chansung's picture
Model save
c93032c 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": 4.679694652557373,
"learning_rate": 3.125e-06,
"loss": 1.3,
"step": 1
},
{
"epoch": 0.0078064012490242,
"grad_norm": 4.700327396392822,
"learning_rate": 1.5625e-05,
"loss": 1.2534,
"step": 5
},
{
"epoch": 0.0156128024980484,
"grad_norm": 4.184088230133057,
"learning_rate": 3.125e-05,
"loss": 1.1414,
"step": 10
},
{
"epoch": 0.0234192037470726,
"grad_norm": 3.3843367099761963,
"learning_rate": 4.6875e-05,
"loss": 0.9791,
"step": 15
},
{
"epoch": 0.0312256049960968,
"grad_norm": 1.38542640209198,
"learning_rate": 6.25e-05,
"loss": 0.8613,
"step": 20
},
{
"epoch": 0.039032006245121,
"grad_norm": 0.8451548218727112,
"learning_rate": 7.8125e-05,
"loss": 0.7475,
"step": 25
},
{
"epoch": 0.0468384074941452,
"grad_norm": 0.6994028091430664,
"learning_rate": 9.375e-05,
"loss": 0.6857,
"step": 30
},
{
"epoch": 0.0546448087431694,
"grad_norm": 0.7693173885345459,
"learning_rate": 0.000109375,
"loss": 0.632,
"step": 35
},
{
"epoch": 0.0624512099921936,
"grad_norm": 1.0630561113357544,
"learning_rate": 0.000125,
"loss": 0.6101,
"step": 40
},
{
"epoch": 0.0702576112412178,
"grad_norm": 0.9208365678787231,
"learning_rate": 0.00014062500000000002,
"loss": 0.5723,
"step": 45
},
{
"epoch": 0.078064012490242,
"grad_norm": 0.8858307003974915,
"learning_rate": 0.00015625,
"loss": 0.5483,
"step": 50
},
{
"epoch": 0.0858704137392662,
"grad_norm": 0.5890257358551025,
"learning_rate": 0.00017187500000000002,
"loss": 0.5341,
"step": 55
},
{
"epoch": 0.0936768149882904,
"grad_norm": 0.8059648275375366,
"learning_rate": 0.0001875,
"loss": 0.5393,
"step": 60
},
{
"epoch": 0.1014832162373146,
"grad_norm": 0.6733461618423462,
"learning_rate": 0.00019999851261394218,
"loss": 0.5324,
"step": 65
},
{
"epoch": 0.1092896174863388,
"grad_norm": 0.5950048565864563,
"learning_rate": 0.00019994645874763658,
"loss": 0.5186,
"step": 70
},
{
"epoch": 0.117096018735363,
"grad_norm": 0.7100915908813477,
"learning_rate": 0.00019982007981886847,
"loss": 0.5022,
"step": 75
},
{
"epoch": 0.1249024199843872,
"grad_norm": 0.38926348090171814,
"learning_rate": 0.00019961946980917456,
"loss": 0.5032,
"step": 80
},
{
"epoch": 0.1327088212334114,
"grad_norm": 0.32585254311561584,
"learning_rate": 0.00019934477790194445,
"loss": 0.4926,
"step": 85
},
{
"epoch": 0.1405152224824356,
"grad_norm": 0.37917089462280273,
"learning_rate": 0.00019899620837148077,
"loss": 0.4963,
"step": 90
},
{
"epoch": 0.1483216237314598,
"grad_norm": 0.4928993582725525,
"learning_rate": 0.0001985740204310909,
"loss": 0.4934,
"step": 95
},
{
"epoch": 0.156128024980484,
"grad_norm": 0.41284865140914917,
"learning_rate": 0.00019807852804032305,
"loss": 0.4988,
"step": 100
},
{
"epoch": 0.16393442622950818,
"grad_norm": 0.3852866291999817,
"learning_rate": 0.00019751009967149087,
"loss": 0.4861,
"step": 105
},
{
"epoch": 0.1717408274785324,
"grad_norm": 0.34940582513809204,
"learning_rate": 0.00019686915803565934,
"loss": 0.4846,
"step": 110
},
{
"epoch": 0.1795472287275566,
"grad_norm": 0.379056453704834,
"learning_rate": 0.0001961561797682962,
"loss": 0.4716,
"step": 115
},
{
"epoch": 0.1873536299765808,
"grad_norm": 0.378898024559021,
"learning_rate": 0.0001953716950748227,
"loss": 0.4862,
"step": 120
},
{
"epoch": 0.195160031225605,
"grad_norm": 0.5000614523887634,
"learning_rate": 0.0001945162873363268,
"loss": 0.4758,
"step": 125
},
{
"epoch": 0.2029664324746292,
"grad_norm": 0.43147751688957214,
"learning_rate": 0.0001935905926757326,
"loss": 0.477,
"step": 130
},
{
"epoch": 0.2107728337236534,
"grad_norm": 0.41404739022254944,
"learning_rate": 0.00019259529948474833,
"loss": 0.4773,
"step": 135
},
{
"epoch": 0.2185792349726776,
"grad_norm": 0.4057549834251404,
"learning_rate": 0.00019153114791194473,
"loss": 0.4777,
"step": 140
},
{
"epoch": 0.2263856362217018,
"grad_norm": 0.3698827624320984,
"learning_rate": 0.00019039892931234435,
"loss": 0.4911,
"step": 145
},
{
"epoch": 0.234192037470726,
"grad_norm": 0.4173194169998169,
"learning_rate": 0.00018919948565893142,
"loss": 0.4722,
"step": 150
},
{
"epoch": 0.2419984387197502,
"grad_norm": 0.5116965174674988,
"learning_rate": 0.00018793370891651972,
"loss": 0.4766,
"step": 155
},
{
"epoch": 0.2498048399687744,
"grad_norm": 0.33461353182792664,
"learning_rate": 0.00018660254037844388,
"loss": 0.4833,
"step": 160
},
{
"epoch": 0.2576112412177986,
"grad_norm": 0.45119625329971313,
"learning_rate": 0.00018520696996656788,
"loss": 0.4633,
"step": 165
},
{
"epoch": 0.2654176424668228,
"grad_norm": 0.5022788047790527,
"learning_rate": 0.0001837480354951308,
"loss": 0.4539,
"step": 170
},
{
"epoch": 0.273224043715847,
"grad_norm": 0.37766632437705994,
"learning_rate": 0.00018222682189897752,
"loss": 0.4615,
"step": 175
},
{
"epoch": 0.2810304449648712,
"grad_norm": 0.34759408235549927,
"learning_rate": 0.00018064446042674828,
"loss": 0.4621,
"step": 180
},
{
"epoch": 0.2888368462138954,
"grad_norm": 0.5189246535301208,
"learning_rate": 0.0001790021277996269,
"loss": 0.4644,
"step": 185
},
{
"epoch": 0.2966432474629196,
"grad_norm": 0.4171973466873169,
"learning_rate": 0.0001773010453362737,
"loss": 0.4733,
"step": 190
},
{
"epoch": 0.3044496487119438,
"grad_norm": 0.3633343577384949,
"learning_rate": 0.00017554247804459316,
"loss": 0.4647,
"step": 195
},
{
"epoch": 0.312256049960968,
"grad_norm": 0.3888181746006012,
"learning_rate": 0.0001737277336810124,
"loss": 0.4666,
"step": 200
},
{
"epoch": 0.3200624512099922,
"grad_norm": 0.35728883743286133,
"learning_rate": 0.0001718581617779698,
"loss": 0.4612,
"step": 205
},
{
"epoch": 0.32786885245901637,
"grad_norm": 0.3724042475223541,
"learning_rate": 0.00016993515264033672,
"loss": 0.4718,
"step": 210
},
{
"epoch": 0.3356752537080406,
"grad_norm": 0.3650490641593933,
"learning_rate": 0.00016796013631151897,
"loss": 0.4519,
"step": 215
},
{
"epoch": 0.3434816549570648,
"grad_norm": 0.3484860062599182,
"learning_rate": 0.00016593458151000688,
"loss": 0.4506,
"step": 220
},
{
"epoch": 0.351288056206089,
"grad_norm": 0.32455822825431824,
"learning_rate": 0.00016385999453716454,
"loss": 0.4456,
"step": 225
},
{
"epoch": 0.3590944574551132,
"grad_norm": 0.4763931334018707,
"learning_rate": 0.00016173791815707051,
"loss": 0.4584,
"step": 230
},
{
"epoch": 0.3669008587041374,
"grad_norm": 0.37389329075813293,
"learning_rate": 0.00015956993044924334,
"loss": 0.4502,
"step": 235
},
{
"epoch": 0.3747072599531616,
"grad_norm": 0.5365679860115051,
"learning_rate": 0.0001573576436351046,
"loss": 0.462,
"step": 240
},
{
"epoch": 0.3825136612021858,
"grad_norm": 0.3317677080631256,
"learning_rate": 0.0001551027028790524,
"loss": 0.4494,
"step": 245
},
{
"epoch": 0.39032006245121,
"grad_norm": 0.3652748763561249,
"learning_rate": 0.0001528067850650368,
"loss": 0.4608,
"step": 250
},
{
"epoch": 0.3981264637002342,
"grad_norm": 0.3600890636444092,
"learning_rate": 0.0001504715975495472,
"loss": 0.4535,
"step": 255
},
{
"epoch": 0.4059328649492584,
"grad_norm": 0.3843066990375519,
"learning_rate": 0.00014809887689193877,
"loss": 0.4471,
"step": 260
},
{
"epoch": 0.4137392661982826,
"grad_norm": 0.42747819423675537,
"learning_rate": 0.00014569038756304207,
"loss": 0.4667,
"step": 265
},
{
"epoch": 0.4215456674473068,
"grad_norm": 0.354117214679718,
"learning_rate": 0.00014324792063301662,
"loss": 0.4481,
"step": 270
},
{
"epoch": 0.42935206869633097,
"grad_norm": 0.3524267375469208,
"learning_rate": 0.00014077329243942369,
"loss": 0.4511,
"step": 275
},
{
"epoch": 0.4371584699453552,
"grad_norm": 0.41965606808662415,
"learning_rate": 0.000138268343236509,
"loss": 0.4519,
"step": 280
},
{
"epoch": 0.4449648711943794,
"grad_norm": 0.338069349527359,
"learning_rate": 0.00013573493582670003,
"loss": 0.4448,
"step": 285
},
{
"epoch": 0.4527712724434036,
"grad_norm": 0.44666725397109985,
"learning_rate": 0.00013317495417533524,
"loss": 0.451,
"step": 290
},
{
"epoch": 0.4605776736924278,
"grad_norm": 0.3880995213985443,
"learning_rate": 0.00013059030200965536,
"loss": 0.4402,
"step": 295
},
{
"epoch": 0.468384074941452,
"grad_norm": 0.39150258898735046,
"learning_rate": 0.00012798290140309923,
"loss": 0.4337,
"step": 300
},
{
"epoch": 0.47619047619047616,
"grad_norm": 0.3439214825630188,
"learning_rate": 0.00012535469134595595,
"loss": 0.4487,
"step": 305
},
{
"epoch": 0.4839968774395004,
"grad_norm": 0.44355058670043945,
"learning_rate": 0.00012270762630343734,
"loss": 0.4448,
"step": 310
},
{
"epoch": 0.4918032786885246,
"grad_norm": 0.36502349376678467,
"learning_rate": 0.00012004367476224206,
"loss": 0.4488,
"step": 315
},
{
"epoch": 0.4996096799375488,
"grad_norm": 0.4983600080013275,
"learning_rate": 0.00011736481776669306,
"loss": 0.4308,
"step": 320
},
{
"epoch": 0.507416081186573,
"grad_norm": 0.5701916217803955,
"learning_rate": 0.00011467304744553618,
"loss": 0.4425,
"step": 325
},
{
"epoch": 0.5152224824355972,
"grad_norm": 0.35758593678474426,
"learning_rate": 0.00011197036553049625,
"loss": 0.4425,
"step": 330
},
{
"epoch": 0.5230288836846214,
"grad_norm": 0.3369368612766266,
"learning_rate": 0.00010925878186769158,
"loss": 0.4444,
"step": 335
},
{
"epoch": 0.5308352849336456,
"grad_norm": 0.35938557982444763,
"learning_rate": 0.00010654031292301432,
"loss": 0.4363,
"step": 340
},
{
"epoch": 0.5386416861826698,
"grad_norm": 0.34216657280921936,
"learning_rate": 0.00010381698028258817,
"loss": 0.4421,
"step": 345
},
{
"epoch": 0.546448087431694,
"grad_norm": 0.36699554324150085,
"learning_rate": 0.00010109080914941824,
"loss": 0.4558,
"step": 350
},
{
"epoch": 0.5542544886807181,
"grad_norm": 0.36921554803848267,
"learning_rate": 9.836382683735132e-05,
"loss": 0.4514,
"step": 355
},
{
"epoch": 0.5620608899297423,
"grad_norm": 0.3594672977924347,
"learning_rate": 9.563806126346642e-05,
"loss": 0.4518,
"step": 360
},
{
"epoch": 0.5698672911787666,
"grad_norm": 0.37948325276374817,
"learning_rate": 9.29155394400166e-05,
"loss": 0.45,
"step": 365
},
{
"epoch": 0.5776736924277908,
"grad_norm": 0.3441675007343292,
"learning_rate": 9.019828596704394e-05,
"loss": 0.4425,
"step": 370
},
{
"epoch": 0.585480093676815,
"grad_norm": 0.3524007499217987,
"learning_rate": 8.74883215267881e-05,
"loss": 0.434,
"step": 375
},
{
"epoch": 0.5932864949258392,
"grad_norm": 0.38553693890571594,
"learning_rate": 8.478766138100834e-05,
"loss": 0.4352,
"step": 380
},
{
"epoch": 0.6010928961748634,
"grad_norm": 0.3395775854587555,
"learning_rate": 8.209831387233676e-05,
"loss": 0.4472,
"step": 385
},
{
"epoch": 0.6088992974238876,
"grad_norm": 0.354542076587677,
"learning_rate": 7.942227893077652e-05,
"loss": 0.4418,
"step": 390
},
{
"epoch": 0.6167056986729118,
"grad_norm": 0.3631182610988617,
"learning_rate": 7.676154658645656e-05,
"loss": 0.4361,
"step": 395
},
{
"epoch": 0.624512099921936,
"grad_norm": 0.35347339510917664,
"learning_rate": 7.411809548974792e-05,
"loss": 0.4348,
"step": 400
},
{
"epoch": 0.6323185011709602,
"grad_norm": 0.4688993990421295,
"learning_rate": 7.149389143984295e-05,
"loss": 0.4464,
"step": 405
},
{
"epoch": 0.6401249024199844,
"grad_norm": 0.3508263826370239,
"learning_rate": 6.889088592289093e-05,
"loss": 0.4428,
"step": 410
},
{
"epoch": 0.6479313036690086,
"grad_norm": 0.3598111867904663,
"learning_rate": 6.6311014660778e-05,
"loss": 0.4442,
"step": 415
},
{
"epoch": 0.6557377049180327,
"grad_norm": 0.35423406958580017,
"learning_rate": 6.375619617162985e-05,
"loss": 0.4442,
"step": 420
},
{
"epoch": 0.663544106167057,
"grad_norm": 0.32423022389411926,
"learning_rate": 6.122833034310793e-05,
"loss": 0.4385,
"step": 425
},
{
"epoch": 0.6713505074160812,
"grad_norm": 0.34300023317337036,
"learning_rate": 5.872929701956054e-05,
"loss": 0.4314,
"step": 430
},
{
"epoch": 0.6791569086651054,
"grad_norm": 0.355337530374527,
"learning_rate": 5.6260954604078585e-05,
"loss": 0.4437,
"step": 435
},
{
"epoch": 0.6869633099141296,
"grad_norm": 0.3726743161678314,
"learning_rate": 5.382513867649663e-05,
"loss": 0.4324,
"step": 440
},
{
"epoch": 0.6947697111631538,
"grad_norm": 0.36385682225227356,
"learning_rate": 5.142366062836599e-05,
"loss": 0.4398,
"step": 445
},
{
"epoch": 0.702576112412178,
"grad_norm": 0.34697508811950684,
"learning_rate": 4.9058306315915826e-05,
"loss": 0.4496,
"step": 450
},
{
"epoch": 0.7103825136612022,
"grad_norm": 0.37256768345832825,
"learning_rate": 4.6730834732003104e-05,
"loss": 0.4432,
"step": 455
},
{
"epoch": 0.7181889149102264,
"grad_norm": 0.36101406812667847,
"learning_rate": 4.444297669803981e-05,
"loss": 0.4506,
"step": 460
},
{
"epoch": 0.7259953161592506,
"grad_norm": 0.3514850437641144,
"learning_rate": 4.219643357686967e-05,
"loss": 0.4321,
"step": 465
},
{
"epoch": 0.7338017174082748,
"grad_norm": 0.36538928747177124,
"learning_rate": 3.999287600755192e-05,
"loss": 0.4278,
"step": 470
},
{
"epoch": 0.741608118657299,
"grad_norm": 0.3825438320636749,
"learning_rate": 3.783394266299228e-05,
"loss": 0.4368,
"step": 475
},
{
"epoch": 0.7494145199063232,
"grad_norm": 0.4204176068305969,
"learning_rate": 3.5721239031346066e-05,
"loss": 0.443,
"step": 480
},
{
"epoch": 0.7572209211553473,
"grad_norm": 0.3882097005844116,
"learning_rate": 3.365633622209891e-05,
"loss": 0.437,
"step": 485
},
{
"epoch": 0.7650273224043715,
"grad_norm": 0.38486871123313904,
"learning_rate": 3.164076979771287e-05,
"loss": 0.4362,
"step": 490
},
{
"epoch": 0.7728337236533958,
"grad_norm": 0.3774687647819519,
"learning_rate": 2.9676038631707593e-05,
"loss": 0.4341,
"step": 495
},
{
"epoch": 0.78064012490242,
"grad_norm": 0.3509750962257385,
"learning_rate": 2.776360379402445e-05,
"loss": 0.437,
"step": 500
},
{
"epoch": 0.7884465261514442,
"grad_norm": 0.4982232451438904,
"learning_rate": 2.5904887464504114e-05,
"loss": 0.4352,
"step": 505
},
{
"epoch": 0.7962529274004684,
"grad_norm": 0.3576488196849823,
"learning_rate": 2.4101271875283817e-05,
"loss": 0.4415,
"step": 510
},
{
"epoch": 0.8040593286494926,
"grad_norm": 0.3655681312084198,
"learning_rate": 2.2354098282902446e-05,
"loss": 0.4302,
"step": 515
},
{
"epoch": 0.8118657298985168,
"grad_norm": 0.34478428959846497,
"learning_rate": 2.0664665970876496e-05,
"loss": 0.4472,
"step": 520
},
{
"epoch": 0.819672131147541,
"grad_norm": 0.34541845321655273,
"learning_rate": 1.903423128348959e-05,
"loss": 0.4425,
"step": 525
},
{
"epoch": 0.8274785323965652,
"grad_norm": 0.3647783398628235,
"learning_rate": 1.7464006691513623e-05,
"loss": 0.4354,
"step": 530
},
{
"epoch": 0.8352849336455894,
"grad_norm": 0.3986523747444153,
"learning_rate": 1.595515989055618e-05,
"loss": 0.4483,
"step": 535
},
{
"epoch": 0.8430913348946136,
"grad_norm": 0.3613293766975403,
"learning_rate": 1.4508812932705363e-05,
"loss": 0.4412,
"step": 540
},
{
"epoch": 0.8508977361436377,
"grad_norm": 0.3343644142150879,
"learning_rate": 1.3126041392116772e-05,
"loss": 0.4445,
"step": 545
},
{
"epoch": 0.8587041373926619,
"grad_norm": 0.34114307165145874,
"learning_rate": 1.1807873565164506e-05,
"loss": 0.4483,
"step": 550
},
{
"epoch": 0.8665105386416861,
"grad_norm": 0.36701440811157227,
"learning_rate": 1.0555289705749483e-05,
"loss": 0.4351,
"step": 555
},
{
"epoch": 0.8743169398907104,
"grad_norm": 0.3947718143463135,
"learning_rate": 9.369221296335006e-06,
"loss": 0.4295,
"step": 560
},
{
"epoch": 0.8821233411397346,
"grad_norm": 0.3417580723762512,
"learning_rate": 8.250550355250875e-06,
"loss": 0.4296,
"step": 565
},
{
"epoch": 0.8899297423887588,
"grad_norm": 0.35511329770088196,
"learning_rate": 7.200108780781556e-06,
"loss": 0.45,
"step": 570
},
{
"epoch": 0.897736143637783,
"grad_norm": 0.3438127636909485,
"learning_rate": 6.218677732526035e-06,
"loss": 0.434,
"step": 575
},
{
"epoch": 0.9055425448868072,
"grad_norm": 0.35285842418670654,
"learning_rate": 5.306987050489442e-06,
"loss": 0.4339,
"step": 580
},
{
"epoch": 0.9133489461358314,
"grad_norm": 0.37091735005378723,
"learning_rate": 4.465714712338398e-06,
"loss": 0.4422,
"step": 585
},
{
"epoch": 0.9211553473848556,
"grad_norm": 0.3490866422653198,
"learning_rate": 3.6954863292237297e-06,
"loss": 0.437,
"step": 590
},
{
"epoch": 0.9289617486338798,
"grad_norm": 0.3426539897918701,
"learning_rate": 2.996874680545603e-06,
"loss": 0.4461,
"step": 595
},
{
"epoch": 0.936768149882904,
"grad_norm": 0.359111487865448,
"learning_rate": 2.3703992880066638e-06,
"loss": 0.4405,
"step": 600
},
{
"epoch": 0.9445745511319282,
"grad_norm": 0.35233959555625916,
"learning_rate": 1.8165260292704711e-06,
"loss": 0.4312,
"step": 605
},
{
"epoch": 0.9523809523809523,
"grad_norm": 0.3432738184928894,
"learning_rate": 1.3356667915121025e-06,
"loss": 0.4395,
"step": 610
},
{
"epoch": 0.9601873536299765,
"grad_norm": 0.3486664295196533,
"learning_rate": 9.281791651187366e-07,
"loss": 0.4326,
"step": 615
},
{
"epoch": 0.9679937548790007,
"grad_norm": 0.35764941573143005,
"learning_rate": 5.943661777680354e-07,
"loss": 0.4345,
"step": 620
},
{
"epoch": 0.975800156128025,
"grad_norm": 0.3644266128540039,
"learning_rate": 3.3447606908196817e-07,
"loss": 0.4317,
"step": 625
},
{
"epoch": 0.9836065573770492,
"grad_norm": 0.34483930468559265,
"learning_rate": 1.487021060236904e-07,
"loss": 0.4345,
"step": 630
},
{
"epoch": 0.9914129586260734,
"grad_norm": 0.36307430267333984,
"learning_rate": 3.7182439174832106e-08,
"loss": 0.4348,
"step": 635
},
{
"epoch": 0.9992193598750976,
"grad_norm": 0.3593035042285919,
"learning_rate": 0.0,
"loss": 0.4351,
"step": 640
},
{
"epoch": 0.9992193598750976,
"eval_loss": 1.0355881452560425,
"eval_runtime": 0.9084,
"eval_samples_per_second": 5.504,
"eval_steps_per_second": 1.101,
"step": 640
},
{
"epoch": 0.9992193598750976,
"step": 640,
"total_flos": 8.955537329204756e+17,
"train_loss": 0.4811625644564629,
"train_runtime": 4514.48,
"train_samples_per_second": 4.539,
"train_steps_per_second": 0.142
}
],
"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": 8.955537329204756e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}