memrag-basecur / trainer_state.json
hyunseoki's picture
checkpoint-846 (inference)
eb681dd 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": 25.660261154174805,
"learning_rate": 2.5000000000000004e-07,
"loss": 1.0588122606277466,
"loss_sft": 1.0588122606277466,
"step": 1,
"token_acc": 0.7147074144816808
},
{
"epoch": 0.0071111111111111115,
"grad_norm": 25.612918853759766,
"learning_rate": 5.000000000000001e-07,
"loss": 1.0494858026504517,
"loss_sft": 1.0494858026504517,
"step": 2,
"token_acc": 0.7185479860765788
},
{
"epoch": 0.010666666666666666,
"grad_norm": 30.56467056274414,
"learning_rate": 7.5e-07,
"loss": 1.0724303722381592,
"loss_sft": 1.0724303722381592,
"step": 3,
"token_acc": 0.7139420667535331
},
{
"epoch": 0.014222222222222223,
"grad_norm": 26.860340118408203,
"learning_rate": 1.0000000000000002e-06,
"loss": 0.9765650033950806,
"loss_sft": 0.9765650033950806,
"step": 4,
"token_acc": 0.7225542623466499
},
{
"epoch": 0.017777777777777778,
"grad_norm": 23.216257095336914,
"learning_rate": 1.25e-06,
"loss": 0.9936501979827881,
"loss_sft": 0.9936501979827881,
"step": 5,
"token_acc": 0.7188734329383479
},
{
"epoch": 0.021333333333333333,
"grad_norm": 15.444840431213379,
"learning_rate": 1.5e-06,
"loss": 0.9497862458229065,
"loss_sft": 0.9497862458229065,
"step": 6,
"token_acc": 0.7257645054255224
},
{
"epoch": 0.024888888888888887,
"grad_norm": 13.837246894836426,
"learning_rate": 1.75e-06,
"loss": 0.9248610734939575,
"loss_sft": 0.9248610734939575,
"step": 7,
"token_acc": 0.7345626477541372
},
{
"epoch": 0.028444444444444446,
"grad_norm": 8.758675575256348,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.8149494528770447,
"loss_sft": 0.8149494528770447,
"step": 8,
"token_acc": 0.7501059591421548
},
{
"epoch": 0.032,
"grad_norm": 8.611343383789062,
"learning_rate": 2.25e-06,
"loss": 0.8490846157073975,
"loss_sft": 0.8490846157073975,
"step": 9,
"token_acc": 0.7477702461648233
},
{
"epoch": 0.035555555555555556,
"grad_norm": 6.641948223114014,
"learning_rate": 2.5e-06,
"loss": 0.7789880037307739,
"loss_sft": 0.7789880037307739,
"step": 10,
"token_acc": 0.7617247895550593
},
{
"epoch": 0.03911111111111111,
"grad_norm": 4.982907772064209,
"learning_rate": 2.7500000000000004e-06,
"loss": 0.7746244668960571,
"loss_sft": 0.7746244668960571,
"step": 11,
"token_acc": 0.765425930552086
},
{
"epoch": 0.042666666666666665,
"grad_norm": 4.527531147003174,
"learning_rate": 3e-06,
"loss": 0.7352659106254578,
"loss_sft": 0.7352659106254578,
"step": 12,
"token_acc": 0.7740326741186586
},
{
"epoch": 0.04622222222222222,
"grad_norm": 4.57786750793457,
"learning_rate": 3.2500000000000002e-06,
"loss": 0.7682478427886963,
"loss_sft": 0.7682478427886963,
"step": 13,
"token_acc": 0.7642631481137335
},
{
"epoch": 0.049777777777777775,
"grad_norm": 4.256239891052246,
"learning_rate": 3.5e-06,
"loss": 0.7320900559425354,
"loss_sft": 0.7320900559425354,
"step": 14,
"token_acc": 0.7747762957960493
},
{
"epoch": 0.05333333333333334,
"grad_norm": 4.044354438781738,
"learning_rate": 3.7500000000000005e-06,
"loss": 0.7242921590805054,
"loss_sft": 0.7242921590805054,
"step": 15,
"token_acc": 0.7773386034255599
},
{
"epoch": 0.05688888888888889,
"grad_norm": 3.8805108070373535,
"learning_rate": 4.000000000000001e-06,
"loss": 0.7318351864814758,
"loss_sft": 0.7318351864814758,
"step": 16,
"token_acc": 0.7774927596193628
},
{
"epoch": 0.060444444444444446,
"grad_norm": 4.007376194000244,
"learning_rate": 4.25e-06,
"loss": 0.6899473667144775,
"loss_sft": 0.6899473667144775,
"step": 17,
"token_acc": 0.7825726141078838
},
{
"epoch": 0.064,
"grad_norm": 3.2923731803894043,
"learning_rate": 4.5e-06,
"loss": 0.71946120262146,
"loss_sft": 0.71946120262146,
"step": 18,
"token_acc": 0.777930507827415
},
{
"epoch": 0.06755555555555555,
"grad_norm": 3.4096949100494385,
"learning_rate": 4.75e-06,
"loss": 0.6855860948562622,
"loss_sft": 0.6855860948562622,
"step": 19,
"token_acc": 0.7849599417334304
},
{
"epoch": 0.07111111111111111,
"grad_norm": 4.166821002960205,
"learning_rate": 5e-06,
"loss": 0.7357954978942871,
"loss_sft": 0.7357954978942871,
"step": 20,
"token_acc": 0.7733848503908707
},
{
"epoch": 0.07466666666666667,
"grad_norm": 3.235140562057495,
"learning_rate": 5.2500000000000006e-06,
"loss": 0.6870613098144531,
"loss_sft": 0.6870613098144531,
"step": 21,
"token_acc": 0.7862441541686824
},
{
"epoch": 0.07822222222222222,
"grad_norm": 3.130876064300537,
"learning_rate": 5.500000000000001e-06,
"loss": 0.6711353659629822,
"loss_sft": 0.6711353659629822,
"step": 22,
"token_acc": 0.7892732219199378
},
{
"epoch": 0.08177777777777778,
"grad_norm": 3.5797438621520996,
"learning_rate": 5.75e-06,
"loss": 0.7389527559280396,
"loss_sft": 0.7389527559280396,
"step": 23,
"token_acc": 0.7724371373307544
},
{
"epoch": 0.08533333333333333,
"grad_norm": 3.3319530487060547,
"learning_rate": 6e-06,
"loss": 0.6563287377357483,
"loss_sft": 0.6563287377357483,
"step": 24,
"token_acc": 0.796186160672142
},
{
"epoch": 0.08888888888888889,
"grad_norm": 3.133545398712158,
"learning_rate": 6.25e-06,
"loss": 0.7274408340454102,
"loss_sft": 0.7274408340454102,
"step": 25,
"token_acc": 0.7783543365455893
},
{
"epoch": 0.09244444444444444,
"grad_norm": 3.0895705223083496,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.7145243883132935,
"loss_sft": 0.7145243883132935,
"step": 26,
"token_acc": 0.7782993799822852
},
{
"epoch": 0.096,
"grad_norm": 3.060349941253662,
"learning_rate": 6.750000000000001e-06,
"loss": 0.7092140913009644,
"loss_sft": 0.7092140913009644,
"step": 27,
"token_acc": 0.7791496320523303
},
{
"epoch": 0.09955555555555555,
"grad_norm": 3.0153110027313232,
"learning_rate": 7e-06,
"loss": 0.7183821201324463,
"loss_sft": 0.7183821201324463,
"step": 28,
"token_acc": 0.7798648871313231
},
{
"epoch": 0.10311111111111111,
"grad_norm": 2.9804539680480957,
"learning_rate": 7.25e-06,
"loss": 0.7196909785270691,
"loss_sft": 0.7196909785270691,
"step": 29,
"token_acc": 0.7789126335637964
},
{
"epoch": 0.10666666666666667,
"grad_norm": 2.9636332988739014,
"learning_rate": 7.500000000000001e-06,
"loss": 0.6773001551628113,
"loss_sft": 0.6773001551628113,
"step": 30,
"token_acc": 0.7822512290218682
},
{
"epoch": 0.11022222222222222,
"grad_norm": 3.155043840408325,
"learning_rate": 7.75e-06,
"loss": 0.7298338413238525,
"loss_sft": 0.7298338413238525,
"step": 31,
"token_acc": 0.7752119790727043
},
{
"epoch": 0.11377777777777778,
"grad_norm": 3.2628629207611084,
"learning_rate": 8.000000000000001e-06,
"loss": 0.7017604112625122,
"loss_sft": 0.7017604112625122,
"step": 32,
"token_acc": 0.7802665679378008
},
{
"epoch": 0.11733333333333333,
"grad_norm": 3.139223098754883,
"learning_rate": 8.25e-06,
"loss": 0.7168850302696228,
"loss_sft": 0.7168850302696228,
"step": 33,
"token_acc": 0.7809480812641083
},
{
"epoch": 0.12088888888888889,
"grad_norm": 3.2499890327453613,
"learning_rate": 8.5e-06,
"loss": 0.6890050172805786,
"loss_sft": 0.6890050172805786,
"step": 34,
"token_acc": 0.7852517637962695
},
{
"epoch": 0.12444444444444444,
"grad_norm": 3.1099562644958496,
"learning_rate": 8.750000000000001e-06,
"loss": 0.6742894053459167,
"loss_sft": 0.6742894053459167,
"step": 35,
"token_acc": 0.7915936952714536
},
{
"epoch": 0.128,
"grad_norm": 2.955266237258911,
"learning_rate": 9e-06,
"loss": 0.7146100997924805,
"loss_sft": 0.7146100997924805,
"step": 36,
"token_acc": 0.7792376317923764
},
{
"epoch": 0.13155555555555556,
"grad_norm": 3.1143367290496826,
"learning_rate": 9.250000000000001e-06,
"loss": 0.6942050457000732,
"loss_sft": 0.6942050457000732,
"step": 37,
"token_acc": 0.7845373891001267
},
{
"epoch": 0.1351111111111111,
"grad_norm": 3.169827938079834,
"learning_rate": 9.5e-06,
"loss": 0.7273025512695312,
"loss_sft": 0.7273025512695312,
"step": 38,
"token_acc": 0.7802168738428987
},
{
"epoch": 0.13866666666666666,
"grad_norm": 3.1556618213653564,
"learning_rate": 9.75e-06,
"loss": 0.7065205574035645,
"loss_sft": 0.7065205574035645,
"step": 39,
"token_acc": 0.7777777777777778
},
{
"epoch": 0.14222222222222222,
"grad_norm": 3.0242624282836914,
"learning_rate": 1e-05,
"loss": 0.7049899101257324,
"loss_sft": 0.7049899101257324,
"step": 40,
"token_acc": 0.779655201989879
},
{
"epoch": 0.14577777777777778,
"grad_norm": 2.929504156112671,
"learning_rate": 9.99996201876164e-06,
"loss": 0.6805373430252075,
"loss_sft": 0.6805373430252075,
"step": 41,
"token_acc": 0.7831908093633929
},
{
"epoch": 0.14933333333333335,
"grad_norm": 2.932039260864258,
"learning_rate": 9.999848075623584e-06,
"loss": 0.682405948638916,
"loss_sft": 0.682405948638916,
"step": 42,
"token_acc": 0.7857377049180327
},
{
"epoch": 0.15288888888888888,
"grad_norm": 3.1063575744628906,
"learning_rate": 9.999658172316915e-06,
"loss": 0.7183451652526855,
"loss_sft": 0.7183451652526855,
"step": 43,
"token_acc": 0.7745383036935705
},
{
"epoch": 0.15644444444444444,
"grad_norm": 2.843851327896118,
"learning_rate": 9.999392311726738e-06,
"loss": 0.7445192337036133,
"loss_sft": 0.7445192337036133,
"step": 44,
"token_acc": 0.7745705545532983
},
{
"epoch": 0.16,
"grad_norm": 3.0106942653656006,
"learning_rate": 9.99905049789214e-06,
"loss": 0.68509840965271,
"loss_sft": 0.68509840965271,
"step": 45,
"token_acc": 0.7884892086330936
},
{
"epoch": 0.16355555555555557,
"grad_norm": 3.096278429031372,
"learning_rate": 9.998632736006124e-06,
"loss": 0.715805172920227,
"loss_sft": 0.715805172920227,
"step": 46,
"token_acc": 0.7725052980742652
},
{
"epoch": 0.1671111111111111,
"grad_norm": 3.3879644870758057,
"learning_rate": 9.998139032415534e-06,
"loss": 0.7081700563430786,
"loss_sft": 0.7081700563430786,
"step": 47,
"token_acc": 0.7791487326637972
},
{
"epoch": 0.17066666666666666,
"grad_norm": 2.87184739112854,
"learning_rate": 9.997569394620965e-06,
"loss": 0.6777009963989258,
"loss_sft": 0.6777009963989258,
"step": 48,
"token_acc": 0.7897079493987194
},
{
"epoch": 0.17422222222222222,
"grad_norm": 2.78963303565979,
"learning_rate": 9.996923831276632e-06,
"loss": 0.6779929399490356,
"loss_sft": 0.6779929399490356,
"step": 49,
"token_acc": 0.78765469363115
},
{
"epoch": 0.17777777777777778,
"grad_norm": 3.3255178928375244,
"learning_rate": 9.996202352190256e-06,
"loss": 0.6960651874542236,
"loss_sft": 0.6960651874542236,
"step": 50,
"token_acc": 0.7835450575625248
},
{
"epoch": 0.18133333333333335,
"grad_norm": 2.92441463470459,
"learning_rate": 9.995404968322902e-06,
"loss": 0.7173489332199097,
"loss_sft": 0.7173489332199097,
"step": 51,
"token_acc": 0.7745397571484528
},
{
"epoch": 0.18488888888888888,
"grad_norm": 3.020625352859497,
"learning_rate": 9.994531691788822e-06,
"loss": 0.6965892314910889,
"loss_sft": 0.6965892314910889,
"step": 52,
"token_acc": 0.7825315107402805
},
{
"epoch": 0.18844444444444444,
"grad_norm": 2.9826860427856445,
"learning_rate": 9.993582535855265e-06,
"loss": 0.713363528251648,
"loss_sft": 0.713363528251648,
"step": 53,
"token_acc": 0.7815787293961732
},
{
"epoch": 0.192,
"grad_norm": 2.809161424636841,
"learning_rate": 9.992557514942278e-06,
"loss": 0.7220005989074707,
"loss_sft": 0.7220005989074707,
"step": 54,
"token_acc": 0.7764980804540144
},
{
"epoch": 0.19555555555555557,
"grad_norm": 3.1863672733306885,
"learning_rate": 9.991456644622489e-06,
"loss": 0.679796576499939,
"loss_sft": 0.679796576499939,
"step": 55,
"token_acc": 0.7857801454768979
},
{
"epoch": 0.1991111111111111,
"grad_norm": 2.885704517364502,
"learning_rate": 9.990279941620861e-06,
"loss": 0.6900360584259033,
"loss_sft": 0.6900360584259033,
"step": 56,
"token_acc": 0.7835435537120122
},
{
"epoch": 0.20266666666666666,
"grad_norm": 2.9646353721618652,
"learning_rate": 9.989027423814454e-06,
"loss": 0.7243561148643494,
"loss_sft": 0.7243561148643494,
"step": 57,
"token_acc": 0.7800967167218121
},
{
"epoch": 0.20622222222222222,
"grad_norm": 2.882827043533325,
"learning_rate": 9.987699110232134e-06,
"loss": 0.7041727304458618,
"loss_sft": 0.7041727304458618,
"step": 58,
"token_acc": 0.7825457641549596
},
{
"epoch": 0.20977777777777779,
"grad_norm": 2.8598296642303467,
"learning_rate": 9.986295021054302e-06,
"loss": 0.6763432025909424,
"loss_sft": 0.6763432025909424,
"step": 59,
"token_acc": 0.7891812865497077
},
{
"epoch": 0.21333333333333335,
"grad_norm": 2.9331629276275635,
"learning_rate": 9.984815177612574e-06,
"loss": 0.7196650505065918,
"loss_sft": 0.7196650505065918,
"step": 60,
"token_acc": 0.7770166741775575
},
{
"epoch": 0.21688888888888888,
"grad_norm": 3.2758922576904297,
"learning_rate": 9.983259602389469e-06,
"loss": 0.7686915397644043,
"loss_sft": 0.7686915397644043,
"step": 61,
"token_acc": 0.7664570642562389
},
{
"epoch": 0.22044444444444444,
"grad_norm": 2.8715407848358154,
"learning_rate": 9.98162831901805e-06,
"loss": 0.7245374917984009,
"loss_sft": 0.7245374917984009,
"step": 62,
"token_acc": 0.7736781165573138
},
{
"epoch": 0.224,
"grad_norm": 2.851677179336548,
"learning_rate": 9.979921352281585e-06,
"loss": 0.7776187658309937,
"loss_sft": 0.7776187658309937,
"step": 63,
"token_acc": 0.7628883291351806
},
{
"epoch": 0.22755555555555557,
"grad_norm": 2.7196009159088135,
"learning_rate": 9.97813872811316e-06,
"loss": 0.6830098628997803,
"loss_sft": 0.6830098628997803,
"step": 64,
"token_acc": 0.7848438382697119
},
{
"epoch": 0.2311111111111111,
"grad_norm": 2.8247342109680176,
"learning_rate": 9.976280473595284e-06,
"loss": 0.7009365558624268,
"loss_sft": 0.7009365558624268,
"step": 65,
"token_acc": 0.7779255319148937
},
{
"epoch": 0.23466666666666666,
"grad_norm": 2.90975284576416,
"learning_rate": 9.974346616959476e-06,
"loss": 0.7488187551498413,
"loss_sft": 0.7488187551498413,
"step": 66,
"token_acc": 0.7765746255188594
},
{
"epoch": 0.23822222222222222,
"grad_norm": 2.9256434440612793,
"learning_rate": 9.972337187585848e-06,
"loss": 0.7200894951820374,
"loss_sft": 0.7200894951820374,
"step": 67,
"token_acc": 0.7783229650398669
},
{
"epoch": 0.24177777777777779,
"grad_norm": 2.661224126815796,
"learning_rate": 9.970252216002647e-06,
"loss": 0.6481496095657349,
"loss_sft": 0.6481496095657349,
"step": 68,
"token_acc": 0.7919801641586868
},
{
"epoch": 0.24533333333333332,
"grad_norm": 3.053274393081665,
"learning_rate": 9.96809173388579e-06,
"loss": 0.687166690826416,
"loss_sft": 0.687166690826416,
"step": 69,
"token_acc": 0.782608695652174
},
{
"epoch": 0.24888888888888888,
"grad_norm": 2.7737700939178467,
"learning_rate": 9.965855774058395e-06,
"loss": 0.7297186851501465,
"loss_sft": 0.7297186851501465,
"step": 70,
"token_acc": 0.7711130531500755
},
{
"epoch": 0.25244444444444447,
"grad_norm": 2.7967190742492676,
"learning_rate": 9.96354437049027e-06,
"loss": 0.7507911920547485,
"loss_sft": 0.7507911920547485,
"step": 71,
"token_acc": 0.7683655812490169
},
{
"epoch": 0.256,
"grad_norm": 2.726029396057129,
"learning_rate": 9.961157558297404e-06,
"loss": 0.6940373182296753,
"loss_sft": 0.6940373182296753,
"step": 72,
"token_acc": 0.7858455148174774
},
{
"epoch": 0.25955555555555554,
"grad_norm": 2.61533522605896,
"learning_rate": 9.958695373741428e-06,
"loss": 0.6708918809890747,
"loss_sft": 0.6708918809890747,
"step": 73,
"token_acc": 0.7878378378378378
},
{
"epoch": 0.26311111111111113,
"grad_norm": 2.7534961700439453,
"learning_rate": 9.956157854229072e-06,
"loss": 0.6960492134094238,
"loss_sft": 0.6960492134094238,
"step": 74,
"token_acc": 0.785272021794766
},
{
"epoch": 0.26666666666666666,
"grad_norm": 2.840034008026123,
"learning_rate": 9.953545038311587e-06,
"loss": 0.715815544128418,
"loss_sft": 0.715815544128418,
"step": 75,
"token_acc": 0.7733375206092487
},
{
"epoch": 0.2702222222222222,
"grad_norm": 2.6020922660827637,
"learning_rate": 9.950856965684167e-06,
"loss": 0.7033058404922485,
"loss_sft": 0.7033058404922485,
"step": 76,
"token_acc": 0.7797749170107684
},
{
"epoch": 0.2737777777777778,
"grad_norm": 2.9179458618164062,
"learning_rate": 9.948093677185345e-06,
"loss": 0.6346369981765747,
"loss_sft": 0.6346369981765747,
"step": 77,
"token_acc": 0.8030289611194756
},
{
"epoch": 0.2773333333333333,
"grad_norm": 2.746721029281616,
"learning_rate": 9.945255214796366e-06,
"loss": 0.7272379398345947,
"loss_sft": 0.7272379398345947,
"step": 78,
"token_acc": 0.7738116591928251
},
{
"epoch": 0.2808888888888889,
"grad_norm": 2.651792526245117,
"learning_rate": 9.942341621640558e-06,
"loss": 0.7153597474098206,
"loss_sft": 0.7153597474098206,
"step": 79,
"token_acc": 0.7756030211971088
},
{
"epoch": 0.28444444444444444,
"grad_norm": 2.5762507915496826,
"learning_rate": 9.939352941982671e-06,
"loss": 0.6582008600234985,
"loss_sft": 0.6582008600234985,
"step": 80,
"token_acc": 0.7950414681489324
},
{
"epoch": 0.288,
"grad_norm": 2.8364334106445312,
"learning_rate": 9.936289221228207e-06,
"loss": 0.7426301836967468,
"loss_sft": 0.7426301836967468,
"step": 81,
"token_acc": 0.773031479798397
},
{
"epoch": 0.29155555555555557,
"grad_norm": 3.6501643657684326,
"learning_rate": 9.93315050592273e-06,
"loss": 0.6809162497520447,
"loss_sft": 0.6809162497520447,
"step": 82,
"token_acc": 0.781474233529028
},
{
"epoch": 0.2951111111111111,
"grad_norm": 2.858762741088867,
"learning_rate": 9.929936843751158e-06,
"loss": 0.7804771661758423,
"loss_sft": 0.7804771661758423,
"step": 83,
"token_acc": 0.7633874969983191
},
{
"epoch": 0.2986666666666667,
"grad_norm": 2.646768093109131,
"learning_rate": 9.926648283537037e-06,
"loss": 0.638466477394104,
"loss_sft": 0.638466477394104,
"step": 84,
"token_acc": 0.7972034280559315
},
{
"epoch": 0.3022222222222222,
"grad_norm": 2.800022840499878,
"learning_rate": 9.923284875241802e-06,
"loss": 0.7209073305130005,
"loss_sft": 0.7209073305130005,
"step": 85,
"token_acc": 0.7787384716732543
},
{
"epoch": 0.30577777777777776,
"grad_norm": 2.8740427494049072,
"learning_rate": 9.91984666996402e-06,
"loss": 0.7440860867500305,
"loss_sft": 0.7440860867500305,
"step": 86,
"token_acc": 0.7676341685272472
},
{
"epoch": 0.30933333333333335,
"grad_norm": 2.6979970932006836,
"learning_rate": 9.916333719938608e-06,
"loss": 0.6785644292831421,
"loss_sft": 0.6785644292831421,
"step": 87,
"token_acc": 0.7824585053500717
},
{
"epoch": 0.3128888888888889,
"grad_norm": 2.7537035942077637,
"learning_rate": 9.912746078536044e-06,
"loss": 0.7008656859397888,
"loss_sft": 0.7008656859397888,
"step": 88,
"token_acc": 0.7769502156017248
},
{
"epoch": 0.3164444444444444,
"grad_norm": 2.941513776779175,
"learning_rate": 9.90908380026155e-06,
"loss": 0.7292702198028564,
"loss_sft": 0.7292702198028564,
"step": 89,
"token_acc": 0.7737024837532271
},
{
"epoch": 0.32,
"grad_norm": 2.751066207885742,
"learning_rate": 9.905346940754274e-06,
"loss": 0.6675152778625488,
"loss_sft": 0.6675152778625488,
"step": 90,
"token_acc": 0.7913983820373122
},
{
"epoch": 0.32355555555555554,
"grad_norm": 2.794391393661499,
"learning_rate": 9.90153555678644e-06,
"loss": 0.7140490412712097,
"loss_sft": 0.7140490412712097,
"step": 91,
"token_acc": 0.7780052102716785
},
{
"epoch": 0.32711111111111113,
"grad_norm": 2.6754798889160156,
"learning_rate": 9.897649706262474e-06,
"loss": 0.7217727899551392,
"loss_sft": 0.7217727899551392,
"step": 92,
"token_acc": 0.7711235614778922
},
{
"epoch": 0.33066666666666666,
"grad_norm": 2.7083091735839844,
"learning_rate": 9.893689448218146e-06,
"loss": 0.7184185981750488,
"loss_sft": 0.7184185981750488,
"step": 93,
"token_acc": 0.7753658144396205
},
{
"epoch": 0.3342222222222222,
"grad_norm": 2.773177146911621,
"learning_rate": 9.889654842819658e-06,
"loss": 0.7029861211776733,
"loss_sft": 0.7029861211776733,
"step": 94,
"token_acc": 0.7819624217118998
},
{
"epoch": 0.3377777777777778,
"grad_norm": 2.678514003753662,
"learning_rate": 9.885545951362733e-06,
"loss": 0.7449573874473572,
"loss_sft": 0.7449573874473572,
"step": 95,
"token_acc": 0.7711228100457409
},
{
"epoch": 0.3413333333333333,
"grad_norm": 2.6051416397094727,
"learning_rate": 9.881362836271686e-06,
"loss": 0.7303364276885986,
"loss_sft": 0.7303364276885986,
"step": 96,
"token_acc": 0.7723043694965941
},
{
"epoch": 0.3448888888888889,
"grad_norm": 2.4825963973999023,
"learning_rate": 9.877105561098473e-06,
"loss": 0.6764535903930664,
"loss_sft": 0.6764535903930664,
"step": 97,
"token_acc": 0.7857082237121277
},
{
"epoch": 0.34844444444444445,
"grad_norm": 2.871222734451294,
"learning_rate": 9.872774190521727e-06,
"loss": 0.6929079294204712,
"loss_sft": 0.6929079294204712,
"step": 98,
"token_acc": 0.7845616198992702
},
{
"epoch": 0.352,
"grad_norm": 2.690645456314087,
"learning_rate": 9.868368790345777e-06,
"loss": 0.7150665521621704,
"loss_sft": 0.7150665521621704,
"step": 99,
"token_acc": 0.779269406392694
},
{
"epoch": 0.35555555555555557,
"grad_norm": 2.6626179218292236,
"learning_rate": 9.863889427499641e-06,
"loss": 0.6892983913421631,
"loss_sft": 0.6892983913421631,
"step": 100,
"token_acc": 0.7853627986042766
},
{
"epoch": 0.3591111111111111,
"grad_norm": 2.556544780731201,
"learning_rate": 9.859336170036022e-06,
"loss": 0.6494540572166443,
"loss_sft": 0.6494540572166443,
"step": 101,
"token_acc": 0.7939301198673808
},
{
"epoch": 0.3626666666666667,
"grad_norm": 2.7808449268341064,
"learning_rate": 9.854709087130261e-06,
"loss": 0.6291214227676392,
"loss_sft": 0.6291214227676392,
"step": 102,
"token_acc": 0.7956445453607184
},
{
"epoch": 0.3662222222222222,
"grad_norm": 3.608428716659546,
"learning_rate": 9.850008249079295e-06,
"loss": 0.6827514171600342,
"loss_sft": 0.6827514171600342,
"step": 103,
"token_acc": 0.7874181971712054
},
{
"epoch": 0.36977777777777776,
"grad_norm": 2.705186128616333,
"learning_rate": 9.84523372730058e-06,
"loss": 0.7219416499137878,
"loss_sft": 0.7219416499137878,
"step": 104,
"token_acc": 0.7723319441927885
},
{
"epoch": 0.37333333333333335,
"grad_norm": 2.686668872833252,
"learning_rate": 9.840385594331022e-06,
"loss": 0.7731634378433228,
"loss_sft": 0.7731634378433228,
"step": 105,
"token_acc": 0.7657402641962601
},
{
"epoch": 0.3768888888888889,
"grad_norm": 2.5533599853515625,
"learning_rate": 9.835463923825854e-06,
"loss": 0.6954440474510193,
"loss_sft": 0.6954440474510193,
"step": 106,
"token_acc": 0.7838794664910258
},
{
"epoch": 0.3804444444444444,
"grad_norm": 2.5891125202178955,
"learning_rate": 9.830468790557536e-06,
"loss": 0.7229180335998535,
"loss_sft": 0.7229180335998535,
"step": 107,
"token_acc": 0.7756319933692499
},
{
"epoch": 0.384,
"grad_norm": 2.5785398483276367,
"learning_rate": 9.825400270414602e-06,
"loss": 0.6533997654914856,
"loss_sft": 0.6533997654914856,
"step": 108,
"token_acc": 0.7918129735723526
},
{
"epoch": 0.38755555555555554,
"grad_norm": 2.687422275543213,
"learning_rate": 9.820258440400525e-06,
"loss": 0.73026442527771,
"loss_sft": 0.73026442527771,
"step": 109,
"token_acc": 0.77734218737499
},
{
"epoch": 0.39111111111111113,
"grad_norm": 2.475904703140259,
"learning_rate": 9.81504337863253e-06,
"loss": 0.6653067469596863,
"loss_sft": 0.6653067469596863,
"step": 110,
"token_acc": 0.7945595445200063
},
{
"epoch": 0.39466666666666667,
"grad_norm": 2.6080358028411865,
"learning_rate": 9.809755164340423e-06,
"loss": 0.6671175956726074,
"loss_sft": 0.6671175956726074,
"step": 111,
"token_acc": 0.7899939826356056
},
{
"epoch": 0.3982222222222222,
"grad_norm": 2.6111488342285156,
"learning_rate": 9.804393877865373e-06,
"loss": 0.6714032888412476,
"loss_sft": 0.6714032888412476,
"step": 112,
"token_acc": 0.7879980324643384
},
{
"epoch": 0.4017777777777778,
"grad_norm": 2.9746458530426025,
"learning_rate": 9.798959600658697e-06,
"loss": 0.669405460357666,
"loss_sft": 0.669405460357666,
"step": 113,
"token_acc": 0.7869058860560277
},
{
"epoch": 0.4053333333333333,
"grad_norm": 2.8806583881378174,
"learning_rate": 9.79345241528063e-06,
"loss": 0.6186861991882324,
"loss_sft": 0.6186861991882324,
"step": 114,
"token_acc": 0.8051972685887708
},
{
"epoch": 0.4088888888888889,
"grad_norm": 2.623242139816284,
"learning_rate": 9.787872405399059e-06,
"loss": 0.7119215726852417,
"loss_sft": 0.7119215726852417,
"step": 115,
"token_acc": 0.7780878717805597
},
{
"epoch": 0.41244444444444445,
"grad_norm": 2.5655598640441895,
"learning_rate": 9.782219655788257e-06,
"loss": 0.6823458671569824,
"loss_sft": 0.6823458671569824,
"step": 116,
"token_acc": 0.788184562311341
},
{
"epoch": 0.416,
"grad_norm": 2.529388666152954,
"learning_rate": 9.776494252327597e-06,
"loss": 0.6849380731582642,
"loss_sft": 0.6849380731582642,
"step": 117,
"token_acc": 0.7850586638997525
},
{
"epoch": 0.41955555555555557,
"grad_norm": 2.6127710342407227,
"learning_rate": 9.770696282000245e-06,
"loss": 0.72218257188797,
"loss_sft": 0.72218257188797,
"step": 118,
"token_acc": 0.7739019121529722
},
{
"epoch": 0.4231111111111111,
"grad_norm": 2.6820690631866455,
"learning_rate": 9.764825832891837e-06,
"loss": 0.7707358598709106,
"loss_sft": 0.7707358598709106,
"step": 119,
"token_acc": 0.7634275069854083
},
{
"epoch": 0.4266666666666667,
"grad_norm": 2.744143009185791,
"learning_rate": 9.758882994189145e-06,
"loss": 0.7200894951820374,
"loss_sft": 0.7200894951820374,
"step": 120,
"token_acc": 0.7773874683440896
},
{
"epoch": 0.43022222222222223,
"grad_norm": 2.542837619781494,
"learning_rate": 9.752867856178719e-06,
"loss": 0.7230944037437439,
"loss_sft": 0.7230944037437439,
"step": 121,
"token_acc": 0.7701384905351143
},
{
"epoch": 0.43377777777777776,
"grad_norm": 2.661562442779541,
"learning_rate": 9.746780510245512e-06,
"loss": 0.7357878088951111,
"loss_sft": 0.7357878088951111,
"step": 122,
"token_acc": 0.7716910808702435
},
{
"epoch": 0.43733333333333335,
"grad_norm": 2.6383490562438965,
"learning_rate": 9.740621048871501e-06,
"loss": 0.695819616317749,
"loss_sft": 0.695819616317749,
"step": 123,
"token_acc": 0.7731165124124211
},
{
"epoch": 0.4408888888888889,
"grad_norm": 2.3929450511932373,
"learning_rate": 9.734389565634277e-06,
"loss": 0.6874669194221497,
"loss_sft": 0.6874669194221497,
"step": 124,
"token_acc": 0.782498463429625
},
{
"epoch": 0.4444444444444444,
"grad_norm": 2.6036064624786377,
"learning_rate": 9.728086155205614e-06,
"loss": 0.7236413359642029,
"loss_sft": 0.7236413359642029,
"step": 125,
"token_acc": 0.7763242894056848
},
{
"epoch": 0.448,
"grad_norm": 2.6495859622955322,
"learning_rate": 9.721710913350048e-06,
"loss": 0.6864535808563232,
"loss_sft": 0.6864535808563232,
"step": 126,
"token_acc": 0.785387718511878
},
{
"epoch": 0.45155555555555554,
"grad_norm": 2.7903456687927246,
"learning_rate": 9.715263936923413e-06,
"loss": 0.7277321219444275,
"loss_sft": 0.7277321219444275,
"step": 127,
"token_acc": 0.7750682990994637
},
{
"epoch": 0.45511111111111113,
"grad_norm": 2.548948049545288,
"learning_rate": 9.708745323871369e-06,
"loss": 0.7505121231079102,
"loss_sft": 0.7505121231079102,
"step": 128,
"token_acc": 0.7673134328358209
},
{
"epoch": 0.45866666666666667,
"grad_norm": 2.751535177230835,
"learning_rate": 9.702155173227911e-06,
"loss": 0.6837215423583984,
"loss_sft": 0.6837215423583984,
"step": 129,
"token_acc": 0.7892829153605015
},
{
"epoch": 0.4622222222222222,
"grad_norm": 2.6494717597961426,
"learning_rate": 9.695493585113873e-06,
"loss": 0.71864914894104,
"loss_sft": 0.71864914894104,
"step": 130,
"token_acc": 0.7745743861750364
},
{
"epoch": 0.4657777777777778,
"grad_norm": 2.5881118774414062,
"learning_rate": 9.688760660735403e-06,
"loss": 0.7104394435882568,
"loss_sft": 0.7104394435882568,
"step": 131,
"token_acc": 0.7815147194498029
},
{
"epoch": 0.4693333333333333,
"grad_norm": 2.616727113723755,
"learning_rate": 9.681956502382423e-06,
"loss": 0.6837850213050842,
"loss_sft": 0.6837850213050842,
"step": 132,
"token_acc": 0.7794079119727535
},
{
"epoch": 0.4728888888888889,
"grad_norm": 2.5975899696350098,
"learning_rate": 9.675081213427076e-06,
"loss": 0.7327243089675903,
"loss_sft": 0.7327243089675903,
"step": 133,
"token_acc": 0.7750108084738435
},
{
"epoch": 0.47644444444444445,
"grad_norm": 2.5211424827575684,
"learning_rate": 9.668134898322157e-06,
"loss": 0.7089502811431885,
"loss_sft": 0.7089502811431885,
"step": 134,
"token_acc": 0.7770884306186648
},
{
"epoch": 0.48,
"grad_norm": 2.2994611263275146,
"learning_rate": 9.661117662599527e-06,
"loss": 0.6976265907287598,
"loss_sft": 0.6976265907287598,
"step": 135,
"token_acc": 0.7797433755979045
},
{
"epoch": 0.48355555555555557,
"grad_norm": 2.6480202674865723,
"learning_rate": 9.654029612868507e-06,
"loss": 0.6763089895248413,
"loss_sft": 0.6763089895248413,
"step": 136,
"token_acc": 0.7934881189914746
},
{
"epoch": 0.4871111111111111,
"grad_norm": 2.634500503540039,
"learning_rate": 9.646870856814259e-06,
"loss": 0.6808346509933472,
"loss_sft": 0.6808346509933472,
"step": 137,
"token_acc": 0.7906399411548364
},
{
"epoch": 0.49066666666666664,
"grad_norm": 2.327711582183838,
"learning_rate": 9.639641503196152e-06,
"loss": 0.6726034283638,
"loss_sft": 0.6726034283638,
"step": 138,
"token_acc": 0.7888091822094692
},
{
"epoch": 0.49422222222222223,
"grad_norm": 2.6230361461639404,
"learning_rate": 9.632341661846107e-06,
"loss": 0.6638027429580688,
"loss_sft": 0.6638027429580688,
"step": 139,
"token_acc": 0.7867574663603545
},
{
"epoch": 0.49777777777777776,
"grad_norm": 2.687469720840454,
"learning_rate": 9.62497144366693e-06,
"loss": 0.7731108069419861,
"loss_sft": 0.7731108069419861,
"step": 140,
"token_acc": 0.767282626629269
},
{
"epoch": 0.5013333333333333,
"grad_norm": 2.6626806259155273,
"learning_rate": 9.617530960630624e-06,
"loss": 0.6771551370620728,
"loss_sft": 0.6771551370620728,
"step": 141,
"token_acc": 0.7875042329834067
},
{
"epoch": 0.5048888888888889,
"grad_norm": 2.50445294380188,
"learning_rate": 9.610020325776694e-06,
"loss": 0.718071699142456,
"loss_sft": 0.718071699142456,
"step": 142,
"token_acc": 0.7773518607251237
},
{
"epoch": 0.5084444444444445,
"grad_norm": 2.5697379112243652,
"learning_rate": 9.602439653210426e-06,
"loss": 0.6578627228736877,
"loss_sft": 0.6578627228736877,
"step": 143,
"token_acc": 0.78983238165408
},
{
"epoch": 0.512,
"grad_norm": 2.3699545860290527,
"learning_rate": 9.594789058101154e-06,
"loss": 0.7090039253234863,
"loss_sft": 0.7090039253234863,
"step": 144,
"token_acc": 0.7785146911390453
},
{
"epoch": 0.5155555555555555,
"grad_norm": 2.567875385284424,
"learning_rate": 9.587068656680506e-06,
"loss": 0.7191736698150635,
"loss_sft": 0.7191736698150635,
"step": 145,
"token_acc": 0.777732060566162
},
{
"epoch": 0.5191111111111111,
"grad_norm": 2.458695650100708,
"learning_rate": 9.579278566240646e-06,
"loss": 0.7034208178520203,
"loss_sft": 0.7034208178520203,
"step": 146,
"token_acc": 0.7787663107947805
},
{
"epoch": 0.5226666666666666,
"grad_norm": 2.5565261840820312,
"learning_rate": 9.571418905132486e-06,
"loss": 0.6433157920837402,
"loss_sft": 0.6433157920837402,
"step": 147,
"token_acc": 0.7952660019788416
},
{
"epoch": 0.5262222222222223,
"grad_norm": 2.815246820449829,
"learning_rate": 9.56348979276389e-06,
"loss": 0.6981197595596313,
"loss_sft": 0.6981197595596313,
"step": 148,
"token_acc": 0.781129900076864
},
{
"epoch": 0.5297777777777778,
"grad_norm": 2.486645460128784,
"learning_rate": 9.555491349597862e-06,
"loss": 0.6674407720565796,
"loss_sft": 0.6674407720565796,
"step": 149,
"token_acc": 0.7885185517922918
},
{
"epoch": 0.5333333333333333,
"grad_norm": 2.4752683639526367,
"learning_rate": 9.547423697150714e-06,
"loss": 0.7166387438774109,
"loss_sft": 0.7166387438774109,
"step": 150,
"token_acc": 0.776538690729548
},
{
"epoch": 0.5368888888888889,
"grad_norm": 2.585681915283203,
"learning_rate": 9.539286957990215e-06,
"loss": 0.6843577027320862,
"loss_sft": 0.6843577027320862,
"step": 151,
"token_acc": 0.7843465984293656
},
{
"epoch": 0.5404444444444444,
"grad_norm": 2.6828017234802246,
"learning_rate": 9.53108125573374e-06,
"loss": 0.6408119201660156,
"loss_sft": 0.6408119201660156,
"step": 152,
"token_acc": 0.7987977672820953
},
{
"epoch": 0.544,
"grad_norm": 2.652644395828247,
"learning_rate": 9.52280671504638e-06,
"loss": 0.7688158750534058,
"loss_sft": 0.7688158750534058,
"step": 153,
"token_acc": 0.7628647631000941
},
{
"epoch": 0.5475555555555556,
"grad_norm": 2.4523842334747314,
"learning_rate": 9.514463461639055e-06,
"loss": 0.7413673400878906,
"loss_sft": 0.7413673400878906,
"step": 154,
"token_acc": 0.7717408648476145
},
{
"epoch": 0.5511111111111111,
"grad_norm": 2.545041799545288,
"learning_rate": 9.506051622266608e-06,
"loss": 0.6747776865959167,
"loss_sft": 0.6747776865959167,
"step": 155,
"token_acc": 0.783630822812364
},
{
"epoch": 0.5546666666666666,
"grad_norm": 2.5072643756866455,
"learning_rate": 9.497571324725865e-06,
"loss": 0.699935793876648,
"loss_sft": 0.699935793876648,
"step": 156,
"token_acc": 0.7836349331235248
},
{
"epoch": 0.5582222222222222,
"grad_norm": 2.595186233520508,
"learning_rate": 9.48902269785371e-06,
"loss": 0.7426058053970337,
"loss_sft": 0.7426058053970337,
"step": 157,
"token_acc": 0.7736556566140099
},
{
"epoch": 0.5617777777777778,
"grad_norm": 2.281616687774658,
"learning_rate": 9.480405871525114e-06,
"loss": 0.6568068861961365,
"loss_sft": 0.6568068861961365,
"step": 158,
"token_acc": 0.7934681181959564
},
{
"epoch": 0.5653333333333334,
"grad_norm": 2.5341670513153076,
"learning_rate": 9.471720976651173e-06,
"loss": 0.6928246021270752,
"loss_sft": 0.6928246021270752,
"step": 159,
"token_acc": 0.7863994518670777
},
{
"epoch": 0.5688888888888889,
"grad_norm": 2.514219045639038,
"learning_rate": 9.462968145177112e-06,
"loss": 0.6754301190376282,
"loss_sft": 0.6754301190376282,
"step": 160,
"token_acc": 0.786493083807974
},
{
"epoch": 0.5724444444444444,
"grad_norm": 2.5935070514678955,
"learning_rate": 9.45414751008028e-06,
"loss": 0.6447234749794006,
"loss_sft": 0.6447234749794006,
"step": 161,
"token_acc": 0.7986577181208053
},
{
"epoch": 0.576,
"grad_norm": 2.6350297927856445,
"learning_rate": 9.445259205368138e-06,
"loss": 0.6928178071975708,
"loss_sft": 0.6928178071975708,
"step": 162,
"token_acc": 0.7820536540240518
},
{
"epoch": 0.5795555555555556,
"grad_norm": 2.571714162826538,
"learning_rate": 9.436303366076213e-06,
"loss": 0.6620484590530396,
"loss_sft": 0.6620484590530396,
"step": 163,
"token_acc": 0.7907332950411641
},
{
"epoch": 0.5831111111111111,
"grad_norm": 2.7014997005462646,
"learning_rate": 9.427280128266049e-06,
"loss": 0.7167201042175293,
"loss_sft": 0.7167201042175293,
"step": 164,
"token_acc": 0.775840597758406
},
{
"epoch": 0.5866666666666667,
"grad_norm": 2.6021246910095215,
"learning_rate": 9.418189629023149e-06,
"loss": 0.6921790838241577,
"loss_sft": 0.6921790838241577,
"step": 165,
"token_acc": 0.7815003641660597
},
{
"epoch": 0.5902222222222222,
"grad_norm": 2.6259853839874268,
"learning_rate": 9.409032006454877e-06,
"loss": 0.7029939889907837,
"loss_sft": 0.7029939889907837,
"step": 166,
"token_acc": 0.7819206984357948
},
{
"epoch": 0.5937777777777777,
"grad_norm": 2.5497219562530518,
"learning_rate": 9.399807399688371e-06,
"loss": 0.6558828353881836,
"loss_sft": 0.6558828353881836,
"step": 167,
"token_acc": 0.7927570711075865
},
{
"epoch": 0.5973333333333334,
"grad_norm": 2.485089063644409,
"learning_rate": 9.39051594886843e-06,
"loss": 0.672088623046875,
"loss_sft": 0.672088623046875,
"step": 168,
"token_acc": 0.7892642377078175
},
{
"epoch": 0.6008888888888889,
"grad_norm": 2.646801710128784,
"learning_rate": 9.381157795155374e-06,
"loss": 0.7501546144485474,
"loss_sft": 0.7501546144485474,
"step": 169,
"token_acc": 0.7674605298367675
},
{
"epoch": 0.6044444444444445,
"grad_norm": 2.616356134414673,
"learning_rate": 9.371733080722911e-06,
"loss": 0.6904166340827942,
"loss_sft": 0.6904166340827942,
"step": 170,
"token_acc": 0.78276578448198
},
{
"epoch": 0.608,
"grad_norm": 2.3303658962249756,
"learning_rate": 9.36224194875597e-06,
"loss": 0.7005062699317932,
"loss_sft": 0.7005062699317932,
"step": 171,
"token_acc": 0.7807158733808538
},
{
"epoch": 0.6115555555555555,
"grad_norm": 2.5834107398986816,
"learning_rate": 9.352684543448532e-06,
"loss": 0.7180582284927368,
"loss_sft": 0.7180582284927368,
"step": 172,
"token_acc": 0.7741369221767115
},
{
"epoch": 0.6151111111111112,
"grad_norm": 2.429469347000122,
"learning_rate": 9.343061010001428e-06,
"loss": 0.723564863204956,
"loss_sft": 0.723564863204956,
"step": 173,
"token_acc": 0.7741130191391179
},
{
"epoch": 0.6186666666666667,
"grad_norm": 2.840566396713257,
"learning_rate": 9.33337149462015e-06,
"loss": 0.6241236925125122,
"loss_sft": 0.6241236925125122,
"step": 174,
"token_acc": 0.7985352243664264
},
{
"epoch": 0.6222222222222222,
"grad_norm": 2.5669333934783936,
"learning_rate": 9.323616144512612e-06,
"loss": 0.7139682769775391,
"loss_sft": 0.7139682769775391,
"step": 175,
"token_acc": 0.7798687089715536
},
{
"epoch": 0.6257777777777778,
"grad_norm": 2.7283575534820557,
"learning_rate": 9.313795107886925e-06,
"loss": 0.6741291284561157,
"loss_sft": 0.6741291284561157,
"step": 176,
"token_acc": 0.7888552281281557
},
{
"epoch": 0.6293333333333333,
"grad_norm": 2.3858468532562256,
"learning_rate": 9.303908533949146e-06,
"loss": 0.6403597593307495,
"loss_sft": 0.6403597593307495,
"step": 177,
"token_acc": 0.7966946266921352
},
{
"epoch": 0.6328888888888888,
"grad_norm": 2.345357894897461,
"learning_rate": 9.293956572900999e-06,
"loss": 0.7029950022697449,
"loss_sft": 0.7029950022697449,
"step": 178,
"token_acc": 0.7785564120532585
},
{
"epoch": 0.6364444444444445,
"grad_norm": 2.474229097366333,
"learning_rate": 9.283939375937609e-06,
"loss": 0.6470286250114441,
"loss_sft": 0.6470286250114441,
"step": 179,
"token_acc": 0.7962058752901625
},
{
"epoch": 0.64,
"grad_norm": 2.503234386444092,
"learning_rate": 9.273857095245192e-06,
"loss": 0.6736735105514526,
"loss_sft": 0.6736735105514526,
"step": 180,
"token_acc": 0.7822236798950476
},
{
"epoch": 0.6435555555555555,
"grad_norm": 2.5394506454467773,
"learning_rate": 9.263709883998753e-06,
"loss": 0.6427879333496094,
"loss_sft": 0.6427879333496094,
"step": 181,
"token_acc": 0.7960655737704918
},
{
"epoch": 0.6471111111111111,
"grad_norm": 2.533243417739868,
"learning_rate": 9.253497896359749e-06,
"loss": 0.7394431233406067,
"loss_sft": 0.7394431233406067,
"step": 182,
"token_acc": 0.7661367249602544
},
{
"epoch": 0.6506666666666666,
"grad_norm": 2.489903450012207,
"learning_rate": 9.243221287473755e-06,
"loss": 0.6519032120704651,
"loss_sft": 0.6519032120704651,
"step": 183,
"token_acc": 0.7911711711711712
},
{
"epoch": 0.6542222222222223,
"grad_norm": 2.3369150161743164,
"learning_rate": 9.232880213468106e-06,
"loss": 0.666367769241333,
"loss_sft": 0.666367769241333,
"step": 184,
"token_acc": 0.7876486382815496
},
{
"epoch": 0.6577777777777778,
"grad_norm": 2.498046875,
"learning_rate": 9.222474831449519e-06,
"loss": 0.7021216154098511,
"loss_sft": 0.7021216154098511,
"step": 185,
"token_acc": 0.7798767309693428
},
{
"epoch": 0.6613333333333333,
"grad_norm": 2.734269618988037,
"learning_rate": 9.212005299501712e-06,
"loss": 0.679713249206543,
"loss_sft": 0.679713249206543,
"step": 186,
"token_acc": 0.7882374871842669
},
{
"epoch": 0.6648888888888889,
"grad_norm": 2.6301252841949463,
"learning_rate": 9.201471776682999e-06,
"loss": 0.6619753837585449,
"loss_sft": 0.6619753837585449,
"step": 187,
"token_acc": 0.790052606408417
},
{
"epoch": 0.6684444444444444,
"grad_norm": 2.7517828941345215,
"learning_rate": 9.19087442302388e-06,
"loss": 0.6709531545639038,
"loss_sft": 0.6709531545639038,
"step": 188,
"token_acc": 0.7872744539411206
},
{
"epoch": 0.672,
"grad_norm": 2.5251734256744385,
"learning_rate": 9.180213399524599e-06,
"loss": 0.6628526449203491,
"loss_sft": 0.6628526449203491,
"step": 189,
"token_acc": 0.7883199464524766
},
{
"epoch": 0.6755555555555556,
"grad_norm": 2.489914655685425,
"learning_rate": 9.169488868152704e-06,
"loss": 0.6828216314315796,
"loss_sft": 0.6828216314315796,
"step": 190,
"token_acc": 0.7839811734155644
},
{
"epoch": 0.6791111111111111,
"grad_norm": 2.675612688064575,
"learning_rate": 9.15870099184059e-06,
"loss": 0.6962224245071411,
"loss_sft": 0.6962224245071411,
"step": 191,
"token_acc": 0.782114403365858
},
{
"epoch": 0.6826666666666666,
"grad_norm": 2.5126285552978516,
"learning_rate": 9.147849934483019e-06,
"loss": 0.6784244179725647,
"loss_sft": 0.6784244179725647,
"step": 192,
"token_acc": 0.7871374527112232
},
{
"epoch": 0.6862222222222222,
"grad_norm": 2.372910261154175,
"learning_rate": 9.136935860934628e-06,
"loss": 0.6575683355331421,
"loss_sft": 0.6575683355331421,
"step": 193,
"token_acc": 0.7935674989420228
},
{
"epoch": 0.6897777777777778,
"grad_norm": 2.3778316974639893,
"learning_rate": 9.125958937007427e-06,
"loss": 0.6807299256324768,
"loss_sft": 0.6807299256324768,
"step": 194,
"token_acc": 0.7886637797711105
},
{
"epoch": 0.6933333333333334,
"grad_norm": 2.543238639831543,
"learning_rate": 9.114919329468283e-06,
"loss": 0.6765571236610413,
"loss_sft": 0.6765571236610413,
"step": 195,
"token_acc": 0.7888258402444347
},
{
"epoch": 0.6968888888888889,
"grad_norm": 2.76015567779541,
"learning_rate": 9.103817206036383e-06,
"loss": 0.7136564254760742,
"loss_sft": 0.7136564254760742,
"step": 196,
"token_acc": 0.779906103286385
},
{
"epoch": 0.7004444444444444,
"grad_norm": 2.7252256870269775,
"learning_rate": 9.092652735380683e-06,
"loss": 0.6603894233703613,
"loss_sft": 0.6603894233703613,
"step": 197,
"token_acc": 0.7899476247358265
},
{
"epoch": 0.704,
"grad_norm": 2.811830997467041,
"learning_rate": 9.081426087117356e-06,
"loss": 0.7277945876121521,
"loss_sft": 0.7277945876121521,
"step": 198,
"token_acc": 0.777927927927928
},
{
"epoch": 0.7075555555555556,
"grad_norm": 2.5671072006225586,
"learning_rate": 9.0701374318072e-06,
"loss": 0.7158607244491577,
"loss_sft": 0.7158607244491577,
"step": 199,
"token_acc": 0.7807679684936003
},
{
"epoch": 0.7111111111111111,
"grad_norm": 2.7316253185272217,
"learning_rate": 9.058786940953057e-06,
"loss": 0.7203642129898071,
"loss_sft": 0.7203642129898071,
"step": 200,
"token_acc": 0.771410803127221
},
{
"epoch": 0.7146666666666667,
"grad_norm": 2.6067488193511963,
"learning_rate": 9.04737478699721e-06,
"loss": 0.709173858165741,
"loss_sft": 0.709173858165741,
"step": 201,
"token_acc": 0.7760461511968314
},
{
"epoch": 0.7182222222222222,
"grad_norm": 2.428915023803711,
"learning_rate": 9.035901143318753e-06,
"loss": 0.6437892913818359,
"loss_sft": 0.6437892913818359,
"step": 202,
"token_acc": 0.7929373996789727
},
{
"epoch": 0.7217777777777777,
"grad_norm": 2.53688645362854,
"learning_rate": 9.024366184230964e-06,
"loss": 0.7296491265296936,
"loss_sft": 0.7296491265296936,
"step": 203,
"token_acc": 0.7743366681165724
},
{
"epoch": 0.7253333333333334,
"grad_norm": 2.4642114639282227,
"learning_rate": 9.012770084978657e-06,
"loss": 0.6877669095993042,
"loss_sft": 0.6877669095993042,
"step": 204,
"token_acc": 0.7817484008528784
},
{
"epoch": 0.7288888888888889,
"grad_norm": 2.520726442337036,
"learning_rate": 9.001113021735512e-06,
"loss": 0.6411060094833374,
"loss_sft": 0.6411060094833374,
"step": 205,
"token_acc": 0.798235347078419
},
{
"epoch": 0.7324444444444445,
"grad_norm": 2.683303117752075,
"learning_rate": 8.989395171601413e-06,
"loss": 0.7191975712776184,
"loss_sft": 0.7191975712776184,
"step": 206,
"token_acc": 0.7790291601187358
},
{
"epoch": 0.736,
"grad_norm": 2.530151605606079,
"learning_rate": 8.97761671259974e-06,
"loss": 0.6858580112457275,
"loss_sft": 0.6858580112457275,
"step": 207,
"token_acc": 0.7843207767688574
},
{
"epoch": 0.7395555555555555,
"grad_norm": 2.556028127670288,
"learning_rate": 8.965777823674679e-06,
"loss": 0.6540775299072266,
"loss_sft": 0.6540775299072266,
"step": 208,
"token_acc": 0.7965488907148727
},
{
"epoch": 0.7431111111111111,
"grad_norm": 2.5358355045318604,
"learning_rate": 8.953878684688492e-06,
"loss": 0.6654298305511475,
"loss_sft": 0.6654298305511475,
"step": 209,
"token_acc": 0.7853901592668386
},
{
"epoch": 0.7466666666666667,
"grad_norm": 2.5297691822052,
"learning_rate": 8.941919476418797e-06,
"loss": 0.6690406203269958,
"loss_sft": 0.6690406203269958,
"step": 210,
"token_acc": 0.7922668688400303
},
{
"epoch": 0.7502222222222222,
"grad_norm": 2.7297613620758057,
"learning_rate": 8.929900380555805e-06,
"loss": 0.763491153717041,
"loss_sft": 0.763491153717041,
"step": 211,
"token_acc": 0.7688363245389133
},
{
"epoch": 0.7537777777777778,
"grad_norm": 2.8712141513824463,
"learning_rate": 8.917821579699578e-06,
"loss": 0.7024750113487244,
"loss_sft": 0.7024750113487244,
"step": 212,
"token_acc": 0.7766816143497758
},
{
"epoch": 0.7573333333333333,
"grad_norm": 2.30037522315979,
"learning_rate": 8.905683257357238e-06,
"loss": 0.674483597278595,
"loss_sft": 0.674483597278595,
"step": 213,
"token_acc": 0.7871611331099604
},
{
"epoch": 0.7608888888888888,
"grad_norm": 2.575424909591675,
"learning_rate": 8.893485597940195e-06,
"loss": 0.668804407119751,
"loss_sft": 0.668804407119751,
"step": 214,
"token_acc": 0.7899144689950107
},
{
"epoch": 0.7644444444444445,
"grad_norm": 2.8385753631591797,
"learning_rate": 8.881228786761329e-06,
"loss": 0.7176224589347839,
"loss_sft": 0.7176224589347839,
"step": 215,
"token_acc": 0.7743912143880312
},
{
"epoch": 0.768,
"grad_norm": 2.399359703063965,
"learning_rate": 8.868913010032188e-06,
"loss": 0.6282802820205688,
"loss_sft": 0.6282802820205688,
"step": 216,
"token_acc": 0.7978705425046476
},
{
"epoch": 0.7715555555555556,
"grad_norm": 2.6174166202545166,
"learning_rate": 8.856538454860155e-06,
"loss": 0.7392973899841309,
"loss_sft": 0.7392973899841309,
"step": 217,
"token_acc": 0.7734328358208955
},
{
"epoch": 0.7751111111111111,
"grad_norm": 2.5575881004333496,
"learning_rate": 8.8441053092456e-06,
"loss": 0.7398823499679565,
"loss_sft": 0.7398823499679565,
"step": 218,
"token_acc": 0.7702180028129395
},
{
"epoch": 0.7786666666666666,
"grad_norm": 2.433011293411255,
"learning_rate": 8.831613762079026e-06,
"loss": 0.6455325484275818,
"loss_sft": 0.6455325484275818,
"step": 219,
"token_acc": 0.797568009977395
},
{
"epoch": 0.7822222222222223,
"grad_norm": 2.456592321395874,
"learning_rate": 8.819064003138211e-06,
"loss": 0.6836599111557007,
"loss_sft": 0.6836599111557007,
"step": 220,
"token_acc": 0.7874347959496778
},
{
"epoch": 0.7857777777777778,
"grad_norm": 2.5440144538879395,
"learning_rate": 8.806456223085308e-06,
"loss": 0.6766096353530884,
"loss_sft": 0.6766096353530884,
"step": 221,
"token_acc": 0.7898030554021719
},
{
"epoch": 0.7893333333333333,
"grad_norm": 2.5348002910614014,
"learning_rate": 8.793790613463956e-06,
"loss": 0.6729533672332764,
"loss_sft": 0.6729533672332764,
"step": 222,
"token_acc": 0.7868811468578096
},
{
"epoch": 0.7928888888888889,
"grad_norm": 2.6501808166503906,
"learning_rate": 8.781067366696368e-06,
"loss": 0.6886971592903137,
"loss_sft": 0.6886971592903137,
"step": 223,
"token_acc": 0.7857977528089888
},
{
"epoch": 0.7964444444444444,
"grad_norm": 2.446477174758911,
"learning_rate": 8.768286676080415e-06,
"loss": 0.6379653215408325,
"loss_sft": 0.6379653215408325,
"step": 224,
"token_acc": 0.7950165815183293
},
{
"epoch": 0.8,
"grad_norm": 2.5404648780822754,
"learning_rate": 8.75544873578668e-06,
"loss": 0.6643590927124023,
"loss_sft": 0.6643590927124023,
"step": 225,
"token_acc": 0.7900611873656358
},
{
"epoch": 0.8035555555555556,
"grad_norm": 2.3945586681365967,
"learning_rate": 8.742553740855507e-06,
"loss": 0.6230128407478333,
"loss_sft": 0.6230128407478333,
"step": 226,
"token_acc": 0.8015783540022547
},
{
"epoch": 0.8071111111111111,
"grad_norm": 2.5670764446258545,
"learning_rate": 8.729601887194048e-06,
"loss": 0.7258001565933228,
"loss_sft": 0.7258001565933228,
"step": 227,
"token_acc": 0.7744393833216539
},
{
"epoch": 0.8106666666666666,
"grad_norm": 2.578752279281616,
"learning_rate": 8.71659337157328e-06,
"loss": 0.6670640707015991,
"loss_sft": 0.6670640707015991,
"step": 228,
"token_acc": 0.7867372623904175
},
{
"epoch": 0.8142222222222222,
"grad_norm": 2.6314897537231445,
"learning_rate": 8.70352839162502e-06,
"loss": 0.6822625398635864,
"loss_sft": 0.6822625398635864,
"step": 229,
"token_acc": 0.7871383203952012
},
{
"epoch": 0.8177777777777778,
"grad_norm": 2.5298144817352295,
"learning_rate": 8.690407145838909e-06,
"loss": 0.7157070636749268,
"loss_sft": 0.7157070636749268,
"step": 230,
"token_acc": 0.7746970253396989
},
{
"epoch": 0.8213333333333334,
"grad_norm": 2.2555112838745117,
"learning_rate": 8.677229833559413e-06,
"loss": 0.6939213275909424,
"loss_sft": 0.6939213275909424,
"step": 231,
"token_acc": 0.7837013904379507
},
{
"epoch": 0.8248888888888889,
"grad_norm": 2.3058218955993652,
"learning_rate": 8.663996654982791e-06,
"loss": 0.7002938985824585,
"loss_sft": 0.7002938985824585,
"step": 232,
"token_acc": 0.7825442633179939
},
{
"epoch": 0.8284444444444444,
"grad_norm": 2.2887587547302246,
"learning_rate": 8.650707811154045e-06,
"loss": 0.6238073110580444,
"loss_sft": 0.6238073110580444,
"step": 233,
"token_acc": 0.8007479561662898
},
{
"epoch": 0.832,
"grad_norm": 2.471907138824463,
"learning_rate": 8.637363503963873e-06,
"loss": 0.7511302828788757,
"loss_sft": 0.7511302828788757,
"step": 234,
"token_acc": 0.7642256902761104
},
{
"epoch": 0.8355555555555556,
"grad_norm": 2.371619939804077,
"learning_rate": 8.6239639361456e-06,
"loss": 0.7046500444412231,
"loss_sft": 0.7046500444412231,
"step": 235,
"token_acc": 0.7786012773003232
},
{
"epoch": 0.8391111111111111,
"grad_norm": 2.411160945892334,
"learning_rate": 8.610509311272099e-06,
"loss": 0.6829012036323547,
"loss_sft": 0.6829012036323547,
"step": 236,
"token_acc": 0.784978026368358
},
{
"epoch": 0.8426666666666667,
"grad_norm": 2.6158106327056885,
"learning_rate": 8.596999833752694e-06,
"loss": 0.6653611063957214,
"loss_sft": 0.6653611063957214,
"step": 237,
"token_acc": 0.7922825878287579
},
{
"epoch": 0.8462222222222222,
"grad_norm": 2.3500356674194336,
"learning_rate": 8.583435708830058e-06,
"loss": 0.7233660221099854,
"loss_sft": 0.7233660221099854,
"step": 238,
"token_acc": 0.7756333830104322
},
{
"epoch": 0.8497777777777777,
"grad_norm": 2.690075397491455,
"learning_rate": 8.569817142577099e-06,
"loss": 0.68476402759552,
"loss_sft": 0.68476402759552,
"step": 239,
"token_acc": 0.7839506172839507
},
{
"epoch": 0.8533333333333334,
"grad_norm": 2.2535135746002197,
"learning_rate": 8.556144341893819e-06,
"loss": 0.6368871331214905,
"loss_sft": 0.6368871331214905,
"step": 240,
"token_acc": 0.7993680884676145
},
{
"epoch": 0.8568888888888889,
"grad_norm": 2.300875663757324,
"learning_rate": 8.54241751450418e-06,
"loss": 0.7263578176498413,
"loss_sft": 0.7263578176498413,
"step": 241,
"token_acc": 0.7760888629055832
},
{
"epoch": 0.8604444444444445,
"grad_norm": 2.47619366645813,
"learning_rate": 8.528636868952944e-06,
"loss": 0.6952390670776367,
"loss_sft": 0.6952390670776367,
"step": 242,
"token_acc": 0.7814606222878774
},
{
"epoch": 0.864,
"grad_norm": 2.4331600666046143,
"learning_rate": 8.514802614602503e-06,
"loss": 0.7081923484802246,
"loss_sft": 0.7081923484802246,
"step": 243,
"token_acc": 0.7771939906411625
},
{
"epoch": 0.8675555555555555,
"grad_norm": 2.4349710941314697,
"learning_rate": 8.5009149616297e-06,
"loss": 0.7285066246986389,
"loss_sft": 0.7285066246986389,
"step": 244,
"token_acc": 0.77128
},
{
"epoch": 0.8711111111111111,
"grad_norm": 2.5287394523620605,
"learning_rate": 8.486974121022642e-06,
"loss": 0.6719340682029724,
"loss_sft": 0.6719340682029724,
"step": 245,
"token_acc": 0.7888494502640464
},
{
"epoch": 0.8746666666666667,
"grad_norm": 2.5344700813293457,
"learning_rate": 8.472980304577483e-06,
"loss": 0.7082866430282593,
"loss_sft": 0.7082866430282593,
"step": 246,
"token_acc": 0.7771880819366853
},
{
"epoch": 0.8782222222222222,
"grad_norm": 2.4744646549224854,
"learning_rate": 8.458933724895216e-06,
"loss": 0.6877163648605347,
"loss_sft": 0.6877163648605347,
"step": 247,
"token_acc": 0.7891390273242825
},
{
"epoch": 0.8817777777777778,
"grad_norm": 2.526578903198242,
"learning_rate": 8.444834595378434e-06,
"loss": 0.6795492172241211,
"loss_sft": 0.6795492172241211,
"step": 248,
"token_acc": 0.7854313759256071
},
{
"epoch": 0.8853333333333333,
"grad_norm": 2.557000160217285,
"learning_rate": 8.4306831302281e-06,
"loss": 0.6654568910598755,
"loss_sft": 0.6654568910598755,
"step": 249,
"token_acc": 0.7897387741164419
},
{
"epoch": 0.8888888888888888,
"grad_norm": 2.4727110862731934,
"learning_rate": 8.416479544440279e-06,
"loss": 0.6629625558853149,
"loss_sft": 0.6629625558853149,
"step": 250,
"token_acc": 0.7905514639841925
},
{
"epoch": 0.8924444444444445,
"grad_norm": 2.6091437339782715,
"learning_rate": 8.402224053802884e-06,
"loss": 0.6646929979324341,
"loss_sft": 0.6646929979324341,
"step": 251,
"token_acc": 0.7899270457941461
},
{
"epoch": 0.896,
"grad_norm": 2.429840564727783,
"learning_rate": 8.38791687489239e-06,
"loss": 0.6530863642692566,
"loss_sft": 0.6530863642692566,
"step": 252,
"token_acc": 0.7912244045131634
},
{
"epoch": 0.8995555555555556,
"grad_norm": 2.3686275482177734,
"learning_rate": 8.373558225070546e-06,
"loss": 0.7298341393470764,
"loss_sft": 0.7298341393470764,
"step": 253,
"token_acc": 0.77202154626109
},
{
"epoch": 0.9031111111111111,
"grad_norm": 2.402869939804077,
"learning_rate": 8.359148322481073e-06,
"loss": 0.6822713613510132,
"loss_sft": 0.6822713613510132,
"step": 254,
"token_acc": 0.7850531960903037
},
{
"epoch": 0.9066666666666666,
"grad_norm": 2.6632466316223145,
"learning_rate": 8.344687386046348e-06,
"loss": 0.6570214629173279,
"loss_sft": 0.6570214629173279,
"step": 255,
"token_acc": 0.7910495301799649
},
{
"epoch": 0.9102222222222223,
"grad_norm": 2.571784257888794,
"learning_rate": 8.330175635464082e-06,
"loss": 0.6397949457168579,
"loss_sft": 0.6397949457168579,
"step": 256,
"token_acc": 0.7965891768292683
},
{
"epoch": 0.9137777777777778,
"grad_norm": 2.5593502521514893,
"learning_rate": 8.315613291203977e-06,
"loss": 0.6897088885307312,
"loss_sft": 0.6897088885307312,
"step": 257,
"token_acc": 0.7862378201258946
},
{
"epoch": 0.9173333333333333,
"grad_norm": 2.330904960632324,
"learning_rate": 8.30100057450438e-06,
"loss": 0.661052942276001,
"loss_sft": 0.661052942276001,
"step": 258,
"token_acc": 0.7922600370102181
},
{
"epoch": 0.9208888888888889,
"grad_norm": 2.5684080123901367,
"learning_rate": 8.286337707368922e-06,
"loss": 0.7451004385948181,
"loss_sft": 0.7451004385948181,
"step": 259,
"token_acc": 0.7741195741195741
},
{
"epoch": 0.9244444444444444,
"grad_norm": 2.4791767597198486,
"learning_rate": 8.271624912563143e-06,
"loss": 0.6460379362106323,
"loss_sft": 0.6460379362106323,
"step": 260,
"token_acc": 0.7945180777206744
},
{
"epoch": 0.928,
"grad_norm": 2.327636957168579,
"learning_rate": 8.256862413611113e-06,
"loss": 0.6593332290649414,
"loss_sft": 0.6593332290649414,
"step": 261,
"token_acc": 0.78983023179889
},
{
"epoch": 0.9315555555555556,
"grad_norm": 2.414180040359497,
"learning_rate": 8.242050434792022e-06,
"loss": 0.6383700966835022,
"loss_sft": 0.6383700966835022,
"step": 262,
"token_acc": 0.7902780335113239
},
{
"epoch": 0.9351111111111111,
"grad_norm": 2.501138210296631,
"learning_rate": 8.227189201136796e-06,
"loss": 0.6696315407752991,
"loss_sft": 0.6696315407752991,
"step": 263,
"token_acc": 0.7905369949274095
},
{
"epoch": 0.9386666666666666,
"grad_norm": 2.500870943069458,
"learning_rate": 8.212278938424654e-06,
"loss": 0.6867767572402954,
"loss_sft": 0.6867767572402954,
"step": 264,
"token_acc": 0.7837143803843605
},
{
"epoch": 0.9422222222222222,
"grad_norm": 2.25700044631958,
"learning_rate": 8.197319873179694e-06,
"loss": 0.6714178323745728,
"loss_sft": 0.6714178323745728,
"step": 265,
"token_acc": 0.7876513317191284
},
{
"epoch": 0.9457777777777778,
"grad_norm": 2.3075528144836426,
"learning_rate": 8.182312232667446e-06,
"loss": 0.6783133745193481,
"loss_sft": 0.6783133745193481,
"step": 266,
"token_acc": 0.7897855295001251
},
{
"epoch": 0.9493333333333334,
"grad_norm": 2.477632999420166,
"learning_rate": 8.167256244891416e-06,
"loss": 0.6760266423225403,
"loss_sft": 0.6760266423225403,
"step": 267,
"token_acc": 0.7865006316549359
},
{
"epoch": 0.9528888888888889,
"grad_norm": 2.432343006134033,
"learning_rate": 8.152152138589633e-06,
"loss": 0.6625715494155884,
"loss_sft": 0.6625715494155884,
"step": 268,
"token_acc": 0.788118478626725
},
{
"epoch": 0.9564444444444444,
"grad_norm": 2.4262497425079346,
"learning_rate": 8.137000143231156e-06,
"loss": 0.6649436354637146,
"loss_sft": 0.6649436354637146,
"step": 269,
"token_acc": 0.7894128334296804
},
{
"epoch": 0.96,
"grad_norm": 2.2512545585632324,
"learning_rate": 8.121800489012608e-06,
"loss": 0.6265535354614258,
"loss_sft": 0.6265535354614258,
"step": 270,
"token_acc": 0.802741935483871
},
{
"epoch": 0.9635555555555556,
"grad_norm": 2.538525342941284,
"learning_rate": 8.106553406854664e-06,
"loss": 0.6871274709701538,
"loss_sft": 0.6871274709701538,
"step": 271,
"token_acc": 0.7819657638833079
},
{
"epoch": 0.9671111111111111,
"grad_norm": 2.6502013206481934,
"learning_rate": 8.091259128398548e-06,
"loss": 0.6586301326751709,
"loss_sft": 0.6586301326751709,
"step": 272,
"token_acc": 0.7929534580252283
},
{
"epoch": 0.9706666666666667,
"grad_norm": 2.541417121887207,
"learning_rate": 8.075917886002514e-06,
"loss": 0.669592559337616,
"loss_sft": 0.669592559337616,
"step": 273,
"token_acc": 0.7868585624946458
},
{
"epoch": 0.9742222222222222,
"grad_norm": 2.4923152923583984,
"learning_rate": 8.060529912738316e-06,
"loss": 0.6454874873161316,
"loss_sft": 0.6454874873161316,
"step": 274,
"token_acc": 0.7938175882472964
},
{
"epoch": 0.9777777777777777,
"grad_norm": 2.464840888977051,
"learning_rate": 8.045095442387666e-06,
"loss": 0.7100995779037476,
"loss_sft": 0.7100995779037476,
"step": 275,
"token_acc": 0.7732960013999475
},
{
"epoch": 0.9813333333333333,
"grad_norm": 2.298234701156616,
"learning_rate": 8.029614709438684e-06,
"loss": 0.696420431137085,
"loss_sft": 0.696420431137085,
"step": 276,
"token_acc": 0.7872523686477175
},
{
"epoch": 0.9848888888888889,
"grad_norm": 2.4526379108428955,
"learning_rate": 8.014087949082333e-06,
"loss": 0.6658641695976257,
"loss_sft": 0.6658641695976257,
"step": 277,
"token_acc": 0.7910529686213521
},
{
"epoch": 0.9884444444444445,
"grad_norm": 2.463639259338379,
"learning_rate": 7.998515397208846e-06,
"loss": 0.668074369430542,
"loss_sft": 0.668074369430542,
"step": 278,
"token_acc": 0.7867846181621231
},
{
"epoch": 0.992,
"grad_norm": 2.703293561935425,
"learning_rate": 7.982897290404146e-06,
"loss": 0.6501681804656982,
"loss_sft": 0.6501681804656982,
"step": 279,
"token_acc": 0.7934236399388246
},
{
"epoch": 0.9955555555555555,
"grad_norm": 2.416779041290283,
"learning_rate": 7.967233865946249e-06,
"loss": 0.6700237989425659,
"loss_sft": 0.6700237989425659,
"step": 280,
"token_acc": 0.7841176470588235
},
{
"epoch": 0.9991111111111111,
"grad_norm": 2.30338978767395,
"learning_rate": 7.951525361801655e-06,
"loss": 0.6401051878929138,
"loss_sft": 0.6401051878929138,
"step": 281,
"token_acc": 0.7948136958710977
},
{
"epoch": 1.0,
"grad_norm": 4.242886066436768,
"learning_rate": 7.935772016621744e-06,
"loss": 0.6590629816055298,
"loss_sft": 0.6590629816055298,
"step": 282,
"token_acc": 0.7991955445544554
},
{
"epoch": 1.0035555555555555,
"grad_norm": 2.2138984203338623,
"learning_rate": 7.919974069739136e-06,
"loss": 0.4501396417617798,
"loss_sft": 0.4501396417617798,
"step": 283,
"token_acc": 0.8530516022002969
},
{
"epoch": 1.007111111111111,
"grad_norm": 2.2157094478607178,
"learning_rate": 7.904131761164068e-06,
"loss": 0.4709664285182953,
"loss_sft": 0.4709664285182953,
"step": 284,
"token_acc": 0.8448854376378103
},
{
"epoch": 1.0106666666666666,
"grad_norm": 2.2698137760162354,
"learning_rate": 7.888245331580737e-06,
"loss": 0.4294776916503906,
"loss_sft": 0.4294776916503906,
"step": 285,
"token_acc": 0.8534061277705346
},
{
"epoch": 1.0142222222222221,
"grad_norm": 2.198092460632324,
"learning_rate": 7.872315022343654e-06,
"loss": 0.4496222138404846,
"loss_sft": 0.4496222138404846,
"step": 286,
"token_acc": 0.8473847212663455
},
{
"epoch": 1.0177777777777777,
"grad_norm": 2.236948013305664,
"learning_rate": 7.856341075473963e-06,
"loss": 0.44401296973228455,
"loss_sft": 0.44401296973228455,
"step": 287,
"token_acc": 0.8510225669957687
},
{
"epoch": 1.0213333333333334,
"grad_norm": 2.32940936088562,
"learning_rate": 7.84032373365578e-06,
"loss": 0.42576032876968384,
"loss_sft": 0.42576032876968384,
"step": 288,
"token_acc": 0.8546143080596031
},
{
"epoch": 1.024888888888889,
"grad_norm": 2.484039306640625,
"learning_rate": 7.824263240232497e-06,
"loss": 0.41227564215660095,
"loss_sft": 0.41227564215660095,
"step": 289,
"token_acc": 0.8595331374469475
},
{
"epoch": 1.0284444444444445,
"grad_norm": 2.745042324066162,
"learning_rate": 7.808159839203085e-06,
"loss": 0.434820294380188,
"loss_sft": 0.434820294380188,
"step": 290,
"token_acc": 0.8525674499564839
},
{
"epoch": 1.032,
"grad_norm": 2.948111057281494,
"learning_rate": 7.792013775218385e-06,
"loss": 0.4367086589336395,
"loss_sft": 0.4367086589336395,
"step": 291,
"token_acc": 0.85426944971537
},
{
"epoch": 1.0355555555555556,
"grad_norm": 2.8025801181793213,
"learning_rate": 7.775825293577407e-06,
"loss": 0.41897547245025635,
"loss_sft": 0.41897547245025635,
"step": 292,
"token_acc": 0.8541594753192958
},
{
"epoch": 1.039111111111111,
"grad_norm": 2.689270257949829,
"learning_rate": 7.759594640223576e-06,
"loss": 0.4060119390487671,
"loss_sft": 0.4060119390487671,
"step": 293,
"token_acc": 0.8614083534691059
},
{
"epoch": 1.0426666666666666,
"grad_norm": 2.5525572299957275,
"learning_rate": 7.743322061741024e-06,
"loss": 0.39905354380607605,
"loss_sft": 0.39905354380607605,
"step": 294,
"token_acc": 0.860615739508473
},
{
"epoch": 1.0462222222222222,
"grad_norm": 2.653327226638794,
"learning_rate": 7.727007805350815e-06,
"loss": 0.4240958094596863,
"loss_sft": 0.4240958094596863,
"step": 295,
"token_acc": 0.8523908523908524
},
{
"epoch": 1.0497777777777777,
"grad_norm": 2.826683759689331,
"learning_rate": 7.71065211890722e-06,
"loss": 0.45821306109428406,
"loss_sft": 0.45821306109428406,
"step": 296,
"token_acc": 0.8489945498966359
},
{
"epoch": 1.0533333333333332,
"grad_norm": 2.46380615234375,
"learning_rate": 7.694255250893927e-06,
"loss": 0.3980301022529602,
"loss_sft": 0.3980301022529602,
"step": 297,
"token_acc": 0.8644877530901646
},
{
"epoch": 1.056888888888889,
"grad_norm": 2.4481019973754883,
"learning_rate": 7.677817450420279e-06,
"loss": 0.3981451392173767,
"loss_sft": 0.3981451392173767,
"step": 298,
"token_acc": 0.8660332541567696
},
{
"epoch": 1.0604444444444445,
"grad_norm": 2.4818520545959473,
"learning_rate": 7.661338967217483e-06,
"loss": 0.4272841811180115,
"loss_sft": 0.4272841811180115,
"step": 299,
"token_acc": 0.8564369310793238
},
{
"epoch": 1.064,
"grad_norm": 2.6314663887023926,
"learning_rate": 7.644820051634813e-06,
"loss": 0.3870023787021637,
"loss_sft": 0.3870023787021637,
"step": 300,
"token_acc": 0.8716049382716049
},
{
"epoch": 1.0675555555555556,
"grad_norm": 2.5124173164367676,
"learning_rate": 7.628260954635822e-06,
"loss": 0.42054182291030884,
"loss_sft": 0.42054182291030884,
"step": 301,
"token_acc": 0.8545139461957418
},
{
"epoch": 1.0711111111111111,
"grad_norm": 2.526641607284546,
"learning_rate": 7.6116619277945135e-06,
"loss": 0.457943856716156,
"loss_sft": 0.457943856716156,
"step": 302,
"token_acc": 0.8472277622656562
},
{
"epoch": 1.0746666666666667,
"grad_norm": 2.6802306175231934,
"learning_rate": 7.595023223291525e-06,
"loss": 0.45375150442123413,
"loss_sft": 0.45375150442123413,
"step": 303,
"token_acc": 0.8503710575139146
},
{
"epoch": 1.0782222222222222,
"grad_norm": 2.5229461193084717,
"learning_rate": 7.578345093910298e-06,
"loss": 0.40317875146865845,
"loss_sft": 0.40317875146865845,
"step": 304,
"token_acc": 0.8623370110330993
},
{
"epoch": 1.0817777777777777,
"grad_norm": 2.7340550422668457,
"learning_rate": 7.561627793033233e-06,
"loss": 0.4202750325202942,
"loss_sft": 0.4202750325202942,
"step": 305,
"token_acc": 0.8546205860255447
},
{
"epoch": 1.0853333333333333,
"grad_norm": 2.594975471496582,
"learning_rate": 7.54487157463785e-06,
"loss": 0.4285189211368561,
"loss_sft": 0.4285189211368561,
"step": 306,
"token_acc": 0.8582362445786206
},
{
"epoch": 1.0888888888888888,
"grad_norm": 2.590012550354004,
"learning_rate": 7.528076693292914e-06,
"loss": 0.496899276971817,
"loss_sft": 0.496899276971817,
"step": 307,
"token_acc": 0.8424388771506188
},
{
"epoch": 1.0924444444444443,
"grad_norm": 2.3477694988250732,
"learning_rate": 7.511243404154586e-06,
"loss": 0.42052632570266724,
"loss_sft": 0.42052632570266724,
"step": 308,
"token_acc": 0.8560564484515876
},
{
"epoch": 1.096,
"grad_norm": 2.42828106880188,
"learning_rate": 7.49437196296253e-06,
"loss": 0.438238263130188,
"loss_sft": 0.438238263130188,
"step": 309,
"token_acc": 0.848504394098202
},
{
"epoch": 1.0995555555555556,
"grad_norm": 2.396926164627075,
"learning_rate": 7.4774626260360384e-06,
"loss": 0.4121718406677246,
"loss_sft": 0.4121718406677246,
"step": 310,
"token_acc": 0.8627187079407806
},
{
"epoch": 1.1031111111111112,
"grad_norm": 2.380232334136963,
"learning_rate": 7.460515650270134e-06,
"loss": 0.39349058270454407,
"loss_sft": 0.39349058270454407,
"step": 311,
"token_acc": 0.8637506736123586
},
{
"epoch": 1.1066666666666667,
"grad_norm": 2.5577590465545654,
"learning_rate": 7.443531293131667e-06,
"loss": 0.4434744715690613,
"loss_sft": 0.4434744715690613,
"step": 312,
"token_acc": 0.8480109739368998
},
{
"epoch": 1.1102222222222222,
"grad_norm": 2.4358432292938232,
"learning_rate": 7.4265098126554065e-06,
"loss": 0.38979750871658325,
"loss_sft": 0.38979750871658325,
"step": 313,
"token_acc": 0.8669380370872908
},
{
"epoch": 1.1137777777777778,
"grad_norm": 2.611020088195801,
"learning_rate": 7.409451467440111e-06,
"loss": 0.38502222299575806,
"loss_sft": 0.38502222299575806,
"step": 314,
"token_acc": 0.8728053062817012
},
{
"epoch": 1.1173333333333333,
"grad_norm": 2.5013318061828613,
"learning_rate": 7.392356516644614e-06,
"loss": 0.4162518382072449,
"loss_sft": 0.4162518382072449,
"step": 315,
"token_acc": 0.8604568165596003
},
{
"epoch": 1.1208888888888888,
"grad_norm": 2.824483871459961,
"learning_rate": 7.375225219983876e-06,
"loss": 0.437302827835083,
"loss_sft": 0.437302827835083,
"step": 316,
"token_acc": 0.8522549019607844
},
{
"epoch": 1.1244444444444444,
"grad_norm": 2.486527681350708,
"learning_rate": 7.358057837725039e-06,
"loss": 0.40140900015830994,
"loss_sft": 0.40140900015830994,
"step": 317,
"token_acc": 0.8607391537225495
},
{
"epoch": 1.1280000000000001,
"grad_norm": 2.7907731533050537,
"learning_rate": 7.340854630683481e-06,
"loss": 0.4416828155517578,
"loss_sft": 0.4416828155517578,
"step": 318,
"token_acc": 0.8543459174714662
},
{
"epoch": 1.1315555555555556,
"grad_norm": 2.831188440322876,
"learning_rate": 7.323615860218844e-06,
"loss": 0.42795419692993164,
"loss_sft": 0.42795419692993164,
"step": 319,
"token_acc": 0.8543726600748776
},
{
"epoch": 1.1351111111111112,
"grad_norm": 2.6764321327209473,
"learning_rate": 7.306341788231067e-06,
"loss": 0.38172268867492676,
"loss_sft": 0.38172268867492676,
"step": 320,
"token_acc": 0.8672308672308673
},
{
"epoch": 1.1386666666666667,
"grad_norm": 2.5762619972229004,
"learning_rate": 7.289032677156408e-06,
"loss": 0.42674219608306885,
"loss_sft": 0.42674219608306885,
"step": 321,
"token_acc": 0.8558388157894737
},
{
"epoch": 1.1422222222222222,
"grad_norm": 2.4600794315338135,
"learning_rate": 7.271688789963458e-06,
"loss": 0.39713525772094727,
"loss_sft": 0.39713525772094727,
"step": 322,
"token_acc": 0.863001010441226
},
{
"epoch": 1.1457777777777778,
"grad_norm": 2.352123737335205,
"learning_rate": 7.254310390149144e-06,
"loss": 0.4158175587654114,
"loss_sft": 0.4158175587654114,
"step": 323,
"token_acc": 0.8563981042654029
},
{
"epoch": 1.1493333333333333,
"grad_norm": 2.269639730453491,
"learning_rate": 7.236897741734721e-06,
"loss": 0.4020949602127075,
"loss_sft": 0.4020949602127075,
"step": 324,
"token_acc": 0.8586649952992792
},
{
"epoch": 1.1528888888888889,
"grad_norm": 2.375946044921875,
"learning_rate": 7.219451109261768e-06,
"loss": 0.3985356092453003,
"loss_sft": 0.3985356092453003,
"step": 325,
"token_acc": 0.8656716417910447
},
{
"epoch": 1.1564444444444444,
"grad_norm": 2.544346809387207,
"learning_rate": 7.201970757788172e-06,
"loss": 0.39826709032058716,
"loss_sft": 0.39826709032058716,
"step": 326,
"token_acc": 0.8638073065902578
},
{
"epoch": 1.16,
"grad_norm": 2.451446533203125,
"learning_rate": 7.184456952884089e-06,
"loss": 0.4372175335884094,
"loss_sft": 0.4372175335884094,
"step": 327,
"token_acc": 0.8512707362915747
},
{
"epoch": 1.1635555555555555,
"grad_norm": 2.293242931365967,
"learning_rate": 7.166909960627918e-06,
"loss": 0.41399064660072327,
"loss_sft": 0.41399064660072327,
"step": 328,
"token_acc": 0.8613854300844426
},
{
"epoch": 1.1671111111111112,
"grad_norm": 2.7794456481933594,
"learning_rate": 7.149330047602258e-06,
"loss": 0.40988582372665405,
"loss_sft": 0.40988582372665405,
"step": 329,
"token_acc": 0.8606184418237534
},
{
"epoch": 1.1706666666666667,
"grad_norm": 2.4466676712036133,
"learning_rate": 7.131717480889854e-06,
"loss": 0.4078211486339569,
"loss_sft": 0.4078211486339569,
"step": 330,
"token_acc": 0.8614050441562574
},
{
"epoch": 1.1742222222222223,
"grad_norm": 2.720423698425293,
"learning_rate": 7.114072528069549e-06,
"loss": 0.4335750937461853,
"loss_sft": 0.4335750937461853,
"step": 331,
"token_acc": 0.8535999313481507
},
{
"epoch": 1.1777777777777778,
"grad_norm": 2.7224743366241455,
"learning_rate": 7.0963954572122e-06,
"loss": 0.3928089439868927,
"loss_sft": 0.3928089439868927,
"step": 332,
"token_acc": 0.8700970054829186
},
{
"epoch": 1.1813333333333333,
"grad_norm": 2.6546199321746826,
"learning_rate": 7.078686536876627e-06,
"loss": 0.429890513420105,
"loss_sft": 0.429890513420105,
"step": 333,
"token_acc": 0.8559940153357023
},
{
"epoch": 1.1848888888888889,
"grad_norm": 2.4845995903015137,
"learning_rate": 7.06094603610552e-06,
"loss": 0.39406970143318176,
"loss_sft": 0.39406970143318176,
"step": 334,
"token_acc": 0.8621738028882696
},
{
"epoch": 1.1884444444444444,
"grad_norm": 2.5234382152557373,
"learning_rate": 7.043174224421353e-06,
"loss": 0.40051430463790894,
"loss_sft": 0.40051430463790894,
"step": 335,
"token_acc": 0.8642507670619455
},
{
"epoch": 1.192,
"grad_norm": 2.4492392539978027,
"learning_rate": 7.025371371822294e-06,
"loss": 0.38803717494010925,
"loss_sft": 0.38803717494010925,
"step": 336,
"token_acc": 0.8675429726996967
},
{
"epoch": 1.1955555555555555,
"grad_norm": 2.594224452972412,
"learning_rate": 7.007537748778097e-06,
"loss": 0.38358014822006226,
"loss_sft": 0.38358014822006226,
"step": 337,
"token_acc": 0.8702528556087084
},
{
"epoch": 1.199111111111111,
"grad_norm": 2.5753533840179443,
"learning_rate": 6.989673626225997e-06,
"loss": 0.45417171716690063,
"loss_sft": 0.45417171716690063,
"step": 338,
"token_acc": 0.8456000677564157
},
{
"epoch": 1.2026666666666666,
"grad_norm": 2.4565317630767822,
"learning_rate": 6.971779275566593e-06,
"loss": 0.3975052237510681,
"loss_sft": 0.3975052237510681,
"step": 339,
"token_acc": 0.8638187817711585
},
{
"epoch": 1.2062222222222223,
"grad_norm": 2.487441062927246,
"learning_rate": 6.953854968659726e-06,
"loss": 0.45879364013671875,
"loss_sft": 0.45879364013671875,
"step": 340,
"token_acc": 0.8462478051759676
},
{
"epoch": 1.2097777777777778,
"grad_norm": 2.475520372390747,
"learning_rate": 6.935900977820341e-06,
"loss": 0.4210956394672394,
"loss_sft": 0.4210956394672394,
"step": 341,
"token_acc": 0.8557509505703422
},
{
"epoch": 1.2133333333333334,
"grad_norm": 2.5081546306610107,
"learning_rate": 6.917917575814364e-06,
"loss": 0.4148748815059662,
"loss_sft": 0.4148748815059662,
"step": 342,
"token_acc": 0.8579606440071557
},
{
"epoch": 1.216888888888889,
"grad_norm": 2.473416805267334,
"learning_rate": 6.899905035854544e-06,
"loss": 0.4137045741081238,
"loss_sft": 0.4137045741081238,
"step": 343,
"token_acc": 0.859068819564195
},
{
"epoch": 1.2204444444444444,
"grad_norm": 2.4610366821289062,
"learning_rate": 6.881863631596313e-06,
"loss": 0.42913228273391724,
"loss_sft": 0.42913228273391724,
"step": 344,
"token_acc": 0.8575004315553254
},
{
"epoch": 1.224,
"grad_norm": 2.566910982131958,
"learning_rate": 6.863793637133618e-06,
"loss": 0.39146167039871216,
"loss_sft": 0.39146167039871216,
"step": 345,
"token_acc": 0.8638019617001401
},
{
"epoch": 1.2275555555555555,
"grad_norm": 2.604145050048828,
"learning_rate": 6.845695326994768e-06,
"loss": 0.404320627450943,
"loss_sft": 0.404320627450943,
"step": 346,
"token_acc": 0.8598987430167597
},
{
"epoch": 1.231111111111111,
"grad_norm": 2.621833086013794,
"learning_rate": 6.827568976138255e-06,
"loss": 0.4568946361541748,
"loss_sft": 0.4568946361541748,
"step": 347,
"token_acc": 0.8485004726304031
},
{
"epoch": 1.2346666666666666,
"grad_norm": 2.6411187648773193,
"learning_rate": 6.809414859948579e-06,
"loss": 0.4159761071205139,
"loss_sft": 0.4159761071205139,
"step": 348,
"token_acc": 0.8545264259549974
},
{
"epoch": 1.2382222222222223,
"grad_norm": 2.5437264442443848,
"learning_rate": 6.791233254232066e-06,
"loss": 0.4021328091621399,
"loss_sft": 0.4021328091621399,
"step": 349,
"token_acc": 0.8650066696309471
},
{
"epoch": 1.2417777777777779,
"grad_norm": 2.5929229259490967,
"learning_rate": 6.773024435212678e-06,
"loss": 0.41667407751083374,
"loss_sft": 0.41667407751083374,
"step": 350,
"token_acc": 0.8595477801617428
},
{
"epoch": 1.2453333333333334,
"grad_norm": 2.8552002906799316,
"learning_rate": 6.7547886795278136e-06,
"loss": 0.42060765624046326,
"loss_sft": 0.42060765624046326,
"step": 351,
"token_acc": 0.8600875787674891
},
{
"epoch": 1.248888888888889,
"grad_norm": 2.6629345417022705,
"learning_rate": 6.736526264224101e-06,
"loss": 0.40468209981918335,
"loss_sft": 0.40468209981918335,
"step": 352,
"token_acc": 0.8591769404410488
},
{
"epoch": 1.2524444444444445,
"grad_norm": 2.8073620796203613,
"learning_rate": 6.718237466753206e-06,
"loss": 0.4348441958427429,
"loss_sft": 0.4348441958427429,
"step": 353,
"token_acc": 0.8536008327813003
},
{
"epoch": 1.256,
"grad_norm": 2.5660197734832764,
"learning_rate": 6.6999225649675955e-06,
"loss": 0.3909097909927368,
"loss_sft": 0.3909097909927368,
"step": 354,
"token_acc": 0.8673011805619109
},
{
"epoch": 1.2595555555555555,
"grad_norm": 2.5229249000549316,
"learning_rate": 6.681581837116331e-06,
"loss": 0.3802889585494995,
"loss_sft": 0.3802889585494995,
"step": 355,
"token_acc": 0.869546621043627
},
{
"epoch": 1.263111111111111,
"grad_norm": 2.6999902725219727,
"learning_rate": 6.6632155618408335e-06,
"loss": 0.39079833030700684,
"loss_sft": 0.39079833030700684,
"step": 356,
"token_acc": 0.8669005164182013
},
{
"epoch": 1.2666666666666666,
"grad_norm": 2.5509564876556396,
"learning_rate": 6.644824018170655e-06,
"loss": 0.38781172037124634,
"loss_sft": 0.38781172037124634,
"step": 357,
"token_acc": 0.8666841736694678
},
{
"epoch": 1.2702222222222221,
"grad_norm": 2.4422037601470947,
"learning_rate": 6.6264074855192385e-06,
"loss": 0.40941157937049866,
"loss_sft": 0.40941157937049866,
"step": 358,
"token_acc": 0.8573615718295483
},
{
"epoch": 1.2737777777777777,
"grad_norm": 2.634253740310669,
"learning_rate": 6.607966243679669e-06,
"loss": 0.3848085105419159,
"loss_sft": 0.3848085105419159,
"step": 359,
"token_acc": 0.8672615727830452
},
{
"epoch": 1.2773333333333334,
"grad_norm": 2.745076894760132,
"learning_rate": 6.589500572820428e-06,
"loss": 0.41631653904914856,
"loss_sft": 0.41631653904914856,
"step": 360,
"token_acc": 0.8584496870486278
},
{
"epoch": 1.280888888888889,
"grad_norm": 2.706134557723999,
"learning_rate": 6.571010753481135e-06,
"loss": 0.4378548562526703,
"loss_sft": 0.4378548562526703,
"step": 361,
"token_acc": 0.849137588426993
},
{
"epoch": 1.2844444444444445,
"grad_norm": 2.681750535964966,
"learning_rate": 6.552497066568282e-06,
"loss": 0.4098965525627136,
"loss_sft": 0.4098965525627136,
"step": 362,
"token_acc": 0.8608999563128004
},
{
"epoch": 1.288,
"grad_norm": 2.974034070968628,
"learning_rate": 6.533959793350974e-06,
"loss": 0.3880460262298584,
"loss_sft": 0.3880460262298584,
"step": 363,
"token_acc": 0.8647777883066426
},
{
"epoch": 1.2915555555555556,
"grad_norm": 2.6866214275360107,
"learning_rate": 6.5153992154566445e-06,
"loss": 0.4445422291755676,
"loss_sft": 0.4445422291755676,
"step": 364,
"token_acc": 0.8493623804463337
},
{
"epoch": 1.295111111111111,
"grad_norm": 2.4886550903320312,
"learning_rate": 6.496815614866792e-06,
"loss": 0.4080396294593811,
"loss_sft": 0.4080396294593811,
"step": 365,
"token_acc": 0.8576906689390712
},
{
"epoch": 1.2986666666666666,
"grad_norm": 2.5003809928894043,
"learning_rate": 6.478209273912675e-06,
"loss": 0.4097267687320709,
"loss_sft": 0.4097267687320709,
"step": 366,
"token_acc": 0.8613749237605646
},
{
"epoch": 1.3022222222222222,
"grad_norm": 2.8297600746154785,
"learning_rate": 6.4595804752710475e-06,
"loss": 0.45503711700439453,
"loss_sft": 0.45503711700439453,
"step": 367,
"token_acc": 0.8461068968523847
},
{
"epoch": 1.3057777777777777,
"grad_norm": 2.4269049167633057,
"learning_rate": 6.440929501959844e-06,
"loss": 0.37815189361572266,
"loss_sft": 0.37815189361572266,
"step": 368,
"token_acc": 0.8695132365499573
},
{
"epoch": 1.3093333333333335,
"grad_norm": 2.3424901962280273,
"learning_rate": 6.42225663733389e-06,
"loss": 0.39403092861175537,
"loss_sft": 0.39403092861175537,
"step": 369,
"token_acc": 0.8659145850120871
},
{
"epoch": 1.3128888888888888,
"grad_norm": 2.885368585586548,
"learning_rate": 6.403562165080594e-06,
"loss": 0.46394628286361694,
"loss_sft": 0.46394628286361694,
"step": 370,
"token_acc": 0.8453311394487865
},
{
"epoch": 1.3164444444444445,
"grad_norm": 2.401287078857422,
"learning_rate": 6.3848463692156396e-06,
"loss": 0.4130973815917969,
"loss_sft": 0.4130973815917969,
"step": 371,
"token_acc": 0.8583201267828843
},
{
"epoch": 1.32,
"grad_norm": 2.3515565395355225,
"learning_rate": 6.366109534078667e-06,
"loss": 0.4309059977531433,
"loss_sft": 0.4309059977531433,
"step": 372,
"token_acc": 0.8521824646342386
},
{
"epoch": 1.3235555555555556,
"grad_norm": 2.8208131790161133,
"learning_rate": 6.347351944328958e-06,
"loss": 0.4134628176689148,
"loss_sft": 0.4134628176689148,
"step": 373,
"token_acc": 0.8594173074840247
},
{
"epoch": 1.3271111111111111,
"grad_norm": 2.6971688270568848,
"learning_rate": 6.328573884941107e-06,
"loss": 0.3830999732017517,
"loss_sft": 0.3830999732017517,
"step": 374,
"token_acc": 0.8690968548619794
},
{
"epoch": 1.3306666666666667,
"grad_norm": 2.6983463764190674,
"learning_rate": 6.309775641200695e-06,
"loss": 0.42556577920913696,
"loss_sft": 0.42556577920913696,
"step": 375,
"token_acc": 0.8545207956600361
},
{
"epoch": 1.3342222222222222,
"grad_norm": 2.6275575160980225,
"learning_rate": 6.290957498699949e-06,
"loss": 0.4060342013835907,
"loss_sft": 0.4060342013835907,
"step": 376,
"token_acc": 0.8579887218045112
},
{
"epoch": 1.3377777777777777,
"grad_norm": 2.5922141075134277,
"learning_rate": 6.272119743333416e-06,
"loss": 0.4227868318557739,
"loss_sft": 0.4227868318557739,
"step": 377,
"token_acc": 0.8538151922582855
},
{
"epoch": 1.3413333333333333,
"grad_norm": 2.8110711574554443,
"learning_rate": 6.2532626612936035e-06,
"loss": 0.42198455333709717,
"loss_sft": 0.42198455333709717,
"step": 378,
"token_acc": 0.8573806881243063
},
{
"epoch": 1.3448888888888888,
"grad_norm": 2.7193284034729004,
"learning_rate": 6.234386539066643e-06,
"loss": 0.4319349527359009,
"loss_sft": 0.4319349527359009,
"step": 379,
"token_acc": 0.8550699146952775
},
{
"epoch": 1.3484444444444446,
"grad_norm": 2.530618190765381,
"learning_rate": 6.215491663427936e-06,
"loss": 0.41811901330947876,
"loss_sft": 0.41811901330947876,
"step": 380,
"token_acc": 0.8571189979123174
},
{
"epoch": 1.3519999999999999,
"grad_norm": 2.4920263290405273,
"learning_rate": 6.1965783214377895e-06,
"loss": 0.4314779043197632,
"loss_sft": 0.4314779043197632,
"step": 381,
"token_acc": 0.8536258583151901
},
{
"epoch": 1.3555555555555556,
"grad_norm": 2.3582730293273926,
"learning_rate": 6.177646800437066e-06,
"loss": 0.3940085172653198,
"loss_sft": 0.3940085172653198,
"step": 382,
"token_acc": 0.8635511214462531
},
{
"epoch": 1.3591111111111112,
"grad_norm": 2.548499822616577,
"learning_rate": 6.158697388042807e-06,
"loss": 0.45402997732162476,
"loss_sft": 0.45402997732162476,
"step": 383,
"token_acc": 0.8483354880413704
},
{
"epoch": 1.3626666666666667,
"grad_norm": 2.5590786933898926,
"learning_rate": 6.139730372143877e-06,
"loss": 0.4340314269065857,
"loss_sft": 0.4340314269065857,
"step": 384,
"token_acc": 0.8588402644130199
},
{
"epoch": 1.3662222222222222,
"grad_norm": 2.5281460285186768,
"learning_rate": 6.120746040896572e-06,
"loss": 0.4090326428413391,
"loss_sft": 0.4090326428413391,
"step": 385,
"token_acc": 0.8633485754484699
},
{
"epoch": 1.3697777777777778,
"grad_norm": 2.380821943283081,
"learning_rate": 6.10174468272026e-06,
"loss": 0.4123620390892029,
"loss_sft": 0.4123620390892029,
"step": 386,
"token_acc": 0.8590783890419117
},
{
"epoch": 1.3733333333333333,
"grad_norm": 2.424027681350708,
"learning_rate": 6.082726586292982e-06,
"loss": 0.4179833233356476,
"loss_sft": 0.4179833233356476,
"step": 387,
"token_acc": 0.8558508965083359
},
{
"epoch": 1.3768888888888888,
"grad_norm": 2.425776243209839,
"learning_rate": 6.063692040547083e-06,
"loss": 0.40507400035858154,
"loss_sft": 0.40507400035858154,
"step": 388,
"token_acc": 0.8637184892024434
},
{
"epoch": 1.3804444444444444,
"grad_norm": 2.8424465656280518,
"learning_rate": 6.044641334664812e-06,
"loss": 0.43018680810928345,
"loss_sft": 0.43018680810928345,
"step": 389,
"token_acc": 0.8571428571428571
},
{
"epoch": 1.384,
"grad_norm": 2.5185177326202393,
"learning_rate": 6.025574758073925e-06,
"loss": 0.42572325468063354,
"loss_sft": 0.42572325468063354,
"step": 390,
"token_acc": 0.8539080258786501
},
{
"epoch": 1.3875555555555557,
"grad_norm": 2.6601829528808594,
"learning_rate": 6.006492600443301e-06,
"loss": 0.40511244535446167,
"loss_sft": 0.40511244535446167,
"step": 391,
"token_acc": 0.8596623597095162
},
{
"epoch": 1.3911111111111112,
"grad_norm": 2.5098254680633545,
"learning_rate": 5.98739515167853e-06,
"loss": 0.4089931547641754,
"loss_sft": 0.4089931547641754,
"step": 392,
"token_acc": 0.8614408176722717
},
{
"epoch": 1.3946666666666667,
"grad_norm": 2.4383413791656494,
"learning_rate": 5.968282701917514e-06,
"loss": 0.38538992404937744,
"loss_sft": 0.38538992404937744,
"step": 393,
"token_acc": 0.8671276241252915
},
{
"epoch": 1.3982222222222223,
"grad_norm": 2.651482105255127,
"learning_rate": 5.949155541526057e-06,
"loss": 0.4013828933238983,
"loss_sft": 0.4013828933238983,
"step": 394,
"token_acc": 0.8612193056731583
},
{
"epoch": 1.4017777777777778,
"grad_norm": 2.64512300491333,
"learning_rate": 5.930013961093454e-06,
"loss": 0.4350872337818146,
"loss_sft": 0.4350872337818146,
"step": 395,
"token_acc": 0.8537820898085344
},
{
"epoch": 1.4053333333333333,
"grad_norm": 2.549133062362671,
"learning_rate": 5.910858251428077e-06,
"loss": 0.4118100106716156,
"loss_sft": 0.4118100106716156,
"step": 396,
"token_acc": 0.8594501718213059
},
{
"epoch": 1.4088888888888889,
"grad_norm": 2.5675790309906006,
"learning_rate": 5.891688703552953e-06,
"loss": 0.4182038903236389,
"loss_sft": 0.4182038903236389,
"step": 397,
"token_acc": 0.8572752548656163
},
{
"epoch": 1.4124444444444444,
"grad_norm": 2.535046100616455,
"learning_rate": 5.872505608701354e-06,
"loss": 0.42980432510375977,
"loss_sft": 0.42980432510375977,
"step": 398,
"token_acc": 0.8507703476647576
},
{
"epoch": 1.416,
"grad_norm": 2.4890763759613037,
"learning_rate": 5.853309258312356e-06,
"loss": 0.4255669116973877,
"loss_sft": 0.4255669116973877,
"step": 399,
"token_acc": 0.856291284034155
},
{
"epoch": 1.4195555555555557,
"grad_norm": 2.818023443222046,
"learning_rate": 5.834099944026422e-06,
"loss": 0.39693740010261536,
"loss_sft": 0.39693740010261536,
"step": 400,
"token_acc": 0.8633419023136247
},
{
"epoch": 1.423111111111111,
"grad_norm": 2.809352159500122,
"learning_rate": 5.814877957680973e-06,
"loss": 0.43299099802970886,
"loss_sft": 0.43299099802970886,
"step": 401,
"token_acc": 0.8575906383867934
},
{
"epoch": 1.4266666666666667,
"grad_norm": 2.6027867794036865,
"learning_rate": 5.795643591305945e-06,
"loss": 0.38124674558639526,
"loss_sft": 0.38124674558639526,
"step": 402,
"token_acc": 0.8678817402990089
},
{
"epoch": 1.4302222222222223,
"grad_norm": 2.6190919876098633,
"learning_rate": 5.776397137119362e-06,
"loss": 0.4276387095451355,
"loss_sft": 0.4276387095451355,
"step": 403,
"token_acc": 0.8545535317636606
},
{
"epoch": 1.4337777777777778,
"grad_norm": 2.648637056350708,
"learning_rate": 5.757138887522884e-06,
"loss": 0.435871958732605,
"loss_sft": 0.435871958732605,
"step": 404,
"token_acc": 0.8525325739925792
},
{
"epoch": 1.4373333333333334,
"grad_norm": 2.671625852584839,
"learning_rate": 5.7378691350973835e-06,
"loss": 0.43491631746292114,
"loss_sft": 0.43491631746292114,
"step": 405,
"token_acc": 0.8505598621877691
},
{
"epoch": 1.4408888888888889,
"grad_norm": 2.5949296951293945,
"learning_rate": 5.7185881725984835e-06,
"loss": 0.4065672755241394,
"loss_sft": 0.4065672755241394,
"step": 406,
"token_acc": 0.8619768477292965
},
{
"epoch": 1.4444444444444444,
"grad_norm": 2.3991339206695557,
"learning_rate": 5.699296292952117e-06,
"loss": 0.3946964740753174,
"loss_sft": 0.3946964740753174,
"step": 407,
"token_acc": 0.8631429525718103
},
{
"epoch": 1.448,
"grad_norm": 2.4032089710235596,
"learning_rate": 5.679993789250075e-06,
"loss": 0.3928908109664917,
"loss_sft": 0.3928908109664917,
"step": 408,
"token_acc": 0.8679259830377795
},
{
"epoch": 1.4515555555555555,
"grad_norm": 2.7371256351470947,
"learning_rate": 5.660680954745554e-06,
"loss": 0.42835211753845215,
"loss_sft": 0.42835211753845215,
"step": 409,
"token_acc": 0.8556916099773243
},
{
"epoch": 1.455111111111111,
"grad_norm": 2.531757116317749,
"learning_rate": 5.641358082848705e-06,
"loss": 0.4172714352607727,
"loss_sft": 0.4172714352607727,
"step": 410,
"token_acc": 0.8580710488566365
},
{
"epoch": 1.4586666666666668,
"grad_norm": 2.7410199642181396,
"learning_rate": 5.622025467122167e-06,
"loss": 0.4458763599395752,
"loss_sft": 0.4458763599395752,
"step": 411,
"token_acc": 0.8489394228931779
},
{
"epoch": 1.462222222222222,
"grad_norm": 2.4596145153045654,
"learning_rate": 5.6026834012766155e-06,
"loss": 0.4035305380821228,
"loss_sft": 0.4035305380821228,
"step": 412,
"token_acc": 0.8620945167133213
},
{
"epoch": 1.4657777777777778,
"grad_norm": 2.424675703048706,
"learning_rate": 5.5833321791662975e-06,
"loss": 0.42571109533309937,
"loss_sft": 0.42571109533309937,
"step": 413,
"token_acc": 0.8522911051212938
},
{
"epoch": 1.4693333333333334,
"grad_norm": 2.5882086753845215,
"learning_rate": 5.563972094784561e-06,
"loss": 0.4010128974914551,
"loss_sft": 0.4010128974914551,
"step": 414,
"token_acc": 0.8646126760563381
},
{
"epoch": 1.472888888888889,
"grad_norm": 2.6831178665161133,
"learning_rate": 5.544603442259401e-06,
"loss": 0.43187621235847473,
"loss_sft": 0.43187621235847473,
"step": 415,
"token_acc": 0.8560774901422938
},
{
"epoch": 1.4764444444444444,
"grad_norm": 2.6013405323028564,
"learning_rate": 5.525226515848979e-06,
"loss": 0.4088302254676819,
"loss_sft": 0.4088302254676819,
"step": 416,
"token_acc": 0.8620597888760756
},
{
"epoch": 1.48,
"grad_norm": 2.4723379611968994,
"learning_rate": 5.505841609937162e-06,
"loss": 0.45564430952072144,
"loss_sft": 0.45564430952072144,
"step": 417,
"token_acc": 0.848135060855909
},
{
"epoch": 1.4835555555555555,
"grad_norm": 2.615095376968384,
"learning_rate": 5.4864490190290386e-06,
"loss": 0.3805040121078491,
"loss_sft": 0.3805040121078491,
"step": 418,
"token_acc": 0.8664813486370158
},
{
"epoch": 1.487111111111111,
"grad_norm": 2.3430473804473877,
"learning_rate": 5.46704903774646e-06,
"loss": 0.38684940338134766,
"loss_sft": 0.38684940338134766,
"step": 419,
"token_acc": 0.869719419924338
},
{
"epoch": 1.4906666666666666,
"grad_norm": 2.3710415363311768,
"learning_rate": 5.447641960823549e-06,
"loss": 0.4247284531593323,
"loss_sft": 0.4247284531593323,
"step": 420,
"token_acc": 0.8568615571039079
},
{
"epoch": 1.4942222222222221,
"grad_norm": 2.661945343017578,
"learning_rate": 5.42822808310223e-06,
"loss": 0.462540864944458,
"loss_sft": 0.462540864944458,
"step": 421,
"token_acc": 0.848578545048527
},
{
"epoch": 1.4977777777777779,
"grad_norm": 2.444784641265869,
"learning_rate": 5.4088076995277564e-06,
"loss": 0.3804017901420593,
"loss_sft": 0.3804017901420593,
"step": 422,
"token_acc": 0.8689524437484546
},
{
"epoch": 1.5013333333333332,
"grad_norm": 2.4105942249298096,
"learning_rate": 5.389381105144208e-06,
"loss": 0.37182724475860596,
"loss_sft": 0.37182724475860596,
"step": 423,
"token_acc": 0.8741408934707904
},
{
"epoch": 1.504888888888889,
"grad_norm": 2.558210611343384,
"learning_rate": 5.369948595090033e-06,
"loss": 0.4048323929309845,
"loss_sft": 0.4048323929309845,
"step": 424,
"token_acc": 0.8646734854445318
},
{
"epoch": 1.5084444444444445,
"grad_norm": 2.9451427459716797,
"learning_rate": 5.350510464593548e-06,
"loss": 0.4208036959171295,
"loss_sft": 0.4208036959171295,
"step": 425,
"token_acc": 0.8575356590479464
},
{
"epoch": 1.512,
"grad_norm": 2.7623283863067627,
"learning_rate": 5.331067008968462e-06,
"loss": 0.4342357814311981,
"loss_sft": 0.4342357814311981,
"step": 426,
"token_acc": 0.8529719365522049
},
{
"epoch": 1.5155555555555555,
"grad_norm": 2.4208507537841797,
"learning_rate": 5.311618523609386e-06,
"loss": 0.3929608464241028,
"loss_sft": 0.3929608464241028,
"step": 427,
"token_acc": 0.8636699689486914
},
{
"epoch": 1.519111111111111,
"grad_norm": 2.496648073196411,
"learning_rate": 5.292165303987336e-06,
"loss": 0.41327792406082153,
"loss_sft": 0.41327792406082153,
"step": 428,
"token_acc": 0.8591252826070008
},
{
"epoch": 1.5226666666666666,
"grad_norm": 2.344733238220215,
"learning_rate": 5.27270764564527e-06,
"loss": 0.40192970633506775,
"loss_sft": 0.40192970633506775,
"step": 429,
"token_acc": 0.8588843738455855
},
{
"epoch": 1.5262222222222221,
"grad_norm": 2.4616682529449463,
"learning_rate": 5.253245844193564e-06,
"loss": 0.40249601006507874,
"loss_sft": 0.40249601006507874,
"step": 430,
"token_acc": 0.8590005099439062
},
{
"epoch": 1.529777777777778,
"grad_norm": 2.75276255607605,
"learning_rate": 5.23378019530555e-06,
"loss": 0.39988404512405396,
"loss_sft": 0.39988404512405396,
"step": 431,
"token_acc": 0.8623655913978494
},
{
"epoch": 1.5333333333333332,
"grad_norm": 2.5461366176605225,
"learning_rate": 5.214310994713008e-06,
"loss": 0.41250938177108765,
"loss_sft": 0.41250938177108765,
"step": 432,
"token_acc": 0.8600363541937159
},
{
"epoch": 1.536888888888889,
"grad_norm": 2.512455940246582,
"learning_rate": 5.194838538201676e-06,
"loss": 0.4173031449317932,
"loss_sft": 0.4173031449317932,
"step": 433,
"token_acc": 0.8548294286888615
},
{
"epoch": 1.5404444444444443,
"grad_norm": 2.4625613689422607,
"learning_rate": 5.175363121606759e-06,
"loss": 0.3964640498161316,
"loss_sft": 0.3964640498161316,
"step": 434,
"token_acc": 0.8646406348633782
},
{
"epoch": 1.544,
"grad_norm": 2.4466757774353027,
"learning_rate": 5.155885040808432e-06,
"loss": 0.4368104636669159,
"loss_sft": 0.4368104636669159,
"step": 435,
"token_acc": 0.8533825338253382
},
{
"epoch": 1.5475555555555556,
"grad_norm": 2.466524362564087,
"learning_rate": 5.1364045917273505e-06,
"loss": 0.4294828772544861,
"loss_sft": 0.4294828772544861,
"step": 436,
"token_acc": 0.8559640522875817
},
{
"epoch": 1.551111111111111,
"grad_norm": 2.567350387573242,
"learning_rate": 5.116922070320144e-06,
"loss": 0.38389402627944946,
"loss_sft": 0.38389402627944946,
"step": 437,
"token_acc": 0.8669265033407573
},
{
"epoch": 1.5546666666666666,
"grad_norm": 2.5159952640533447,
"learning_rate": 5.097437772574927e-06,
"loss": 0.4189486503601074,
"loss_sft": 0.4189486503601074,
"step": 438,
"token_acc": 0.8579682233991334
},
{
"epoch": 1.5582222222222222,
"grad_norm": 2.547276258468628,
"learning_rate": 5.077951994506805e-06,
"loss": 0.4077380299568176,
"loss_sft": 0.4077380299568176,
"step": 439,
"token_acc": 0.8617021276595744
},
{
"epoch": 1.561777777777778,
"grad_norm": 2.531675338745117,
"learning_rate": 5.058465032153368e-06,
"loss": 0.366779088973999,
"loss_sft": 0.366779088973999,
"step": 440,
"token_acc": 0.871260535333889
},
{
"epoch": 1.5653333333333332,
"grad_norm": 2.456918716430664,
"learning_rate": 5.038977181570204e-06,
"loss": 0.42157694697380066,
"loss_sft": 0.42157694697380066,
"step": 441,
"token_acc": 0.8602448602448602
},
{
"epoch": 1.568888888888889,
"grad_norm": 2.604818105697632,
"learning_rate": 5.0194887388263895e-06,
"loss": 0.43773597478866577,
"loss_sft": 0.43773597478866577,
"step": 442,
"token_acc": 0.8529053583855254
},
{
"epoch": 1.5724444444444443,
"grad_norm": 2.756160259246826,
"learning_rate": 5e-06,
"loss": 0.41446825861930847,
"loss_sft": 0.41446825861930847,
"step": 443,
"token_acc": 0.8608258011503698
},
{
"epoch": 1.576,
"grad_norm": 2.5138373374938965,
"learning_rate": 4.980511261173613e-06,
"loss": 0.4151393175125122,
"loss_sft": 0.4151393175125122,
"step": 444,
"token_acc": 0.8602368866328257
},
{
"epoch": 1.5795555555555556,
"grad_norm": 2.480863094329834,
"learning_rate": 4.961022818429798e-06,
"loss": 0.4205665588378906,
"loss_sft": 0.4205665588378906,
"step": 445,
"token_acc": 0.8562118806917871
},
{
"epoch": 1.5831111111111111,
"grad_norm": 2.442974090576172,
"learning_rate": 4.9415349678466335e-06,
"loss": 0.4180167019367218,
"loss_sft": 0.4180167019367218,
"step": 446,
"token_acc": 0.8568026670900143
},
{
"epoch": 1.5866666666666667,
"grad_norm": 2.6823389530181885,
"learning_rate": 4.922048005493196e-06,
"loss": 0.38717252016067505,
"loss_sft": 0.38717252016067505,
"step": 447,
"token_acc": 0.8674515761488796
},
{
"epoch": 1.5902222222222222,
"grad_norm": 2.6289291381835938,
"learning_rate": 4.902562227425075e-06,
"loss": 0.36047160625457764,
"loss_sft": 0.36047160625457764,
"step": 448,
"token_acc": 0.8735471873547187
},
{
"epoch": 1.5937777777777777,
"grad_norm": 2.593087673187256,
"learning_rate": 4.883077929679859e-06,
"loss": 0.440936803817749,
"loss_sft": 0.440936803817749,
"step": 449,
"token_acc": 0.851396822912591
},
{
"epoch": 1.5973333333333333,
"grad_norm": 2.720933437347412,
"learning_rate": 4.86359540827265e-06,
"loss": 0.3769437074661255,
"loss_sft": 0.3769437074661255,
"step": 450,
"token_acc": 0.8693536540580242
},
{
"epoch": 1.600888888888889,
"grad_norm": 2.5147266387939453,
"learning_rate": 4.844114959191569e-06,
"loss": 0.3978460431098938,
"loss_sft": 0.3978460431098938,
"step": 451,
"token_acc": 0.8653257180696963
},
{
"epoch": 1.6044444444444443,
"grad_norm": 2.4179396629333496,
"learning_rate": 4.824636878393243e-06,
"loss": 0.39086681604385376,
"loss_sft": 0.39086681604385376,
"step": 452,
"token_acc": 0.8661042106085463
},
{
"epoch": 1.608,
"grad_norm": 2.7889676094055176,
"learning_rate": 4.805161461798326e-06,
"loss": 0.43365710973739624,
"loss_sft": 0.43365710973739624,
"step": 453,
"token_acc": 0.8518981688253685
},
{
"epoch": 1.6115555555555554,
"grad_norm": 2.6129744052886963,
"learning_rate": 4.785689005286995e-06,
"loss": 0.4446248412132263,
"loss_sft": 0.4446248412132263,
"step": 454,
"token_acc": 0.8481868953040195
},
{
"epoch": 1.6151111111111112,
"grad_norm": 2.55051589012146,
"learning_rate": 4.766219804694451e-06,
"loss": 0.40924376249313354,
"loss_sft": 0.40924376249313354,
"step": 455,
"token_acc": 0.8595682755275285
},
{
"epoch": 1.6186666666666667,
"grad_norm": 2.6745641231536865,
"learning_rate": 4.746754155806437e-06,
"loss": 0.43278616666793823,
"loss_sft": 0.43278616666793823,
"step": 456,
"token_acc": 0.8521298882681564
},
{
"epoch": 1.6222222222222222,
"grad_norm": 2.548082113265991,
"learning_rate": 4.727292354354731e-06,
"loss": 0.39443960785865784,
"loss_sft": 0.39443960785865784,
"step": 457,
"token_acc": 0.8657091561938959
},
{
"epoch": 1.6257777777777778,
"grad_norm": 2.7090344429016113,
"learning_rate": 4.7078346960126645e-06,
"loss": 0.4248617887496948,
"loss_sft": 0.4248617887496948,
"step": 458,
"token_acc": 0.8565815324165029
},
{
"epoch": 1.6293333333333333,
"grad_norm": 2.9535295963287354,
"learning_rate": 4.688381476390617e-06,
"loss": 0.4150468111038208,
"loss_sft": 0.4150468111038208,
"step": 459,
"token_acc": 0.861900097943193
},
{
"epoch": 1.6328888888888888,
"grad_norm": 2.417861223220825,
"learning_rate": 4.668932991031538e-06,
"loss": 0.39712393283843994,
"loss_sft": 0.39712393283843994,
"step": 460,
"token_acc": 0.8630221672701581
},
{
"epoch": 1.6364444444444444,
"grad_norm": 2.5491838455200195,
"learning_rate": 4.649489535406454e-06,
"loss": 0.3735419809818268,
"loss_sft": 0.3735419809818268,
"step": 461,
"token_acc": 0.8694835680751174
},
{
"epoch": 1.6400000000000001,
"grad_norm": 2.537912368774414,
"learning_rate": 4.6300514049099694e-06,
"loss": 0.41379135847091675,
"loss_sft": 0.41379135847091675,
"step": 462,
"token_acc": 0.8609911932308755
},
{
"epoch": 1.6435555555555554,
"grad_norm": 2.6269140243530273,
"learning_rate": 4.6106188948557935e-06,
"loss": 0.4184068441390991,
"loss_sft": 0.4184068441390991,
"step": 463,
"token_acc": 0.855824133993148
},
{
"epoch": 1.6471111111111112,
"grad_norm": 2.4479029178619385,
"learning_rate": 4.591192300472247e-06,
"loss": 0.4207981824874878,
"loss_sft": 0.4207981824874878,
"step": 464,
"token_acc": 0.8615294998751768
},
{
"epoch": 1.6506666666666665,
"grad_norm": 2.5414154529571533,
"learning_rate": 4.5717719168977696e-06,
"loss": 0.43636447191238403,
"loss_sft": 0.43636447191238403,
"step": 465,
"token_acc": 0.8569161376545272
},
{
"epoch": 1.6542222222222223,
"grad_norm": 2.60005259513855,
"learning_rate": 4.552358039176453e-06,
"loss": 0.38482415676116943,
"loss_sft": 0.38482415676116943,
"step": 466,
"token_acc": 0.8657809110629068
},
{
"epoch": 1.6577777777777778,
"grad_norm": 2.4949734210968018,
"learning_rate": 4.532950962253543e-06,
"loss": 0.37001559138298035,
"loss_sft": 0.37001559138298035,
"step": 467,
"token_acc": 0.8748182984181274
},
{
"epoch": 1.6613333333333333,
"grad_norm": 2.392667293548584,
"learning_rate": 4.513550980970962e-06,
"loss": 0.38974517583847046,
"loss_sft": 0.38974517583847046,
"step": 468,
"token_acc": 0.8665720961281709
},
{
"epoch": 1.6648888888888889,
"grad_norm": 2.4590888023376465,
"learning_rate": 4.49415839006284e-06,
"loss": 0.43116486072540283,
"loss_sft": 0.43116486072540283,
"step": 469,
"token_acc": 0.8557722803679035
},
{
"epoch": 1.6684444444444444,
"grad_norm": 2.672375440597534,
"learning_rate": 4.474773484151021e-06,
"loss": 0.43402379751205444,
"loss_sft": 0.43402379751205444,
"step": 470,
"token_acc": 0.8526143509346197
},
{
"epoch": 1.6720000000000002,
"grad_norm": 2.5787241458892822,
"learning_rate": 4.4553965577406006e-06,
"loss": 0.41687723994255066,
"loss_sft": 0.41687723994255066,
"step": 471,
"token_acc": 0.8605815831987076
},
{
"epoch": 1.6755555555555555,
"grad_norm": 2.4987120628356934,
"learning_rate": 4.4360279052154406e-06,
"loss": 0.4203779101371765,
"loss_sft": 0.4203779101371765,
"step": 472,
"token_acc": 0.8560694998826015
},
{
"epoch": 1.6791111111111112,
"grad_norm": 2.4112040996551514,
"learning_rate": 4.416667820833704e-06,
"loss": 0.3573368787765503,
"loss_sft": 0.3573368787765503,
"step": 473,
"token_acc": 0.8743016759776536
},
{
"epoch": 1.6826666666666665,
"grad_norm": 2.3794875144958496,
"learning_rate": 4.397316598723385e-06,
"loss": 0.40012288093566895,
"loss_sft": 0.40012288093566895,
"step": 474,
"token_acc": 0.8595118171251452
},
{
"epoch": 1.6862222222222223,
"grad_norm": 2.5853352546691895,
"learning_rate": 4.377974532877834e-06,
"loss": 0.4105216860771179,
"loss_sft": 0.4105216860771179,
"step": 475,
"token_acc": 0.861875
},
{
"epoch": 1.6897777777777778,
"grad_norm": 2.388425588607788,
"learning_rate": 4.358641917151297e-06,
"loss": 0.4228043854236603,
"loss_sft": 0.4228043854236603,
"step": 476,
"token_acc": 0.8589354219948849
},
{
"epoch": 1.6933333333333334,
"grad_norm": 2.5801262855529785,
"learning_rate": 4.339319045254448e-06,
"loss": 0.42797642946243286,
"loss_sft": 0.42797642946243286,
"step": 477,
"token_acc": 0.8498199179160734
},
{
"epoch": 1.696888888888889,
"grad_norm": 2.3542237281799316,
"learning_rate": 4.320006210749928e-06,
"loss": 0.3988221287727356,
"loss_sft": 0.3988221287727356,
"step": 478,
"token_acc": 0.8647803538743136
},
{
"epoch": 1.7004444444444444,
"grad_norm": 2.5886404514312744,
"learning_rate": 4.3007037070478855e-06,
"loss": 0.42260903120040894,
"loss_sft": 0.42260903120040894,
"step": 479,
"token_acc": 0.8583894936927596
},
{
"epoch": 1.704,
"grad_norm": 2.749100685119629,
"learning_rate": 4.281411827401517e-06,
"loss": 0.3879903256893158,
"loss_sft": 0.3879903256893158,
"step": 480,
"token_acc": 0.8657464212678937
},
{
"epoch": 1.7075555555555555,
"grad_norm": 2.617905378341675,
"learning_rate": 4.262130864902617e-06,
"loss": 0.3977535367012024,
"loss_sft": 0.3977535367012024,
"step": 481,
"token_acc": 0.8632277210439033
},
{
"epoch": 1.7111111111111112,
"grad_norm": 2.494044303894043,
"learning_rate": 4.2428611124771184e-06,
"loss": 0.40377452969551086,
"loss_sft": 0.40377452969551086,
"step": 482,
"token_acc": 0.8611560070807358
},
{
"epoch": 1.7146666666666666,
"grad_norm": 2.3884387016296387,
"learning_rate": 4.22360286288064e-06,
"loss": 0.4131760001182556,
"loss_sft": 0.4131760001182556,
"step": 483,
"token_acc": 0.861328125
},
{
"epoch": 1.7182222222222223,
"grad_norm": 2.458975076675415,
"learning_rate": 4.2043564086940565e-06,
"loss": 0.38575589656829834,
"loss_sft": 0.38575589656829834,
"step": 484,
"token_acc": 0.8655936991695916
},
{
"epoch": 1.7217777777777776,
"grad_norm": 2.457685947418213,
"learning_rate": 4.185122042319027e-06,
"loss": 0.46455228328704834,
"loss_sft": 0.46455228328704834,
"step": 485,
"token_acc": 0.8457469318662717
},
{
"epoch": 1.7253333333333334,
"grad_norm": 2.3860831260681152,
"learning_rate": 4.165900055973579e-06,
"loss": 0.34921085834503174,
"loss_sft": 0.34921085834503174,
"step": 486,
"token_acc": 0.8770872362955119
},
{
"epoch": 1.728888888888889,
"grad_norm": 2.5879619121551514,
"learning_rate": 4.1466907416876466e-06,
"loss": 0.4380263686180115,
"loss_sft": 0.4380263686180115,
"step": 487,
"token_acc": 0.8497033618984838
},
{
"epoch": 1.7324444444444445,
"grad_norm": 2.5727882385253906,
"learning_rate": 4.127494391298647e-06,
"loss": 0.37577131390571594,
"loss_sft": 0.37577131390571594,
"step": 488,
"token_acc": 0.8680820177826166
},
{
"epoch": 1.736,
"grad_norm": 2.5273501873016357,
"learning_rate": 4.108311296447048e-06,
"loss": 0.3887605667114258,
"loss_sft": 0.3887605667114258,
"step": 489,
"token_acc": 0.865
},
{
"epoch": 1.7395555555555555,
"grad_norm": 2.6509921550750732,
"learning_rate": 4.089141748571926e-06,
"loss": 0.4401734173297882,
"loss_sft": 0.4401734173297882,
"step": 490,
"token_acc": 0.8529917328313007
},
{
"epoch": 1.743111111111111,
"grad_norm": 2.369041919708252,
"learning_rate": 4.069986038906547e-06,
"loss": 0.35022294521331787,
"loss_sft": 0.35022294521331787,
"step": 491,
"token_acc": 0.8783783783783784
},
{
"epoch": 1.7466666666666666,
"grad_norm": 2.507014274597168,
"learning_rate": 4.050844458473945e-06,
"loss": 0.40764015913009644,
"loss_sft": 0.40764015913009644,
"step": 492,
"token_acc": 0.8578667124699416
},
{
"epoch": 1.7502222222222223,
"grad_norm": 2.4932072162628174,
"learning_rate": 4.031717298082487e-06,
"loss": 0.40207499265670776,
"loss_sft": 0.40207499265670776,
"step": 493,
"token_acc": 0.857003286259881
},
{
"epoch": 1.7537777777777777,
"grad_norm": 2.454225540161133,
"learning_rate": 4.012604848321471e-06,
"loss": 0.42095011472702026,
"loss_sft": 0.42095011472702026,
"step": 494,
"token_acc": 0.8550191946418362
},
{
"epoch": 1.7573333333333334,
"grad_norm": 2.616231918334961,
"learning_rate": 3.993507399556699e-06,
"loss": 0.3955419063568115,
"loss_sft": 0.3955419063568115,
"step": 495,
"token_acc": 0.863795825293808
},
{
"epoch": 1.7608888888888887,
"grad_norm": 2.4657962322235107,
"learning_rate": 3.974425241926076e-06,
"loss": 0.3853417634963989,
"loss_sft": 0.3853417634963989,
"step": 496,
"token_acc": 0.8658023236235056
},
{
"epoch": 1.7644444444444445,
"grad_norm": 2.6933951377868652,
"learning_rate": 3.95535866533519e-06,
"loss": 0.4267171621322632,
"loss_sft": 0.4267171621322632,
"step": 497,
"token_acc": 0.853969216771552
},
{
"epoch": 1.768,
"grad_norm": 2.646475315093994,
"learning_rate": 3.936307959452917e-06,
"loss": 0.3965246081352234,
"loss_sft": 0.3965246081352234,
"step": 498,
"token_acc": 0.869446218410729
},
{
"epoch": 1.7715555555555556,
"grad_norm": 2.550901412963867,
"learning_rate": 3.91727341370702e-06,
"loss": 0.4151879549026489,
"loss_sft": 0.4151879549026489,
"step": 499,
"token_acc": 0.8619210977701544
},
{
"epoch": 1.775111111111111,
"grad_norm": 2.4510843753814697,
"learning_rate": 3.898255317279744e-06,
"loss": 0.38885295391082764,
"loss_sft": 0.38885295391082764,
"step": 500,
"token_acc": 0.8655998738767144
},
{
"epoch": 1.7786666666666666,
"grad_norm": 2.737840414047241,
"learning_rate": 3.879253959103429e-06,
"loss": 0.4123331904411316,
"loss_sft": 0.4123331904411316,
"step": 501,
"token_acc": 0.8572189191587541
},
{
"epoch": 1.7822222222222224,
"grad_norm": 2.5538578033447266,
"learning_rate": 3.860269627856126e-06,
"loss": 0.4307522177696228,
"loss_sft": 0.4307522177696228,
"step": 502,
"token_acc": 0.8538455394936507
},
{
"epoch": 1.7857777777777777,
"grad_norm": 2.624786376953125,
"learning_rate": 3.841302611957193e-06,
"loss": 0.40709376335144043,
"loss_sft": 0.40709376335144043,
"step": 503,
"token_acc": 0.8608233962921055
},
{
"epoch": 1.7893333333333334,
"grad_norm": 2.4077212810516357,
"learning_rate": 3.822353199562936e-06,
"loss": 0.3671134114265442,
"loss_sft": 0.3671134114265442,
"step": 504,
"token_acc": 0.8733694408421695
},
{
"epoch": 1.7928888888888888,
"grad_norm": 2.671459913253784,
"learning_rate": 3.803421678562213e-06,
"loss": 0.41062527894973755,
"loss_sft": 0.41062527894973755,
"step": 505,
"token_acc": 0.8615925862364854
},
{
"epoch": 1.7964444444444445,
"grad_norm": 2.5792226791381836,
"learning_rate": 3.784508336572066e-06,
"loss": 0.4470058083534241,
"loss_sft": 0.4470058083534241,
"step": 506,
"token_acc": 0.8527843844829001
},
{
"epoch": 1.8,
"grad_norm": 2.5291571617126465,
"learning_rate": 3.7656134609333576e-06,
"loss": 0.4011933207511902,
"loss_sft": 0.4011933207511902,
"step": 507,
"token_acc": 0.8615126050420168
},
{
"epoch": 1.8035555555555556,
"grad_norm": 2.8376951217651367,
"learning_rate": 3.7467373387063973e-06,
"loss": 0.4378824830055237,
"loss_sft": 0.4378824830055237,
"step": 508,
"token_acc": 0.851415313225058
},
{
"epoch": 1.8071111111111111,
"grad_norm": 2.7484447956085205,
"learning_rate": 3.727880256666586e-06,
"loss": 0.4337455630302429,
"loss_sft": 0.4337455630302429,
"step": 509,
"token_acc": 0.8563588028470972
},
{
"epoch": 1.8106666666666666,
"grad_norm": 2.6020312309265137,
"learning_rate": 3.709042501300052e-06,
"loss": 0.414725124835968,
"loss_sft": 0.414725124835968,
"step": 510,
"token_acc": 0.8600836545558928
},
{
"epoch": 1.8142222222222222,
"grad_norm": 2.5890231132507324,
"learning_rate": 3.6902243587993068e-06,
"loss": 0.3962720036506653,
"loss_sft": 0.3962720036506653,
"step": 511,
"token_acc": 0.8600308516684257
},
{
"epoch": 1.8177777777777777,
"grad_norm": 2.611687421798706,
"learning_rate": 3.6714261150588947e-06,
"loss": 0.4270366132259369,
"loss_sft": 0.4270366132259369,
"step": 512,
"token_acc": 0.8564540686855016
},
{
"epoch": 1.8213333333333335,
"grad_norm": 2.5532143115997314,
"learning_rate": 3.652648055671043e-06,
"loss": 0.419534832239151,
"loss_sft": 0.419534832239151,
"step": 513,
"token_acc": 0.8571548397919813
},
{
"epoch": 1.8248888888888888,
"grad_norm": 2.776819944381714,
"learning_rate": 3.6338904659213335e-06,
"loss": 0.4017767310142517,
"loss_sft": 0.4017767310142517,
"step": 514,
"token_acc": 0.8612167300380228
},
{
"epoch": 1.8284444444444445,
"grad_norm": 2.4380135536193848,
"learning_rate": 3.6151536307843625e-06,
"loss": 0.41459983587265015,
"loss_sft": 0.41459983587265015,
"step": 515,
"token_acc": 0.8595163678236553
},
{
"epoch": 1.8319999999999999,
"grad_norm": 2.491868257522583,
"learning_rate": 3.5964378349194075e-06,
"loss": 0.3903500437736511,
"loss_sft": 0.3903500437736511,
"step": 516,
"token_acc": 0.8669014084507042
},
{
"epoch": 1.8355555555555556,
"grad_norm": 2.5479695796966553,
"learning_rate": 3.577743362666112e-06,
"loss": 0.42270946502685547,
"loss_sft": 0.42270946502685547,
"step": 517,
"token_acc": 0.8562976171178474
},
{
"epoch": 1.8391111111111111,
"grad_norm": 2.8537967205047607,
"learning_rate": 3.5590704980401564e-06,
"loss": 0.42291441559791565,
"loss_sft": 0.42291441559791565,
"step": 518,
"token_acc": 0.8621459750279159
},
{
"epoch": 1.8426666666666667,
"grad_norm": 2.5061566829681396,
"learning_rate": 3.540419524728954e-06,
"loss": 0.41545170545578003,
"loss_sft": 0.41545170545578003,
"step": 519,
"token_acc": 0.8601197801296251
},
{
"epoch": 1.8462222222222222,
"grad_norm": 2.4618139266967773,
"learning_rate": 3.5217907260873265e-06,
"loss": 0.3710988163948059,
"loss_sft": 0.3710988163948059,
"step": 520,
"token_acc": 0.8672281397077037
},
{
"epoch": 1.8497777777777777,
"grad_norm": 2.471956491470337,
"learning_rate": 3.5031843851332105e-06,
"loss": 0.3849433660507202,
"loss_sft": 0.3849433660507202,
"step": 521,
"token_acc": 0.87468050499574
},
{
"epoch": 1.8533333333333335,
"grad_norm": 2.5956432819366455,
"learning_rate": 3.4846007845433568e-06,
"loss": 0.3915981650352478,
"loss_sft": 0.3915981650352478,
"step": 522,
"token_acc": 0.8684422200375369
},
{
"epoch": 1.8568888888888888,
"grad_norm": 2.5695784091949463,
"learning_rate": 3.4660402066490274e-06,
"loss": 0.3776896595954895,
"loss_sft": 0.3776896595954895,
"step": 523,
"token_acc": 0.8682107909148038
},
{
"epoch": 1.8604444444444446,
"grad_norm": 2.5689573287963867,
"learning_rate": 3.4475029334317195e-06,
"loss": 0.4091906249523163,
"loss_sft": 0.4091906249523163,
"step": 524,
"token_acc": 0.860189772441011
},
{
"epoch": 1.8639999999999999,
"grad_norm": 2.4982361793518066,
"learning_rate": 3.428989246518867e-06,
"loss": 0.39030933380126953,
"loss_sft": 0.39030933380126953,
"step": 525,
"token_acc": 0.8675128227418935
},
{
"epoch": 1.8675555555555556,
"grad_norm": 2.327930450439453,
"learning_rate": 3.410499427179572e-06,
"loss": 0.3827740550041199,
"loss_sft": 0.3827740550041199,
"step": 526,
"token_acc": 0.8652714192282538
},
{
"epoch": 1.871111111111111,
"grad_norm": 2.6272811889648438,
"learning_rate": 3.392033756320333e-06,
"loss": 0.4008473753929138,
"loss_sft": 0.4008473753929138,
"step": 527,
"token_acc": 0.8613265928702171
},
{
"epoch": 1.8746666666666667,
"grad_norm": 2.569789409637451,
"learning_rate": 3.3735925144807623e-06,
"loss": 0.40779390931129456,
"loss_sft": 0.40779390931129456,
"step": 528,
"token_acc": 0.8595163678236553
},
{
"epoch": 1.8782222222222222,
"grad_norm": 2.738184690475464,
"learning_rate": 3.355175981829346e-06,
"loss": 0.40250498056411743,
"loss_sft": 0.40250498056411743,
"step": 529,
"token_acc": 0.8613896530304375
},
{
"epoch": 1.8817777777777778,
"grad_norm": 2.5826187133789062,
"learning_rate": 3.33678443815917e-06,
"loss": 0.3996252417564392,
"loss_sft": 0.3996252417564392,
"step": 530,
"token_acc": 0.8631181966229536
},
{
"epoch": 1.8853333333333333,
"grad_norm": 2.29195237159729,
"learning_rate": 3.318418162883671e-06,
"loss": 0.3649451732635498,
"loss_sft": 0.3649451732635498,
"step": 531,
"token_acc": 0.874463995268372
},
{
"epoch": 1.8888888888888888,
"grad_norm": 2.5420398712158203,
"learning_rate": 3.300077435032406e-06,
"loss": 0.39636877179145813,
"loss_sft": 0.39636877179145813,
"step": 532,
"token_acc": 0.8676768495259704
},
{
"epoch": 1.8924444444444446,
"grad_norm": 2.588834047317505,
"learning_rate": 3.281762533246794e-06,
"loss": 0.41558629274368286,
"loss_sft": 0.41558629274368286,
"step": 533,
"token_acc": 0.8585076593641904
},
{
"epoch": 1.896,
"grad_norm": 3.0065157413482666,
"learning_rate": 3.2634737357758994e-06,
"loss": 0.4321349263191223,
"loss_sft": 0.4321349263191223,
"step": 534,
"token_acc": 0.8535313791068266
},
{
"epoch": 1.8995555555555557,
"grad_norm": 2.603196382522583,
"learning_rate": 3.245211320472189e-06,
"loss": 0.37932658195495605,
"loss_sft": 0.37932658195495605,
"step": 535,
"token_acc": 0.8697860962566845
},
{
"epoch": 1.903111111111111,
"grad_norm": 2.599184513092041,
"learning_rate": 3.226975564787322e-06,
"loss": 0.3946661353111267,
"loss_sft": 0.3946661353111267,
"step": 536,
"token_acc": 0.8630727288716912
},
{
"epoch": 1.9066666666666667,
"grad_norm": 2.8424394130706787,
"learning_rate": 3.208766745767935e-06,
"loss": 0.411368191242218,
"loss_sft": 0.411368191242218,
"step": 537,
"token_acc": 0.8622562809195848
},
{
"epoch": 1.9102222222222223,
"grad_norm": 2.7322208881378174,
"learning_rate": 3.1905851400514232e-06,
"loss": 0.40226468443870544,
"loss_sft": 0.40226468443870544,
"step": 538,
"token_acc": 0.8596026490066225
},
{
"epoch": 1.9137777777777778,
"grad_norm": 2.5031557083129883,
"learning_rate": 3.1724310238617475e-06,
"loss": 0.38684508204460144,
"loss_sft": 0.38684508204460144,
"step": 539,
"token_acc": 0.8673512687770517
},
{
"epoch": 1.9173333333333333,
"grad_norm": 2.576122999191284,
"learning_rate": 3.154304673005235e-06,
"loss": 0.3792768120765686,
"loss_sft": 0.3792768120765686,
"step": 540,
"token_acc": 0.8707482993197279
},
{
"epoch": 1.9208888888888889,
"grad_norm": 2.7705307006835938,
"learning_rate": 3.1362063628663836e-06,
"loss": 0.4563199281692505,
"loss_sft": 0.4563199281692505,
"step": 541,
"token_acc": 0.846728647145712
},
{
"epoch": 1.9244444444444444,
"grad_norm": 2.4811346530914307,
"learning_rate": 3.118136368403689e-06,
"loss": 0.39493077993392944,
"loss_sft": 0.39493077993392944,
"step": 542,
"token_acc": 0.865824211204121
},
{
"epoch": 1.928,
"grad_norm": 2.6122243404388428,
"learning_rate": 3.100094964145458e-06,
"loss": 0.3636088967323303,
"loss_sft": 0.3636088967323303,
"step": 543,
"token_acc": 0.8783938814531549
},
{
"epoch": 1.9315555555555557,
"grad_norm": 2.5847508907318115,
"learning_rate": 3.0820824241856377e-06,
"loss": 0.38992923498153687,
"loss_sft": 0.38992923498153687,
"step": 544,
"token_acc": 0.8673680424422264
},
{
"epoch": 1.935111111111111,
"grad_norm": 2.5277059078216553,
"learning_rate": 3.064099022179661e-06,
"loss": 0.38417667150497437,
"loss_sft": 0.38417667150497437,
"step": 545,
"token_acc": 0.8669001751313485
},
{
"epoch": 1.9386666666666668,
"grad_norm": 2.436102867126465,
"learning_rate": 3.046145031340275e-06,
"loss": 0.4031416177749634,
"loss_sft": 0.4031416177749634,
"step": 546,
"token_acc": 0.8660896546092921
},
{
"epoch": 1.942222222222222,
"grad_norm": 2.589829206466675,
"learning_rate": 3.0282207244334084e-06,
"loss": 0.43223583698272705,
"loss_sft": 0.43223583698272705,
"step": 547,
"token_acc": 0.8521997430956969
},
{
"epoch": 1.9457777777777778,
"grad_norm": 2.7186975479125977,
"learning_rate": 3.010326373774004e-06,
"loss": 0.40543848276138306,
"loss_sft": 0.40543848276138306,
"step": 548,
"token_acc": 0.8586903483946521
},
{
"epoch": 1.9493333333333334,
"grad_norm": 2.5322623252868652,
"learning_rate": 2.9924622512219037e-06,
"loss": 0.3990423083305359,
"loss_sft": 0.3990423083305359,
"step": 549,
"token_acc": 0.8634534786253143
},
{
"epoch": 1.952888888888889,
"grad_norm": 2.4155116081237793,
"learning_rate": 2.9746286281777075e-06,
"loss": 0.37820863723754883,
"loss_sft": 0.37820863723754883,
"step": 550,
"token_acc": 0.8695297856735392
},
{
"epoch": 1.9564444444444444,
"grad_norm": 2.7682383060455322,
"learning_rate": 2.9568257755786474e-06,
"loss": 0.41545629501342773,
"loss_sft": 0.41545629501342773,
"step": 551,
"token_acc": 0.8606777386344573
},
{
"epoch": 1.96,
"grad_norm": 2.7402219772338867,
"learning_rate": 2.939053963894481e-06,
"loss": 0.42262837290763855,
"loss_sft": 0.42262837290763855,
"step": 552,
"token_acc": 0.8544308845816384
},
{
"epoch": 1.9635555555555557,
"grad_norm": 2.637162685394287,
"learning_rate": 2.921313463123375e-06,
"loss": 0.39897477626800537,
"loss_sft": 0.39897477626800537,
"step": 553,
"token_acc": 0.8650591845681719
},
{
"epoch": 1.967111111111111,
"grad_norm": 2.5800321102142334,
"learning_rate": 2.9036045427878e-06,
"loss": 0.37637901306152344,
"loss_sft": 0.37637901306152344,
"step": 554,
"token_acc": 0.8683005912645432
},
{
"epoch": 1.9706666666666668,
"grad_norm": 2.5375280380249023,
"learning_rate": 2.8859274719304543e-06,
"loss": 0.4282844662666321,
"loss_sft": 0.4282844662666321,
"step": 555,
"token_acc": 0.8550245789346442
},
{
"epoch": 1.974222222222222,
"grad_norm": 2.6395108699798584,
"learning_rate": 2.8682825191101447e-06,
"loss": 0.3765285611152649,
"loss_sft": 0.3765285611152649,
"step": 556,
"token_acc": 0.8716730038022814
},
{
"epoch": 1.9777777777777779,
"grad_norm": 2.5628502368927,
"learning_rate": 2.850669952397743e-06,
"loss": 0.40104517340660095,
"loss_sft": 0.40104517340660095,
"step": 557,
"token_acc": 0.8631203373337658
},
{
"epoch": 1.9813333333333332,
"grad_norm": 2.6946022510528564,
"learning_rate": 2.8330900393720846e-06,
"loss": 0.44506531953811646,
"loss_sft": 0.44506531953811646,
"step": 558,
"token_acc": 0.8495327849213321
},
{
"epoch": 1.984888888888889,
"grad_norm": 2.3483667373657227,
"learning_rate": 2.8155430471159118e-06,
"loss": 0.3790060579776764,
"loss_sft": 0.3790060579776764,
"step": 559,
"token_acc": 0.8737014314928425
},
{
"epoch": 1.9884444444444445,
"grad_norm": 2.7915592193603516,
"learning_rate": 2.7980292422118282e-06,
"loss": 0.4087577760219574,
"loss_sft": 0.4087577760219574,
"step": 560,
"token_acc": 0.8632608489222297
},
{
"epoch": 1.992,
"grad_norm": 2.545546770095825,
"learning_rate": 2.7805488907382316e-06,
"loss": 0.42341354489326477,
"loss_sft": 0.42341354489326477,
"step": 561,
"token_acc": 0.8561505532671028
},
{
"epoch": 1.9955555555555555,
"grad_norm": 2.4651565551757812,
"learning_rate": 2.7631022582652808e-06,
"loss": 0.3653791844844818,
"loss_sft": 0.3653791844844818,
"step": 562,
"token_acc": 0.8774062816616008
},
{
"epoch": 1.999111111111111,
"grad_norm": 2.655350923538208,
"learning_rate": 2.745689609850859e-06,
"loss": 0.3831256628036499,
"loss_sft": 0.3831256628036499,
"step": 563,
"token_acc": 0.8677510608203678
},
{
"epoch": 2.0,
"grad_norm": 4.539825916290283,
"learning_rate": 2.728311210036542e-06,
"loss": 0.38500434160232544,
"loss_sft": 0.38500434160232544,
"step": 564,
"token_acc": 0.8604507564062982
},
{
"epoch": 2.0035555555555558,
"grad_norm": 2.1747772693634033,
"learning_rate": 2.7109673228435925e-06,
"loss": 0.2346683144569397,
"loss_sft": 0.2346683144569397,
"step": 565,
"token_acc": 0.9234685631162114
},
{
"epoch": 2.007111111111111,
"grad_norm": 2.291053533554077,
"learning_rate": 2.6936582117689347e-06,
"loss": 0.2536627948284149,
"loss_sft": 0.2536627948284149,
"step": 566,
"token_acc": 0.9201904918301995
},
{
"epoch": 2.010666666666667,
"grad_norm": 2.315946578979492,
"learning_rate": 2.6763841397811576e-06,
"loss": 0.22049154341220856,
"loss_sft": 0.22049154341220856,
"step": 567,
"token_acc": 0.9287589051368579
},
{
"epoch": 2.014222222222222,
"grad_norm": 2.2529373168945312,
"learning_rate": 2.6591453693165204e-06,
"loss": 0.23101037740707397,
"loss_sft": 0.23101037740707397,
"step": 568,
"token_acc": 0.9222426163857834
},
{
"epoch": 2.017777777777778,
"grad_norm": 2.3223443031311035,
"learning_rate": 2.641942162274962e-06,
"loss": 0.20342367887496948,
"loss_sft": 0.20342367887496948,
"step": 569,
"token_acc": 0.931980153360397
},
{
"epoch": 2.021333333333333,
"grad_norm": 2.28841495513916,
"learning_rate": 2.624774780016126e-06,
"loss": 0.18857476115226746,
"loss_sft": 0.18857476115226746,
"step": 570,
"token_acc": 0.9337654266181301
},
{
"epoch": 2.024888888888889,
"grad_norm": 2.4593770503997803,
"learning_rate": 2.607643483355387e-06,
"loss": 0.18936146795749664,
"loss_sft": 0.18936146795749664,
"step": 571,
"token_acc": 0.9346041590660343
},
{
"epoch": 2.0284444444444443,
"grad_norm": 2.7651491165161133,
"learning_rate": 2.59054853255989e-06,
"loss": 0.19923639297485352,
"loss_sft": 0.19923639297485352,
"step": 572,
"token_acc": 0.9318448883666275
},
{
"epoch": 2.032,
"grad_norm": 3.157787322998047,
"learning_rate": 2.573490187344596e-06,
"loss": 0.1960778832435608,
"loss_sft": 0.1960778832435608,
"step": 573,
"token_acc": 0.9305275537634409
},
{
"epoch": 2.0355555555555553,
"grad_norm": 3.9161415100097656,
"learning_rate": 2.5564687068683335e-06,
"loss": 0.204880952835083,
"loss_sft": 0.204880952835083,
"step": 574,
"token_acc": 0.9320308200330215
},
{
"epoch": 2.039111111111111,
"grad_norm": 4.0761942863464355,
"learning_rate": 2.539484349729868e-06,
"loss": 0.2095157504081726,
"loss_sft": 0.2095157504081726,
"step": 575,
"token_acc": 0.9317512021095083
},
{
"epoch": 2.042666666666667,
"grad_norm": 3.5518853664398193,
"learning_rate": 2.5225373739639637e-06,
"loss": 0.2153884768486023,
"loss_sft": 0.2153884768486023,
"step": 576,
"token_acc": 0.9263014398909432
},
{
"epoch": 2.046222222222222,
"grad_norm": 3.358978509902954,
"learning_rate": 2.5056280370374725e-06,
"loss": 0.22358079254627228,
"loss_sft": 0.22358079254627228,
"step": 577,
"token_acc": 0.9219449225473322
},
{
"epoch": 2.049777777777778,
"grad_norm": 3.4067559242248535,
"learning_rate": 2.488756595845417e-06,
"loss": 0.22681552171707153,
"loss_sft": 0.22681552171707153,
"step": 578,
"token_acc": 0.9220222793487575
},
{
"epoch": 2.0533333333333332,
"grad_norm": 3.2785274982452393,
"learning_rate": 2.471923306707086e-06,
"loss": 0.19896847009658813,
"loss_sft": 0.19896847009658813,
"step": 579,
"token_acc": 0.928486174656853
},
{
"epoch": 2.056888888888889,
"grad_norm": 2.7540602684020996,
"learning_rate": 2.455128425362153e-06,
"loss": 0.21246746182441711,
"loss_sft": 0.21246746182441711,
"step": 580,
"token_acc": 0.9233880376448628
},
{
"epoch": 2.0604444444444443,
"grad_norm": 2.681729555130005,
"learning_rate": 2.4383722069667683e-06,
"loss": 0.18616808950901031,
"loss_sft": 0.18616808950901031,
"step": 581,
"token_acc": 0.9335436797475843
},
{
"epoch": 2.064,
"grad_norm": 2.627020835876465,
"learning_rate": 2.4216549060897026e-06,
"loss": 0.20557956397533417,
"loss_sft": 0.20557956397533417,
"step": 582,
"token_acc": 0.9285327558202491
},
{
"epoch": 2.0675555555555554,
"grad_norm": 2.660526752471924,
"learning_rate": 2.404976776708477e-06,
"loss": 0.18170322477817535,
"loss_sft": 0.18170322477817535,
"step": 583,
"token_acc": 0.9394750937332619
},
{
"epoch": 2.071111111111111,
"grad_norm": 2.5148396492004395,
"learning_rate": 2.3883380722054865e-06,
"loss": 0.21018336713314056,
"loss_sft": 0.21018336713314056,
"step": 584,
"token_acc": 0.9279256654947262
},
{
"epoch": 2.074666666666667,
"grad_norm": 2.5572638511657715,
"learning_rate": 2.371739045364178e-06,
"loss": 0.22585470974445343,
"loss_sft": 0.22585470974445343,
"step": 585,
"token_acc": 0.9224207307258974
},
{
"epoch": 2.078222222222222,
"grad_norm": 2.5098631381988525,
"learning_rate": 2.3551799483651894e-06,
"loss": 0.17506377398967743,
"loss_sft": 0.17506377398967743,
"step": 586,
"token_acc": 0.9401693320790216
},
{
"epoch": 2.081777777777778,
"grad_norm": 2.5438649654388428,
"learning_rate": 2.3386610327825194e-06,
"loss": 0.19875019788742065,
"loss_sft": 0.19875019788742065,
"step": 587,
"token_acc": 0.9339853300733496
},
{
"epoch": 2.0853333333333333,
"grad_norm": 2.9650871753692627,
"learning_rate": 2.3221825495797213e-06,
"loss": 0.18974845111370087,
"loss_sft": 0.18974845111370087,
"step": 588,
"token_acc": 0.9348122202763184
},
{
"epoch": 2.088888888888889,
"grad_norm": 2.8284084796905518,
"learning_rate": 2.3057447491060725e-06,
"loss": 0.18660187721252441,
"loss_sft": 0.18660187721252441,
"step": 589,
"token_acc": 0.9372904752514297
},
{
"epoch": 2.0924444444444443,
"grad_norm": 2.7909176349639893,
"learning_rate": 2.28934788109278e-06,
"loss": 0.199959397315979,
"loss_sft": 0.199959397315979,
"step": 590,
"token_acc": 0.9320824782566207
},
{
"epoch": 2.096,
"grad_norm": 2.8060808181762695,
"learning_rate": 2.2729921946491855e-06,
"loss": 0.1844910979270935,
"loss_sft": 0.1844910979270935,
"step": 591,
"token_acc": 0.9359400998336106
},
{
"epoch": 2.0995555555555554,
"grad_norm": 2.581951379776001,
"learning_rate": 2.2566779382589788e-06,
"loss": 0.19697339832782745,
"loss_sft": 0.19697339832782745,
"step": 592,
"token_acc": 0.931214421252372
},
{
"epoch": 2.103111111111111,
"grad_norm": 2.7985644340515137,
"learning_rate": 2.240405359776424e-06,
"loss": 0.2013942003250122,
"loss_sft": 0.2013942003250122,
"step": 593,
"token_acc": 0.9309628645011215
},
{
"epoch": 2.1066666666666665,
"grad_norm": 2.9956350326538086,
"learning_rate": 2.2241747064225942e-06,
"loss": 0.20478421449661255,
"loss_sft": 0.20478421449661255,
"step": 594,
"token_acc": 0.9299380220964699
},
{
"epoch": 2.110222222222222,
"grad_norm": 2.737675189971924,
"learning_rate": 2.2079862247816148e-06,
"loss": 0.19903981685638428,
"loss_sft": 0.19903981685638428,
"step": 595,
"token_acc": 0.9320081897287152
},
{
"epoch": 2.113777777777778,
"grad_norm": 2.749505043029785,
"learning_rate": 2.191840160796918e-06,
"loss": 0.1944810003042221,
"loss_sft": 0.1944810003042221,
"step": 596,
"token_acc": 0.9343329057087877
},
{
"epoch": 2.1173333333333333,
"grad_norm": 2.669701337814331,
"learning_rate": 2.175736759767505e-06,
"loss": 0.2074553668498993,
"loss_sft": 0.2074553668498993,
"step": 597,
"token_acc": 0.9262592768040423
},
{
"epoch": 2.120888888888889,
"grad_norm": 2.8744184970855713,
"learning_rate": 2.159676266344222e-06,
"loss": 0.19187158346176147,
"loss_sft": 0.19187158346176147,
"step": 598,
"token_acc": 0.9342606538495615
},
{
"epoch": 2.1244444444444444,
"grad_norm": 3.8266401290893555,
"learning_rate": 2.1436589245260375e-06,
"loss": 0.1988949179649353,
"loss_sft": 0.1988949179649353,
"step": 599,
"token_acc": 0.9295262267343486
},
{
"epoch": 2.128,
"grad_norm": 2.9030916690826416,
"learning_rate": 2.127684977656349e-06,
"loss": 0.19787614047527313,
"loss_sft": 0.19787614047527313,
"step": 600,
"token_acc": 0.9304194925898016
},
{
"epoch": 2.1315555555555554,
"grad_norm": 2.7274010181427,
"learning_rate": 2.111754668419264e-06,
"loss": 0.20285405218601227,
"loss_sft": 0.20285405218601227,
"step": 601,
"token_acc": 0.9287454835728091
},
{
"epoch": 2.135111111111111,
"grad_norm": 2.7807695865631104,
"learning_rate": 2.095868238835932e-06,
"loss": 0.23012775182724,
"loss_sft": 0.23012775182724,
"step": 602,
"token_acc": 0.9183338799606429
},
{
"epoch": 2.1386666666666665,
"grad_norm": 2.499314308166504,
"learning_rate": 2.0800259302608656e-06,
"loss": 0.19873420894145966,
"loss_sft": 0.19873420894145966,
"step": 603,
"token_acc": 0.9298311444652908
},
{
"epoch": 2.1422222222222222,
"grad_norm": 2.659306287765503,
"learning_rate": 2.064227983378256e-06,
"loss": 0.1933509111404419,
"loss_sft": 0.1933509111404419,
"step": 604,
"token_acc": 0.9342578325629173
},
{
"epoch": 2.145777777777778,
"grad_norm": 2.6356680393218994,
"learning_rate": 2.048474638198347e-06,
"loss": 0.18095850944519043,
"loss_sft": 0.18095850944519043,
"step": 605,
"token_acc": 0.9373908487600419
},
{
"epoch": 2.1493333333333333,
"grad_norm": 2.6587841510772705,
"learning_rate": 2.0327661340537536e-06,
"loss": 0.20657497644424438,
"loss_sft": 0.20657497644424438,
"step": 606,
"token_acc": 0.927768233640481
},
{
"epoch": 2.152888888888889,
"grad_norm": 2.4159650802612305,
"learning_rate": 2.0171027095958543e-06,
"loss": 0.19735893607139587,
"loss_sft": 0.19735893607139587,
"step": 607,
"token_acc": 0.9310721247563353
},
{
"epoch": 2.1564444444444444,
"grad_norm": 2.6386470794677734,
"learning_rate": 2.001484602791157e-06,
"loss": 0.201791450381279,
"loss_sft": 0.201791450381279,
"step": 608,
"token_acc": 0.9330642448705012
},
{
"epoch": 2.16,
"grad_norm": 2.64615535736084,
"learning_rate": 1.9859120509176706e-06,
"loss": 0.1922120451927185,
"loss_sft": 0.1922120451927185,
"step": 609,
"token_acc": 0.931741734350644
},
{
"epoch": 2.1635555555555555,
"grad_norm": 2.568408727645874,
"learning_rate": 1.970385290561317e-06,
"loss": 0.2095409482717514,
"loss_sft": 0.2095409482717514,
"step": 610,
"token_acc": 0.9257908507095853
},
{
"epoch": 2.167111111111111,
"grad_norm": 2.612868309020996,
"learning_rate": 1.954904557612337e-06,
"loss": 0.19083017110824585,
"loss_sft": 0.19083017110824585,
"step": 611,
"token_acc": 0.9329383691269645
},
{
"epoch": 2.1706666666666665,
"grad_norm": 2.8778908252716064,
"learning_rate": 1.9394700872616856e-06,
"loss": 0.2021474540233612,
"loss_sft": 0.2021474540233612,
"step": 612,
"token_acc": 0.9299855142443264
},
{
"epoch": 2.1742222222222223,
"grad_norm": 2.961771011352539,
"learning_rate": 1.924082113997488e-06,
"loss": 0.19557005167007446,
"loss_sft": 0.19557005167007446,
"step": 613,
"token_acc": 0.9307933934542507
},
{
"epoch": 2.1777777777777776,
"grad_norm": 2.8893849849700928,
"learning_rate": 1.9087408716014562e-06,
"loss": 0.2092362344264984,
"loss_sft": 0.2092362344264984,
"step": 614,
"token_acc": 0.9264607118871726
},
{
"epoch": 2.1813333333333333,
"grad_norm": 2.4951133728027344,
"learning_rate": 1.8934465931453378e-06,
"loss": 0.19276288151741028,
"loss_sft": 0.19276288151741028,
"step": 615,
"token_acc": 0.9343396787626412
},
{
"epoch": 2.1848888888888887,
"grad_norm": 2.80733060836792,
"learning_rate": 1.8781995109873929e-06,
"loss": 0.20329850912094116,
"loss_sft": 0.20329850912094116,
"step": 616,
"token_acc": 0.9298230470435909
},
{
"epoch": 2.1884444444444444,
"grad_norm": 2.657226324081421,
"learning_rate": 1.8629998567688445e-06,
"loss": 0.1850716769695282,
"loss_sft": 0.1850716769695282,
"step": 617,
"token_acc": 0.936
},
{
"epoch": 2.192,
"grad_norm": 2.5907394886016846,
"learning_rate": 1.8478478614103684e-06,
"loss": 0.1955103576183319,
"loss_sft": 0.1955103576183319,
"step": 618,
"token_acc": 0.9318181818181818
},
{
"epoch": 2.1955555555555555,
"grad_norm": 2.5809783935546875,
"learning_rate": 1.8327437551085842e-06,
"loss": 0.1801038682460785,
"loss_sft": 0.1801038682460785,
"step": 619,
"token_acc": 0.9363152643831061
},
{
"epoch": 2.1991111111111112,
"grad_norm": 2.790290594100952,
"learning_rate": 1.8176877673325555e-06,
"loss": 0.1863715648651123,
"loss_sft": 0.1863715648651123,
"step": 620,
"token_acc": 0.937380298109751
},
{
"epoch": 2.2026666666666666,
"grad_norm": 2.7557408809661865,
"learning_rate": 1.802680126820307e-06,
"loss": 0.2073497474193573,
"loss_sft": 0.2073497474193573,
"step": 621,
"token_acc": 0.9279541446208113
},
{
"epoch": 2.2062222222222223,
"grad_norm": 2.673795461654663,
"learning_rate": 1.7877210615753477e-06,
"loss": 0.1762343943119049,
"loss_sft": 0.1762343943119049,
"step": 622,
"token_acc": 0.9372890888638921
},
{
"epoch": 2.2097777777777776,
"grad_norm": 2.782465696334839,
"learning_rate": 1.772810798863206e-06,
"loss": 0.1905660331249237,
"loss_sft": 0.1905660331249237,
"step": 623,
"token_acc": 0.9312704839020628
},
{
"epoch": 2.2133333333333334,
"grad_norm": 2.7434780597686768,
"learning_rate": 1.7579495652079786e-06,
"loss": 0.1992965042591095,
"loss_sft": 0.1992965042591095,
"step": 624,
"token_acc": 0.9305287194800632
},
{
"epoch": 2.2168888888888887,
"grad_norm": 2.718320608139038,
"learning_rate": 1.74313758638889e-06,
"loss": 0.20575201511383057,
"loss_sft": 0.20575201511383057,
"step": 625,
"token_acc": 0.9312709166598645
},
{
"epoch": 2.2204444444444444,
"grad_norm": 3.1262898445129395,
"learning_rate": 1.7283750874368577e-06,
"loss": 0.18603146076202393,
"loss_sft": 0.18603146076202393,
"step": 626,
"token_acc": 0.936985751555288
},
{
"epoch": 2.224,
"grad_norm": 2.9450271129608154,
"learning_rate": 1.713662292631078e-06,
"loss": 0.19103722274303436,
"loss_sft": 0.19103722274303436,
"step": 627,
"token_acc": 0.9349270814103747
},
{
"epoch": 2.2275555555555555,
"grad_norm": 2.9555110931396484,
"learning_rate": 1.6989994254956222e-06,
"loss": 0.21993732452392578,
"loss_sft": 0.21993732452392578,
"step": 628,
"token_acc": 0.9234069175082671
},
{
"epoch": 2.2311111111111113,
"grad_norm": 2.8966495990753174,
"learning_rate": 1.6843867087960252e-06,
"loss": 0.19336821138858795,
"loss_sft": 0.19336821138858795,
"step": 629,
"token_acc": 0.9320323014804845
},
{
"epoch": 2.2346666666666666,
"grad_norm": 2.541254758834839,
"learning_rate": 1.669824364535918e-06,
"loss": 0.17367511987686157,
"loss_sft": 0.17367511987686157,
"step": 630,
"token_acc": 0.9399025852379168
},
{
"epoch": 2.2382222222222223,
"grad_norm": 2.7811059951782227,
"learning_rate": 1.6553126139536534e-06,
"loss": 0.19320498406887054,
"loss_sft": 0.19320498406887054,
"step": 631,
"token_acc": 0.9331630481055768
},
{
"epoch": 2.2417777777777776,
"grad_norm": 2.595618963241577,
"learning_rate": 1.640851677518927e-06,
"loss": 0.19643685221672058,
"loss_sft": 0.19643685221672058,
"step": 632,
"token_acc": 0.9366061074603789
},
{
"epoch": 2.2453333333333334,
"grad_norm": 2.6902763843536377,
"learning_rate": 1.6264417749294543e-06,
"loss": 0.18876558542251587,
"loss_sft": 0.18876558542251587,
"step": 633,
"token_acc": 0.9352716279896861
},
{
"epoch": 2.2488888888888887,
"grad_norm": 2.7905678749084473,
"learning_rate": 1.6120831251076118e-06,
"loss": 0.21057236194610596,
"loss_sft": 0.21057236194610596,
"step": 634,
"token_acc": 0.9273299122661811
},
{
"epoch": 2.2524444444444445,
"grad_norm": 2.4698686599731445,
"learning_rate": 1.597775946197117e-06,
"loss": 0.16951118409633636,
"loss_sft": 0.16951118409633636,
"step": 635,
"token_acc": 0.9397039254823686
},
{
"epoch": 2.2560000000000002,
"grad_norm": 2.92950177192688,
"learning_rate": 1.5835204555597227e-06,
"loss": 0.20030246675014496,
"loss_sft": 0.20030246675014496,
"step": 636,
"token_acc": 0.9321304419798047
},
{
"epoch": 2.2595555555555555,
"grad_norm": 2.7453880310058594,
"learning_rate": 1.5693168697719024e-06,
"loss": 0.2218300700187683,
"loss_sft": 0.2218300700187683,
"step": 637,
"token_acc": 0.9244853189335134
},
{
"epoch": 2.2631111111111113,
"grad_norm": 2.5229060649871826,
"learning_rate": 1.555165404621567e-06,
"loss": 0.190987229347229,
"loss_sft": 0.190987229347229,
"step": 638,
"token_acc": 0.934009360374415
},
{
"epoch": 2.2666666666666666,
"grad_norm": 3.1500985622406006,
"learning_rate": 1.5410662751047855e-06,
"loss": 0.20426036417484283,
"loss_sft": 0.20426036417484283,
"step": 639,
"token_acc": 0.9306131857999078
},
{
"epoch": 2.2702222222222224,
"grad_norm": 2.6293997764587402,
"learning_rate": 1.5270196954225175e-06,
"loss": 0.20575830340385437,
"loss_sft": 0.20575830340385437,
"step": 640,
"token_acc": 0.9277999335327352
},
{
"epoch": 2.2737777777777777,
"grad_norm": 2.619807481765747,
"learning_rate": 1.5130258789773583e-06,
"loss": 0.20222562551498413,
"loss_sft": 0.20222562551498413,
"step": 641,
"token_acc": 0.9308885754583921
},
{
"epoch": 2.2773333333333334,
"grad_norm": 2.701447010040283,
"learning_rate": 1.4990850383703005e-06,
"loss": 0.1890687644481659,
"loss_sft": 0.1890687644481659,
"step": 642,
"token_acc": 0.9356768272697477
},
{
"epoch": 2.2808888888888887,
"grad_norm": 2.6472737789154053,
"learning_rate": 1.4851973853974987e-06,
"loss": 0.16918960213661194,
"loss_sft": 0.16918960213661194,
"step": 643,
"token_acc": 0.9412237516326736
},
{
"epoch": 2.2844444444444445,
"grad_norm": 2.9728641510009766,
"learning_rate": 1.4713631310470571e-06,
"loss": 0.21068130433559418,
"loss_sft": 0.21068130433559418,
"step": 644,
"token_acc": 0.9284918464499139
},
{
"epoch": 2.288,
"grad_norm": 2.6223902702331543,
"learning_rate": 1.457582485495821e-06,
"loss": 0.18204444646835327,
"loss_sft": 0.18204444646835327,
"step": 645,
"token_acc": 0.9377307408318976
},
{
"epoch": 2.2915555555555556,
"grad_norm": 2.8263981342315674,
"learning_rate": 1.4438556581061819e-06,
"loss": 0.19484788179397583,
"loss_sft": 0.19484788179397583,
"step": 646,
"token_acc": 0.93337360835419
},
{
"epoch": 2.295111111111111,
"grad_norm": 2.9367198944091797,
"learning_rate": 1.4301828574229026e-06,
"loss": 0.18675702810287476,
"loss_sft": 0.18675702810287476,
"step": 647,
"token_acc": 0.9357450473729544
},
{
"epoch": 2.2986666666666666,
"grad_norm": 2.730617046356201,
"learning_rate": 1.4165642911699435e-06,
"loss": 0.18839356303215027,
"loss_sft": 0.18839356303215027,
"step": 648,
"token_acc": 0.9342382830413648
},
{
"epoch": 2.3022222222222224,
"grad_norm": 2.550629138946533,
"learning_rate": 1.4030001662473086e-06,
"loss": 0.19302262365818024,
"loss_sft": 0.19302262365818024,
"step": 649,
"token_acc": 0.9325825362542425
},
{
"epoch": 2.3057777777777777,
"grad_norm": 2.7511205673217773,
"learning_rate": 1.389490688727903e-06,
"loss": 0.1946883201599121,
"loss_sft": 0.1946883201599121,
"step": 650,
"token_acc": 0.9317655136898121
},
{
"epoch": 2.3093333333333335,
"grad_norm": 2.7703752517700195,
"learning_rate": 1.3760360638544012e-06,
"loss": 0.19888922572135925,
"loss_sft": 0.19888922572135925,
"step": 651,
"token_acc": 0.93048128342246
},
{
"epoch": 2.3128888888888888,
"grad_norm": 2.983177900314331,
"learning_rate": 1.3626364960361282e-06,
"loss": 0.18993470072746277,
"loss_sft": 0.18993470072746277,
"step": 652,
"token_acc": 0.9341640319880975
},
{
"epoch": 2.3164444444444445,
"grad_norm": 2.552877187728882,
"learning_rate": 1.3492921888459566e-06,
"loss": 0.17532691359519958,
"loss_sft": 0.17532691359519958,
"step": 653,
"token_acc": 0.9385906040268457
},
{
"epoch": 2.32,
"grad_norm": 2.6213417053222656,
"learning_rate": 1.3360033450172106e-06,
"loss": 0.19140905141830444,
"loss_sft": 0.19140905141830444,
"step": 654,
"token_acc": 0.931980003029844
},
{
"epoch": 2.3235555555555556,
"grad_norm": 3.0324511528015137,
"learning_rate": 1.3227701664405868e-06,
"loss": 0.18983525037765503,
"loss_sft": 0.18983525037765503,
"step": 655,
"token_acc": 0.9348777092861403
},
{
"epoch": 2.327111111111111,
"grad_norm": 2.9595119953155518,
"learning_rate": 1.3095928541610936e-06,
"loss": 0.1786542534828186,
"loss_sft": 0.1786542534828186,
"step": 656,
"token_acc": 0.9380457380457381
},
{
"epoch": 2.3306666666666667,
"grad_norm": 2.466749429702759,
"learning_rate": 1.2964716083749829e-06,
"loss": 0.1687704622745514,
"loss_sft": 0.1687704622745514,
"step": 657,
"token_acc": 0.9421686746987952
},
{
"epoch": 2.3342222222222224,
"grad_norm": 3.9528796672821045,
"learning_rate": 1.2834066284267189e-06,
"loss": 0.20943708717823029,
"loss_sft": 0.20943708717823029,
"step": 658,
"token_acc": 0.9294945198756748
},
{
"epoch": 2.3377777777777777,
"grad_norm": 2.7764971256256104,
"learning_rate": 1.2703981128059534e-06,
"loss": 0.20439909398555756,
"loss_sft": 0.20439909398555756,
"step": 659,
"token_acc": 0.933202838267678
},
{
"epoch": 2.3413333333333335,
"grad_norm": 2.8635764122009277,
"learning_rate": 1.257446259144494e-06,
"loss": 0.1944856345653534,
"loss_sft": 0.1944856345653534,
"step": 660,
"token_acc": 0.9355538780902057
},
{
"epoch": 2.344888888888889,
"grad_norm": 2.821960926055908,
"learning_rate": 1.2445512642133218e-06,
"loss": 0.20042800903320312,
"loss_sft": 0.20042800903320312,
"step": 661,
"token_acc": 0.9305123914256116
},
{
"epoch": 2.3484444444444446,
"grad_norm": 2.6095001697540283,
"learning_rate": 1.2317133239195866e-06,
"loss": 0.1816886067390442,
"loss_sft": 0.1816886067390442,
"step": 662,
"token_acc": 0.9372599231754162
},
{
"epoch": 2.352,
"grad_norm": 2.772376775741577,
"learning_rate": 1.2189326333036323e-06,
"loss": 0.17986129224300385,
"loss_sft": 0.17986129224300385,
"step": 663,
"token_acc": 0.9380500726309493
},
{
"epoch": 2.3555555555555556,
"grad_norm": 2.7254045009613037,
"learning_rate": 1.2062093865360458e-06,
"loss": 0.1874287873506546,
"loss_sft": 0.1874287873506546,
"step": 664,
"token_acc": 0.9368523316062176
},
{
"epoch": 2.359111111111111,
"grad_norm": 2.65085506439209,
"learning_rate": 1.193543776914693e-06,
"loss": 0.17799338698387146,
"loss_sft": 0.17799338698387146,
"step": 665,
"token_acc": 0.9395775209246712
},
{
"epoch": 2.3626666666666667,
"grad_norm": 2.751464605331421,
"learning_rate": 1.1809359968617894e-06,
"loss": 0.20440709590911865,
"loss_sft": 0.20440709590911865,
"step": 666,
"token_acc": 0.9277301498277382
},
{
"epoch": 2.3662222222222224,
"grad_norm": 2.7445507049560547,
"learning_rate": 1.1683862379209747e-06,
"loss": 0.17293334007263184,
"loss_sft": 0.17293334007263184,
"step": 667,
"token_acc": 0.9389204545454546
},
{
"epoch": 2.3697777777777778,
"grad_norm": 2.7575273513793945,
"learning_rate": 1.1558946907544034e-06,
"loss": 0.2130010575056076,
"loss_sft": 0.2130010575056076,
"step": 668,
"token_acc": 0.9253860020377772
},
{
"epoch": 2.3733333333333335,
"grad_norm": 2.9588608741760254,
"learning_rate": 1.1434615451398467e-06,
"loss": 0.20575463771820068,
"loss_sft": 0.20575463771820068,
"step": 669,
"token_acc": 0.9272522733664573
},
{
"epoch": 2.376888888888889,
"grad_norm": 2.7114598751068115,
"learning_rate": 1.1310869899678122e-06,
"loss": 0.19226443767547607,
"loss_sft": 0.19226443767547607,
"step": 670,
"token_acc": 0.93167104111986
},
{
"epoch": 2.3804444444444446,
"grad_norm": 2.6829819679260254,
"learning_rate": 1.1187712132386723e-06,
"loss": 0.19152238965034485,
"loss_sft": 0.19152238965034485,
"step": 671,
"token_acc": 0.9335521575534385
},
{
"epoch": 2.384,
"grad_norm": 2.774921417236328,
"learning_rate": 1.1065144020598067e-06,
"loss": 0.19084396958351135,
"loss_sft": 0.19084396958351135,
"step": 672,
"token_acc": 0.9325548141086749
},
{
"epoch": 2.3875555555555557,
"grad_norm": 2.6978862285614014,
"learning_rate": 1.0943167426427625e-06,
"loss": 0.17401374876499176,
"loss_sft": 0.17401374876499176,
"step": 673,
"token_acc": 0.9382892742886008
},
{
"epoch": 2.391111111111111,
"grad_norm": 2.6622302532196045,
"learning_rate": 1.0821784203004238e-06,
"loss": 0.17525947093963623,
"loss_sft": 0.17525947093963623,
"step": 674,
"token_acc": 0.9387893134934054
},
{
"epoch": 2.3946666666666667,
"grad_norm": 2.814655303955078,
"learning_rate": 1.0700996194441944e-06,
"loss": 0.18369118869304657,
"loss_sft": 0.18369118869304657,
"step": 675,
"token_acc": 0.9327033218785796
},
{
"epoch": 2.398222222222222,
"grad_norm": 2.710477352142334,
"learning_rate": 1.0580805235812042e-06,
"loss": 0.19263188540935516,
"loss_sft": 0.19263188540935516,
"step": 676,
"token_acc": 0.9323791566678094
},
{
"epoch": 2.401777777777778,
"grad_norm": 2.7136497497558594,
"learning_rate": 1.046121315311508e-06,
"loss": 0.17976251244544983,
"loss_sft": 0.17976251244544983,
"step": 677,
"token_acc": 0.9372756640344581
},
{
"epoch": 2.405333333333333,
"grad_norm": 2.6189379692077637,
"learning_rate": 1.0342221763253207e-06,
"loss": 0.18855607509613037,
"loss_sft": 0.18855607509613037,
"step": 678,
"token_acc": 0.9368050555955524
},
{
"epoch": 2.408888888888889,
"grad_norm": 2.8044228553771973,
"learning_rate": 1.0223832874002603e-06,
"loss": 0.19502097368240356,
"loss_sft": 0.19502097368240356,
"step": 679,
"token_acc": 0.9328933836133013
},
{
"epoch": 2.4124444444444446,
"grad_norm": 2.778628349304199,
"learning_rate": 1.010604828398587e-06,
"loss": 0.18156121671199799,
"loss_sft": 0.18156121671199799,
"step": 680,
"token_acc": 0.9354981028853789
},
{
"epoch": 2.416,
"grad_norm": 2.847377300262451,
"learning_rate": 9.988869782644872e-07,
"loss": 0.1790425032377243,
"loss_sft": 0.1790425032377243,
"step": 681,
"token_acc": 0.9367146239300674
},
{
"epoch": 2.4195555555555557,
"grad_norm": 2.8271334171295166,
"learning_rate": 9.872299150213455e-07,
"loss": 0.21350498497486115,
"loss_sft": 0.21350498497486115,
"step": 682,
"token_acc": 0.9258656153582447
},
{
"epoch": 2.423111111111111,
"grad_norm": 2.682698965072632,
"learning_rate": 9.75633815769036e-07,
"loss": 0.20161369442939758,
"loss_sft": 0.20161369442939758,
"step": 683,
"token_acc": 0.9298965936739659
},
{
"epoch": 2.4266666666666667,
"grad_norm": 2.5782878398895264,
"learning_rate": 9.640988566812475e-07,
"loss": 0.19973236322402954,
"loss_sft": 0.19973236322402954,
"step": 684,
"token_acc": 0.9316569954867827
},
{
"epoch": 2.430222222222222,
"grad_norm": 2.8129353523254395,
"learning_rate": 9.526252130027919e-07,
"loss": 0.18847820162773132,
"loss_sft": 0.18847820162773132,
"step": 685,
"token_acc": 0.9332250341997264
},
{
"epoch": 2.433777777777778,
"grad_norm": 2.9322848320007324,
"learning_rate": 9.412130590469438e-07,
"loss": 0.19337347149848938,
"loss_sft": 0.19337347149848938,
"step": 686,
"token_acc": 0.9344031531531531
},
{
"epoch": 2.437333333333333,
"grad_norm": 2.8443686962127686,
"learning_rate": 9.298625681928031e-07,
"loss": 0.2070617824792862,
"loss_sft": 0.2070617824792862,
"step": 687,
"token_acc": 0.9286177205814277
},
{
"epoch": 2.440888888888889,
"grad_norm": 2.7011122703552246,
"learning_rate": 9.185739128826454e-07,
"loss": 0.1802581548690796,
"loss_sft": 0.1802581548690796,
"step": 688,
"token_acc": 0.9387495456197746
},
{
"epoch": 2.4444444444444446,
"grad_norm": 3.021794319152832,
"learning_rate": 9.073472646193171e-07,
"loss": 0.20768821239471436,
"loss_sft": 0.20768821239471436,
"step": 689,
"token_acc": 0.9260020554984584
},
{
"epoch": 2.448,
"grad_norm": 2.528430461883545,
"learning_rate": 8.961827939636198e-07,
"loss": 0.183142751455307,
"loss_sft": 0.183142751455307,
"step": 690,
"token_acc": 0.9344323437266896
},
{
"epoch": 2.4515555555555557,
"grad_norm": 2.7379767894744873,
"learning_rate": 8.850806705317183e-07,
"loss": 0.18841874599456787,
"loss_sft": 0.18841874599456787,
"step": 691,
"token_acc": 0.9357790094132137
},
{
"epoch": 2.455111111111111,
"grad_norm": 2.766420841217041,
"learning_rate": 8.740410629925744e-07,
"loss": 0.2205144762992859,
"loss_sft": 0.2205144762992859,
"step": 692,
"token_acc": 0.9251700680272109
},
{
"epoch": 2.458666666666667,
"grad_norm": 2.7139368057250977,
"learning_rate": 8.630641390653743e-07,
"loss": 0.20389555394649506,
"loss_sft": 0.20389555394649506,
"step": 693,
"token_acc": 0.9301208809405531
},
{
"epoch": 2.462222222222222,
"grad_norm": 2.7454299926757812,
"learning_rate": 8.521500655169823e-07,
"loss": 0.17874322831630707,
"loss_sft": 0.17874322831630707,
"step": 694,
"token_acc": 0.9370383095688026
},
{
"epoch": 2.465777777777778,
"grad_norm": 2.737285852432251,
"learning_rate": 8.412990081594102e-07,
"loss": 0.20597487688064575,
"loss_sft": 0.20597487688064575,
"step": 695,
"token_acc": 0.9287361452056833
},
{
"epoch": 2.469333333333333,
"grad_norm": 2.6146230697631836,
"learning_rate": 8.30511131847297e-07,
"loss": 0.18359915912151337,
"loss_sft": 0.18359915912151337,
"step": 696,
"token_acc": 0.9348174532502226
},
{
"epoch": 2.472888888888889,
"grad_norm": 3.2805652618408203,
"learning_rate": 8.197866004754029e-07,
"loss": 0.18914654850959778,
"loss_sft": 0.18914654850959778,
"step": 697,
"token_acc": 0.9331138590011241
},
{
"epoch": 2.4764444444444447,
"grad_norm": 2.928779125213623,
"learning_rate": 8.091255769761213e-07,
"loss": 0.1942884624004364,
"loss_sft": 0.1942884624004364,
"step": 698,
"token_acc": 0.9339291093483282
},
{
"epoch": 2.48,
"grad_norm": 2.5084075927734375,
"learning_rate": 7.985282233170011e-07,
"loss": 0.17477425932884216,
"loss_sft": 0.17477425932884216,
"step": 699,
"token_acc": 0.9388229634381013
},
{
"epoch": 2.4835555555555557,
"grad_norm": 2.519057512283325,
"learning_rate": 7.879947004982896e-07,
"loss": 0.1848813146352768,
"loss_sft": 0.1848813146352768,
"step": 700,
"token_acc": 0.9370736567793363
},
{
"epoch": 2.487111111111111,
"grad_norm": 2.5513229370117188,
"learning_rate": 7.775251685504826e-07,
"loss": 0.17141607403755188,
"loss_sft": 0.17141607403755188,
"step": 701,
"token_acc": 0.9405192244495564
},
{
"epoch": 2.490666666666667,
"grad_norm": 2.668494701385498,
"learning_rate": 7.671197865318952e-07,
"loss": 0.19210359454154968,
"loss_sft": 0.19210359454154968,
"step": 702,
"token_acc": 0.9327320867118981
},
{
"epoch": 2.494222222222222,
"grad_norm": 2.592811346054077,
"learning_rate": 7.567787125262449e-07,
"loss": 0.18700769543647766,
"loss_sft": 0.18700769543647766,
"step": 703,
"token_acc": 0.9360098176718092
},
{
"epoch": 2.497777777777778,
"grad_norm": 3.0305991172790527,
"learning_rate": 7.465021036402531e-07,
"loss": 0.18039140105247498,
"loss_sft": 0.18039140105247498,
"step": 704,
"token_acc": 0.9347649918962723
},
{
"epoch": 2.501333333333333,
"grad_norm": 2.5868568420410156,
"learning_rate": 7.362901160012492e-07,
"loss": 0.1872612088918686,
"loss_sft": 0.1872612088918686,
"step": 705,
"token_acc": 0.9344091201027617
},
{
"epoch": 2.504888888888889,
"grad_norm": 2.8019750118255615,
"learning_rate": 7.261429047548085e-07,
"loss": 0.192620187997818,
"loss_sft": 0.192620187997818,
"step": 706,
"token_acc": 0.9334470989761092
},
{
"epoch": 2.5084444444444447,
"grad_norm": 2.791964292526245,
"learning_rate": 7.16060624062393e-07,
"loss": 0.18782371282577515,
"loss_sft": 0.18782371282577515,
"step": 707,
"token_acc": 0.933403342959657
},
{
"epoch": 2.512,
"grad_norm": 2.7080771923065186,
"learning_rate": 7.060434270990013e-07,
"loss": 0.1883229911327362,
"loss_sft": 0.1883229911327362,
"step": 708,
"token_acc": 0.9338287611668906
},
{
"epoch": 2.5155555555555553,
"grad_norm": 3.0514519214630127,
"learning_rate": 6.96091466050855e-07,
"loss": 0.16829772293567657,
"loss_sft": 0.16829772293567657,
"step": 709,
"token_acc": 0.942652329749104
},
{
"epoch": 2.519111111111111,
"grad_norm": 2.782956838607788,
"learning_rate": 6.86204892113076e-07,
"loss": 0.17585113644599915,
"loss_sft": 0.17585113644599915,
"step": 710,
"token_acc": 0.9417410338440815
},
{
"epoch": 2.522666666666667,
"grad_norm": 2.830371379852295,
"learning_rate": 6.763838554873892e-07,
"loss": 0.1984882950782776,
"loss_sft": 0.1984882950782776,
"step": 711,
"token_acc": 0.9317339918856942
},
{
"epoch": 2.526222222222222,
"grad_norm": 2.846914052963257,
"learning_rate": 6.66628505379851e-07,
"loss": 0.2091943919658661,
"loss_sft": 0.2091943919658661,
"step": 712,
"token_acc": 0.9267665304887275
},
{
"epoch": 2.529777777777778,
"grad_norm": 2.6870155334472656,
"learning_rate": 6.569389899985723e-07,
"loss": 0.19458621740341187,
"loss_sft": 0.19458621740341187,
"step": 713,
"token_acc": 0.9330594857751408
},
{
"epoch": 2.533333333333333,
"grad_norm": 2.7646307945251465,
"learning_rate": 6.473154565514695e-07,
"loss": 0.19579321146011353,
"loss_sft": 0.19579321146011353,
"step": 714,
"token_acc": 0.9300840336134454
},
{
"epoch": 2.536888888888889,
"grad_norm": 2.7302122116088867,
"learning_rate": 6.377580512440301e-07,
"loss": 0.18625396490097046,
"loss_sft": 0.18625396490097046,
"step": 715,
"token_acc": 0.9341757515692104
},
{
"epoch": 2.5404444444444443,
"grad_norm": 2.4824278354644775,
"learning_rate": 6.282669192770896e-07,
"loss": 0.17889976501464844,
"loss_sft": 0.17889976501464844,
"step": 716,
"token_acc": 0.9392045010549348
},
{
"epoch": 2.544,
"grad_norm": 2.8821957111358643,
"learning_rate": 6.188422048446263e-07,
"loss": 0.20137062668800354,
"loss_sft": 0.20137062668800354,
"step": 717,
"token_acc": 0.9299661126351461
},
{
"epoch": 2.5475555555555554,
"grad_norm": 2.9125261306762695,
"learning_rate": 6.094840511315703e-07,
"loss": 0.16783937811851501,
"loss_sft": 0.16783937811851501,
"step": 718,
"token_acc": 0.9404943545926152
},
{
"epoch": 2.551111111111111,
"grad_norm": 2.764349937438965,
"learning_rate": 6.001926003116282e-07,
"loss": 0.19386354088783264,
"loss_sft": 0.19386354088783264,
"step": 719,
"token_acc": 0.9315595859354949
},
{
"epoch": 2.554666666666667,
"grad_norm": 2.8563075065612793,
"learning_rate": 5.909679935451235e-07,
"loss": 0.19786891341209412,
"loss_sft": 0.19786891341209412,
"step": 720,
"token_acc": 0.9314799213704016
},
{
"epoch": 2.558222222222222,
"grad_norm": 2.663599729537964,
"learning_rate": 5.818103709768519e-07,
"loss": 0.18172010779380798,
"loss_sft": 0.18172010779380798,
"step": 721,
"token_acc": 0.9373431555087833
},
{
"epoch": 2.561777777777778,
"grad_norm": 2.8229429721832275,
"learning_rate": 5.727198717339511e-07,
"loss": 0.18148422241210938,
"loss_sft": 0.18148422241210938,
"step": 722,
"token_acc": 0.9378488317093936
},
{
"epoch": 2.5653333333333332,
"grad_norm": 2.638422966003418,
"learning_rate": 5.636966339237882e-07,
"loss": 0.18546408414840698,
"loss_sft": 0.18546408414840698,
"step": 723,
"token_acc": 0.9357721540131964
},
{
"epoch": 2.568888888888889,
"grad_norm": 3.2908942699432373,
"learning_rate": 5.547407946318628e-07,
"loss": 0.21257361769676208,
"loss_sft": 0.21257361769676208,
"step": 724,
"token_acc": 0.9265181458857348
},
{
"epoch": 2.5724444444444443,
"grad_norm": 2.886324167251587,
"learning_rate": 5.458524899197204e-07,
"loss": 0.21434122323989868,
"loss_sft": 0.21434122323989868,
"step": 725,
"token_acc": 0.927958526993207
},
{
"epoch": 2.576,
"grad_norm": 2.847505569458008,
"learning_rate": 5.370318548228886e-07,
"loss": 0.21782180666923523,
"loss_sft": 0.21782180666923523,
"step": 726,
"token_acc": 0.9258269820067573
},
{
"epoch": 2.5795555555555554,
"grad_norm": 2.929295778274536,
"learning_rate": 5.282790233488283e-07,
"loss": 0.20660823583602905,
"loss_sft": 0.20660823583602905,
"step": 727,
"token_acc": 0.9290664841276632
},
{
"epoch": 2.583111111111111,
"grad_norm": 2.6943576335906982,
"learning_rate": 5.195941284748879e-07,
"loss": 0.1793433427810669,
"loss_sft": 0.1793433427810669,
"step": 728,
"token_acc": 0.9385319601111656
},
{
"epoch": 2.586666666666667,
"grad_norm": 2.690903425216675,
"learning_rate": 5.109773021462921e-07,
"loss": 0.17306211590766907,
"loss_sft": 0.17306211590766907,
"step": 729,
"token_acc": 0.9419840937630808
},
{
"epoch": 2.590222222222222,
"grad_norm": 2.677744150161743,
"learning_rate": 5.024286752741364e-07,
"loss": 0.21147862076759338,
"loss_sft": 0.21147862076759338,
"step": 730,
"token_acc": 0.9255679072015467
},
{
"epoch": 2.5937777777777775,
"grad_norm": 2.9246270656585693,
"learning_rate": 4.939483777333931e-07,
"loss": 0.18388132750988007,
"loss_sft": 0.18388132750988007,
"step": 731,
"token_acc": 0.9372676136905329
},
{
"epoch": 2.5973333333333333,
"grad_norm": 2.6163809299468994,
"learning_rate": 4.855365383609457e-07,
"loss": 0.17831042408943176,
"loss_sft": 0.17831042408943176,
"step": 732,
"token_acc": 0.9371385655276004
},
{
"epoch": 2.600888888888889,
"grad_norm": 2.5188536643981934,
"learning_rate": 4.77193284953622e-07,
"loss": 0.18781685829162598,
"loss_sft": 0.18781685829162598,
"step": 733,
"token_acc": 0.93570617804443
},
{
"epoch": 2.6044444444444443,
"grad_norm": 2.7731027603149414,
"learning_rate": 4.6891874426626125e-07,
"loss": 0.20185299217700958,
"loss_sft": 0.20185299217700958,
"step": 734,
"token_acc": 0.9310315210856308
},
{
"epoch": 2.608,
"grad_norm": 2.7207086086273193,
"learning_rate": 4.607130420097866e-07,
"loss": 0.18952256441116333,
"loss_sft": 0.18952256441116333,
"step": 735,
"token_acc": 0.9361941915908106
},
{
"epoch": 2.6115555555555554,
"grad_norm": 2.6794168949127197,
"learning_rate": 4.525763028492869e-07,
"loss": 0.20114415884017944,
"loss_sft": 0.20114415884017944,
"step": 736,
"token_acc": 0.9304799604156359
},
{
"epoch": 2.615111111111111,
"grad_norm": 2.967855930328369,
"learning_rate": 4.445086504021384e-07,
"loss": 0.20481745898723602,
"loss_sft": 0.20481745898723602,
"step": 737,
"token_acc": 0.9287511230907457
},
{
"epoch": 2.618666666666667,
"grad_norm": 2.5926387310028076,
"learning_rate": 4.365102072361116e-07,
"loss": 0.18741759657859802,
"loss_sft": 0.18741759657859802,
"step": 738,
"token_acc": 0.9360535440690267
},
{
"epoch": 2.6222222222222222,
"grad_norm": 2.84592604637146,
"learning_rate": 4.285810948675156e-07,
"loss": 0.1873520016670227,
"loss_sft": 0.1873520016670227,
"step": 739,
"token_acc": 0.9371667537898588
},
{
"epoch": 2.6257777777777775,
"grad_norm": 2.9166107177734375,
"learning_rate": 4.207214337593557e-07,
"loss": 0.1775439828634262,
"loss_sft": 0.1775439828634262,
"step": 740,
"token_acc": 0.935292937399679
},
{
"epoch": 2.6293333333333333,
"grad_norm": 2.845803737640381,
"learning_rate": 4.129313433194948e-07,
"loss": 0.20385167002677917,
"loss_sft": 0.20385167002677917,
"step": 741,
"token_acc": 0.9301189118748981
},
{
"epoch": 2.632888888888889,
"grad_norm": 2.8998193740844727,
"learning_rate": 4.05210941898847e-07,
"loss": 0.19685634970664978,
"loss_sft": 0.19685634970664978,
"step": 742,
"token_acc": 0.933920704845815
},
{
"epoch": 2.6364444444444444,
"grad_norm": 3.0047550201416016,
"learning_rate": 3.975603467895739e-07,
"loss": 0.18327781558036804,
"loss_sft": 0.18327781558036804,
"step": 743,
"token_acc": 0.934224049331963
},
{
"epoch": 2.64,
"grad_norm": 2.775958776473999,
"learning_rate": 3.8997967422330693e-07,
"loss": 0.17468193173408508,
"loss_sft": 0.17468193173408508,
"step": 744,
"token_acc": 0.9375679973219516
},
{
"epoch": 2.6435555555555554,
"grad_norm": 2.726837396621704,
"learning_rate": 3.8246903936937806e-07,
"loss": 0.1818585842847824,
"loss_sft": 0.1818585842847824,
"step": 745,
"token_acc": 0.9362288307268455
},
{
"epoch": 2.647111111111111,
"grad_norm": 3.9764692783355713,
"learning_rate": 3.7502855633307246e-07,
"loss": 0.19163265824317932,
"loss_sft": 0.19163265824317932,
"step": 746,
"token_acc": 0.9312363564199246
},
{
"epoch": 2.6506666666666665,
"grad_norm": 2.7666332721710205,
"learning_rate": 3.676583381538945e-07,
"loss": 0.20780622959136963,
"loss_sft": 0.20780622959136963,
"step": 747,
"token_acc": 0.9287303413885692
},
{
"epoch": 2.6542222222222223,
"grad_norm": 2.96278977394104,
"learning_rate": 3.603584968038487e-07,
"loss": 0.2147180736064911,
"loss_sft": 0.2147180736064911,
"step": 748,
"token_acc": 0.9272157261481105
},
{
"epoch": 2.6577777777777776,
"grad_norm": 2.855870246887207,
"learning_rate": 3.53129143185742e-07,
"loss": 0.20830529928207397,
"loss_sft": 0.20830529928207397,
"step": 749,
"token_acc": 0.9286667779075588
},
{
"epoch": 2.6613333333333333,
"grad_norm": 2.569326400756836,
"learning_rate": 3.4597038713149455e-07,
"loss": 0.18940183520317078,
"loss_sft": 0.18940183520317078,
"step": 750,
"token_acc": 0.9371379364333803
},
{
"epoch": 2.664888888888889,
"grad_norm": 2.692840099334717,
"learning_rate": 3.388823374004735e-07,
"loss": 0.16750888526439667,
"loss_sft": 0.16750888526439667,
"step": 751,
"token_acc": 0.9431868228662028
},
{
"epoch": 2.6684444444444444,
"grad_norm": 2.9402012825012207,
"learning_rate": 3.3186510167784456e-07,
"loss": 0.18632622063159943,
"loss_sft": 0.18632622063159943,
"step": 752,
"token_acc": 0.9334459776588754
},
{
"epoch": 2.672,
"grad_norm": 2.7272469997406006,
"learning_rate": 3.2491878657292643e-07,
"loss": 0.19793246686458588,
"loss_sft": 0.19793246686458588,
"step": 753,
"token_acc": 0.9352339881808018
},
{
"epoch": 2.6755555555555555,
"grad_norm": 2.654890775680542,
"learning_rate": 3.180434976175784e-07,
"loss": 0.2020973265171051,
"loss_sft": 0.2020973265171051,
"step": 754,
"token_acc": 0.9319294809010774
},
{
"epoch": 2.679111111111111,
"grad_norm": 2.805154800415039,
"learning_rate": 3.112393392645985e-07,
"loss": 0.18373054265975952,
"loss_sft": 0.18373054265975952,
"step": 755,
"token_acc": 0.9395501405810684
},
{
"epoch": 2.6826666666666665,
"grad_norm": 3.0275561809539795,
"learning_rate": 3.045064148861282e-07,
"loss": 0.2097509205341339,
"loss_sft": 0.2097509205341339,
"step": 756,
"token_acc": 0.9267343646199108
},
{
"epoch": 2.6862222222222223,
"grad_norm": 2.7499921321868896,
"learning_rate": 2.9784482677209013e-07,
"loss": 0.1804673969745636,
"loss_sft": 0.1804673969745636,
"step": 757,
"token_acc": 0.9370858218113393
},
{
"epoch": 2.6897777777777776,
"grad_norm": 2.8307371139526367,
"learning_rate": 2.912546761286333e-07,
"loss": 0.1866697371006012,
"loss_sft": 0.1866697371006012,
"step": 758,
"token_acc": 0.9342129208371247
},
{
"epoch": 2.6933333333333334,
"grad_norm": 2.9989938735961914,
"learning_rate": 2.847360630765866e-07,
"loss": 0.18947447836399078,
"loss_sft": 0.18947447836399078,
"step": 759,
"token_acc": 0.9367885238315594
},
{
"epoch": 2.696888888888889,
"grad_norm": 2.519296884536743,
"learning_rate": 2.7828908664995216e-07,
"loss": 0.1839563250541687,
"loss_sft": 0.1839563250541687,
"step": 760,
"token_acc": 0.9373976477594164
},
{
"epoch": 2.7004444444444444,
"grad_norm": 2.6605520248413086,
"learning_rate": 2.719138447943881e-07,
"loss": 0.20287805795669556,
"loss_sft": 0.20287805795669556,
"step": 761,
"token_acc": 0.9286905122764474
},
{
"epoch": 2.7039999999999997,
"grad_norm": 2.9450504779815674,
"learning_rate": 2.656104343657251e-07,
"loss": 0.2017054259777069,
"loss_sft": 0.2017054259777069,
"step": 762,
"token_acc": 0.932173752990981
},
{
"epoch": 2.7075555555555555,
"grad_norm": 2.801025629043579,
"learning_rate": 2.5937895112849886e-07,
"loss": 0.17514121532440186,
"loss_sft": 0.17514121532440186,
"step": 763,
"token_acc": 0.9412087912087912
},
{
"epoch": 2.7111111111111112,
"grad_norm": 2.824176549911499,
"learning_rate": 2.532194897544882e-07,
"loss": 0.19747674465179443,
"loss_sft": 0.19747674465179443,
"step": 764,
"token_acc": 0.9332013428596023
},
{
"epoch": 2.7146666666666666,
"grad_norm": 2.6715922355651855,
"learning_rate": 2.4713214382128204e-07,
"loss": 0.18756616115570068,
"loss_sft": 0.18756616115570068,
"step": 765,
"token_acc": 0.9358335518839059
},
{
"epoch": 2.7182222222222223,
"grad_norm": 2.69205641746521,
"learning_rate": 2.411170058108558e-07,
"loss": 0.20186126232147217,
"loss_sft": 0.20186126232147217,
"step": 766,
"token_acc": 0.9306946688206785
},
{
"epoch": 2.7217777777777776,
"grad_norm": 2.844322681427002,
"learning_rate": 2.351741671081631e-07,
"loss": 0.18071135878562927,
"loss_sft": 0.18071135878562927,
"step": 767,
"token_acc": 0.934114071160379
},
{
"epoch": 2.7253333333333334,
"grad_norm": 2.8722591400146484,
"learning_rate": 2.2930371799975593e-07,
"loss": 0.21717549860477448,
"loss_sft": 0.21717549860477448,
"step": 768,
"token_acc": 0.9251712477239227
},
{
"epoch": 2.728888888888889,
"grad_norm": 2.714641809463501,
"learning_rate": 2.2350574767240395e-07,
"loss": 0.1820165514945984,
"loss_sft": 0.1820165514945984,
"step": 769,
"token_acc": 0.9367315573770492
},
{
"epoch": 2.7324444444444445,
"grad_norm": 2.665795087814331,
"learning_rate": 2.1778034421174433e-07,
"loss": 0.1903153359889984,
"loss_sft": 0.1903153359889984,
"step": 770,
"token_acc": 0.9333983898511832
},
{
"epoch": 2.7359999999999998,
"grad_norm": 2.605334758758545,
"learning_rate": 2.1212759460094268e-07,
"loss": 0.19885686039924622,
"loss_sft": 0.19885686039924622,
"step": 771,
"token_acc": 0.9309717059849807
},
{
"epoch": 2.7395555555555555,
"grad_norm": 2.639838695526123,
"learning_rate": 2.0654758471937098e-07,
"loss": 0.1776367723941803,
"loss_sft": 0.1776367723941803,
"step": 772,
"token_acc": 0.9394416732300691
},
{
"epoch": 2.7431111111111113,
"grad_norm": 2.723078489303589,
"learning_rate": 2.0104039934130348e-07,
"loss": 0.18507392704486847,
"loss_sft": 0.18507392704486847,
"step": 773,
"token_acc": 0.9337137103540862
},
{
"epoch": 2.7466666666666666,
"grad_norm": 2.5900092124938965,
"learning_rate": 1.9560612213462837e-07,
"loss": 0.17871883511543274,
"loss_sft": 0.17871883511543274,
"step": 774,
"token_acc": 0.9373716632443532
},
{
"epoch": 2.7502222222222223,
"grad_norm": 3.064812421798706,
"learning_rate": 1.9024483565957707e-07,
"loss": 0.2114132046699524,
"loss_sft": 0.2114132046699524,
"step": 775,
"token_acc": 0.9298800959232614
},
{
"epoch": 2.7537777777777777,
"grad_norm": 2.871481418609619,
"learning_rate": 1.8495662136746916e-07,
"loss": 0.18899960815906525,
"loss_sft": 0.18899960815906525,
"step": 776,
"token_acc": 0.9341864716636198
},
{
"epoch": 2.7573333333333334,
"grad_norm": 2.795487642288208,
"learning_rate": 1.7974155959947614e-07,
"loss": 0.20108124613761902,
"loss_sft": 0.20108124613761902,
"step": 777,
"token_acc": 0.9338042049934296
},
{
"epoch": 2.7608888888888887,
"grad_norm": 2.7187371253967285,
"learning_rate": 1.745997295853985e-07,
"loss": 0.17069584131240845,
"loss_sft": 0.17069584131240845,
"step": 778,
"token_acc": 0.9388874603582755
},
{
"epoch": 2.7644444444444445,
"grad_norm": 2.919107675552368,
"learning_rate": 1.6953120944246503e-07,
"loss": 0.202523335814476,
"loss_sft": 0.202523335814476,
"step": 779,
"token_acc": 0.9330868428172062
},
{
"epoch": 2.768,
"grad_norm": 2.810739517211914,
"learning_rate": 1.6453607617414603e-07,
"loss": 0.1683150827884674,
"loss_sft": 0.1683150827884674,
"step": 780,
"token_acc": 0.941493391699638
},
{
"epoch": 2.7715555555555556,
"grad_norm": 2.7224628925323486,
"learning_rate": 1.5961440566897913e-07,
"loss": 0.1763046532869339,
"loss_sft": 0.1763046532869339,
"step": 781,
"token_acc": 0.9370500719884819
},
{
"epoch": 2.7751111111111113,
"grad_norm": 2.6987104415893555,
"learning_rate": 1.5476627269941925e-07,
"loss": 0.1834906041622162,
"loss_sft": 0.1834906041622162,
"step": 782,
"token_acc": 0.9371736567289877
},
{
"epoch": 2.7786666666666666,
"grad_norm": 2.9807426929473877,
"learning_rate": 1.4999175092070716e-07,
"loss": 0.19927339255809784,
"loss_sft": 0.19927339255809784,
"step": 783,
"token_acc": 0.9314265090679419
},
{
"epoch": 2.7822222222222224,
"grad_norm": 2.8939366340637207,
"learning_rate": 1.4529091286973994e-07,
"loss": 0.20292577147483826,
"loss_sft": 0.20292577147483826,
"step": 784,
"token_acc": 0.9291538798772468
},
{
"epoch": 2.7857777777777777,
"grad_norm": 2.788940191268921,
"learning_rate": 1.4066382996397898e-07,
"loss": 0.2093643695116043,
"loss_sft": 0.2093643695116043,
"step": 785,
"token_acc": 0.9280611816119002
},
{
"epoch": 2.7893333333333334,
"grad_norm": 2.5868608951568604,
"learning_rate": 1.361105725003603e-07,
"loss": 0.19405758380889893,
"loss_sft": 0.19405758380889893,
"step": 786,
"token_acc": 0.9336200156372165
},
{
"epoch": 2.7928888888888888,
"grad_norm": 2.4726169109344482,
"learning_rate": 1.3163120965422494e-07,
"loss": 0.17949256300926208,
"loss_sft": 0.17949256300926208,
"step": 787,
"token_acc": 0.9376190476190476
},
{
"epoch": 2.7964444444444445,
"grad_norm": 2.833343982696533,
"learning_rate": 1.272258094782741e-07,
"loss": 0.18183688819408417,
"loss_sft": 0.18183688819408417,
"step": 788,
"token_acc": 0.9366845433442728
},
{
"epoch": 2.8,
"grad_norm": 2.521583080291748,
"learning_rate": 1.2289443890152852e-07,
"loss": 0.19447685778141022,
"loss_sft": 0.19447685778141022,
"step": 789,
"token_acc": 0.9328912048002527
},
{
"epoch": 2.8035555555555556,
"grad_norm": 3.2106516361236572,
"learning_rate": 1.1863716372831513e-07,
"loss": 0.1812889724969864,
"loss_sft": 0.1812889724969864,
"step": 790,
"token_acc": 0.9357969025681239
},
{
"epoch": 2.8071111111111113,
"grad_norm": 2.707206964492798,
"learning_rate": 1.1445404863726806e-07,
"loss": 0.19075971841812134,
"loss_sft": 0.19075971841812134,
"step": 791,
"token_acc": 0.9316420014094433
},
{
"epoch": 2.8106666666666666,
"grad_norm": 2.9277496337890625,
"learning_rate": 1.1034515718034321e-07,
"loss": 0.2353270947933197,
"loss_sft": 0.2353270947933197,
"step": 792,
"token_acc": 0.9219467401285583
},
{
"epoch": 2.814222222222222,
"grad_norm": 2.632071018218994,
"learning_rate": 1.0631055178185523e-07,
"loss": 0.1600821614265442,
"loss_sft": 0.1600821614265442,
"step": 793,
"token_acc": 0.944143966958403
},
{
"epoch": 2.8177777777777777,
"grad_norm": 2.7210655212402344,
"learning_rate": 1.0235029373752758e-07,
"loss": 0.19867268204689026,
"loss_sft": 0.19867268204689026,
"step": 794,
"token_acc": 0.931993350460934
},
{
"epoch": 2.8213333333333335,
"grad_norm": 3.018519401550293,
"learning_rate": 9.84644432135623e-08,
"loss": 0.20142260193824768,
"loss_sft": 0.20142260193824768,
"step": 795,
"token_acc": 0.9318513119533528
},
{
"epoch": 2.824888888888889,
"grad_norm": 2.829655408859253,
"learning_rate": 9.465305924572566e-08,
"loss": 0.1699664294719696,
"loss_sft": 0.1699664294719696,
"step": 796,
"token_acc": 0.9421344736602675
},
{
"epoch": 2.8284444444444445,
"grad_norm": 2.6472973823547363,
"learning_rate": 9.091619973845056e-08,
"loss": 0.1955169141292572,
"loss_sft": 0.1955169141292572,
"step": 797,
"token_acc": 0.9337710765239948
},
{
"epoch": 2.832,
"grad_norm": 2.8763346672058105,
"learning_rate": 8.725392146395728e-08,
"loss": 0.19352325797080994,
"loss_sft": 0.19352325797080994,
"step": 798,
"token_acc": 0.9315969481715338
},
{
"epoch": 2.8355555555555556,
"grad_norm": 2.7860326766967773,
"learning_rate": 8.366628006139243e-08,
"loss": 0.20639920234680176,
"loss_sft": 0.20639920234680176,
"step": 799,
"token_acc": 0.9274193548387096
},
{
"epoch": 2.8391111111111114,
"grad_norm": 2.87343692779541,
"learning_rate": 8.015333003598025e-08,
"loss": 0.19217310845851898,
"loss_sft": 0.19217310845851898,
"step": 800,
"token_acc": 0.9353986332574031
},
{
"epoch": 2.8426666666666667,
"grad_norm": 2.9104435443878174,
"learning_rate": 7.671512475819876e-08,
"loss": 0.20616498589515686,
"loss_sft": 0.20616498589515686,
"step": 801,
"token_acc": 0.9278471845011769
},
{
"epoch": 2.846222222222222,
"grad_norm": 2.6660847663879395,
"learning_rate": 7.335171646296436e-08,
"loss": 0.1993139386177063,
"loss_sft": 0.1993139386177063,
"step": 802,
"token_acc": 0.9321827057182706
},
{
"epoch": 2.8497777777777777,
"grad_norm": 2.833153247833252,
"learning_rate": 7.006315624884296e-08,
"loss": 0.20246970653533936,
"loss_sft": 0.20246970653533936,
"step": 803,
"token_acc": 0.9332608513841264
},
{
"epoch": 2.8533333333333335,
"grad_norm": 2.801199197769165,
"learning_rate": 6.684949407727015e-08,
"loss": 0.16902843117713928,
"loss_sft": 0.16902843117713928,
"step": 804,
"token_acc": 0.9412232720039782
},
{
"epoch": 2.856888888888889,
"grad_norm": 2.775937557220459,
"learning_rate": 6.371077877179277e-08,
"loss": 0.18457172811031342,
"loss_sft": 0.18457172811031342,
"step": 805,
"token_acc": 0.93581726593712
},
{
"epoch": 2.8604444444444446,
"grad_norm": 2.9067697525024414,
"learning_rate": 6.064705801732962e-08,
"loss": 0.1850523054599762,
"loss_sft": 0.1850523054599762,
"step": 806,
"token_acc": 0.9370332654446707
},
{
"epoch": 2.864,
"grad_norm": 2.657909393310547,
"learning_rate": 5.7658378359443104e-08,
"loss": 0.17264163494110107,
"loss_sft": 0.17264163494110107,
"step": 807,
"token_acc": 0.9411325141149119
},
{
"epoch": 2.8675555555555556,
"grad_norm": 2.8361856937408447,
"learning_rate": 5.4744785203635355e-08,
"loss": 0.1799735128879547,
"loss_sft": 0.1799735128879547,
"step": 808,
"token_acc": 0.9396403290606467
},
{
"epoch": 2.871111111111111,
"grad_norm": 2.8313121795654297,
"learning_rate": 5.190632281465713e-08,
"loss": 0.19137787818908691,
"loss_sft": 0.19137787818908691,
"step": 809,
"token_acc": 0.9330813158529719
},
{
"epoch": 2.8746666666666667,
"grad_norm": 2.651089906692505,
"learning_rate": 4.914303431583389e-08,
"loss": 0.19388604164123535,
"loss_sft": 0.19388604164123535,
"step": 810,
"token_acc": 0.9318434865595674
},
{
"epoch": 2.878222222222222,
"grad_norm": 2.729330539703369,
"learning_rate": 4.645496168841468e-08,
"loss": 0.18136849999427795,
"loss_sft": 0.18136849999427795,
"step": 811,
"token_acc": 0.9400897334389021
},
{
"epoch": 2.8817777777777778,
"grad_norm": 2.9207730293273926,
"learning_rate": 4.384214577092927e-08,
"loss": 0.1663387417793274,
"loss_sft": 0.1663387417793274,
"step": 812,
"token_acc": 0.9428848015488868
},
{
"epoch": 2.8853333333333335,
"grad_norm": 2.8292951583862305,
"learning_rate": 4.130462625857257e-08,
"loss": 0.18162667751312256,
"loss_sft": 0.18162667751312256,
"step": 813,
"token_acc": 0.9367977528089888
},
{
"epoch": 2.888888888888889,
"grad_norm": 3.020155906677246,
"learning_rate": 3.884244170259732e-08,
"loss": 0.18117643892765045,
"loss_sft": 0.18117643892765045,
"step": 814,
"token_acc": 0.9369983948635634
},
{
"epoch": 2.8924444444444446,
"grad_norm": 2.9712772369384766,
"learning_rate": 3.645562950973014e-08,
"loss": 0.22095483541488647,
"loss_sft": 0.22095483541488647,
"step": 815,
"token_acc": 0.9236930117936983
},
{
"epoch": 2.896,
"grad_norm": 2.7452125549316406,
"learning_rate": 3.414422594160527e-08,
"loss": 0.1681252419948578,
"loss_sft": 0.1681252419948578,
"step": 816,
"token_acc": 0.9404548457110503
},
{
"epoch": 2.8995555555555557,
"grad_norm": 2.7096595764160156,
"learning_rate": 3.1908266114210054e-08,
"loss": 0.18273943662643433,
"loss_sft": 0.18273943662643433,
"step": 817,
"token_acc": 0.9357065997130559
},
{
"epoch": 2.903111111111111,
"grad_norm": 2.9406182765960693,
"learning_rate": 2.974778399735423e-08,
"loss": 0.21434247493743896,
"loss_sft": 0.21434247493743896,
"step": 818,
"token_acc": 0.9239790401730017
},
{
"epoch": 2.9066666666666667,
"grad_norm": 2.7685508728027344,
"learning_rate": 2.766281241415203e-08,
"loss": 0.1793900579214096,
"loss_sft": 0.1793900579214096,
"step": 819,
"token_acc": 0.9384469696969697
},
{
"epoch": 2.910222222222222,
"grad_norm": 2.727780342102051,
"learning_rate": 2.5653383040524228e-08,
"loss": 0.20031699538230896,
"loss_sft": 0.20031699538230896,
"step": 820,
"token_acc": 0.9293945788021425
},
{
"epoch": 2.913777777777778,
"grad_norm": 2.7438018321990967,
"learning_rate": 2.3719526404717975e-08,
"loss": 0.1880510300397873,
"loss_sft": 0.1880510300397873,
"step": 821,
"token_acc": 0.937181397537448
},
{
"epoch": 2.9173333333333336,
"grad_norm": 2.5647518634796143,
"learning_rate": 2.1861271886840508e-08,
"loss": 0.18202780187129974,
"loss_sft": 0.18202780187129974,
"step": 822,
"token_acc": 0.9355675835990213
},
{
"epoch": 2.920888888888889,
"grad_norm": 2.702347993850708,
"learning_rate": 2.0078647718415058e-08,
"loss": 0.19995903968811035,
"loss_sft": 0.19995903968811035,
"step": 823,
"token_acc": 0.9315763627434932
},
{
"epoch": 2.924444444444444,
"grad_norm": 2.7479679584503174,
"learning_rate": 1.837168098195119e-08,
"loss": 0.18104644119739532,
"loss_sft": 0.18104644119739532,
"step": 824,
"token_acc": 0.9399856682192762
},
{
"epoch": 2.928,
"grad_norm": 2.707202911376953,
"learning_rate": 1.6740397610533477e-08,
"loss": 0.16002894937992096,
"loss_sft": 0.16002894937992096,
"step": 825,
"token_acc": 0.9443000371333086
},
{
"epoch": 2.9315555555555557,
"grad_norm": 2.6256890296936035,
"learning_rate": 1.5184822387426246e-08,
"loss": 0.1906948983669281,
"loss_sft": 0.1906948983669281,
"step": 826,
"token_acc": 0.9340266751349635
},
{
"epoch": 2.935111111111111,
"grad_norm": 2.681226968765259,
"learning_rate": 1.3704978945698888e-08,
"loss": 0.1949460506439209,
"loss_sft": 0.1949460506439209,
"step": 827,
"token_acc": 0.9310911224090731
},
{
"epoch": 2.9386666666666668,
"grad_norm": 2.974146842956543,
"learning_rate": 1.2300889767866697e-08,
"loss": 0.20629552006721497,
"loss_sft": 0.20629552006721497,
"step": 828,
"token_acc": 0.9255199792871321
},
{
"epoch": 2.942222222222222,
"grad_norm": 2.7185864448547363,
"learning_rate": 1.0972576185547256e-08,
"loss": 0.1920526623725891,
"loss_sft": 0.1920526623725891,
"step": 829,
"token_acc": 0.9345727565248481
},
{
"epoch": 2.945777777777778,
"grad_norm": 2.8740179538726807,
"learning_rate": 9.720058379138474e-09,
"loss": 0.20426131784915924,
"loss_sft": 0.20426131784915924,
"step": 830,
"token_acc": 0.9299219427580225
},
{
"epoch": 2.9493333333333336,
"grad_norm": 2.6651880741119385,
"learning_rate": 8.543355377512164e-09,
"loss": 0.19474563002586365,
"loss_sft": 0.19474563002586365,
"step": 831,
"token_acc": 0.9322077040556592
},
{
"epoch": 2.952888888888889,
"grad_norm": 2.575183391571045,
"learning_rate": 7.442485057722048e-09,
"loss": 0.1748010814189911,
"loss_sft": 0.1748010814189911,
"step": 832,
"token_acc": 0.9400151285930408
},
{
"epoch": 2.956444444444444,
"grad_norm": 2.7950997352600098,
"learning_rate": 6.417464144736208e-09,
"loss": 0.20314106345176697,
"loss_sft": 0.20314106345176697,
"step": 833,
"token_acc": 0.93076792388719
},
{
"epoch": 2.96,
"grad_norm": 3.1104612350463867,
"learning_rate": 5.468308211179496e-09,
"loss": 0.21240943670272827,
"loss_sft": 0.21240943670272827,
"step": 834,
"token_acc": 0.929423170140572
},
{
"epoch": 2.9635555555555557,
"grad_norm": 2.799894094467163,
"learning_rate": 4.595031677099293e-09,
"loss": 0.1968020796775818,
"loss_sft": 0.1968020796775818,
"step": 835,
"token_acc": 0.9320062259359384
},
{
"epoch": 2.967111111111111,
"grad_norm": 2.55328631401062,
"learning_rate": 3.7976478097451196e-09,
"loss": 0.17850297689437866,
"loss_sft": 0.17850297689437866,
"step": 836,
"token_acc": 0.9400752060164813
},
{
"epoch": 2.970666666666667,
"grad_norm": 2.897758960723877,
"learning_rate": 3.0761687233682445e-09,
"loss": 0.18851017951965332,
"loss_sft": 0.18851017951965332,
"step": 837,
"token_acc": 0.9328469719609739
},
{
"epoch": 2.974222222222222,
"grad_norm": 2.586577892303467,
"learning_rate": 2.430605379035722e-09,
"loss": 0.19424137473106384,
"loss_sft": 0.19424137473106384,
"step": 838,
"token_acc": 0.9317819572840293
},
{
"epoch": 2.977777777777778,
"grad_norm": 2.721605062484741,
"learning_rate": 1.8609675844655228e-09,
"loss": 0.18937301635742188,
"loss_sft": 0.18937301635742188,
"step": 839,
"token_acc": 0.9358484559057625
},
{
"epoch": 2.981333333333333,
"grad_norm": 2.817483425140381,
"learning_rate": 1.3672639938777655e-09,
"loss": 0.18685343861579895,
"loss_sft": 0.18685343861579895,
"step": 840,
"token_acc": 0.9351392277277698
},
{
"epoch": 2.984888888888889,
"grad_norm": 2.697528123855591,
"learning_rate": 9.495021078614885e-10,
"loss": 0.1892673373222351,
"loss_sft": 0.1892673373222351,
"step": 841,
"token_acc": 0.934172918899607
},
{
"epoch": 2.9884444444444442,
"grad_norm": 2.8409249782562256,
"learning_rate": 6.076882732625189e-10,
"loss": 0.20392055809497833,
"loss_sft": 0.20392055809497833,
"step": 842,
"token_acc": 0.9311359623307828
},
{
"epoch": 2.992,
"grad_norm": 2.7260000705718994,
"learning_rate": 3.4182768308577143e-10,
"loss": 0.1813220977783203,
"loss_sft": 0.1813220977783203,
"step": 843,
"token_acc": 0.9374895954719494
},
{
"epoch": 2.9955555555555557,
"grad_norm": 2.629594087600708,
"learning_rate": 1.5192437641753376e-10,
"loss": 0.1754237413406372,
"loss_sft": 0.1754237413406372,
"step": 844,
"token_acc": 0.9376775007888924
},
{
"epoch": 2.999111111111111,
"grad_norm": 3.008458137512207,
"learning_rate": 3.798123836162848e-11,
"loss": 0.20497778058052063,
"loss_sft": 0.20497778058052063,
"step": 845,
"token_acc": 0.9318758198513336
},
{
"epoch": 3.0,
"grad_norm": 5.721316814422607,
"learning_rate": 0.0,
"loss": 0.13550405204296112,
"loss_sft": 0.13550405204296112,
"step": 846,
"token_acc": 0.9548813691170751
}
],
"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": 8.61323408896426e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}