Qwen2Math_Pretrain / trainer_state.json
pepoo20's picture
Upload trainer_state.json with huggingface_hub
0095c57 verified
raw
history blame
41.4 kB
{
"best_metric": 0.48741263151168823,
"best_model_checkpoint": "/home/ray/default/save/checkpoint-1000",
"epoch": 1.0,
"eval_steps": 500,
"global_step": 1154,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.004332755632582322,
"grad_norm": 0.10313185304403305,
"learning_rate": 8.333333333333333e-07,
"loss": 1.0706,
"step": 5
},
{
"epoch": 0.008665511265164644,
"grad_norm": 0.10810094326734543,
"learning_rate": 1.6666666666666667e-06,
"loss": 1.0694,
"step": 10
},
{
"epoch": 0.012998266897746967,
"grad_norm": 0.09731286019086838,
"learning_rate": 2.5e-06,
"loss": 1.0706,
"step": 15
},
{
"epoch": 0.01733102253032929,
"grad_norm": 0.11459868401288986,
"learning_rate": 3.3333333333333333e-06,
"loss": 1.0772,
"step": 20
},
{
"epoch": 0.021663778162911613,
"grad_norm": 0.10845116525888443,
"learning_rate": 4.166666666666667e-06,
"loss": 1.0809,
"step": 25
},
{
"epoch": 0.025996533795493933,
"grad_norm": 0.12091381102800369,
"learning_rate": 5e-06,
"loss": 1.0737,
"step": 30
},
{
"epoch": 0.030329289428076257,
"grad_norm": 0.12315661460161209,
"learning_rate": 5.833333333333334e-06,
"loss": 1.0675,
"step": 35
},
{
"epoch": 0.03466204506065858,
"grad_norm": 0.13054250180721283,
"learning_rate": 6.666666666666667e-06,
"loss": 1.0638,
"step": 40
},
{
"epoch": 0.0389948006932409,
"grad_norm": 0.13414187729358673,
"learning_rate": 7.5e-06,
"loss": 1.0588,
"step": 45
},
{
"epoch": 0.043327556325823226,
"grad_norm": 0.12992985546588898,
"learning_rate": 8.333333333333334e-06,
"loss": 1.0512,
"step": 50
},
{
"epoch": 0.047660311958405546,
"grad_norm": 0.11500700563192368,
"learning_rate": 9.166666666666666e-06,
"loss": 1.0545,
"step": 55
},
{
"epoch": 0.05199306759098787,
"grad_norm": 0.10958714783191681,
"learning_rate": 1e-05,
"loss": 1.0357,
"step": 60
},
{
"epoch": 0.05632582322357019,
"grad_norm": 0.10853663086891174,
"learning_rate": 1.0833333333333334e-05,
"loss": 1.0268,
"step": 65
},
{
"epoch": 0.060658578856152515,
"grad_norm": 0.12498235702514648,
"learning_rate": 1.1666666666666668e-05,
"loss": 1.0135,
"step": 70
},
{
"epoch": 0.06499133448873484,
"grad_norm": 0.11716682463884354,
"learning_rate": 1.25e-05,
"loss": 1.0014,
"step": 75
},
{
"epoch": 0.06932409012131716,
"grad_norm": 0.09530466794967651,
"learning_rate": 1.3333333333333333e-05,
"loss": 0.9824,
"step": 80
},
{
"epoch": 0.07365684575389948,
"grad_norm": 0.09298353642225266,
"learning_rate": 1.4166666666666668e-05,
"loss": 0.9676,
"step": 85
},
{
"epoch": 0.0779896013864818,
"grad_norm": 0.09433547407388687,
"learning_rate": 1.5e-05,
"loss": 0.9463,
"step": 90
},
{
"epoch": 0.08232235701906412,
"grad_norm": 0.09255563467741013,
"learning_rate": 1.5833333333333333e-05,
"loss": 0.9341,
"step": 95
},
{
"epoch": 0.08665511265164645,
"grad_norm": 0.09951213002204895,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.9209,
"step": 100
},
{
"epoch": 0.09098786828422877,
"grad_norm": 0.10951012372970581,
"learning_rate": 1.75e-05,
"loss": 0.8927,
"step": 105
},
{
"epoch": 0.09532062391681109,
"grad_norm": 0.110866479575634,
"learning_rate": 1.8333333333333333e-05,
"loss": 0.8755,
"step": 110
},
{
"epoch": 0.09965337954939342,
"grad_norm": 0.11648208647966385,
"learning_rate": 1.9166666666666667e-05,
"loss": 0.8453,
"step": 115
},
{
"epoch": 0.10398613518197573,
"grad_norm": 0.12016862630844116,
"learning_rate": 2e-05,
"loss": 0.8081,
"step": 120
},
{
"epoch": 0.10831889081455806,
"grad_norm": 0.12387488037347794,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.7784,
"step": 125
},
{
"epoch": 0.11265164644714037,
"grad_norm": 0.12779255211353302,
"learning_rate": 2.1666666666666667e-05,
"loss": 0.7353,
"step": 130
},
{
"epoch": 0.1169844020797227,
"grad_norm": 0.12649372220039368,
"learning_rate": 2.25e-05,
"loss": 0.7085,
"step": 135
},
{
"epoch": 0.12131715771230503,
"grad_norm": 0.1445430964231491,
"learning_rate": 2.3333333333333336e-05,
"loss": 0.6753,
"step": 140
},
{
"epoch": 0.12564991334488734,
"grad_norm": 0.1329505294561386,
"learning_rate": 2.4166666666666667e-05,
"loss": 0.6448,
"step": 145
},
{
"epoch": 0.12998266897746968,
"grad_norm": 0.13544394075870514,
"learning_rate": 2.5e-05,
"loss": 0.621,
"step": 150
},
{
"epoch": 0.134315424610052,
"grad_norm": 0.1351090520620346,
"learning_rate": 2.5833333333333336e-05,
"loss": 0.5997,
"step": 155
},
{
"epoch": 0.1386481802426343,
"grad_norm": 0.127303346991539,
"learning_rate": 2.6666666666666667e-05,
"loss": 0.5791,
"step": 160
},
{
"epoch": 0.14298093587521662,
"grad_norm": 0.1476767510175705,
"learning_rate": 2.7500000000000004e-05,
"loss": 0.57,
"step": 165
},
{
"epoch": 0.14731369150779897,
"grad_norm": 0.13856437802314758,
"learning_rate": 2.8333333333333335e-05,
"loss": 0.5645,
"step": 170
},
{
"epoch": 0.15164644714038128,
"grad_norm": 0.1533508449792862,
"learning_rate": 2.916666666666667e-05,
"loss": 0.5583,
"step": 175
},
{
"epoch": 0.1559792027729636,
"grad_norm": 0.13325001299381256,
"learning_rate": 3e-05,
"loss": 0.555,
"step": 180
},
{
"epoch": 0.16031195840554593,
"grad_norm": 0.13416974246501923,
"learning_rate": 3.0833333333333335e-05,
"loss": 0.5439,
"step": 185
},
{
"epoch": 0.16464471403812825,
"grad_norm": 0.1278882771730423,
"learning_rate": 3.1666666666666666e-05,
"loss": 0.537,
"step": 190
},
{
"epoch": 0.16897746967071056,
"grad_norm": 0.14047101140022278,
"learning_rate": 3.2500000000000004e-05,
"loss": 0.5381,
"step": 195
},
{
"epoch": 0.1733102253032929,
"grad_norm": 0.13340455293655396,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.5344,
"step": 200
},
{
"epoch": 0.17764298093587522,
"grad_norm": 0.13049094378948212,
"learning_rate": 3.4166666666666666e-05,
"loss": 0.5291,
"step": 205
},
{
"epoch": 0.18197573656845753,
"grad_norm": 0.16296444833278656,
"learning_rate": 3.5e-05,
"loss": 0.5342,
"step": 210
},
{
"epoch": 0.18630849220103987,
"grad_norm": 0.1682613343000412,
"learning_rate": 3.5833333333333335e-05,
"loss": 0.5285,
"step": 215
},
{
"epoch": 0.19064124783362218,
"grad_norm": 0.1439386010169983,
"learning_rate": 3.6666666666666666e-05,
"loss": 0.5268,
"step": 220
},
{
"epoch": 0.1949740034662045,
"grad_norm": 0.15248768031597137,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.5252,
"step": 225
},
{
"epoch": 0.19930675909878684,
"grad_norm": 0.1604214906692505,
"learning_rate": 3.8333333333333334e-05,
"loss": 0.5201,
"step": 230
},
{
"epoch": 0.20363951473136915,
"grad_norm": 0.16192543506622314,
"learning_rate": 3.9166666666666665e-05,
"loss": 0.5222,
"step": 235
},
{
"epoch": 0.20797227036395147,
"grad_norm": 0.14945088326931,
"learning_rate": 4e-05,
"loss": 0.5158,
"step": 240
},
{
"epoch": 0.2123050259965338,
"grad_norm": 0.16204427182674408,
"learning_rate": 4.0833333333333334e-05,
"loss": 0.5127,
"step": 245
},
{
"epoch": 0.21663778162911612,
"grad_norm": 0.1618306040763855,
"learning_rate": 4.166666666666667e-05,
"loss": 0.5181,
"step": 250
},
{
"epoch": 0.22097053726169844,
"grad_norm": 0.14193885028362274,
"learning_rate": 4.25e-05,
"loss": 0.5164,
"step": 255
},
{
"epoch": 0.22530329289428075,
"grad_norm": 0.12552691996097565,
"learning_rate": 4.3333333333333334e-05,
"loss": 0.5149,
"step": 260
},
{
"epoch": 0.2296360485268631,
"grad_norm": 0.14471225440502167,
"learning_rate": 4.4166666666666665e-05,
"loss": 0.5137,
"step": 265
},
{
"epoch": 0.2339688041594454,
"grad_norm": 0.13988590240478516,
"learning_rate": 4.5e-05,
"loss": 0.5066,
"step": 270
},
{
"epoch": 0.23830155979202772,
"grad_norm": 0.13964875042438507,
"learning_rate": 4.5833333333333334e-05,
"loss": 0.5116,
"step": 275
},
{
"epoch": 0.24263431542461006,
"grad_norm": 0.12847208976745605,
"learning_rate": 4.666666666666667e-05,
"loss": 0.5095,
"step": 280
},
{
"epoch": 0.24696707105719237,
"grad_norm": 0.13142219185829163,
"learning_rate": 4.75e-05,
"loss": 0.5077,
"step": 285
},
{
"epoch": 0.2512998266897747,
"grad_norm": 0.19098567962646484,
"learning_rate": 4.8333333333333334e-05,
"loss": 0.5099,
"step": 290
},
{
"epoch": 0.255632582322357,
"grad_norm": 0.1430283784866333,
"learning_rate": 4.9166666666666665e-05,
"loss": 0.5036,
"step": 295
},
{
"epoch": 0.25996533795493937,
"grad_norm": 0.12594453990459442,
"learning_rate": 5e-05,
"loss": 0.5085,
"step": 300
},
{
"epoch": 0.26429809358752165,
"grad_norm": 0.14377984404563904,
"learning_rate": 4.999577115486055e-05,
"loss": 0.4978,
"step": 305
},
{
"epoch": 0.268630849220104,
"grad_norm": 0.12468158453702927,
"learning_rate": 4.998308605009268e-05,
"loss": 0.5095,
"step": 310
},
{
"epoch": 0.2729636048526863,
"grad_norm": 0.12854167819023132,
"learning_rate": 4.996194897716389e-05,
"loss": 0.5037,
"step": 315
},
{
"epoch": 0.2772963604852686,
"grad_norm": 0.13529527187347412,
"learning_rate": 4.993236708690683e-05,
"loss": 0.5058,
"step": 320
},
{
"epoch": 0.28162911611785096,
"grad_norm": 0.1377994418144226,
"learning_rate": 4.9894350387100126e-05,
"loss": 0.4998,
"step": 325
},
{
"epoch": 0.28596187175043325,
"grad_norm": 0.14942322671413422,
"learning_rate": 4.984791173908267e-05,
"loss": 0.5007,
"step": 330
},
{
"epoch": 0.2902946273830156,
"grad_norm": 0.1366725116968155,
"learning_rate": 4.9793066853402536e-05,
"loss": 0.5038,
"step": 335
},
{
"epoch": 0.29462738301559793,
"grad_norm": 0.13133087754249573,
"learning_rate": 4.9729834284501995e-05,
"loss": 0.5062,
"step": 340
},
{
"epoch": 0.2989601386481802,
"grad_norm": 0.11725670844316483,
"learning_rate": 4.965823542444037e-05,
"loss": 0.5025,
"step": 345
},
{
"epoch": 0.30329289428076256,
"grad_norm": 0.10628046840429306,
"learning_rate": 4.9578294495656965e-05,
"loss": 0.4999,
"step": 350
},
{
"epoch": 0.3076256499133449,
"grad_norm": 0.13826170563697815,
"learning_rate": 4.949003854277644e-05,
"loss": 0.4978,
"step": 355
},
{
"epoch": 0.3119584055459272,
"grad_norm": 0.1305851936340332,
"learning_rate": 4.9393497423459376e-05,
"loss": 0.4997,
"step": 360
},
{
"epoch": 0.31629116117850953,
"grad_norm": 0.11465763300657272,
"learning_rate": 4.928870379830124e-05,
"loss": 0.5037,
"step": 365
},
{
"epoch": 0.32062391681109187,
"grad_norm": 0.15975706279277802,
"learning_rate": 4.9175693119783013e-05,
"loss": 0.4982,
"step": 370
},
{
"epoch": 0.32495667244367415,
"grad_norm": 0.15360799431800842,
"learning_rate": 4.905450362027738e-05,
"loss": 0.5013,
"step": 375
},
{
"epoch": 0.3292894280762565,
"grad_norm": 0.14006198942661285,
"learning_rate": 4.8925176299114416e-05,
"loss": 0.5008,
"step": 380
},
{
"epoch": 0.33362218370883884,
"grad_norm": 0.14255651831626892,
"learning_rate": 4.878775490871121e-05,
"loss": 0.4975,
"step": 385
},
{
"epoch": 0.3379549393414211,
"grad_norm": 0.12012791633605957,
"learning_rate": 4.864228593977006e-05,
"loss": 0.5047,
"step": 390
},
{
"epoch": 0.34228769497400346,
"grad_norm": 0.11433300375938416,
"learning_rate": 4.848881860555035e-05,
"loss": 0.4986,
"step": 395
},
{
"epoch": 0.3466204506065858,
"grad_norm": 0.11102011054754257,
"learning_rate": 4.832740482521931e-05,
"loss": 0.4981,
"step": 400
},
{
"epoch": 0.3509532062391681,
"grad_norm": 0.12340573221445084,
"learning_rate": 4.815809920628738e-05,
"loss": 0.4984,
"step": 405
},
{
"epoch": 0.35528596187175043,
"grad_norm": 0.1148650050163269,
"learning_rate": 4.7980959026134044e-05,
"loss": 0.4942,
"step": 410
},
{
"epoch": 0.3596187175043328,
"grad_norm": 0.10840712487697601,
"learning_rate": 4.7796044212630486e-05,
"loss": 0.4903,
"step": 415
},
{
"epoch": 0.36395147313691506,
"grad_norm": 0.11093516647815704,
"learning_rate": 4.7603417323865547e-05,
"loss": 0.4957,
"step": 420
},
{
"epoch": 0.3682842287694974,
"grad_norm": 0.1237047016620636,
"learning_rate": 4.74031435269818e-05,
"loss": 0.499,
"step": 425
},
{
"epoch": 0.37261698440207974,
"grad_norm": 0.11613244563341141,
"learning_rate": 4.7195290576129034e-05,
"loss": 0.4959,
"step": 430
},
{
"epoch": 0.37694974003466203,
"grad_norm": 0.10890854150056839,
"learning_rate": 4.697992878954255e-05,
"loss": 0.4944,
"step": 435
},
{
"epoch": 0.38128249566724437,
"grad_norm": 0.11364572495222092,
"learning_rate": 4.6757131025753886e-05,
"loss": 0.4909,
"step": 440
},
{
"epoch": 0.3856152512998267,
"grad_norm": 0.12619757652282715,
"learning_rate": 4.652697265894228e-05,
"loss": 0.4966,
"step": 445
},
{
"epoch": 0.389948006932409,
"grad_norm": 0.11669816076755524,
"learning_rate": 4.628953155343499e-05,
"loss": 0.4956,
"step": 450
},
{
"epoch": 0.39428076256499134,
"grad_norm": 0.13808482885360718,
"learning_rate": 4.604488803736523e-05,
"loss": 0.4973,
"step": 455
},
{
"epoch": 0.3986135181975737,
"grad_norm": 0.11171045899391174,
"learning_rate": 4.579312487549649e-05,
"loss": 0.4903,
"step": 460
},
{
"epoch": 0.40294627383015597,
"grad_norm": 0.11149395257234573,
"learning_rate": 4.553432724122265e-05,
"loss": 0.4999,
"step": 465
},
{
"epoch": 0.4072790294627383,
"grad_norm": 0.11308333277702332,
"learning_rate": 4.526858268775313e-05,
"loss": 0.4967,
"step": 470
},
{
"epoch": 0.41161178509532065,
"grad_norm": 0.10610105097293854,
"learning_rate": 4.499598111849299e-05,
"loss": 0.4936,
"step": 475
},
{
"epoch": 0.41594454072790293,
"grad_norm": 0.11356962472200394,
"learning_rate": 4.471661475662792e-05,
"loss": 0.493,
"step": 480
},
{
"epoch": 0.4202772963604853,
"grad_norm": 0.10310888290405273,
"learning_rate": 4.443057811392445e-05,
"loss": 0.5002,
"step": 485
},
{
"epoch": 0.4246100519930676,
"grad_norm": 0.11358631402254105,
"learning_rate": 4.413796795875586e-05,
"loss": 0.4983,
"step": 490
},
{
"epoch": 0.4289428076256499,
"grad_norm": 0.12575574219226837,
"learning_rate": 4.383888328336476e-05,
"loss": 0.4949,
"step": 495
},
{
"epoch": 0.43327556325823224,
"grad_norm": 0.09878399968147278,
"learning_rate": 4.3533425270373216e-05,
"loss": 0.4953,
"step": 500
},
{
"epoch": 0.43327556325823224,
"eval_loss": 0.4938061535358429,
"eval_runtime": 140.4236,
"eval_samples_per_second": 2.685,
"eval_steps_per_second": 0.677,
"step": 500
},
{
"epoch": 0.4376083188908146,
"grad_norm": 0.11693672835826874,
"learning_rate": 4.3221697258551906e-05,
"loss": 0.4934,
"step": 505
},
{
"epoch": 0.44194107452339687,
"grad_norm": 0.11737816035747528,
"learning_rate": 4.2903804707859835e-05,
"loss": 0.4934,
"step": 510
},
{
"epoch": 0.4462738301559792,
"grad_norm": 0.11542918533086777,
"learning_rate": 4.257985516376644e-05,
"loss": 0.4927,
"step": 515
},
{
"epoch": 0.4506065857885615,
"grad_norm": 0.10959002375602722,
"learning_rate": 4.224995822086812e-05,
"loss": 0.4935,
"step": 520
},
{
"epoch": 0.45493934142114384,
"grad_norm": 0.10852424800395966,
"learning_rate": 4.191422548581154e-05,
"loss": 0.4947,
"step": 525
},
{
"epoch": 0.4592720970537262,
"grad_norm": 0.11614954471588135,
"learning_rate": 4.157277053953631e-05,
"loss": 0.4935,
"step": 530
},
{
"epoch": 0.46360485268630847,
"grad_norm": 0.11209428310394287,
"learning_rate": 4.1225708898849594e-05,
"loss": 0.4975,
"step": 535
},
{
"epoch": 0.4679376083188908,
"grad_norm": 0.11714442819356918,
"learning_rate": 4.0873157977346e-05,
"loss": 0.4923,
"step": 540
},
{
"epoch": 0.47227036395147315,
"grad_norm": 0.1179489716887474,
"learning_rate": 4.051523704568557e-05,
"loss": 0.4939,
"step": 545
},
{
"epoch": 0.47660311958405543,
"grad_norm": 0.11872310936450958,
"learning_rate": 4.0152067191243696e-05,
"loss": 0.4927,
"step": 550
},
{
"epoch": 0.4809358752166378,
"grad_norm": 0.1182030588388443,
"learning_rate": 3.978377127714628e-05,
"loss": 0.4908,
"step": 555
},
{
"epoch": 0.4852686308492201,
"grad_norm": 0.11203285306692123,
"learning_rate": 3.941047390070419e-05,
"loss": 0.4898,
"step": 560
},
{
"epoch": 0.4896013864818024,
"grad_norm": 0.10341402143239975,
"learning_rate": 3.903230135126113e-05,
"loss": 0.4899,
"step": 565
},
{
"epoch": 0.49393414211438474,
"grad_norm": 0.11410869657993317,
"learning_rate": 3.864938156746891e-05,
"loss": 0.4914,
"step": 570
},
{
"epoch": 0.4982668977469671,
"grad_norm": 0.11390075832605362,
"learning_rate": 3.8261844094004815e-05,
"loss": 0.4922,
"step": 575
},
{
"epoch": 0.5025996533795494,
"grad_norm": 0.10876569896936417,
"learning_rate": 3.7869820037745776e-05,
"loss": 0.4964,
"step": 580
},
{
"epoch": 0.5069324090121318,
"grad_norm": 0.11602895706892014,
"learning_rate": 3.747344202341373e-05,
"loss": 0.4945,
"step": 585
},
{
"epoch": 0.511265164644714,
"grad_norm": 0.12035666406154633,
"learning_rate": 3.707284414870786e-05,
"loss": 0.499,
"step": 590
},
{
"epoch": 0.5155979202772963,
"grad_norm": 0.1848023533821106,
"learning_rate": 3.666816193893817e-05,
"loss": 0.4947,
"step": 595
},
{
"epoch": 0.5199306759098787,
"grad_norm": 0.12017575651407242,
"learning_rate": 3.6259532301176335e-05,
"loss": 0.4905,
"step": 600
},
{
"epoch": 0.524263431542461,
"grad_norm": 0.10070477426052094,
"learning_rate": 3.5847093477938956e-05,
"loss": 0.4974,
"step": 605
},
{
"epoch": 0.5285961871750433,
"grad_norm": 0.11581992357969284,
"learning_rate": 3.543098500041906e-05,
"loss": 0.4927,
"step": 610
},
{
"epoch": 0.5329289428076257,
"grad_norm": 0.09777580201625824,
"learning_rate": 3.501134764128167e-05,
"loss": 0.4907,
"step": 615
},
{
"epoch": 0.537261698440208,
"grad_norm": 0.10121456533670425,
"learning_rate": 3.458832336703929e-05,
"loss": 0.491,
"step": 620
},
{
"epoch": 0.5415944540727903,
"grad_norm": 0.10312589257955551,
"learning_rate": 3.416205529002363e-05,
"loss": 0.4942,
"step": 625
},
{
"epoch": 0.5459272097053726,
"grad_norm": 0.10299069434404373,
"learning_rate": 3.37326876199695e-05,
"loss": 0.493,
"step": 630
},
{
"epoch": 0.550259965337955,
"grad_norm": 0.11771389842033386,
"learning_rate": 3.3300365615227685e-05,
"loss": 0.4942,
"step": 635
},
{
"epoch": 0.5545927209705372,
"grad_norm": 0.1121087372303009,
"learning_rate": 3.286523553362287e-05,
"loss": 0.4946,
"step": 640
},
{
"epoch": 0.5589254766031195,
"grad_norm": 0.10604743659496307,
"learning_rate": 3.242744458297348e-05,
"loss": 0.4907,
"step": 645
},
{
"epoch": 0.5632582322357019,
"grad_norm": 0.10975582152605057,
"learning_rate": 3.1987140871290236e-05,
"loss": 0.4889,
"step": 650
},
{
"epoch": 0.5675909878682842,
"grad_norm": 0.10636895149946213,
"learning_rate": 3.154447335667001e-05,
"loss": 0.4935,
"step": 655
},
{
"epoch": 0.5719237435008665,
"grad_norm": 0.10076680779457092,
"learning_rate": 3.1099591796902215e-05,
"loss": 0.4925,
"step": 660
},
{
"epoch": 0.5762564991334489,
"grad_norm": 0.10885344445705414,
"learning_rate": 3.0652646698804585e-05,
"loss": 0.4892,
"step": 665
},
{
"epoch": 0.5805892547660312,
"grad_norm": 0.11372784525156021,
"learning_rate": 3.0203789267305567e-05,
"loss": 0.4918,
"step": 670
},
{
"epoch": 0.5849220103986135,
"grad_norm": 0.11742879450321198,
"learning_rate": 2.975317135429056e-05,
"loss": 0.4885,
"step": 675
},
{
"epoch": 0.5892547660311959,
"grad_norm": 0.09908663481473923,
"learning_rate": 2.930094540722927e-05,
"loss": 0.4889,
"step": 680
},
{
"epoch": 0.5935875216637782,
"grad_norm": 0.11032566428184509,
"learning_rate": 2.884726441760155e-05,
"loss": 0.4936,
"step": 685
},
{
"epoch": 0.5979202772963604,
"grad_norm": 0.11033419519662857,
"learning_rate": 2.8392281869139213e-05,
"loss": 0.4925,
"step": 690
},
{
"epoch": 0.6022530329289428,
"grad_norm": 0.12115441262722015,
"learning_rate": 2.7936151685901373e-05,
"loss": 0.49,
"step": 695
},
{
"epoch": 0.6065857885615251,
"grad_norm": 0.13125091791152954,
"learning_rate": 2.747902818020067e-05,
"loss": 0.4919,
"step": 700
},
{
"epoch": 0.6109185441941074,
"grad_norm": 0.11565785109996796,
"learning_rate": 2.7021066000398344e-05,
"loss": 0.4928,
"step": 705
},
{
"epoch": 0.6152512998266898,
"grad_norm": 0.12640348076820374,
"learning_rate": 2.6562420078585433e-05,
"loss": 0.489,
"step": 710
},
{
"epoch": 0.6195840554592721,
"grad_norm": 0.12297876179218292,
"learning_rate": 2.6103245578168106e-05,
"loss": 0.4919,
"step": 715
},
{
"epoch": 0.6239168110918544,
"grad_norm": 0.10511300712823868,
"learning_rate": 2.564369784137472e-05,
"loss": 0.4883,
"step": 720
},
{
"epoch": 0.6282495667244368,
"grad_norm": 0.10803347080945969,
"learning_rate": 2.5183932336702297e-05,
"loss": 0.4909,
"step": 725
},
{
"epoch": 0.6325823223570191,
"grad_norm": 0.11415662616491318,
"learning_rate": 2.4724104606320445e-05,
"loss": 0.4878,
"step": 730
},
{
"epoch": 0.6369150779896013,
"grad_norm": 0.10404873639345169,
"learning_rate": 2.426437021345015e-05,
"loss": 0.4901,
"step": 735
},
{
"epoch": 0.6412478336221837,
"grad_norm": 0.11284226924180984,
"learning_rate": 2.3804884689735642e-05,
"loss": 0.4906,
"step": 740
},
{
"epoch": 0.645580589254766,
"grad_norm": 0.1138242855668068,
"learning_rate": 2.3345803482626797e-05,
"loss": 0.494,
"step": 745
},
{
"epoch": 0.6499133448873483,
"grad_norm": 0.18223117291927338,
"learning_rate": 2.288728190279014e-05,
"loss": 0.4886,
"step": 750
},
{
"epoch": 0.6542461005199307,
"grad_norm": 0.1119702160358429,
"learning_rate": 2.2429475071565987e-05,
"loss": 0.4908,
"step": 755
},
{
"epoch": 0.658578856152513,
"grad_norm": 0.11668991297483444,
"learning_rate": 2.1972537868489797e-05,
"loss": 0.4896,
"step": 760
},
{
"epoch": 0.6629116117850953,
"grad_norm": 0.10587523132562637,
"learning_rate": 2.151662487889518e-05,
"loss": 0.4885,
"step": 765
},
{
"epoch": 0.6672443674176777,
"grad_norm": 0.10824766755104065,
"learning_rate": 2.1061890341616558e-05,
"loss": 0.4873,
"step": 770
},
{
"epoch": 0.67157712305026,
"grad_norm": 0.11111487448215485,
"learning_rate": 2.060848809680893e-05,
"loss": 0.4853,
"step": 775
},
{
"epoch": 0.6759098786828422,
"grad_norm": 0.10754924267530441,
"learning_rate": 2.0156571533902627e-05,
"loss": 0.4849,
"step": 780
},
{
"epoch": 0.6802426343154246,
"grad_norm": 0.10407605022192001,
"learning_rate": 1.97062935397105e-05,
"loss": 0.488,
"step": 785
},
{
"epoch": 0.6845753899480069,
"grad_norm": 0.1077013909816742,
"learning_rate": 1.9257806446705116e-05,
"loss": 0.4867,
"step": 790
},
{
"epoch": 0.6889081455805892,
"grad_norm": 0.11459195613861084,
"learning_rate": 1.8811261981483548e-05,
"loss": 0.4911,
"step": 795
},
{
"epoch": 0.6932409012131716,
"grad_norm": 0.12345458567142487,
"learning_rate": 1.8366811213437092e-05,
"loss": 0.4888,
"step": 800
},
{
"epoch": 0.6975736568457539,
"grad_norm": 0.10806908458471298,
"learning_rate": 1.7924604503643367e-05,
"loss": 0.4903,
"step": 805
},
{
"epoch": 0.7019064124783362,
"grad_norm": 0.12911449372768402,
"learning_rate": 1.7484791453998006e-05,
"loss": 0.4874,
"step": 810
},
{
"epoch": 0.7062391681109186,
"grad_norm": 0.11117111146450043,
"learning_rate": 1.7047520856603183e-05,
"loss": 0.4895,
"step": 815
},
{
"epoch": 0.7105719237435009,
"grad_norm": 0.10993503779172897,
"learning_rate": 1.6612940643430138e-05,
"loss": 0.4909,
"step": 820
},
{
"epoch": 0.7149046793760832,
"grad_norm": 0.10971593856811523,
"learning_rate": 1.618119783627263e-05,
"loss": 0.4898,
"step": 825
},
{
"epoch": 0.7192374350086655,
"grad_norm": 0.12124701589345932,
"learning_rate": 1.5752438497008405e-05,
"loss": 0.4886,
"step": 830
},
{
"epoch": 0.7235701906412478,
"grad_norm": 0.10207706689834595,
"learning_rate": 1.5326807678185267e-05,
"loss": 0.4893,
"step": 835
},
{
"epoch": 0.7279029462738301,
"grad_norm": 0.10885774344205856,
"learning_rate": 1.490444937394879e-05,
"loss": 0.4891,
"step": 840
},
{
"epoch": 0.7322357019064125,
"grad_norm": 0.10901923477649689,
"learning_rate": 1.4485506471327914e-05,
"loss": 0.4968,
"step": 845
},
{
"epoch": 0.7365684575389948,
"grad_norm": 0.11790922284126282,
"learning_rate": 1.407012070189524e-05,
"loss": 0.4861,
"step": 850
},
{
"epoch": 0.7409012131715771,
"grad_norm": 0.11575620621442795,
"learning_rate": 1.3658432593818149e-05,
"loss": 0.4888,
"step": 855
},
{
"epoch": 0.7452339688041595,
"grad_norm": 0.11284902691841125,
"learning_rate": 1.325058142431701e-05,
"loss": 0.4879,
"step": 860
},
{
"epoch": 0.7495667244367418,
"grad_norm": 0.11264630407094955,
"learning_rate": 1.2846705172546675e-05,
"loss": 0.4909,
"step": 865
},
{
"epoch": 0.7538994800693241,
"grad_norm": 0.11645620316267014,
"learning_rate": 1.2446940472917099e-05,
"loss": 0.485,
"step": 870
},
{
"epoch": 0.7582322357019065,
"grad_norm": 0.10355979204177856,
"learning_rate": 1.2051422568868833e-05,
"loss": 0.4891,
"step": 875
},
{
"epoch": 0.7625649913344887,
"grad_norm": 0.10600101947784424,
"learning_rate": 1.1660285267119167e-05,
"loss": 0.4903,
"step": 880
},
{
"epoch": 0.766897746967071,
"grad_norm": 0.11851372569799423,
"learning_rate": 1.1273660892394297e-05,
"loss": 0.4895,
"step": 885
},
{
"epoch": 0.7712305025996534,
"grad_norm": 0.11533911526203156,
"learning_rate": 1.0891680242662835e-05,
"loss": 0.4843,
"step": 890
},
{
"epoch": 0.7755632582322357,
"grad_norm": 0.09746929258108139,
"learning_rate": 1.051447254488591e-05,
"loss": 0.4894,
"step": 895
},
{
"epoch": 0.779896013864818,
"grad_norm": 0.1193188726902008,
"learning_rate": 1.0142165411298662e-05,
"loss": 0.4884,
"step": 900
},
{
"epoch": 0.7842287694974004,
"grad_norm": 0.11370085179805756,
"learning_rate": 9.774884796238085e-06,
"loss": 0.49,
"step": 905
},
{
"epoch": 0.7885615251299827,
"grad_norm": 0.1011599749326706,
"learning_rate": 9.412754953531663e-06,
"loss": 0.4884,
"step": 910
},
{
"epoch": 0.792894280762565,
"grad_norm": 0.11001604050397873,
"learning_rate": 9.055898394461423e-06,
"loss": 0.4883,
"step": 915
},
{
"epoch": 0.7972270363951474,
"grad_norm": 0.11630392074584961,
"learning_rate": 8.704435846317386e-06,
"loss": 0.4869,
"step": 920
},
{
"epoch": 0.8015597920277296,
"grad_norm": 0.10578346997499466,
"learning_rate": 8.358486211554637e-06,
"loss": 0.4887,
"step": 925
},
{
"epoch": 0.8058925476603119,
"grad_norm": 0.11845128238201141,
"learning_rate": 8.018166527567672e-06,
"loss": 0.4852,
"step": 930
},
{
"epoch": 0.8102253032928943,
"grad_norm": 0.10032304376363754,
"learning_rate": 7.683591927095824e-06,
"loss": 0.4865,
"step": 935
},
{
"epoch": 0.8145580589254766,
"grad_norm": 0.10029744356870651,
"learning_rate": 7.354875599272928e-06,
"loss": 0.4902,
"step": 940
},
{
"epoch": 0.8188908145580589,
"grad_norm": 0.10729606449604034,
"learning_rate": 7.0321287513346074e-06,
"loss": 0.4879,
"step": 945
},
{
"epoch": 0.8232235701906413,
"grad_norm": 0.11077585816383362,
"learning_rate": 6.715460570995988e-06,
"loss": 0.4905,
"step": 950
},
{
"epoch": 0.8275563258232236,
"grad_norm": 0.112645223736763,
"learning_rate": 6.404978189512681e-06,
"loss": 0.492,
"step": 955
},
{
"epoch": 0.8318890814558059,
"grad_norm": 0.11094575375318527,
"learning_rate": 6.100786645437481e-06,
"loss": 0.4855,
"step": 960
},
{
"epoch": 0.8362218370883883,
"grad_norm": 0.10790159553289413,
"learning_rate": 5.8029888490850005e-06,
"loss": 0.4942,
"step": 965
},
{
"epoch": 0.8405545927209706,
"grad_norm": 0.11253953725099564,
"learning_rate": 5.511685547716328e-06,
"loss": 0.4829,
"step": 970
},
{
"epoch": 0.8448873483535528,
"grad_norm": 0.11395300179719925,
"learning_rate": 5.226975291455477e-06,
"loss": 0.4856,
"step": 975
},
{
"epoch": 0.8492201039861352,
"grad_norm": 0.1128508597612381,
"learning_rate": 4.9489543999491045e-06,
"loss": 0.4852,
"step": 980
},
{
"epoch": 0.8535528596187175,
"grad_norm": 0.11124628782272339,
"learning_rate": 4.67771692978087e-06,
"loss": 0.4885,
"step": 985
},
{
"epoch": 0.8578856152512998,
"grad_norm": 0.10993985086679459,
"learning_rate": 4.413354642651369e-06,
"loss": 0.4867,
"step": 990
},
{
"epoch": 0.8622183708838822,
"grad_norm": 0.1036507710814476,
"learning_rate": 4.1559569743344405e-06,
"loss": 0.4839,
"step": 995
},
{
"epoch": 0.8665511265164645,
"grad_norm": 0.10577788203954697,
"learning_rate": 3.90561100442036e-06,
"loss": 0.4909,
"step": 1000
},
{
"epoch": 0.8665511265164645,
"eval_loss": 0.48741263151168823,
"eval_runtime": 139.1562,
"eval_samples_per_second": 2.709,
"eval_steps_per_second": 0.683,
"step": 1000
},
{
"epoch": 0.8708838821490468,
"grad_norm": 0.10687436163425446,
"learning_rate": 3.662401426856177e-06,
"loss": 0.4912,
"step": 1005
},
{
"epoch": 0.8752166377816292,
"grad_norm": 0.10468524694442749,
"learning_rate": 3.4264105212930915e-06,
"loss": 0.4882,
"step": 1010
},
{
"epoch": 0.8795493934142115,
"grad_norm": 0.11409106850624084,
"learning_rate": 3.197718125250618e-06,
"loss": 0.4834,
"step": 1015
},
{
"epoch": 0.8838821490467937,
"grad_norm": 0.1069907397031784,
"learning_rate": 2.9764016071069434e-06,
"loss": 0.4881,
"step": 1020
},
{
"epoch": 0.8882149046793761,
"grad_norm": 0.10976656526327133,
"learning_rate": 2.7625358399246376e-06,
"loss": 0.4852,
"step": 1025
},
{
"epoch": 0.8925476603119584,
"grad_norm": 0.11071466654539108,
"learning_rate": 2.5561931761205082e-06,
"loss": 0.4879,
"step": 1030
},
{
"epoch": 0.8968804159445407,
"grad_norm": 0.10794007033109665,
"learning_rate": 2.3574434229882145e-06,
"loss": 0.4865,
"step": 1035
},
{
"epoch": 0.901213171577123,
"grad_norm": 0.11333485692739487,
"learning_rate": 2.166353819081968e-06,
"loss": 0.4855,
"step": 1040
},
{
"epoch": 0.9055459272097054,
"grad_norm": 0.10130605846643448,
"learning_rate": 1.982989011469172e-06,
"loss": 0.4868,
"step": 1045
},
{
"epoch": 0.9098786828422877,
"grad_norm": 0.10454142093658447,
"learning_rate": 1.8074110338598682e-06,
"loss": 0.487,
"step": 1050
},
{
"epoch": 0.91421143847487,
"grad_norm": 0.10157745331525803,
"learning_rate": 1.6396792856202298e-06,
"loss": 0.4881,
"step": 1055
},
{
"epoch": 0.9185441941074524,
"grad_norm": 0.11562693864107132,
"learning_rate": 1.479850511677322e-06,
"loss": 0.4877,
"step": 1060
},
{
"epoch": 0.9228769497400346,
"grad_norm": 0.12777337431907654,
"learning_rate": 1.3279787833218537e-06,
"loss": 0.4886,
"step": 1065
},
{
"epoch": 0.9272097053726169,
"grad_norm": 0.10074026137590408,
"learning_rate": 1.1841154799154374e-06,
"loss": 0.4856,
"step": 1070
},
{
"epoch": 0.9315424610051993,
"grad_norm": 0.11187005788087845,
"learning_rate": 1.0483092715085879e-06,
"loss": 0.4872,
"step": 1075
},
{
"epoch": 0.9358752166377816,
"grad_norm": 0.10271576046943665,
"learning_rate": 9.206061023752516e-07,
"loss": 0.4867,
"step": 1080
},
{
"epoch": 0.9402079722703639,
"grad_norm": 0.11686773598194122,
"learning_rate": 8.010491754695177e-07,
"loss": 0.4868,
"step": 1085
},
{
"epoch": 0.9445407279029463,
"grad_norm": 0.12068697810173035,
"learning_rate": 6.896789378097179e-07,
"loss": 0.4911,
"step": 1090
},
{
"epoch": 0.9488734835355286,
"grad_norm": 0.10469616949558258,
"learning_rate": 5.865330667949115e-07,
"loss": 0.485,
"step": 1095
},
{
"epoch": 0.9532062391681109,
"grad_norm": 0.11302938312292099,
"learning_rate": 4.916464574583251e-07,
"loss": 0.4864,
"step": 1100
},
{
"epoch": 0.9575389948006933,
"grad_norm": 0.11449886858463287,
"learning_rate": 4.050512106620913e-07,
"loss": 0.4873,
"step": 1105
},
{
"epoch": 0.9618717504332756,
"grad_norm": 0.1062023788690567,
"learning_rate": 3.267766222372931e-07,
"loss": 0.4885,
"step": 1110
},
{
"epoch": 0.9662045060658578,
"grad_norm": 0.11473377794027328,
"learning_rate": 2.568491730729539e-07,
"loss": 0.4869,
"step": 1115
},
{
"epoch": 0.9705372616984402,
"grad_norm": 0.10777969658374786,
"learning_rate": 1.95292520157353e-07,
"loss": 0.489,
"step": 1120
},
{
"epoch": 0.9748700173310225,
"grad_norm": 0.10438723862171173,
"learning_rate": 1.4212748857468926e-07,
"loss": 0.4871,
"step": 1125
},
{
"epoch": 0.9792027729636048,
"grad_norm": 0.10945618152618408,
"learning_rate": 9.737206445979696e-08,
"loss": 0.4875,
"step": 1130
},
{
"epoch": 0.9835355285961872,
"grad_norm": 0.10462047904729843,
"learning_rate": 6.104138891329659e-08,
"loss": 0.4895,
"step": 1135
},
{
"epoch": 0.9878682842287695,
"grad_norm": 0.1181010752916336,
"learning_rate": 3.314775287923677e-08,
"loss": 0.4857,
"step": 1140
},
{
"epoch": 0.9922010398613518,
"grad_norm": 0.10257267206907272,
"learning_rate": 1.3700592986998217e-08,
"loss": 0.4866,
"step": 1145
},
{
"epoch": 0.9965337954939342,
"grad_norm": 0.10488082468509674,
"learning_rate": 2.7064883587807345e-09,
"loss": 0.4893,
"step": 1150
},
{
"epoch": 1.0,
"step": 1154,
"total_flos": 8.066906520991826e+17,
"train_loss": 0.5547944433354423,
"train_runtime": 22221.3478,
"train_samples_per_second": 0.831,
"train_steps_per_second": 0.052
}
],
"logging_steps": 5,
"max_steps": 1154,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 8.066906520991826e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}