_norm": 2.2926671504974365,
"learning_rate": 6.001926003116282e-07,
"loss": 0.21795794367790222,
"loss_sft": 0.21795794367790222,
"step": 719,
"token_acc": 0.9260166305957398
},
{
"epoch": 2.554666666666667,
"grad_norm": 2.3727362155914307,
"learning_rate": 5.909679935451235e-07,
"loss": 0.22051924467086792,
"loss_sft": 0.22051924467086792,
"step": 720,
"token_acc": 0.9287703446995482
},
{
"epoch": 2.558222222222222,
"grad_norm": 2.2904484272003174,
"learning_rate": 5.818103709768519e-07,
"loss": 0.22298826277256012,
"loss_sft": 0.22298826277256012,
"step": 721,
"token_acc": 0.9279592058153412
},
{
"epoch": 2.561777777777778,
"grad_norm": 2.3771870136260986,
"learning_rate": 5.727198717339511e-07,
"loss": 0.21877795457839966,
"loss_sft": 0.21877795457839966,
"step": 722,
"token_acc": 0.9308510638297872
},
{
"epoch": 2.5653333333333332,
"grad_norm": 2.2592902183532715,
"learning_rate": 5.636966339237882e-07,
"loss": 0.2152928113937378,
"loss_sft": 0.2152928113937378,
"step": 723,
"token_acc": 0.9320640176600441
},
{
"epoch": 2.568888888888889,
"grad_norm": 2.4092557430267334,
"learning_rate": 5.547407946318628e-07,
"loss": 0.2329767346382141,
"loss_sft": 0.2329767346382141,
"step": 724,
"token_acc": 0.9240171593112769
},
{
"epoch": 2.5724444444444443,
"grad_norm": 2.309614658355713,
"learning_rate": 5.458524899197204e-07,
"loss": 0.23667675256729126,
"loss_sft": 0.23667675256729126,
"step": 725,
"token_acc": 0.9242958764675264
},
{
"epoch": 2.576,
"grad_norm": 2.278935670852661,
"learning_rate": 5.370318548228886e-07,
"loss": 0.25069692730903625,
"loss_sft": 0.25069692730903625,
"step": 726,
"token_acc": 0.9199272065514104
},
{
"epoch": 2.5795555555555554,
"grad_norm": 2.422684907913208,
"learning_rate": 5.282790233488283e-07,
"loss": 0.23523519933223724,
"loss_sft": 0.23523519933223724,
"step": 727,
"token_acc": 0.9237819025522042
},
{
"epoch": 2.583111111111111,
"grad_norm": 2.313244581222534,
"learning_rate": 5.195941284748879e-07,
"loss": 0.20533347129821777,
"loss_sft": 0.20533347129821777,
"step": 728,
"token_acc": 0.9342806394316163
},
{
"epoch": 2.586666666666667,
"grad_norm": 2.234987497329712,
"learning_rate": 5.109773021462921e-07,
"loss": 0.2120046466588974,
"loss_sft": 0.2120046466588974,
"step": 729,
"token_acc": 0.931488432020615
},
{
"epoch": 2.590222222222222,
"grad_norm": 2.2962193489074707,
"learning_rate": 5.024286752741364e-07,
"loss": 0.24171319603919983,
"loss_sft": 0.24171319603919983,
"step": 730,
"token_acc": 0.9203016493055556
},
{
"epoch": 2.5937777777777775,
"grad_norm": 2.368356227874756,
"learning_rate": 4.939483777333931e-07,
"loss": 0.209232896566391,
"loss_sft": 0.209232896566391,
"step": 731,
"token_acc": 0.9306137771382094
},
{
"epoch": 2.5973333333333333,
"grad_norm": 2.0591397285461426,
"learning_rate": 4.855365383609457e-07,
"loss": 0.21489405632019043,
"loss_sft": 0.21489405632019043,
"step": 732,
"token_acc": 0.9299463875414858
},
{
"epoch": 2.600888888888889,
"grad_norm": 2.2085916996002197,
"learning_rate": 4.77193284953622e-07,
"loss": 0.22096888720989227,
"loss_sft": 0.22096888720989227,
"step": 733,
"token_acc": 0.9290647955268063
},
{
"epoch": 2.6044444444444443,
"grad_norm": 2.2806551456451416,
"learning_rate": 4.6891874426626125e-07,
"loss": 0.22931520640850067,
"loss_sft": 0.22931520640850067,
"step": 734,
"token_acc": 0.9252925292529253
},
{
"epoch": 2.608,
"grad_norm": 2.2529654502868652,
"learning_rate": 4.607130420097866e-07,
"loss": 0.22546866536140442,
"loss_sft": 0.22546866536140442,
"step": 735,
"token_acc": 0.9270790897980562
},
{
"epoch": 2.6115555555555554,
"grad_norm": 2.296017646789551,
"learning_rate": 4.525763028492869e-07,
"loss": 0.23669913411140442,
"loss_sft": 0.23669913411140442,
"step": 736,
"token_acc": 0.9209030006192648
},
{
"epoch": 2.615111111111111,
"grad_norm": 2.2661802768707275,
"learning_rate": 4.445086504021384e-07,
"loss": 0.22535058856010437,
"loss_sft": 0.22535058856010437,
"step": 737,
"token_acc": 0.9256261814744802
},
{
"epoch": 2.618666666666667,
"grad_norm": 2.268094301223755,
"learning_rate": 4.365102072361116e-07,
"loss": 0.2173900306224823,
"loss_sft": 0.2173900306224823,
"step": 738,
"token_acc": 0.929536467486819
},
{
"epoch": 2.6222222222222222,
"grad_norm": 2.3296658992767334,
"learning_rate": 4.285810948675156e-07,
"loss": 0.21424409747123718,
"loss_sft": 0.21424409747123718,
"step": 739,
"token_acc": 0.9289404817821915
},
{
"epoch": 2.6257777777777775,
"grad_norm": 2.3846795558929443,
"learning_rate": 4.207214337593557e-07,
"loss": 0.20750558376312256,
"loss_sft": 0.20750558376312256,
"step": 740,
"token_acc": 0.9336318145904662
},
{
"epoch": 2.6293333333333333,
"grad_norm": 2.3596928119659424,
"learning_rate": 4.129313433194948e-07,
"loss": 0.24149560928344727,
"loss_sft": 0.24149560928344727,
"step": 741,
"token_acc": 0.9210886588905317
},
{
"epoch": 2.632888888888889,
"grad_norm": 2.3027002811431885,
"learning_rate": 4.05210941898847e-07,
"loss": 0.2225145399570465,
"loss_sft": 0.2225145399570465,
"step": 742,
"token_acc": 0.9274283587644213
},
{
"epoch": 2.6364444444444444,
"grad_norm": 2.292609930038452,
"learning_rate": 3.975603467895739e-07,
"loss": 0.20992445945739746,
"loss_sft": 0.20992445945739746,
"step": 743,
"token_acc": 0.9301441275682306
},
{
"epoch": 2.64,
"grad_norm": 2.343275308609009,
"learning_rate": 3.8997967422330693e-07,
"loss": 0.21266596019268036,
"loss_sft": 0.21266596019268036,
"step": 744,
"token_acc": 0.9301039593010396
},
{
"epoch": 2.6435555555555554,
"grad_norm": 2.3408119678497314,
"learning_rate": 3.8246903936937806e-07,
"loss": 0.21324416995048523,
"loss_sft": 0.21324416995048523,
"step": 745,
"token_acc": 0.9282095045069071
},
{
"epoch": 2.647111111111111,
"grad_norm": 2.4059996604919434,
"learning_rate": 3.7502855633307246e-07,
"loss": 0.20570334792137146,
"loss_sft": 0.20570334792137146,
"step": 746,
"token_acc": 0.932125248127041
},
{
"epoch": 2.6506666666666665,
"grad_norm": 2.2163541316986084,
"learning_rate": 3.676583381538945e-07,
"loss": 0.2277967929840088,
"loss_sft": 0.2277967929840088,
"step": 747,
"token_acc": 0.9255352317605487
},
{
"epoch": 2.6542222222222223,
"grad_norm": 2.3716914653778076,
"learning_rate": 3.603584968038487e-07,
"loss": 0.23870588839054108,
"loss_sft": 0.23870588839054108,
"step": 748,
"token_acc": 0.9237633149294517
},
{
"epoch": 2.6577777777777776,
"grad_norm": 2.5130834579467773,
"learning_rate": 3.53129143185742e-07,
"loss": 0.24331116676330566,
"loss_sft": 0.24331116676330566,
"step": 749,
"token_acc": 0.9216343694948003
},
{
"epoch": 2.6613333333333333,
"grad_norm": 2.2302961349487305,
"learning_rate": 3.4597038713149455e-07,
"loss": 0.23555251955986023,
"loss_sft": 0.23555251955986023,
"step": 750,
"token_acc": 0.9260173273825151
},
{
"epoch": 2.664888888888889,
"grad_norm": 2.2986302375793457,
"learning_rate": 3.388823374004735e-07,
"loss": 0.2061428725719452,
"loss_sft": 0.2061428725719452,
"step": 751,
"token_acc": 0.930930582918067
},
{
"epoch": 2.6684444444444444,
"grad_norm": 2.349370002746582,
"learning_rate": 3.3186510167784456e-07,
"loss": 0.2080996334552765,
"loss_sft": 0.2080996334552765,
"step": 752,
"token_acc": 0.9297761615148286
},
{
"epoch": 2.672,
"grad_norm": 2.522251844406128,
"learning_rate": 3.2491878657292643e-07,
"loss": 0.23804304003715515,
"loss_sft": 0.23804304003715515,
"step": 753,
"token_acc": 0.9252241129421869
},
{
"epoch": 2.6755555555555555,
"grad_norm": 2.31113338470459,
"learning_rate": 3.180434976175784e-07,
"loss": 0.23362340033054352,
"loss_sft": 0.23362340033054352,
"step": 754,
"token_acc": 0.9241837299391257
},
{
"epoch": 2.679111111111111,
"grad_norm": 2.3328609466552734,
"learning_rate": 3.112393392645985e-07,
"loss": 0.21051113307476044,
"loss_sft": 0.21051113307476044,
"step": 755,
"token_acc": 0.9341561246528849
},
{
"epoch": 2.6826666666666665,
"grad_norm": 2.349313259124756,
"learning_rate": 3.045064148861282e-07,
"loss": 0.22120699286460876,
"loss_sft": 0.22120699286460876,
"step": 756,
"token_acc": 0.9275686420673376
},
{
"epoch": 2.6862222222222223,
"grad_norm": 2.149635076522827,
"learning_rate": 2.9784482677209013e-07,
"loss": 0.2023593634366989,
"loss_sft": 0.2023593634366989,
"step": 757,
"token_acc": 0.9338634494104744
},
{
"epoch": 2.6897777777777776,
"grad_norm": 2.431436538696289,
"learning_rate": 2.912546761286333e-07,
"loss": 0.2163369357585907,
"loss_sft": 0.2163369357585907,
"step": 758,
"token_acc": 0.9290658462994734
},
{
"epoch": 2.6933333333333334,
"grad_norm": 2.3487656116485596,
"learning_rate": 2.847360630765866e-07,
"loss": 0.20875662565231323,
"loss_sft": 0.20875662565231323,
"step": 759,
"token_acc": 0.9318041300853221
},
{
"epoch": 2.696888888888889,
"grad_norm": 2.1448185443878174,
"learning_rate": 2.7828908664995216e-07,
"loss": 0.22066381573677063,
"loss_sft": 0.22066381573677063,
"step": 760,
"token_acc": 0.9296571226581831
},
{
"epoch": 2.7004444444444444,
"grad_norm": 2.216562509536743,
"learning_rate": 2.719138447943881e-07,
"loss": 0.2316487431526184,
"loss_sft": 0.2316487431526184,
"step": 761,
"token_acc": 0.92334820039415
},
{
"epoch": 2.7039999999999997,
"grad_norm": 2.428004264831543,
"learning_rate": 2.656104343657251e-07,
"loss": 0.22632446885108948,
"loss_sft": 0.22632446885108948,
"step": 762,
"token_acc": 0.924120759414877
},
{
"epoch": 2.7075555555555555,
"grad_norm": 2.2970967292785645,
"learning_rate": 2.5937895112849886e-07,
"loss": 0.20677095651626587,
"loss_sft": 0.20677095651626587,
"step": 763,
"token_acc": 0.9348162245747591
},
{
"epoch": 2.7111111111111112,
"grad_norm": 2.3385844230651855,
"learning_rate": 2.532194897544882e-07,
"loss": 0.22836467623710632,
"loss_sft": 0.22836467623710632,
"step": 764,
"token_acc": 0.924868390936141
},
{
"epoch": 2.7146666666666666,
"grad_norm": 2.3681743144989014,
"learning_rate": 2.4713214382128204e-07,
"loss": 0.22776830196380615,
"loss_sft": 0.22776830196380615,
"step": 765,
"token_acc": 0.9279363593823117
},
{
"epoch": 2.7182222222222223,
"grad_norm": 2.2631309032440186,
"learning_rate": 2.411170058108558e-07,
"loss": 0.22913134098052979,
"loss_sft": 0.22913134098052979,
"step": 766,
"token_acc": 0.9268838325967153
},
{
"epoch": 2.7217777777777776,
"grad_norm": 2.5855839252471924,
"learning_rate": 2.351741671081631e-07,
"loss": 0.2211557924747467,
"loss_sft": 0.2211557924747467,
"step": 767,
"token_acc": 0.9285923153400784
},
{
"epoch": 2.7253333333333334,
"grad_norm": 2.4138174057006836,
"learning_rate": 2.2930371799975593e-07,
"loss": 0.23430970311164856,
"loss_sft": 0.23430970311164856,
"step": 768,
"token_acc": 0.9237984944991314
},
{
"epoch": 2.728888888888889,
"grad_norm": 2.3260719776153564,
"learning_rate": 2.2350574767240395e-07,
"loss": 0.21351540088653564,
"loss_sft": 0.21351540088653564,
"step": 769,
"token_acc": 0.9312820512820513
},
{
"epoch": 2.7324444444444445,
"grad_norm": 2.256317615509033,
"learning_rate": 2.1778034421174433e-07,
"loss": 0.21965140104293823,
"loss_sft": 0.21965140104293823,
"step": 770,
"token_acc": 0.9284398548702205
},
{
"epoch": 2.7359999999999998,
"grad_norm": 2.286078453063965,
"learning_rate": 2.1212759460094268e-07,
"loss": 0.22792696952819824,
"loss_sft": 0.22792696952819824,
"step": 771,
"token_acc": 0.9244815029816877
},
{
"epoch": 2.7395555555555555,
"grad_norm": 2.300149440765381,
"learning_rate": 2.0654758471937098e-07,
"loss": 0.21655958890914917,
"loss_sft": 0.21655958890914917,
"step": 772,
"token_acc": 0.9309116393155406
},
{
"epoch": 2.7431111111111113,
"grad_norm": 2.226922035217285,
"learning_rate": 2.0104039934130348e-07,
"loss": 0.20887339115142822,
"loss_sft": 0.20887339115142822,
"step": 773,
"token_acc": 0.9290796288917981
},
{
"epoch": 2.7466666666666666,
"grad_norm": 2.213681936264038,
"learning_rate": 1.9560612213462837e-07,
"loss": 0.21002738177776337,
"loss_sft": 0.21002738177776337,
"step": 774,
"token_acc": 0.9307831506987031
},
{
"epoch": 2.7502222222222223,
"grad_norm": 2.404092788696289,
"learning_rate": 1.9024483565957707e-07,
"loss": 0.23099403083324432,
"loss_sft": 0.23099403083324432,
"step": 775,
"token_acc": 0.9232992988067413
},
{
"epoch": 2.7537777777777777,
"grad_norm": 2.4170093536376953,
"learning_rate": 1.8495662136746916e-07,
"loss": 0.22152146697044373,
"loss_sft": 0.22152146697044373,
"step": 776,
"token_acc": 0.9289992513102071
},
{
"epoch": 2.7573333333333334,
"grad_norm": 2.267754554748535,
"learning_rate": 1.7974155959947614e-07,
"loss": 0.2264728844165802,
"loss_sft": 0.2264728844165802,
"step": 777,
"token_acc": 0.92746567474427
},
{
"epoch": 2.7608888888888887,
"grad_norm": 2.233266592025757,
"learning_rate": 1.745997295853985e-07,
"loss": 0.2043970674276352,
"loss_sft": 0.2043970674276352,
"step": 778,
"token_acc": 0.9293123543123543
},
{
"epoch": 2.7644444444444445,
"grad_norm": 2.415250778198242,
"learning_rate": 1.6953120944246503e-07,
"loss": 0.23126524686813354,
"loss_sft": 0.23126524686813354,
"step": 779,
"token_acc": 0.9249924081384756
},
{
"epoch": 2.768,
"grad_norm": 2.40570068359375,
"learning_rate": 1.6453607617414603e-07,
"loss": 0.20348799228668213,
"loss_sft": 0.20348799228668213,
"step": 780,
"token_acc": 0.9335650390291151
},
{
"epoch": 2.7715555555555556,
"grad_norm": 2.249272584915161,
"learning_rate": 1.5961440566897913e-07,
"loss": 0.2084093987941742,
"loss_sft": 0.2084093987941742,
"step": 781,
"token_acc": 0.9301868761945211
},
{
"epoch": 2.7751111111111113,
"grad_norm": 2.3114168643951416,
"learning_rate": 1.5476627269941925e-07,
"loss": 0.21748429536819458,
"loss_sft": 0.21748429536819458,
"step": 782,
"token_acc": 0.9285472018089316
},
{
"epoch": 2.7786666666666666,
"grad_norm": 2.3742172718048096,
"learning_rate": 1.4999175092070716e-07,
"loss": 0.22353892028331757,
"loss_sft": 0.22353892028331757,
"step": 783,
"token_acc": 0.9282139878470154
},
{
"epoch": 2.7822222222222224,
"grad_norm": 2.2798521518707275,
"learning_rate": 1.4529091286973994e-07,
"loss": 0.22898119688034058,
"loss_sft": 0.22898119688034058,
"step": 784,
"token_acc": 0.9265151515151515
},
{
"epoch": 2.7857777777777777,
"grad_norm": 2.3060412406921387,
"learning_rate": 1.4066382996397898e-07,
"loss": 0.2372695505619049,
"loss_sft": 0.2372695505619049,
"step": 785,
"token_acc": 0.9228634850166482
},
{
"epoch": 2.7893333333333334,
"grad_norm": 2.255695343017578,
"learning_rate": 1.361105725003603e-07,
"loss": 0.2329796999692917,
"loss_sft": 0.2329796999692917,
"step": 786,
"token_acc": 0.9260407905803196
},
{
"epoch": 2.7928888888888888,
"grad_norm": 2.193325996398926,
"learning_rate": 1.3163120965422494e-07,
"loss": 0.22065503895282745,
"loss_sft": 0.22065503895282745,
"step": 787,
"token_acc": 0.9283380643779122
},
{
"epoch": 2.7964444444444445,
"grad_norm": 2.213216543197632,
"learning_rate": 1.272258094782741e-07,
"loss": 0.21991030871868134,
"loss_sft": 0.21991030871868134,
"step": 788,
"token_acc": 0.929769780365176
},
{
"epoch": 2.8,
"grad_norm": 2.1995456218719482,
"learning_rate": 1.2289443890152852e-07,
"loss": 0.22797149419784546,
"loss_sft": 0.22797149419784546,
"step": 789,
"token_acc": 0.9256340142619699
},
{
"epoch": 2.8035555555555556,
"grad_norm": 2.3130643367767334,
"learning_rate": 1.1863716372831513e-07,
"loss": 0.21125397086143494,
"loss_sft": 0.21125397086143494,
"step": 790,
"token_acc": 0.9319351266853384
},
{
"epoch": 2.8071111111111113,
"grad_norm": 2.210423707962036,
"learning_rate": 1.1445404863726806e-07,
"loss": 0.2174803614616394,
"loss_sft": 0.2174803614616394,
"step": 791,
"token_acc": 0.9273788674839463
},
{
"epoch": 2.8106666666666666,
"grad_norm": 2.4450199604034424,
"learning_rate": 1.1034515718034321e-07,
"loss": 0.2683491110801697,
"loss_sft": 0.2683491110801697,
"step": 792,
"token_acc": 0.9150261372483595
},
{
"epoch": 2.814222222222222,
"grad_norm": 2.3486268520355225,
"learning_rate": 1.0631055178185523e-07,
"loss": 0.19007007777690887,
"loss_sft": 0.19007007777690887,
"step": 793,
"token_acc": 0.9364460105713567
},
{
"epoch": 2.8177777777777777,
"grad_norm": 2.1782171726226807,
"learning_rate": 1.0235029373752758e-07,
"loss": 0.22465762495994568,
"loss_sft": 0.22465762495994568,
"step": 794,
"token_acc": 0.9269854383971909
},
{
"epoch": 2.8213333333333335,
"grad_norm": 2.3150432109832764,
"learning_rate": 9.84644432135623e-08,
"loss": 0.22664430737495422,
"loss_sft": 0.22664430737495422,
"step": 795,
"token_acc": 0.9271921301918873
},
{
"epoch": 2.824888888888889,
"grad_norm": 2.2939445972442627,
"learning_rate": 9.465305924572566e-08,
"loss": 0.19834165275096893,
"loss_sft": 0.19834165275096893,
"step": 796,
"token_acc": 0.934943351222421
},
{
"epoch": 2.8284444444444445,
"grad_norm": 2.296612024307251,
"learning_rate": 9.091619973845056e-08,
"loss": 0.23483601212501526,
"loss_sft": 0.23483601212501526,
"step": 797,
"token_acc": 0.9233495066237802
},
{
"epoch": 2.832,
"grad_norm": 2.2161712646484375,
"learning_rate": 8.725392146395728e-08,
"loss": 0.2190060019493103,
"loss_sft": 0.2190060019493103,
"step": 798,
"token_acc": 0.9307522637566752
},
{
"epoch": 2.8355555555555556,
"grad_norm": 2.42097544670105,
"learning_rate": 8.366628006139243e-08,
"loss": 0.24028503894805908,
"loss_sft": 0.24028503894805908,
"step": 799,
"token_acc": 0.92346355951565
},
{
"epoch": 2.8391111111111114,
"grad_norm": 2.343261480331421,
"learning_rate": 8.015333003598025e-08,
"loss": 0.2228238731622696,
"loss_sft": 0.2228238731622696,
"step": 800,
"token_acc": 0.9280197900325812
},
{
"epoch": 2.8426666666666667,
"grad_norm": 2.398681879043579,
"learning_rate": 7.671512475819876e-08,
"loss": 0.2346479892730713,
"loss_sft": 0.2346479892730713,
"step": 801,
"token_acc": 0.9225679542203148
},
{
"epoch": 2.846222222222222,
"grad_norm": 2.2917404174804688,
"learning_rate": 7.335171646296436e-08,
"loss": 0.238242045044899,
"loss_sft": 0.238242045044899,
"step": 802,
"token_acc": 0.9228965112870126
},
{
"epoch": 2.8497777777777777,
"grad_norm": 2.2950029373168945,
"learning_rate": 7.006315624884296e-08,
"loss": 0.23262113332748413,
"loss_sft": 0.23262113332748413,
"step": 803,
"token_acc": 0.9256820844724656
},
{
"epoch": 2.8533333333333335,
"grad_norm": 2.3436965942382812,
"learning_rate": 6.684949407727015e-08,
"loss": 0.20629116892814636,
"loss_sft": 0.20629116892814636,
"step": 804,
"token_acc": 0.9343418088187962
},
{
"epoch": 2.856888888888889,
"grad_norm": 2.2757763862609863,
"learning_rate": 6.371077877179277e-08,
"loss": 0.20900432765483856,
"loss_sft": 0.20900432765483856,
"step": 805,
"token_acc": 0.9325597426470589
},
{
"epoch": 2.8604444444444446,
"grad_norm": 2.3043203353881836,
"learning_rate": 6.064705801732962e-08,
"loss": 0.20802822709083557,
"loss_sft": 0.20802822709083557,
"step": 806,
"token_acc": 0.932649060085342
},
{
"epoch": 2.864,
"grad_norm": 2.304035186767578,
"learning_rate": 5.7658378359443104e-08,
"loss": 0.2138587236404419,
"loss_sft": 0.2138587236404419,
"step": 807,
"token_acc": 0.9323274657725622
},
{
"epoch": 2.8675555555555556,
"grad_norm": 2.3305411338806152,
"learning_rate": 5.4744785203635355e-08,
"loss": 0.2058597058057785,
"loss_sft": 0.2058597058057785,
"step": 808,
"token_acc": 0.9340346534653465
},
{
"epoch": 2.871111111111111,
"grad_norm": 2.3507184982299805,
"learning_rate": 5.190632281465713e-08,
"loss": 0.22053909301757812,
"loss_sft": 0.22053909301757812,
"step": 809,
"token_acc": 0.928410403035135
},
{
"epoch": 2.8746666666666667,
"grad_norm": 2.180372476577759,
"learning_rate": 4.914303431583389e-08,
"loss": 0.22321942448616028,
"loss_sft": 0.22321942448616028,
"step": 810,
"token_acc": 0.9263622974963182
},
{
"epoch": 2.878222222222222,
"grad_norm": 2.254753828048706,
"learning_rate": 4.645496168841468e-08,
"loss": 0.20938748121261597,
"loss_sft": 0.20938748121261597,
"step": 811,
"token_acc": 0.9334228815044092
},
{
"epoch": 2.8817777777777778,
"grad_norm": 2.412715435028076,
"learning_rate": 4.384214577092927e-08,
"loss": 0.21015211939811707,
"loss_sft": 0.21015211939811707,
"step": 812,
"token_acc": 0.9329512893982808
},
{
"epoch": 2.8853333333333335,
"grad_norm": 2.338899850845337,
"learning_rate": 4.130462625857257e-08,
"loss": 0.21445578336715698,
"loss_sft": 0.21445578336715698,
"step": 813,
"token_acc": 0.929355578986541
},
{
"epoch": 2.888888888888889,
"grad_norm": 2.4105489253997803,
"learning_rate": 3.884244170259732e-08,
"loss": 0.20582976937294006,
"loss_sft": 0.20582976937294006,
"step": 814,
"token_acc": 0.9323293691830403
},
{
"epoch": 2.8924444444444446,
"grad_norm": 2.3862078189849854,
"learning_rate": 3.645562950973014e-08,
"loss": 0.23894110321998596,
"loss_sft": 0.23894110321998596,
"step": 815,
"token_acc": 0.9224041791395199
},
{
"epoch": 2.896,
"grad_norm": 2.391489267349243,
"learning_rate": 3.414422594160527e-08,
"loss": 0.19547885656356812,
"loss_sft": 0.19547885656356812,
"step": 816,
"token_acc": 0.936228287841191
},
{
"epoch": 2.8995555555555557,
"grad_norm": 2.266090154647827,
"learning_rate": 3.1908266114210054e-08,
"loss": 0.20504418015480042,
"loss_sft": 0.20504418015480042,
"step": 817,
"token_acc": 0.9329081172152356
},
{
"epoch": 2.903111111111111,
"grad_norm": 2.308250665664673,
"learning_rate": 2.974778399735423e-08,
"loss": 0.23431923985481262,
"loss_sft": 0.23431923985481262,
"step": 818,
"token_acc": 0.9227270174650418
},
{
"epoch": 2.9066666666666667,
"grad_norm": 2.2525670528411865,
"learning_rate": 2.766281241415203e-08,
"loss": 0.21195153892040253,
"loss_sft": 0.21195153892040253,
"step": 819,
"token_acc": 0.9317767653758542
},
{
"epoch": 2.910222222222222,
"grad_norm": 2.2136147022247314,
"learning_rate": 2.5653383040524228e-08,
"loss": 0.22890228033065796,
"loss_sft": 0.22890228033065796,
"step": 820,
"token_acc": 0.9258076944245693
},
{
"epoch": 2.913777777777778,
"grad_norm": 2.5726301670074463,
"learning_rate": 2.3719526404717975e-08,
"loss": 0.22015199065208435,
"loss_sft": 0.22015199065208435,
"step": 821,
"token_acc": 0.9294488357188635
},
{
"epoch": 2.9173333333333336,
"grad_norm": 2.196023941040039,
"learning_rate": 2.1861271886840508e-08,
"loss": 0.21100088953971863,
"loss_sft": 0.21100088953971863,
"step": 822,
"token_acc": 0.929148665819568
},
{
"epoch": 2.920888888888889,
"grad_norm": 2.1497766971588135,
"learning_rate": 2.0078647718415058e-08,
"loss": 0.21922646462917328,
"loss_sft": 0.21922646462917328,
"step": 823,
"token_acc": 0.9285208003925631
},
{
"epoch": 2.924444444444444,
"grad_norm": 2.3242197036743164,
"learning_rate": 1.837168098195119e-08,
"loss": 0.2184222787618637,
"loss_sft": 0.2184222787618637,
"step": 824,
"token_acc": 0.928899219262173
},
{
"epoch": 2.928,
"grad_norm": 2.0465757846832275,
"learning_rate": 1.6740397610533477e-08,
"loss": 0.19837015867233276,
"loss_sft": 0.19837015867233276,
"step": 825,
"token_acc": 0.9343531241897848
},
{
"epoch": 2.9315555555555557,
"grad_norm": 2.2778143882751465,
"learning_rate": 1.5184822387426246e-08,
"loss": 0.22872623801231384,
"loss_sft": 0.22872623801231384,
"step": 826,
"token_acc": 0.9277395590704729
},
{
"epoch": 2.935111111111111,
"grad_norm": 2.2200214862823486,
"learning_rate": 1.3704978945698888e-08,
"loss": 0.22263160347938538,
"loss_sft": 0.22263160347938538,
"step": 827,
"token_acc": 0.9256627783669141
},
{
"epoch": 2.9386666666666668,
"grad_norm": 2.3505682945251465,
"learning_rate": 1.2300889767866697e-08,
"loss": 0.225810706615448,
"loss_sft": 0.225810706615448,
"step": 828,
"token_acc": 0.9258873255746627
},
{
"epoch": 2.942222222222222,
"grad_norm": 2.3278541564941406,
"learning_rate": 1.0972576185547256e-08,
"loss": 0.22645622491836548,
"loss_sft": 0.22645622491836548,
"step": 829,
"token_acc": 0.9283620588580138
},
{
"epoch": 2.945777777777778,
"grad_norm": 2.3490898609161377,
"learning_rate": 9.720058379138474e-09,
"loss": 0.2338504046201706,
"loss_sft": 0.2338504046201706,
"step": 830,
"token_acc": 0.9253002355915647
},
{
"epoch": 2.9493333333333336,
"grad_norm": 2.386215925216675,
"learning_rate": 8.543355377512164e-09,
"loss": 0.23085765540599823,
"loss_sft": 0.23085765540599823,
"step": 831,
"token_acc": 0.9244747302668939
},
{
"epoch": 2.952888888888889,
"grad_norm": 2.120513677597046,
"learning_rate": 7.442485057722048e-09,
"loss": 0.20780342817306519,
"loss_sft": 0.20780342817306519,
"step": 832,
"token_acc": 0.933859255002351
},
{
"epoch": 2.956444444444444,
"grad_norm": 2.2752957344055176,
"learning_rate": 6.417464144736208e-09,
"loss": 0.23323115706443787,
"loss_sft": 0.23323115706443787,
"step": 833,
"token_acc": 0.9243964782732178
},
{
"epoch": 2.96,
"grad_norm": 2.382864475250244,
"learning_rate": 5.468308211179496e-09,
"loss": 0.23151040077209473,
"loss_sft": 0.23151040077209473,
"step": 834,
"token_acc": 0.923560077884555
},
{
"epoch": 2.9635555555555557,
"grad_norm": 2.242945432662964,
"learning_rate": 4.595031677099293e-09,
"loss": 0.2328961342573166,
"loss_sft": 0.2328961342573166,
"step": 835,
"token_acc": 0.9240381991814461
},
{
"epoch": 2.967111111111111,
"grad_norm": 2.267942428588867,
"learning_rate": 3.7976478097451196e-09,
"loss": 0.2135615199804306,
"loss_sft": 0.2135615199804306,
"step": 836,
"token_acc": 0.9303689486031137
},
{
"epoch": 2.970666666666667,
"grad_norm": 2.3810133934020996,
"learning_rate": 3.0761687233682445e-09,
"loss": 0.21753928065299988,
"loss_sft": 0.21753928065299988,
"step": 837,
"token_acc": 0.9273073153859457
},
{
"epoch": 2.974222222222222,
"grad_norm": 2.2142817974090576,
"learning_rate": 2.430605379035722e-09,
"loss": 0.21645505726337433,
"loss_sft": 0.21645505726337433,
"step": 838,
"token_acc": 0.9296912881849039
},
{
"epoch": 2.977777777777778,
"grad_norm": 2.308995246887207,
"learning_rate": 1.8609675844655228e-09,
"loss": 0.21676571667194366,
"loss_sft": 0.21676571667194366,
"step": 839,
"token_acc": 0.9290442149854382
},
{
"epoch": 2.981333333333333,
"grad_norm": 2.2542054653167725,
"learning_rate": 1.3672639938777655e-09,
"loss": 0.20977747440338135,
"loss_sft": 0.20977747440338135,
"step": 840,
"token_acc": 0.9288383152173914
},
{
"epoch": 2.984888888888889,
"grad_norm": 2.3006591796875,
"learning_rate": 9.495021078614885e-10,
"loss": 0.21558210253715515,
"loss_sft": 0.21558210253715515,
"step": 841,
"token_acc": 0.9286012028821533
},
{
"epoch": 2.9884444444444442,
"grad_norm": 2.261469841003418,
"learning_rate": 6.076882732625189e-10,
"loss": 0.2320500910282135,
"loss_sft": 0.2320500910282135,
"step": 842,
"token_acc": 0.9270064956512165
},
{
"epoch": 2.992,
"grad_norm": 2.270570755004883,
"learning_rate": 3.4182768308577143e-10,
"loss": 0.21072573959827423,
"loss_sft": 0.21072573959827423,
"step": 843,
"token_acc": 0.932830857902904
},
{
"epoch": 2.9955555555555557,
"grad_norm": 2.236737012863159,
"learning_rate": 1.5192437641753376e-10,
"loss": 0.21560460329055786,
"loss_sft": 0.21560460329055786,
"step": 844,
"token_acc": 0.9281679633379516
},
{
"epoch": 2.999111111111111,
"grad_norm": 2.4227771759033203,
"learning_rate": 3.798123836162848e-11,
"loss": 0.23297986388206482,
"loss_sft": 0.23297986388206482,
"step": 845,
"token_acc": 0.9250929899856939
},
{
"epoch": 3.0,
"grad_norm": 4.790007591247559,
"learning_rate": 0.0,
"loss": 0.16432157158851624,
"loss_sft": 0.16432157158851624,
"step": 846,
"token_acc": 0.945184812672869
}
],
"logging_steps": 1,
"max_steps": 846,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.6070116485544018e+18,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}