chansung's picture
Model save
dd3aafc verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9992193598750976,
"eval_steps": 500,
"global_step": 640,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00156128024980484,
"grad_norm": 4.341845512390137,
"learning_rate": 3.125e-06,
"loss": 1.3019,
"step": 1
},
{
"epoch": 0.0078064012490242,
"grad_norm": 4.292470455169678,
"learning_rate": 1.5625e-05,
"loss": 1.2557,
"step": 5
},
{
"epoch": 0.0156128024980484,
"grad_norm": 3.9200799465179443,
"learning_rate": 3.125e-05,
"loss": 1.1486,
"step": 10
},
{
"epoch": 0.0234192037470726,
"grad_norm": 3.1646339893341064,
"learning_rate": 4.6875e-05,
"loss": 0.9901,
"step": 15
},
{
"epoch": 0.0312256049960968,
"grad_norm": 1.539257526397705,
"learning_rate": 6.25e-05,
"loss": 0.8708,
"step": 20
},
{
"epoch": 0.039032006245121,
"grad_norm": 0.8220561742782593,
"learning_rate": 7.8125e-05,
"loss": 0.7545,
"step": 25
},
{
"epoch": 0.0468384074941452,
"grad_norm": 0.6788249015808105,
"learning_rate": 9.375e-05,
"loss": 0.6926,
"step": 30
},
{
"epoch": 0.0546448087431694,
"grad_norm": 0.8923994302749634,
"learning_rate": 0.000109375,
"loss": 0.6369,
"step": 35
},
{
"epoch": 0.0624512099921936,
"grad_norm": 0.9089710712432861,
"learning_rate": 0.000125,
"loss": 0.614,
"step": 40
},
{
"epoch": 0.0702576112412178,
"grad_norm": 0.9337915778160095,
"learning_rate": 0.00014062500000000002,
"loss": 0.5766,
"step": 45
},
{
"epoch": 0.078064012490242,
"grad_norm": 0.7455429434776306,
"learning_rate": 0.00015625,
"loss": 0.5516,
"step": 50
},
{
"epoch": 0.0858704137392662,
"grad_norm": 0.49435052275657654,
"learning_rate": 0.00017187500000000002,
"loss": 0.536,
"step": 55
},
{
"epoch": 0.0936768149882904,
"grad_norm": 0.9767287969589233,
"learning_rate": 0.0001875,
"loss": 0.5417,
"step": 60
},
{
"epoch": 0.1014832162373146,
"grad_norm": 0.8675209879875183,
"learning_rate": 0.00019999851261394218,
"loss": 0.5344,
"step": 65
},
{
"epoch": 0.1092896174863388,
"grad_norm": 0.6101036667823792,
"learning_rate": 0.00019994645874763658,
"loss": 0.5209,
"step": 70
},
{
"epoch": 0.117096018735363,
"grad_norm": 0.4980247914791107,
"learning_rate": 0.00019982007981886847,
"loss": 0.5043,
"step": 75
},
{
"epoch": 0.1249024199843872,
"grad_norm": 0.34284278750419617,
"learning_rate": 0.00019961946980917456,
"loss": 0.5064,
"step": 80
},
{
"epoch": 0.1327088212334114,
"grad_norm": 0.3613043427467346,
"learning_rate": 0.00019934477790194445,
"loss": 0.496,
"step": 85
},
{
"epoch": 0.1405152224824356,
"grad_norm": 0.40784284472465515,
"learning_rate": 0.00019899620837148077,
"loss": 0.4996,
"step": 90
},
{
"epoch": 0.1483216237314598,
"grad_norm": 0.3546779453754425,
"learning_rate": 0.0001985740204310909,
"loss": 0.4969,
"step": 95
},
{
"epoch": 0.156128024980484,
"grad_norm": 0.35221970081329346,
"learning_rate": 0.00019807852804032305,
"loss": 0.5014,
"step": 100
},
{
"epoch": 0.16393442622950818,
"grad_norm": 0.45173051953315735,
"learning_rate": 0.00019751009967149087,
"loss": 0.4895,
"step": 105
},
{
"epoch": 0.1717408274785324,
"grad_norm": 0.3956238329410553,
"learning_rate": 0.00019686915803565934,
"loss": 0.4877,
"step": 110
},
{
"epoch": 0.1795472287275566,
"grad_norm": 0.3475348949432373,
"learning_rate": 0.0001961561797682962,
"loss": 0.4746,
"step": 115
},
{
"epoch": 0.1873536299765808,
"grad_norm": 0.36577436327934265,
"learning_rate": 0.0001953716950748227,
"loss": 0.4888,
"step": 120
},
{
"epoch": 0.195160031225605,
"grad_norm": 0.3727911412715912,
"learning_rate": 0.0001945162873363268,
"loss": 0.4785,
"step": 125
},
{
"epoch": 0.2029664324746292,
"grad_norm": 0.3759865462779999,
"learning_rate": 0.0001935905926757326,
"loss": 0.4798,
"step": 130
},
{
"epoch": 0.2107728337236534,
"grad_norm": 0.40641459822654724,
"learning_rate": 0.00019259529948474833,
"loss": 0.4795,
"step": 135
},
{
"epoch": 0.2185792349726776,
"grad_norm": 0.3547082543373108,
"learning_rate": 0.00019153114791194473,
"loss": 0.4809,
"step": 140
},
{
"epoch": 0.2263856362217018,
"grad_norm": 0.3588828146457672,
"learning_rate": 0.00019039892931234435,
"loss": 0.4933,
"step": 145
},
{
"epoch": 0.234192037470726,
"grad_norm": 0.4158973693847656,
"learning_rate": 0.00018919948565893142,
"loss": 0.4746,
"step": 150
},
{
"epoch": 0.2419984387197502,
"grad_norm": 0.5559388399124146,
"learning_rate": 0.00018793370891651972,
"loss": 0.4791,
"step": 155
},
{
"epoch": 0.2498048399687744,
"grad_norm": 0.3747630715370178,
"learning_rate": 0.00018660254037844388,
"loss": 0.4864,
"step": 160
},
{
"epoch": 0.2576112412177986,
"grad_norm": 0.4632702171802521,
"learning_rate": 0.00018520696996656788,
"loss": 0.4658,
"step": 165
},
{
"epoch": 0.2654176424668228,
"grad_norm": 0.5524964332580566,
"learning_rate": 0.0001837480354951308,
"loss": 0.4561,
"step": 170
},
{
"epoch": 0.273224043715847,
"grad_norm": 0.4398689866065979,
"learning_rate": 0.00018222682189897752,
"loss": 0.4633,
"step": 175
},
{
"epoch": 0.2810304449648712,
"grad_norm": 0.3512100577354431,
"learning_rate": 0.00018064446042674828,
"loss": 0.4641,
"step": 180
},
{
"epoch": 0.2888368462138954,
"grad_norm": 0.414985716342926,
"learning_rate": 0.0001790021277996269,
"loss": 0.4664,
"step": 185
},
{
"epoch": 0.2966432474629196,
"grad_norm": 0.3512822687625885,
"learning_rate": 0.0001773010453362737,
"loss": 0.4753,
"step": 190
},
{
"epoch": 0.3044496487119438,
"grad_norm": 0.3715687692165375,
"learning_rate": 0.00017554247804459316,
"loss": 0.4667,
"step": 195
},
{
"epoch": 0.312256049960968,
"grad_norm": 0.4506210684776306,
"learning_rate": 0.0001737277336810124,
"loss": 0.4694,
"step": 200
},
{
"epoch": 0.3200624512099922,
"grad_norm": 0.36245930194854736,
"learning_rate": 0.0001718581617779698,
"loss": 0.4634,
"step": 205
},
{
"epoch": 0.32786885245901637,
"grad_norm": 0.37324413657188416,
"learning_rate": 0.00016993515264033672,
"loss": 0.4739,
"step": 210
},
{
"epoch": 0.3356752537080406,
"grad_norm": 0.3662189841270447,
"learning_rate": 0.00016796013631151897,
"loss": 0.4548,
"step": 215
},
{
"epoch": 0.3434816549570648,
"grad_norm": 0.3321515917778015,
"learning_rate": 0.00016593458151000688,
"loss": 0.4533,
"step": 220
},
{
"epoch": 0.351288056206089,
"grad_norm": 0.32358649373054504,
"learning_rate": 0.00016385999453716454,
"loss": 0.448,
"step": 225
},
{
"epoch": 0.3590944574551132,
"grad_norm": 0.41246530413627625,
"learning_rate": 0.00016173791815707051,
"loss": 0.46,
"step": 230
},
{
"epoch": 0.3669008587041374,
"grad_norm": 0.3504014313220978,
"learning_rate": 0.00015956993044924334,
"loss": 0.4524,
"step": 235
},
{
"epoch": 0.3747072599531616,
"grad_norm": 0.6306522488594055,
"learning_rate": 0.0001573576436351046,
"loss": 0.4649,
"step": 240
},
{
"epoch": 0.3825136612021858,
"grad_norm": 0.3309260308742523,
"learning_rate": 0.0001551027028790524,
"loss": 0.452,
"step": 245
},
{
"epoch": 0.39032006245121,
"grad_norm": 0.3508729338645935,
"learning_rate": 0.0001528067850650368,
"loss": 0.4631,
"step": 250
},
{
"epoch": 0.3981264637002342,
"grad_norm": 0.3504168391227722,
"learning_rate": 0.0001504715975495472,
"loss": 0.4553,
"step": 255
},
{
"epoch": 0.4059328649492584,
"grad_norm": 0.3634853959083557,
"learning_rate": 0.00014809887689193877,
"loss": 0.4488,
"step": 260
},
{
"epoch": 0.4137392661982826,
"grad_norm": 0.3699380159378052,
"learning_rate": 0.00014569038756304207,
"loss": 0.4681,
"step": 265
},
{
"epoch": 0.4215456674473068,
"grad_norm": 0.3571968376636505,
"learning_rate": 0.00014324792063301662,
"loss": 0.4508,
"step": 270
},
{
"epoch": 0.42935206869633097,
"grad_norm": 0.37507084012031555,
"learning_rate": 0.00014077329243942369,
"loss": 0.4542,
"step": 275
},
{
"epoch": 0.4371584699453552,
"grad_norm": 0.3596557378768921,
"learning_rate": 0.000138268343236509,
"loss": 0.4542,
"step": 280
},
{
"epoch": 0.4449648711943794,
"grad_norm": 0.3457932472229004,
"learning_rate": 0.00013573493582670003,
"loss": 0.4476,
"step": 285
},
{
"epoch": 0.4527712724434036,
"grad_norm": 0.4396996796131134,
"learning_rate": 0.00013317495417533524,
"loss": 0.4538,
"step": 290
},
{
"epoch": 0.4605776736924278,
"grad_norm": 0.3878016173839569,
"learning_rate": 0.00013059030200965536,
"loss": 0.4432,
"step": 295
},
{
"epoch": 0.468384074941452,
"grad_norm": 0.38895392417907715,
"learning_rate": 0.00012798290140309923,
"loss": 0.4362,
"step": 300
},
{
"epoch": 0.47619047619047616,
"grad_norm": 0.34619084000587463,
"learning_rate": 0.00012535469134595595,
"loss": 0.451,
"step": 305
},
{
"epoch": 0.4839968774395004,
"grad_norm": 0.46733614802360535,
"learning_rate": 0.00012270762630343734,
"loss": 0.4474,
"step": 310
},
{
"epoch": 0.4918032786885246,
"grad_norm": 0.39834851026535034,
"learning_rate": 0.00012004367476224206,
"loss": 0.4508,
"step": 315
},
{
"epoch": 0.4996096799375488,
"grad_norm": 0.5216445922851562,
"learning_rate": 0.00011736481776669306,
"loss": 0.4334,
"step": 320
},
{
"epoch": 0.507416081186573,
"grad_norm": 0.5899478793144226,
"learning_rate": 0.00011467304744553618,
"loss": 0.4448,
"step": 325
},
{
"epoch": 0.5152224824355972,
"grad_norm": 0.34121420979499817,
"learning_rate": 0.00011197036553049625,
"loss": 0.4446,
"step": 330
},
{
"epoch": 0.5230288836846214,
"grad_norm": 0.3324553072452545,
"learning_rate": 0.00010925878186769158,
"loss": 0.4464,
"step": 335
},
{
"epoch": 0.5308352849336456,
"grad_norm": 0.3622628152370453,
"learning_rate": 0.00010654031292301432,
"loss": 0.4389,
"step": 340
},
{
"epoch": 0.5386416861826698,
"grad_norm": 0.33782732486724854,
"learning_rate": 0.00010381698028258817,
"loss": 0.4446,
"step": 345
},
{
"epoch": 0.546448087431694,
"grad_norm": 0.36627140641212463,
"learning_rate": 0.00010109080914941824,
"loss": 0.4582,
"step": 350
},
{
"epoch": 0.5542544886807181,
"grad_norm": 0.37063345313072205,
"learning_rate": 9.836382683735132e-05,
"loss": 0.4536,
"step": 355
},
{
"epoch": 0.5620608899297423,
"grad_norm": 0.3683343827724457,
"learning_rate": 9.563806126346642e-05,
"loss": 0.4546,
"step": 360
},
{
"epoch": 0.5698672911787666,
"grad_norm": 0.40577879548072815,
"learning_rate": 9.29155394400166e-05,
"loss": 0.4528,
"step": 365
},
{
"epoch": 0.5776736924277908,
"grad_norm": 0.3519943952560425,
"learning_rate": 9.019828596704394e-05,
"loss": 0.4452,
"step": 370
},
{
"epoch": 0.585480093676815,
"grad_norm": 0.3319181501865387,
"learning_rate": 8.74883215267881e-05,
"loss": 0.4365,
"step": 375
},
{
"epoch": 0.5932864949258392,
"grad_norm": 0.37883397936820984,
"learning_rate": 8.478766138100834e-05,
"loss": 0.4378,
"step": 380
},
{
"epoch": 0.6010928961748634,
"grad_norm": 0.3303097188472748,
"learning_rate": 8.209831387233676e-05,
"loss": 0.4491,
"step": 385
},
{
"epoch": 0.6088992974238876,
"grad_norm": 0.3626279830932617,
"learning_rate": 7.942227893077652e-05,
"loss": 0.4449,
"step": 390
},
{
"epoch": 0.6167056986729118,
"grad_norm": 0.38927000761032104,
"learning_rate": 7.676154658645656e-05,
"loss": 0.4386,
"step": 395
},
{
"epoch": 0.624512099921936,
"grad_norm": 0.34341347217559814,
"learning_rate": 7.411809548974792e-05,
"loss": 0.4367,
"step": 400
},
{
"epoch": 0.6323185011709602,
"grad_norm": 0.48884400725364685,
"learning_rate": 7.149389143984295e-05,
"loss": 0.4481,
"step": 405
},
{
"epoch": 0.6401249024199844,
"grad_norm": 0.3529874384403229,
"learning_rate": 6.889088592289093e-05,
"loss": 0.4452,
"step": 410
},
{
"epoch": 0.6479313036690086,
"grad_norm": 0.35880032181739807,
"learning_rate": 6.6311014660778e-05,
"loss": 0.4466,
"step": 415
},
{
"epoch": 0.6557377049180327,
"grad_norm": 0.3469119369983673,
"learning_rate": 6.375619617162985e-05,
"loss": 0.4465,
"step": 420
},
{
"epoch": 0.663544106167057,
"grad_norm": 0.33010825514793396,
"learning_rate": 6.122833034310793e-05,
"loss": 0.4418,
"step": 425
},
{
"epoch": 0.6713505074160812,
"grad_norm": 0.32382553815841675,
"learning_rate": 5.872929701956054e-05,
"loss": 0.4344,
"step": 430
},
{
"epoch": 0.6791569086651054,
"grad_norm": 0.38468924164772034,
"learning_rate": 5.6260954604078585e-05,
"loss": 0.446,
"step": 435
},
{
"epoch": 0.6869633099141296,
"grad_norm": 0.38541513681411743,
"learning_rate": 5.382513867649663e-05,
"loss": 0.435,
"step": 440
},
{
"epoch": 0.6947697111631538,
"grad_norm": 0.3638440668582916,
"learning_rate": 5.142366062836599e-05,
"loss": 0.4421,
"step": 445
},
{
"epoch": 0.702576112412178,
"grad_norm": 0.3527115285396576,
"learning_rate": 4.9058306315915826e-05,
"loss": 0.452,
"step": 450
},
{
"epoch": 0.7103825136612022,
"grad_norm": 0.3580182194709778,
"learning_rate": 4.6730834732003104e-05,
"loss": 0.4457,
"step": 455
},
{
"epoch": 0.7181889149102264,
"grad_norm": 0.3419717848300934,
"learning_rate": 4.444297669803981e-05,
"loss": 0.4529,
"step": 460
},
{
"epoch": 0.7259953161592506,
"grad_norm": 0.3361879885196686,
"learning_rate": 4.219643357686967e-05,
"loss": 0.4338,
"step": 465
},
{
"epoch": 0.7338017174082748,
"grad_norm": 0.35652631521224976,
"learning_rate": 3.999287600755192e-05,
"loss": 0.4301,
"step": 470
},
{
"epoch": 0.741608118657299,
"grad_norm": 0.34089410305023193,
"learning_rate": 3.783394266299228e-05,
"loss": 0.4392,
"step": 475
},
{
"epoch": 0.7494145199063232,
"grad_norm": 0.3883989751338959,
"learning_rate": 3.5721239031346066e-05,
"loss": 0.4447,
"step": 480
},
{
"epoch": 0.7572209211553473,
"grad_norm": 0.3701310455799103,
"learning_rate": 3.365633622209891e-05,
"loss": 0.4399,
"step": 485
},
{
"epoch": 0.7650273224043715,
"grad_norm": 0.37045201659202576,
"learning_rate": 3.164076979771287e-05,
"loss": 0.4386,
"step": 490
},
{
"epoch": 0.7728337236533958,
"grad_norm": 0.3776572346687317,
"learning_rate": 2.9676038631707593e-05,
"loss": 0.4359,
"step": 495
},
{
"epoch": 0.78064012490242,
"grad_norm": 0.35069331526756287,
"learning_rate": 2.776360379402445e-05,
"loss": 0.439,
"step": 500
},
{
"epoch": 0.7884465261514442,
"grad_norm": 0.33275192975997925,
"learning_rate": 2.5904887464504114e-05,
"loss": 0.4375,
"step": 505
},
{
"epoch": 0.7962529274004684,
"grad_norm": 0.34463104605674744,
"learning_rate": 2.4101271875283817e-05,
"loss": 0.4432,
"step": 510
},
{
"epoch": 0.8040593286494926,
"grad_norm": 0.3554835617542267,
"learning_rate": 2.2354098282902446e-05,
"loss": 0.4327,
"step": 515
},
{
"epoch": 0.8118657298985168,
"grad_norm": 0.3450513482093811,
"learning_rate": 2.0664665970876496e-05,
"loss": 0.4498,
"step": 520
},
{
"epoch": 0.819672131147541,
"grad_norm": 0.34782862663269043,
"learning_rate": 1.903423128348959e-05,
"loss": 0.4455,
"step": 525
},
{
"epoch": 0.8274785323965652,
"grad_norm": 0.3532538115978241,
"learning_rate": 1.7464006691513623e-05,
"loss": 0.4377,
"step": 530
},
{
"epoch": 0.8352849336455894,
"grad_norm": 0.37945592403411865,
"learning_rate": 1.595515989055618e-05,
"loss": 0.4499,
"step": 535
},
{
"epoch": 0.8430913348946136,
"grad_norm": 0.3615630269050598,
"learning_rate": 1.4508812932705363e-05,
"loss": 0.4427,
"step": 540
},
{
"epoch": 0.8508977361436377,
"grad_norm": 0.3308858275413513,
"learning_rate": 1.3126041392116772e-05,
"loss": 0.4466,
"step": 545
},
{
"epoch": 0.8587041373926619,
"grad_norm": 0.3379354178905487,
"learning_rate": 1.1807873565164506e-05,
"loss": 0.4508,
"step": 550
},
{
"epoch": 0.8665105386416861,
"grad_norm": 0.36346766352653503,
"learning_rate": 1.0555289705749483e-05,
"loss": 0.4378,
"step": 555
},
{
"epoch": 0.8743169398907104,
"grad_norm": 0.37977978587150574,
"learning_rate": 9.369221296335006e-06,
"loss": 0.432,
"step": 560
},
{
"epoch": 0.8821233411397346,
"grad_norm": 0.3398190438747406,
"learning_rate": 8.250550355250875e-06,
"loss": 0.4317,
"step": 565
},
{
"epoch": 0.8899297423887588,
"grad_norm": 0.3510373532772064,
"learning_rate": 7.200108780781556e-06,
"loss": 0.4526,
"step": 570
},
{
"epoch": 0.897736143637783,
"grad_norm": 0.341723769903183,
"learning_rate": 6.218677732526035e-06,
"loss": 0.4366,
"step": 575
},
{
"epoch": 0.9055425448868072,
"grad_norm": 0.36046266555786133,
"learning_rate": 5.306987050489442e-06,
"loss": 0.436,
"step": 580
},
{
"epoch": 0.9133489461358314,
"grad_norm": 0.3662625551223755,
"learning_rate": 4.465714712338398e-06,
"loss": 0.444,
"step": 585
},
{
"epoch": 0.9211553473848556,
"grad_norm": 0.3435019254684448,
"learning_rate": 3.6954863292237297e-06,
"loss": 0.4393,
"step": 590
},
{
"epoch": 0.9289617486338798,
"grad_norm": 0.3347768783569336,
"learning_rate": 2.996874680545603e-06,
"loss": 0.4482,
"step": 595
},
{
"epoch": 0.936768149882904,
"grad_norm": 0.34706735610961914,
"learning_rate": 2.3703992880066638e-06,
"loss": 0.4433,
"step": 600
},
{
"epoch": 0.9445745511319282,
"grad_norm": 0.33762818574905396,
"learning_rate": 1.8165260292704711e-06,
"loss": 0.4342,
"step": 605
},
{
"epoch": 0.9523809523809523,
"grad_norm": 0.3355940580368042,
"learning_rate": 1.3356667915121025e-06,
"loss": 0.4419,
"step": 610
},
{
"epoch": 0.9601873536299765,
"grad_norm": 0.34287044405937195,
"learning_rate": 9.281791651187366e-07,
"loss": 0.4345,
"step": 615
},
{
"epoch": 0.9679937548790007,
"grad_norm": 0.3489275872707367,
"learning_rate": 5.943661777680354e-07,
"loss": 0.4362,
"step": 620
},
{
"epoch": 0.975800156128025,
"grad_norm": 0.3551251292228699,
"learning_rate": 3.3447606908196817e-07,
"loss": 0.4331,
"step": 625
},
{
"epoch": 0.9836065573770492,
"grad_norm": 0.34148380160331726,
"learning_rate": 1.487021060236904e-07,
"loss": 0.4369,
"step": 630
},
{
"epoch": 0.9914129586260734,
"grad_norm": 0.35507649183273315,
"learning_rate": 3.7182439174832106e-08,
"loss": 0.4372,
"step": 635
},
{
"epoch": 0.9992193598750976,
"grad_norm": 0.35683757066726685,
"learning_rate": 0.0,
"loss": 0.438,
"step": 640
},
{
"epoch": 0.9992193598750976,
"eval_loss": 1.0797678232192993,
"eval_runtime": 0.9282,
"eval_samples_per_second": 5.387,
"eval_steps_per_second": 1.077,
"step": 640
},
{
"epoch": 0.9992193598750976,
"step": 640,
"total_flos": 8.955542824078541e+17,
"train_loss": 0.4838556606322527,
"train_runtime": 4573.5964,
"train_samples_per_second": 4.48,
"train_steps_per_second": 0.14
}
],
"logging_steps": 5,
"max_steps": 640,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"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": 8.955542824078541e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}