jdqqjr's picture
init
6827b89
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 8.0,
"eval_steps": 400,
"global_step": 1712,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04672897196261682,
"grad_norm": 0.8108399564287991,
"learning_rate": 9.999158178436007e-05,
"loss": 1.1115,
"step": 10
},
{
"epoch": 0.09345794392523364,
"grad_norm": 0.19874272531557952,
"learning_rate": 9.996632997209443e-05,
"loss": 0.1905,
"step": 20
},
{
"epoch": 0.14018691588785046,
"grad_norm": 0.12703843810365584,
"learning_rate": 9.992425306621115e-05,
"loss": 0.0789,
"step": 30
},
{
"epoch": 0.18691588785046728,
"grad_norm": 0.10355469925054432,
"learning_rate": 9.986536523520889e-05,
"loss": 0.0924,
"step": 40
},
{
"epoch": 0.2336448598130841,
"grad_norm": 0.1852892166670333,
"learning_rate": 9.978968630830607e-05,
"loss": 0.065,
"step": 50
},
{
"epoch": 0.2803738317757009,
"grad_norm": 0.09323366928517218,
"learning_rate": 9.969724176876373e-05,
"loss": 0.0531,
"step": 60
},
{
"epoch": 0.32710280373831774,
"grad_norm": 0.08829483418959966,
"learning_rate": 9.95880627453046e-05,
"loss": 0.0782,
"step": 70
},
{
"epoch": 0.37383177570093457,
"grad_norm": 0.15247331956402047,
"learning_rate": 9.94621860016312e-05,
"loss": 0.0633,
"step": 80
},
{
"epoch": 0.4205607476635514,
"grad_norm": 0.14725457723867078,
"learning_rate": 9.931965392404641e-05,
"loss": 0.0501,
"step": 90
},
{
"epoch": 0.4672897196261682,
"grad_norm": 0.0922740352810748,
"learning_rate": 9.916051450718084e-05,
"loss": 0.0508,
"step": 100
},
{
"epoch": 0.514018691588785,
"grad_norm": 0.0938141317411703,
"learning_rate": 9.89848213378316e-05,
"loss": 0.0356,
"step": 110
},
{
"epoch": 0.5607476635514018,
"grad_norm": 0.12421407254599728,
"learning_rate": 9.879263357691814e-05,
"loss": 0.0416,
"step": 120
},
{
"epoch": 0.6074766355140186,
"grad_norm": 0.237166690677984,
"learning_rate": 9.858401593956104e-05,
"loss": 0.0407,
"step": 130
},
{
"epoch": 0.6542056074766355,
"grad_norm": 0.08880528008550467,
"learning_rate": 9.83590386732906e-05,
"loss": 0.0338,
"step": 140
},
{
"epoch": 0.7009345794392523,
"grad_norm": 0.11986338307776492,
"learning_rate": 9.811777753439248e-05,
"loss": 0.0341,
"step": 150
},
{
"epoch": 0.7476635514018691,
"grad_norm": 0.08032330195379932,
"learning_rate": 9.786031376239842e-05,
"loss": 0.024,
"step": 160
},
{
"epoch": 0.794392523364486,
"grad_norm": 0.07663400363911589,
"learning_rate": 9.758673405273046e-05,
"loss": 0.0242,
"step": 170
},
{
"epoch": 0.8411214953271028,
"grad_norm": 0.14946426405800525,
"learning_rate": 9.729713052750826e-05,
"loss": 0.0237,
"step": 180
},
{
"epoch": 0.8878504672897196,
"grad_norm": 0.0759393683447156,
"learning_rate": 9.699160070452882e-05,
"loss": 0.0202,
"step": 190
},
{
"epoch": 0.9345794392523364,
"grad_norm": 0.12858143092057014,
"learning_rate": 9.667024746442952e-05,
"loss": 0.019,
"step": 200
},
{
"epoch": 0.9813084112149533,
"grad_norm": 0.10495732835434407,
"learning_rate": 9.633317901604523e-05,
"loss": 0.0183,
"step": 210
},
{
"epoch": 1.02803738317757,
"grad_norm": 0.06980923471022675,
"learning_rate": 9.59805088599713e-05,
"loss": 0.015,
"step": 220
},
{
"epoch": 1.074766355140187,
"grad_norm": 0.06932653950015637,
"learning_rate": 9.561235575034469e-05,
"loss": 0.0131,
"step": 230
},
{
"epoch": 1.1214953271028036,
"grad_norm": 0.058548744277500064,
"learning_rate": 9.522884365485598e-05,
"loss": 0.0118,
"step": 240
},
{
"epoch": 1.1682242990654206,
"grad_norm": 0.03726047410080607,
"learning_rate": 9.483010171300602e-05,
"loss": 0.0135,
"step": 250
},
{
"epoch": 1.2149532710280373,
"grad_norm": 0.05545270603496205,
"learning_rate": 9.441626419262084e-05,
"loss": 0.0121,
"step": 260
},
{
"epoch": 1.2616822429906542,
"grad_norm": 0.05893605931281947,
"learning_rate": 9.398747044463992e-05,
"loss": 0.012,
"step": 270
},
{
"epoch": 1.308411214953271,
"grad_norm": 0.052627420151130956,
"learning_rate": 9.354386485619264e-05,
"loss": 0.0136,
"step": 280
},
{
"epoch": 1.355140186915888,
"grad_norm": 0.051128701449751776,
"learning_rate": 9.308559680197914e-05,
"loss": 0.0137,
"step": 290
},
{
"epoch": 1.4018691588785046,
"grad_norm": 0.09981668657519205,
"learning_rate": 9.261282059397145e-05,
"loss": 0.0116,
"step": 300
},
{
"epoch": 1.4485981308411215,
"grad_norm": 0.04456617572462679,
"learning_rate": 9.212569542945234e-05,
"loss": 0.0128,
"step": 310
},
{
"epoch": 1.4953271028037383,
"grad_norm": 0.06065053274675731,
"learning_rate": 9.162438533740892e-05,
"loss": 0.0113,
"step": 320
},
{
"epoch": 1.542056074766355,
"grad_norm": 0.08108720647504085,
"learning_rate": 9.110905912329949e-05,
"loss": 0.012,
"step": 330
},
{
"epoch": 1.588785046728972,
"grad_norm": 0.04189447531626791,
"learning_rate": 9.057989031221188e-05,
"loss": 0.0112,
"step": 340
},
{
"epoch": 1.6355140186915889,
"grad_norm": 0.06292392241512725,
"learning_rate": 9.003705709043253e-05,
"loss": 0.0101,
"step": 350
},
{
"epoch": 1.6822429906542056,
"grad_norm": 0.16839424797262825,
"learning_rate": 8.948074224544614e-05,
"loss": 0.0113,
"step": 360
},
{
"epoch": 1.7289719626168223,
"grad_norm": 0.03631611864356132,
"learning_rate": 8.891113310438587e-05,
"loss": 0.0117,
"step": 370
},
{
"epoch": 1.7757009345794392,
"grad_norm": 0.02998579815737736,
"learning_rate": 8.832842147095495e-05,
"loss": 0.0109,
"step": 380
},
{
"epoch": 1.8224299065420562,
"grad_norm": 0.04479035362698208,
"learning_rate": 8.773280356084077e-05,
"loss": 0.0098,
"step": 390
},
{
"epoch": 1.8691588785046729,
"grad_norm": 0.05286540264614212,
"learning_rate": 8.712447993564361e-05,
"loss": 0.0109,
"step": 400
},
{
"epoch": 1.8691588785046729,
"eval_loss": 0.01072569377720356,
"eval_runtime": 3.437,
"eval_samples_per_second": 70.993,
"eval_steps_per_second": 2.037,
"step": 400
},
{
"epoch": 1.9158878504672896,
"grad_norm": 0.05068443093268522,
"learning_rate": 8.650365543534168e-05,
"loss": 0.01,
"step": 410
},
{
"epoch": 1.9626168224299065,
"grad_norm": 0.034340227946281374,
"learning_rate": 8.587053910931575e-05,
"loss": 0.0093,
"step": 420
},
{
"epoch": 2.0093457943925235,
"grad_norm": 0.035620502174738417,
"learning_rate": 8.522534414595609e-05,
"loss": 0.0091,
"step": 430
},
{
"epoch": 2.05607476635514,
"grad_norm": 0.03143044172853539,
"learning_rate": 8.456828780087598e-05,
"loss": 0.008,
"step": 440
},
{
"epoch": 2.102803738317757,
"grad_norm": 0.05199083018672625,
"learning_rate": 8.38995913237554e-05,
"loss": 0.0078,
"step": 450
},
{
"epoch": 2.149532710280374,
"grad_norm": 0.03405283371576451,
"learning_rate": 8.321947988384005e-05,
"loss": 0.0089,
"step": 460
},
{
"epoch": 2.196261682242991,
"grad_norm": 0.050263061552753986,
"learning_rate": 8.252818249412037e-05,
"loss": 0.0084,
"step": 470
},
{
"epoch": 2.2429906542056073,
"grad_norm": 0.032967778518522166,
"learning_rate": 8.182593193421625e-05,
"loss": 0.0082,
"step": 480
},
{
"epoch": 2.289719626168224,
"grad_norm": 0.035321987909812884,
"learning_rate": 8.111296467199356e-05,
"loss": 0.008,
"step": 490
},
{
"epoch": 2.336448598130841,
"grad_norm": 0.029919394297831105,
"learning_rate": 8.038952078393862e-05,
"loss": 0.008,
"step": 500
},
{
"epoch": 2.383177570093458,
"grad_norm": 0.037839397396561614,
"learning_rate": 7.96558438743175e-05,
"loss": 0.0081,
"step": 510
},
{
"epoch": 2.4299065420560746,
"grad_norm": 0.03447748806162398,
"learning_rate": 7.89121809931477e-05,
"loss": 0.0082,
"step": 520
},
{
"epoch": 2.4766355140186915,
"grad_norm": 0.041992049585078906,
"learning_rate": 7.815878255300901e-05,
"loss": 0.0082,
"step": 530
},
{
"epoch": 2.5233644859813085,
"grad_norm": 0.034025872963113706,
"learning_rate": 7.739590224472275e-05,
"loss": 0.0078,
"step": 540
},
{
"epoch": 2.5700934579439254,
"grad_norm": 0.028042564915330366,
"learning_rate": 7.662379695192663e-05,
"loss": 0.0078,
"step": 550
},
{
"epoch": 2.616822429906542,
"grad_norm": 0.041319137226956335,
"learning_rate": 7.58427266645747e-05,
"loss": 0.0078,
"step": 560
},
{
"epoch": 2.663551401869159,
"grad_norm": 0.03522762783198424,
"learning_rate": 7.505295439139133e-05,
"loss": 0.0079,
"step": 570
},
{
"epoch": 2.710280373831776,
"grad_norm": 0.033750692975752054,
"learning_rate": 7.425474607130858e-05,
"loss": 0.0076,
"step": 580
},
{
"epoch": 2.7570093457943923,
"grad_norm": 0.030318374748605468,
"learning_rate": 7.344837048391702e-05,
"loss": 0.0081,
"step": 590
},
{
"epoch": 2.803738317757009,
"grad_norm": 0.0269893097675881,
"learning_rate": 7.263409915895992e-05,
"loss": 0.0075,
"step": 600
},
{
"epoch": 2.850467289719626,
"grad_norm": 0.080809904610704,
"learning_rate": 7.181220628490135e-05,
"loss": 0.0072,
"step": 610
},
{
"epoch": 2.897196261682243,
"grad_norm": 0.0479816994697567,
"learning_rate": 7.098296861659925e-05,
"loss": 0.0078,
"step": 620
},
{
"epoch": 2.94392523364486,
"grad_norm": 0.024727139983622894,
"learning_rate": 7.014666538211391e-05,
"loss": 0.0075,
"step": 630
},
{
"epoch": 2.9906542056074765,
"grad_norm": 0.035477577514644594,
"learning_rate": 6.930357818868409e-05,
"loss": 0.0076,
"step": 640
},
{
"epoch": 3.0373831775700935,
"grad_norm": 0.03489962140746076,
"learning_rate": 6.845399092790168e-05,
"loss": 0.0067,
"step": 650
},
{
"epoch": 3.0841121495327104,
"grad_norm": 0.028458398477187792,
"learning_rate": 6.75981896801173e-05,
"loss": 0.0069,
"step": 660
},
{
"epoch": 3.130841121495327,
"grad_norm": 0.029455198773575362,
"learning_rate": 6.673646261810897e-05,
"loss": 0.0073,
"step": 670
},
{
"epoch": 3.177570093457944,
"grad_norm": 0.021331692675162994,
"learning_rate": 6.586909991004586e-05,
"loss": 0.0063,
"step": 680
},
{
"epoch": 3.2242990654205608,
"grad_norm": 0.030623459310690392,
"learning_rate": 6.499639362178057e-05,
"loss": 0.0074,
"step": 690
},
{
"epoch": 3.2710280373831777,
"grad_norm": 0.033777185932132234,
"learning_rate": 6.41186376185021e-05,
"loss": 0.0073,
"step": 700
},
{
"epoch": 3.317757009345794,
"grad_norm": 0.03878683167456107,
"learning_rate": 6.323612746578306e-05,
"loss": 0.0071,
"step": 710
},
{
"epoch": 3.364485981308411,
"grad_norm": 0.0326477494007077,
"learning_rate": 6.234916033005421e-05,
"loss": 0.0073,
"step": 720
},
{
"epoch": 3.411214953271028,
"grad_norm": 0.03373929782607459,
"learning_rate": 6.145803487854015e-05,
"loss": 0.0067,
"step": 730
},
{
"epoch": 3.457943925233645,
"grad_norm": 0.023188946944247713,
"learning_rate": 6.056305117868939e-05,
"loss": 0.0065,
"step": 740
},
{
"epoch": 3.5046728971962615,
"grad_norm": 0.023508542077636894,
"learning_rate": 5.966451059713311e-05,
"loss": 0.0065,
"step": 750
},
{
"epoch": 3.5514018691588785,
"grad_norm": 0.02989633738957656,
"learning_rate": 5.8762715698206385e-05,
"loss": 0.0068,
"step": 760
},
{
"epoch": 3.5981308411214954,
"grad_norm": 0.02608338564335597,
"learning_rate": 5.78579701420661e-05,
"loss": 0.0065,
"step": 770
},
{
"epoch": 3.6448598130841123,
"grad_norm": 0.02843620648372604,
"learning_rate": 5.695057858243988e-05,
"loss": 0.0069,
"step": 780
},
{
"epoch": 3.691588785046729,
"grad_norm": 0.026903369236640558,
"learning_rate": 5.604084656404052e-05,
"loss": 0.0064,
"step": 790
},
{
"epoch": 3.7383177570093458,
"grad_norm": 0.03087535604554872,
"learning_rate": 5.512908041968018e-05,
"loss": 0.0068,
"step": 800
},
{
"epoch": 3.7383177570093458,
"eval_loss": 0.007949975319206715,
"eval_runtime": 3.3099,
"eval_samples_per_second": 73.719,
"eval_steps_per_second": 2.115,
"step": 800
},
{
"epoch": 3.7850467289719627,
"grad_norm": 0.027092999369925153,
"learning_rate": 5.4215587167119554e-05,
"loss": 0.0066,
"step": 810
},
{
"epoch": 3.831775700934579,
"grad_norm": 0.022172386959882197,
"learning_rate": 5.330067440568605e-05,
"loss": 0.0063,
"step": 820
},
{
"epoch": 3.878504672897196,
"grad_norm": 0.02880464575204624,
"learning_rate": 5.238465021269637e-05,
"loss": 0.0072,
"step": 830
},
{
"epoch": 3.925233644859813,
"grad_norm": 0.026027691897489803,
"learning_rate": 5.1467823039718044e-05,
"loss": 0.0068,
"step": 840
},
{
"epoch": 3.97196261682243,
"grad_norm": 0.027074409669940073,
"learning_rate": 5.0550501608704926e-05,
"loss": 0.0064,
"step": 850
},
{
"epoch": 4.018691588785047,
"grad_norm": 0.022224303104632814,
"learning_rate": 4.9632994808041724e-05,
"loss": 0.0066,
"step": 860
},
{
"epoch": 4.065420560747664,
"grad_norm": 0.029720109312162384,
"learning_rate": 4.871561158853241e-05,
"loss": 0.0056,
"step": 870
},
{
"epoch": 4.11214953271028,
"grad_norm": 0.021302383660758806,
"learning_rate": 4.7798660859367615e-05,
"loss": 0.006,
"step": 880
},
{
"epoch": 4.158878504672897,
"grad_norm": 0.019221061294488036,
"learning_rate": 4.688245138410612e-05,
"loss": 0.006,
"step": 890
},
{
"epoch": 4.205607476635514,
"grad_norm": 0.025917557422674823,
"learning_rate": 4.5967291676705286e-05,
"loss": 0.0063,
"step": 900
},
{
"epoch": 4.252336448598131,
"grad_norm": 0.020256663017665554,
"learning_rate": 4.5053489897635585e-05,
"loss": 0.0066,
"step": 910
},
{
"epoch": 4.299065420560748,
"grad_norm": 0.017745913007780147,
"learning_rate": 4.414135375011416e-05,
"loss": 0.0063,
"step": 920
},
{
"epoch": 4.345794392523365,
"grad_norm": 0.02011370658432465,
"learning_rate": 4.323119037649232e-05,
"loss": 0.0057,
"step": 930
},
{
"epoch": 4.392523364485982,
"grad_norm": 0.024844301511041878,
"learning_rate": 4.2323306254831934e-05,
"loss": 0.0064,
"step": 940
},
{
"epoch": 4.4392523364485985,
"grad_norm": 0.022813937850513024,
"learning_rate": 4.14180070957055e-05,
"loss": 0.0067,
"step": 950
},
{
"epoch": 4.485981308411215,
"grad_norm": 0.020230098188474713,
"learning_rate": 4.051559773925462e-05,
"loss": 0.0062,
"step": 960
},
{
"epoch": 4.5327102803738315,
"grad_norm": 0.023811479597101767,
"learning_rate": 3.961638205254161e-05,
"loss": 0.0061,
"step": 970
},
{
"epoch": 4.579439252336448,
"grad_norm": 0.02740260066991885,
"learning_rate": 3.872066282722877e-05,
"loss": 0.0064,
"step": 980
},
{
"epoch": 4.626168224299065,
"grad_norm": 0.02317216889617009,
"learning_rate": 3.782874167761977e-05,
"loss": 0.006,
"step": 990
},
{
"epoch": 4.672897196261682,
"grad_norm": 0.012363068654073208,
"learning_rate": 3.694091893909746e-05,
"loss": 0.0062,
"step": 1000
},
{
"epoch": 4.719626168224299,
"grad_norm": 0.03140973721401593,
"learning_rate": 3.605749356699235e-05,
"loss": 0.0066,
"step": 1010
},
{
"epoch": 4.766355140186916,
"grad_norm": 0.02856466113066483,
"learning_rate": 3.5178763035915804e-05,
"loss": 0.0058,
"step": 1020
},
{
"epoch": 4.813084112149532,
"grad_norm": 0.023964636381919646,
"learning_rate": 3.430502323959185e-05,
"loss": 0.0061,
"step": 1030
},
{
"epoch": 4.859813084112149,
"grad_norm": 0.025128922566591057,
"learning_rate": 3.343656839122121e-05,
"loss": 0.0059,
"step": 1040
},
{
"epoch": 4.906542056074766,
"grad_norm": 0.0186144001471827,
"learning_rate": 3.257369092441137e-05,
"loss": 0.0064,
"step": 1050
},
{
"epoch": 4.953271028037383,
"grad_norm": 0.024922675182496204,
"learning_rate": 3.171668139470578e-05,
"loss": 0.0067,
"step": 1060
},
{
"epoch": 5.0,
"grad_norm": 0.023268694442631153,
"learning_rate": 3.086582838174551e-05,
"loss": 0.0066,
"step": 1070
},
{
"epoch": 5.046728971962617,
"grad_norm": 0.022858435420561946,
"learning_rate": 3.0021418392096213e-05,
"loss": 0.0056,
"step": 1080
},
{
"epoch": 5.093457943925234,
"grad_norm": 0.03037878789593789,
"learning_rate": 2.9183735762773124e-05,
"loss": 0.0058,
"step": 1090
},
{
"epoch": 5.140186915887851,
"grad_norm": 0.027831633677455617,
"learning_rate": 2.8353062565496713e-05,
"loss": 0.0057,
"step": 1100
},
{
"epoch": 5.186915887850467,
"grad_norm": 0.02234327350971588,
"learning_rate": 2.7529678511711036e-05,
"loss": 0.0058,
"step": 1110
},
{
"epoch": 5.233644859813084,
"grad_norm": 0.02420276983311316,
"learning_rate": 2.671386085839682e-05,
"loss": 0.0056,
"step": 1120
},
{
"epoch": 5.280373831775701,
"grad_norm": 0.026279801398446302,
"learning_rate": 2.5905884314711238e-05,
"loss": 0.0059,
"step": 1130
},
{
"epoch": 5.327102803738318,
"grad_norm": 0.024113471920291712,
"learning_rate": 2.5106020949485348e-05,
"loss": 0.0055,
"step": 1140
},
{
"epoch": 5.373831775700935,
"grad_norm": 0.02757405539371541,
"learning_rate": 2.4314540099610812e-05,
"loss": 0.0064,
"step": 1150
},
{
"epoch": 5.420560747663552,
"grad_norm": 0.023071247903649506,
"learning_rate": 2.353170827934635e-05,
"loss": 0.006,
"step": 1160
},
{
"epoch": 5.4672897196261685,
"grad_norm": 0.0288737767028276,
"learning_rate": 2.27577890905749e-05,
"loss": 0.0058,
"step": 1170
},
{
"epoch": 5.5140186915887845,
"grad_norm": 0.029369556474304507,
"learning_rate": 2.1993043134041214e-05,
"loss": 0.0058,
"step": 1180
},
{
"epoch": 5.5607476635514015,
"grad_norm": 0.021962327780855752,
"learning_rate": 2.1237727921600194e-05,
"loss": 0.0062,
"step": 1190
},
{
"epoch": 5.607476635514018,
"grad_norm": 0.027762289058911312,
"learning_rate": 2.0492097789505178e-05,
"loss": 0.0059,
"step": 1200
},
{
"epoch": 5.607476635514018,
"eval_loss": 0.007473748177289963,
"eval_runtime": 3.3345,
"eval_samples_per_second": 73.174,
"eval_steps_per_second": 2.099,
"step": 1200
},
{
"epoch": 5.654205607476635,
"grad_norm": 0.02469372448375762,
"learning_rate": 1.9756403812765763e-05,
"loss": 0.0056,
"step": 1210
},
{
"epoch": 5.700934579439252,
"grad_norm": 0.025899196843355222,
"learning_rate": 1.9030893720603605e-05,
"loss": 0.0058,
"step": 1220
},
{
"epoch": 5.747663551401869,
"grad_norm": 0.0265685769061,
"learning_rate": 1.831581181303489e-05,
"loss": 0.006,
"step": 1230
},
{
"epoch": 5.794392523364486,
"grad_norm": 0.0236407517735495,
"learning_rate": 1.7611398878607544e-05,
"loss": 0.0056,
"step": 1240
},
{
"epoch": 5.841121495327103,
"grad_norm": 0.028633962957475903,
"learning_rate": 1.6917892113320826e-05,
"loss": 0.0056,
"step": 1250
},
{
"epoch": 5.88785046728972,
"grad_norm": 0.025134468654909488,
"learning_rate": 1.6235525040754667e-05,
"loss": 0.0059,
"step": 1260
},
{
"epoch": 5.934579439252336,
"grad_norm": 0.018750521917658348,
"learning_rate": 1.5564527433435565e-05,
"loss": 0.0054,
"step": 1270
},
{
"epoch": 5.981308411214953,
"grad_norm": 0.02286869577795185,
"learning_rate": 1.4905125235465589e-05,
"loss": 0.0056,
"step": 1280
},
{
"epoch": 6.02803738317757,
"grad_norm": 0.02111534299450912,
"learning_rate": 1.4257540486440596e-05,
"loss": 0.0052,
"step": 1290
},
{
"epoch": 6.074766355140187,
"grad_norm": 0.020056682398156774,
"learning_rate": 1.362199124668309e-05,
"loss": 0.0052,
"step": 1300
},
{
"epoch": 6.121495327102804,
"grad_norm": 0.023191711564030777,
"learning_rate": 1.2998691523815043e-05,
"loss": 0.0053,
"step": 1310
},
{
"epoch": 6.168224299065421,
"grad_norm": 0.026466885727658793,
"learning_rate": 1.23878512006955e-05,
"loss": 0.0051,
"step": 1320
},
{
"epoch": 6.214953271028038,
"grad_norm": 0.02292053543721861,
"learning_rate": 1.178967596474691e-05,
"loss": 0.0048,
"step": 1330
},
{
"epoch": 6.261682242990654,
"grad_norm": 0.020924799561897002,
"learning_rate": 1.1204367238694358e-05,
"loss": 0.0056,
"step": 1340
},
{
"epoch": 6.308411214953271,
"grad_norm": 0.02136313146143031,
"learning_rate": 1.06321221127407e-05,
"loss": 0.0051,
"step": 1350
},
{
"epoch": 6.355140186915888,
"grad_norm": 0.023684372680794725,
"learning_rate": 1.0073133278200703e-05,
"loss": 0.0053,
"step": 1360
},
{
"epoch": 6.401869158878505,
"grad_norm": 0.02285715840701787,
"learning_rate": 9.527588962616352e-06,
"loss": 0.0058,
"step": 1370
},
{
"epoch": 6.4485981308411215,
"grad_norm": 0.026613967435669312,
"learning_rate": 8.995672866375237e-06,
"loss": 0.0058,
"step": 1380
},
{
"epoch": 6.4953271028037385,
"grad_norm": 0.02754174252021364,
"learning_rate": 8.47756410085338e-06,
"loss": 0.0055,
"step": 1390
},
{
"epoch": 6.542056074766355,
"grad_norm": 0.02170488478202453,
"learning_rate": 7.973437128103306e-06,
"loss": 0.0055,
"step": 1400
},
{
"epoch": 6.588785046728972,
"grad_norm": 0.02730782213919483,
"learning_rate": 7.4834617021076695e-06,
"loss": 0.0055,
"step": 1410
},
{
"epoch": 6.635514018691588,
"grad_norm": 0.02739226459077928,
"learning_rate": 7.007802811618258e-06,
"loss": 0.0056,
"step": 1420
},
{
"epoch": 6.682242990654205,
"grad_norm": 0.021614604070826122,
"learning_rate": 6.546620624599497e-06,
"loss": 0.0055,
"step": 1430
},
{
"epoch": 6.728971962616822,
"grad_norm": 0.02122813708773405,
"learning_rate": 6.100070434295379e-06,
"loss": 0.0055,
"step": 1440
},
{
"epoch": 6.775700934579439,
"grad_norm": 0.023837078920304434,
"learning_rate": 5.6683026069377535e-06,
"loss": 0.0052,
"step": 1450
},
{
"epoch": 6.822429906542056,
"grad_norm": 0.01936888685997085,
"learning_rate": 5.251462531113704e-06,
"loss": 0.0053,
"step": 1460
},
{
"epoch": 6.869158878504673,
"grad_norm": 0.027794129758199313,
"learning_rate": 4.84969056880904e-06,
"loss": 0.0056,
"step": 1470
},
{
"epoch": 6.91588785046729,
"grad_norm": 0.02333499254194896,
"learning_rate": 4.4631220081444495e-06,
"loss": 0.0052,
"step": 1480
},
{
"epoch": 6.962616822429906,
"grad_norm": 0.022923436069210098,
"learning_rate": 4.091887017820051e-06,
"loss": 0.0056,
"step": 1490
},
{
"epoch": 7.009345794392523,
"grad_norm": 0.0225875119677355,
"learning_rate": 3.7361106032839264e-06,
"loss": 0.0055,
"step": 1500
},
{
"epoch": 7.05607476635514,
"grad_norm": 0.01938821336041444,
"learning_rate": 3.3959125646391476e-06,
"loss": 0.0052,
"step": 1510
},
{
"epoch": 7.102803738317757,
"grad_norm": 0.03088082827099114,
"learning_rate": 3.0714074563037044e-06,
"loss": 0.0051,
"step": 1520
},
{
"epoch": 7.149532710280374,
"grad_norm": 0.021927820725786244,
"learning_rate": 2.7627045484367232e-06,
"loss": 0.005,
"step": 1530
},
{
"epoch": 7.196261682242991,
"grad_norm": 0.02690920708848648,
"learning_rate": 2.4699077901440883e-06,
"loss": 0.0049,
"step": 1540
},
{
"epoch": 7.242990654205608,
"grad_norm": 0.025630634718450354,
"learning_rate": 2.193115774475807e-06,
"loss": 0.0048,
"step": 1550
},
{
"epoch": 7.289719626168225,
"grad_norm": 0.025670435693965685,
"learning_rate": 1.9324217052268835e-06,
"loss": 0.005,
"step": 1560
},
{
"epoch": 7.336448598130841,
"grad_norm": 0.025967263979272275,
"learning_rate": 1.6879133655529622e-06,
"loss": 0.0052,
"step": 1570
},
{
"epoch": 7.383177570093458,
"grad_norm": 0.018831317858244562,
"learning_rate": 1.4596730884112008e-06,
"loss": 0.0052,
"step": 1580
},
{
"epoch": 7.429906542056075,
"grad_norm": 0.025607378308549714,
"learning_rate": 1.2477777288364355e-06,
"loss": 0.0051,
"step": 1590
},
{
"epoch": 7.4766355140186915,
"grad_norm": 0.025037526298953592,
"learning_rate": 1.0522986380618605e-06,
"loss": 0.0053,
"step": 1600
},
{
"epoch": 7.4766355140186915,
"eval_loss": 0.007368447724729776,
"eval_runtime": 3.3243,
"eval_samples_per_second": 73.4,
"eval_steps_per_second": 2.106,
"step": 1600
},
{
"epoch": 7.5233644859813085,
"grad_norm": 0.029090444318040674,
"learning_rate": 8.733016394930571e-07,
"loss": 0.0049,
"step": 1610
},
{
"epoch": 7.570093457943925,
"grad_norm": 0.02787466565009468,
"learning_rate": 7.108470065433193e-07,
"loss": 0.0054,
"step": 1620
},
{
"epoch": 7.616822429906542,
"grad_norm": 0.025684874852680496,
"learning_rate": 5.649894423379376e-07,
"loss": 0.0053,
"step": 1630
},
{
"epoch": 7.663551401869158,
"grad_norm": 0.02409464783376482,
"learning_rate": 4.357780612940343e-07,
"loss": 0.005,
"step": 1640
},
{
"epoch": 7.710280373831775,
"grad_norm": 0.024379690680637514,
"learning_rate": 3.232563725823645e-07,
"loss": 0.005,
"step": 1650
},
{
"epoch": 7.757009345794392,
"grad_norm": 0.027633151279458295,
"learning_rate": 2.274622654765135e-07,
"loss": 0.0052,
"step": 1660
},
{
"epoch": 7.803738317757009,
"grad_norm": 0.0303886393499085,
"learning_rate": 1.484279965945079e-07,
"loss": 0.0054,
"step": 1670
},
{
"epoch": 7.850467289719626,
"grad_norm": 0.028466293354854937,
"learning_rate": 8.618017903708197e-08,
"loss": 0.0053,
"step": 1680
},
{
"epoch": 7.897196261682243,
"grad_norm": 0.021774487133470112,
"learning_rate": 4.073977342629598e-08,
"loss": 0.0052,
"step": 1690
},
{
"epoch": 7.94392523364486,
"grad_norm": 0.021060893164801635,
"learning_rate": 1.2122080847470906e-08,
"loss": 0.0051,
"step": 1700
},
{
"epoch": 7.990654205607477,
"grad_norm": 0.02474060027341605,
"learning_rate": 3.367376968932412e-10,
"loss": 0.0051,
"step": 1710
},
{
"epoch": 8.0,
"step": 1712,
"total_flos": 30760457011200.0,
"train_loss": 0.01869776981195139,
"train_runtime": 4171.5414,
"train_samples_per_second": 22.925,
"train_steps_per_second": 0.41
}
],
"logging_steps": 10,
"max_steps": 1712,
"num_input_tokens_seen": 0,
"num_train_epochs": 8,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 30760457011200.0,
"train_batch_size": 14,
"trial_name": null,
"trial_params": null
}