mistral_poe_small / trainer_state.json
terry69's picture
Model save
3af7bb2 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9994869163673679,
"eval_steps": 500,
"global_step": 974,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"grad_norm": 3.351157385916466,
"learning_rate": 2.040816326530612e-06,
"loss": 1.6096,
"step": 1
},
{
"epoch": 0.01,
"grad_norm": 2.2833707241529,
"learning_rate": 1.0204081632653061e-05,
"loss": 1.307,
"step": 5
},
{
"epoch": 0.01,
"grad_norm": 1.2912917729665443,
"learning_rate": 2.0408163265306123e-05,
"loss": 1.0807,
"step": 10
},
{
"epoch": 0.02,
"grad_norm": 0.9989739565689935,
"learning_rate": 3.061224489795919e-05,
"loss": 1.0011,
"step": 15
},
{
"epoch": 0.02,
"grad_norm": 1.079400758509174,
"learning_rate": 4.0816326530612245e-05,
"loss": 0.6897,
"step": 20
},
{
"epoch": 0.03,
"grad_norm": 0.6210959789194382,
"learning_rate": 5.102040816326531e-05,
"loss": 0.6209,
"step": 25
},
{
"epoch": 0.03,
"grad_norm": 0.604422455243622,
"learning_rate": 6.122448979591838e-05,
"loss": 0.6185,
"step": 30
},
{
"epoch": 0.04,
"grad_norm": 0.5170498833821362,
"learning_rate": 7.142857142857143e-05,
"loss": 0.5902,
"step": 35
},
{
"epoch": 0.04,
"grad_norm": 0.2650153905272352,
"learning_rate": 8.163265306122449e-05,
"loss": 0.5035,
"step": 40
},
{
"epoch": 0.05,
"grad_norm": 0.3188138819665904,
"learning_rate": 9.183673469387756e-05,
"loss": 0.5295,
"step": 45
},
{
"epoch": 0.05,
"grad_norm": 0.36979118306554615,
"learning_rate": 0.00010204081632653062,
"loss": 0.5734,
"step": 50
},
{
"epoch": 0.06,
"grad_norm": 0.3973602747604989,
"learning_rate": 0.00011224489795918367,
"loss": 0.594,
"step": 55
},
{
"epoch": 0.06,
"grad_norm": 0.30107519885544926,
"learning_rate": 0.00012244897959183676,
"loss": 0.5777,
"step": 60
},
{
"epoch": 0.07,
"grad_norm": 0.35967504356035196,
"learning_rate": 0.0001326530612244898,
"loss": 0.547,
"step": 65
},
{
"epoch": 0.07,
"grad_norm": 0.4161565630483454,
"learning_rate": 0.00014285714285714287,
"loss": 0.6266,
"step": 70
},
{
"epoch": 0.08,
"grad_norm": 0.46021023477325246,
"learning_rate": 0.0001530612244897959,
"loss": 0.5249,
"step": 75
},
{
"epoch": 0.08,
"grad_norm": 0.5486875740290827,
"learning_rate": 0.00016326530612244898,
"loss": 0.6381,
"step": 80
},
{
"epoch": 0.09,
"grad_norm": 0.3715332784286634,
"learning_rate": 0.00017346938775510205,
"loss": 0.5778,
"step": 85
},
{
"epoch": 0.09,
"grad_norm": 0.25669849660669547,
"learning_rate": 0.00018367346938775512,
"loss": 0.5461,
"step": 90
},
{
"epoch": 0.1,
"grad_norm": 0.30206213988264413,
"learning_rate": 0.00019387755102040816,
"loss": 0.4736,
"step": 95
},
{
"epoch": 0.1,
"grad_norm": 0.33277431668095503,
"learning_rate": 0.0001999974277115551,
"loss": 0.5328,
"step": 100
},
{
"epoch": 0.11,
"grad_norm": 0.3336885386817361,
"learning_rate": 0.00019996849098629418,
"loss": 0.62,
"step": 105
},
{
"epoch": 0.11,
"grad_norm": 0.3468751205649051,
"learning_rate": 0.00019990741151022301,
"loss": 0.5247,
"step": 110
},
{
"epoch": 0.12,
"grad_norm": 0.3115320559980584,
"learning_rate": 0.0001998142089221534,
"loss": 0.4816,
"step": 115
},
{
"epoch": 0.12,
"grad_norm": 1.7970263609657833,
"learning_rate": 0.0001996889131894033,
"loss": 0.5282,
"step": 120
},
{
"epoch": 0.13,
"grad_norm": 0.420052571262797,
"learning_rate": 0.00019953156459816179,
"loss": 0.5812,
"step": 125
},
{
"epoch": 0.13,
"grad_norm": 0.2997887688631509,
"learning_rate": 0.0001993422137405354,
"loss": 0.5722,
"step": 130
},
{
"epoch": 0.14,
"grad_norm": 0.4538126671892769,
"learning_rate": 0.00019912092149828174,
"loss": 0.4534,
"step": 135
},
{
"epoch": 0.14,
"grad_norm": 0.33254418592286095,
"learning_rate": 0.00019886775902323405,
"loss": 0.5745,
"step": 140
},
{
"epoch": 0.15,
"grad_norm": 0.3406819289335851,
"learning_rate": 0.00019858280771442385,
"loss": 0.4937,
"step": 145
},
{
"epoch": 0.15,
"grad_norm": 0.44532117964605517,
"learning_rate": 0.00019826615919190887,
"loss": 0.5034,
"step": 150
},
{
"epoch": 0.16,
"grad_norm": 0.26124089824170305,
"learning_rate": 0.00019791791526731445,
"loss": 0.4308,
"step": 155
},
{
"epoch": 0.16,
"grad_norm": 0.41699002376176875,
"learning_rate": 0.00019753818791109828,
"loss": 0.5843,
"step": 160
},
{
"epoch": 0.17,
"grad_norm": 0.30966016172554256,
"learning_rate": 0.0001971270992165486,
"loss": 0.6161,
"step": 165
},
{
"epoch": 0.17,
"grad_norm": 0.35777456904642235,
"learning_rate": 0.00019668478136052774,
"loss": 0.5352,
"step": 170
},
{
"epoch": 0.18,
"grad_norm": 0.30456414186376835,
"learning_rate": 0.0001962113765609735,
"loss": 0.5787,
"step": 175
},
{
"epoch": 0.18,
"grad_norm": 0.42315137043021306,
"learning_rate": 0.0001957070370311717,
"loss": 0.5959,
"step": 180
},
{
"epoch": 0.19,
"grad_norm": 0.3294546389618734,
"learning_rate": 0.00019517192493081565,
"loss": 0.4895,
"step": 185
},
{
"epoch": 0.19,
"grad_norm": 0.3169197670504051,
"learning_rate": 0.00019460621231386676,
"loss": 0.5021,
"step": 190
},
{
"epoch": 0.2,
"grad_norm": 0.2834179242800674,
"learning_rate": 0.00019401008107323455,
"loss": 0.6374,
"step": 195
},
{
"epoch": 0.21,
"grad_norm": 0.47305112179270714,
"learning_rate": 0.0001933837228822925,
"loss": 0.6208,
"step": 200
},
{
"epoch": 0.21,
"grad_norm": 0.28157269728650575,
"learning_rate": 0.0001927273391332499,
"loss": 0.5011,
"step": 205
},
{
"epoch": 0.22,
"grad_norm": 0.34262878952387293,
"learning_rate": 0.00019204114087239806,
"loss": 0.6011,
"step": 210
},
{
"epoch": 0.22,
"grad_norm": 0.2720473437585886,
"learning_rate": 0.00019132534873225323,
"loss": 0.5448,
"step": 215
},
{
"epoch": 0.23,
"grad_norm": 0.23695943062888203,
"learning_rate": 0.00019058019286061665,
"loss": 0.5199,
"step": 220
},
{
"epoch": 0.23,
"grad_norm": 0.38061349949148005,
"learning_rate": 0.00018980591284657535,
"loss": 0.5167,
"step": 225
},
{
"epoch": 0.24,
"grad_norm": 0.31515458071917996,
"learning_rate": 0.00018900275764346768,
"loss": 0.5201,
"step": 230
},
{
"epoch": 0.24,
"grad_norm": 0.3592571813640925,
"learning_rate": 0.0001881709854888372,
"loss": 0.5638,
"step": 235
},
{
"epoch": 0.25,
"grad_norm": 0.2780598506887593,
"learning_rate": 0.00018731086382140226,
"loss": 0.5052,
"step": 240
},
{
"epoch": 0.25,
"grad_norm": 0.34554750179980265,
"learning_rate": 0.00018642266919506644,
"loss": 0.6022,
"step": 245
},
{
"epoch": 0.26,
"grad_norm": 0.35137223950131713,
"learning_rate": 0.00018550668718999872,
"loss": 0.5534,
"step": 250
},
{
"epoch": 0.26,
"grad_norm": 0.27529845988300145,
"learning_rate": 0.0001845632123208111,
"loss": 0.5037,
"step": 255
},
{
"epoch": 0.27,
"grad_norm": 0.3706754129199105,
"learning_rate": 0.0001835925479418637,
"loss": 0.5209,
"step": 260
},
{
"epoch": 0.27,
"grad_norm": 0.3098529350262172,
"learning_rate": 0.0001825950061497276,
"loss": 0.5844,
"step": 265
},
{
"epoch": 0.28,
"grad_norm": 0.3548275391577921,
"learning_rate": 0.00018157090768283678,
"loss": 0.5311,
"step": 270
},
{
"epoch": 0.28,
"grad_norm": 0.3284638012761491,
"learning_rate": 0.00018052058181836151,
"loss": 0.5675,
"step": 275
},
{
"epoch": 0.29,
"grad_norm": 0.38114338073859594,
"learning_rate": 0.00017944436626633623,
"loss": 0.5932,
"step": 280
},
{
"epoch": 0.29,
"grad_norm": 0.3299526601897124,
"learning_rate": 0.00017834260706107595,
"loss": 0.5857,
"step": 285
},
{
"epoch": 0.3,
"grad_norm": 0.327976106509182,
"learning_rate": 0.00017721565844991643,
"loss": 0.5068,
"step": 290
},
{
"epoch": 0.3,
"grad_norm": 0.35435043972661756,
"learning_rate": 0.00017606388277931328,
"loss": 0.5429,
"step": 295
},
{
"epoch": 0.31,
"grad_norm": 0.38328377811579767,
"learning_rate": 0.0001748876503783373,
"loss": 0.6278,
"step": 300
},
{
"epoch": 0.31,
"grad_norm": 0.3261327232654444,
"learning_rate": 0.00017368733943960276,
"loss": 0.5533,
"step": 305
},
{
"epoch": 0.32,
"grad_norm": 0.3009811720038025,
"learning_rate": 0.00017246333589766787,
"loss": 0.5083,
"step": 310
},
{
"epoch": 0.32,
"grad_norm": 0.3342684613566798,
"learning_rate": 0.00017121603330494544,
"loss": 0.6145,
"step": 315
},
{
"epoch": 0.33,
"grad_norm": 0.31521187564836095,
"learning_rate": 0.0001699458327051647,
"loss": 0.5734,
"step": 320
},
{
"epoch": 0.33,
"grad_norm": 0.3840993018659151,
"learning_rate": 0.00016865314250442398,
"loss": 0.5061,
"step": 325
},
{
"epoch": 0.34,
"grad_norm": 0.4041050776935621,
"learning_rate": 0.00016733837833987633,
"loss": 0.5413,
"step": 330
},
{
"epoch": 0.34,
"grad_norm": 0.402488525664061,
"learning_rate": 0.00016600196294609045,
"loss": 0.5306,
"step": 335
},
{
"epoch": 0.35,
"grad_norm": 0.34505184031782365,
"learning_rate": 0.00016464432601912912,
"loss": 0.4892,
"step": 340
},
{
"epoch": 0.35,
"grad_norm": 0.4416727448065266,
"learning_rate": 0.0001632659040783897,
"loss": 0.6394,
"step": 345
},
{
"epoch": 0.36,
"grad_norm": 0.2902721265803442,
"learning_rate": 0.00016186714032625035,
"loss": 0.422,
"step": 350
},
{
"epoch": 0.36,
"grad_norm": 0.34038881357193446,
"learning_rate": 0.00016044848450556787,
"loss": 0.4897,
"step": 355
},
{
"epoch": 0.37,
"grad_norm": 0.3537467402249508,
"learning_rate": 0.00015901039275507245,
"loss": 0.5271,
"step": 360
},
{
"epoch": 0.37,
"grad_norm": 0.3096464093822923,
"learning_rate": 0.00015755332746270572,
"loss": 0.4851,
"step": 365
},
{
"epoch": 0.38,
"grad_norm": 0.33349578622268794,
"learning_rate": 0.00015607775711694977,
"loss": 0.4433,
"step": 370
},
{
"epoch": 0.38,
"grad_norm": 0.3984031583488789,
"learning_rate": 0.00015458415615619484,
"loss": 0.5227,
"step": 375
},
{
"epoch": 0.39,
"grad_norm": 0.3791504242897034,
"learning_rate": 0.00015307300481619333,
"loss": 0.5553,
"step": 380
},
{
"epoch": 0.4,
"grad_norm": 0.3624413301072706,
"learning_rate": 0.00015154478897565045,
"loss": 0.6114,
"step": 385
},
{
"epoch": 0.4,
"grad_norm": 0.3660206278643239,
"learning_rate": 0.00015000000000000001,
"loss": 0.5811,
"step": 390
},
{
"epoch": 0.41,
"grad_norm": 0.3465368400413121,
"learning_rate": 0.00014843913458341645,
"loss": 0.5336,
"step": 395
},
{
"epoch": 0.41,
"grad_norm": 0.3962794628289249,
"learning_rate": 0.00014686269458911332,
"loss": 0.5364,
"step": 400
},
{
"epoch": 0.42,
"grad_norm": 0.3039345955854238,
"learning_rate": 0.00014527118688797963,
"loss": 0.5545,
"step": 405
},
{
"epoch": 0.42,
"grad_norm": 0.3884998390679994,
"learning_rate": 0.0001436651231956064,
"loss": 0.6274,
"step": 410
},
{
"epoch": 0.43,
"grad_norm": 0.43141743992301423,
"learning_rate": 0.00014204501990775533,
"loss": 0.6474,
"step": 415
},
{
"epoch": 0.43,
"grad_norm": 0.31938982955473066,
"learning_rate": 0.00014041139793432274,
"loss": 0.5153,
"step": 420
},
{
"epoch": 0.44,
"grad_norm": 0.3760731825930333,
"learning_rate": 0.00013876478253185183,
"loss": 0.5613,
"step": 425
},
{
"epoch": 0.44,
"grad_norm": 0.2586169946555733,
"learning_rate": 0.00013710570313464778,
"loss": 0.61,
"step": 430
},
{
"epoch": 0.45,
"grad_norm": 0.2412655626604391,
"learning_rate": 0.0001354346931845492,
"loss": 0.4844,
"step": 435
},
{
"epoch": 0.45,
"grad_norm": 0.3116268309859188,
"learning_rate": 0.00013375228995941133,
"loss": 0.5836,
"step": 440
},
{
"epoch": 0.46,
"grad_norm": 0.3358278406427287,
"learning_rate": 0.0001320590344003557,
"loss": 0.5687,
"step": 445
},
{
"epoch": 0.46,
"grad_norm": 0.3678466240784056,
"learning_rate": 0.00013035547093784186,
"loss": 0.5203,
"step": 450
},
{
"epoch": 0.47,
"grad_norm": 0.30727587441358567,
"learning_rate": 0.00012864214731661742,
"loss": 0.5356,
"step": 455
},
{
"epoch": 0.47,
"grad_norm": 0.3488513387485165,
"learning_rate": 0.00012691961441960238,
"loss": 0.5693,
"step": 460
},
{
"epoch": 0.48,
"grad_norm": 0.36921204919306894,
"learning_rate": 0.00012518842609076413,
"loss": 0.6275,
"step": 465
},
{
"epoch": 0.48,
"grad_norm": 0.23691846553937704,
"learning_rate": 0.00012344913895704097,
"loss": 0.5006,
"step": 470
},
{
"epoch": 0.49,
"grad_norm": 0.3452667363717209,
"learning_rate": 0.00012170231224937032,
"loss": 0.4681,
"step": 475
},
{
"epoch": 0.49,
"grad_norm": 0.3547204515478599,
"learning_rate": 0.00011994850762287989,
"loss": 0.4838,
"step": 480
},
{
"epoch": 0.5,
"grad_norm": 0.29663657465334964,
"learning_rate": 0.0001181882889762994,
"loss": 0.5503,
"step": 485
},
{
"epoch": 0.5,
"grad_norm": 0.4673130847127176,
"learning_rate": 0.00011642222227065089,
"loss": 0.4547,
"step": 490
},
{
"epoch": 0.51,
"grad_norm": 0.32632856811462063,
"learning_rate": 0.00011465087534727587,
"loss": 0.639,
"step": 495
},
{
"epoch": 0.51,
"grad_norm": 0.2884311338934753,
"learning_rate": 0.0001128748177452581,
"loss": 0.5277,
"step": 500
},
{
"epoch": 0.52,
"grad_norm": 0.31893330070055553,
"learning_rate": 0.00011109462051830017,
"loss": 0.607,
"step": 505
},
{
"epoch": 0.52,
"grad_norm": 0.30741665764408194,
"learning_rate": 0.00010931085605111354,
"loss": 0.53,
"step": 510
},
{
"epoch": 0.53,
"grad_norm": 0.3697277160270557,
"learning_rate": 0.00010752409787538,
"loss": 0.5399,
"step": 515
},
{
"epoch": 0.53,
"grad_norm": 0.2544618124964993,
"learning_rate": 0.00010573492048534515,
"loss": 0.5838,
"step": 520
},
{
"epoch": 0.54,
"grad_norm": 0.293117155180105,
"learning_rate": 0.00010394389915310149,
"loss": 0.6265,
"step": 525
},
{
"epoch": 0.54,
"grad_norm": 0.253262019800845,
"learning_rate": 0.00010215160974362223,
"loss": 0.458,
"step": 530
},
{
"epoch": 0.55,
"grad_norm": 0.3086878679373377,
"learning_rate": 0.00010035862852960387,
"loss": 0.558,
"step": 535
},
{
"epoch": 0.55,
"grad_norm": 0.26554052796001515,
"learning_rate": 9.856553200617805e-05,
"loss": 0.4748,
"step": 540
},
{
"epoch": 0.56,
"grad_norm": 0.3957125840673903,
"learning_rate": 9.677289670555169e-05,
"loss": 0.4904,
"step": 545
},
{
"epoch": 0.56,
"grad_norm": 0.3311345580462952,
"learning_rate": 9.49812990116353e-05,
"loss": 0.5189,
"step": 550
},
{
"epoch": 0.57,
"grad_norm": 0.3452796566826953,
"learning_rate": 9.319131497471894e-05,
"loss": 0.5287,
"step": 555
},
{
"epoch": 0.57,
"grad_norm": 0.3026098766078991,
"learning_rate": 9.140352012625537e-05,
"loss": 0.6037,
"step": 560
},
{
"epoch": 0.58,
"grad_norm": 0.33840602079777765,
"learning_rate": 8.961848929381026e-05,
"loss": 0.5232,
"step": 565
},
{
"epoch": 0.58,
"grad_norm": 0.26523037796888355,
"learning_rate": 8.783679641623845e-05,
"loss": 0.5282,
"step": 570
},
{
"epoch": 0.59,
"grad_norm": 0.30471264337335136,
"learning_rate": 8.605901435914607e-05,
"loss": 0.4891,
"step": 575
},
{
"epoch": 0.6,
"grad_norm": 0.4626790163374621,
"learning_rate": 8.428571473069775e-05,
"loss": 0.4732,
"step": 580
},
{
"epoch": 0.6,
"grad_norm": 0.36758855634493254,
"learning_rate": 8.25174676978282e-05,
"loss": 0.6205,
"step": 585
},
{
"epoch": 0.61,
"grad_norm": 0.40307304885776646,
"learning_rate": 8.075484180291701e-05,
"loss": 0.5508,
"step": 590
},
{
"epoch": 0.61,
"grad_norm": 0.2696699367170565,
"learning_rate": 7.899840378098588e-05,
"loss": 0.444,
"step": 595
},
{
"epoch": 0.62,
"grad_norm": 0.30148215688959473,
"learning_rate": 7.724871837747707e-05,
"loss": 0.5485,
"step": 600
},
{
"epoch": 0.62,
"grad_norm": 0.3348894175707125,
"learning_rate": 7.550634816667142e-05,
"loss": 0.5062,
"step": 605
},
{
"epoch": 0.63,
"grad_norm": 0.3337832447029819,
"learning_rate": 7.377185337080442e-05,
"loss": 0.6384,
"step": 610
},
{
"epoch": 0.63,
"grad_norm": 0.32562651669263704,
"learning_rate": 7.204579167993881e-05,
"loss": 0.5165,
"step": 615
},
{
"epoch": 0.64,
"grad_norm": 0.3488615725937723,
"learning_rate": 7.032871807265096e-05,
"loss": 0.5548,
"step": 620
},
{
"epoch": 0.64,
"grad_norm": 0.2911657658627874,
"learning_rate": 6.862118463758943e-05,
"loss": 0.4718,
"step": 625
},
{
"epoch": 0.65,
"grad_norm": 0.3027034627321655,
"learning_rate": 6.69237403959624e-05,
"loss": 0.5289,
"step": 630
},
{
"epoch": 0.65,
"grad_norm": 0.28279412370215407,
"learning_rate": 6.52369311250116e-05,
"loss": 0.6282,
"step": 635
},
{
"epoch": 0.66,
"grad_norm": 0.325397208023914,
"learning_rate": 6.356129918252927e-05,
"loss": 0.5032,
"step": 640
},
{
"epoch": 0.66,
"grad_norm": 0.4468881362131618,
"learning_rate": 6.189738333247432e-05,
"loss": 0.548,
"step": 645
},
{
"epoch": 0.67,
"grad_norm": 0.32616567931810037,
"learning_rate": 6.024571857174443e-05,
"loss": 0.4915,
"step": 650
},
{
"epoch": 0.67,
"grad_norm": 0.29764758497214333,
"learning_rate": 5.860683595815893e-05,
"loss": 0.572,
"step": 655
},
{
"epoch": 0.68,
"grad_norm": 0.3463962850271051,
"learning_rate": 5.698126243970845e-05,
"loss": 0.5181,
"step": 660
},
{
"epoch": 0.68,
"grad_norm": 0.346392219127237,
"learning_rate": 5.536952068512608e-05,
"loss": 0.6312,
"step": 665
},
{
"epoch": 0.69,
"grad_norm": 0.27525269488138604,
"learning_rate": 5.3772128915834184e-05,
"loss": 0.5446,
"step": 670
},
{
"epoch": 0.69,
"grad_norm": 0.3098672923380687,
"learning_rate": 5.218960073932122e-05,
"loss": 0.4164,
"step": 675
},
{
"epoch": 0.7,
"grad_norm": 0.30953891133733785,
"learning_rate": 5.062244498400228e-05,
"loss": 0.5787,
"step": 680
},
{
"epoch": 0.7,
"grad_norm": 0.29936249031132967,
"learning_rate": 4.907116553561607e-05,
"loss": 0.5776,
"step": 685
},
{
"epoch": 0.71,
"grad_norm": 0.267987992377413,
"learning_rate": 4.753626117521103e-05,
"loss": 0.5566,
"step": 690
},
{
"epoch": 0.71,
"grad_norm": 0.33079494946575233,
"learning_rate": 4.601822541877291e-05,
"loss": 0.5182,
"step": 695
},
{
"epoch": 0.72,
"grad_norm": 0.25868748114720685,
"learning_rate": 4.451754635854517e-05,
"loss": 0.559,
"step": 700
},
{
"epoch": 0.72,
"grad_norm": 0.3290802188517225,
"learning_rate": 4.303470650609325e-05,
"loss": 0.5881,
"step": 705
},
{
"epoch": 0.73,
"grad_norm": 0.32058655048671514,
"learning_rate": 4.1570182637163155e-05,
"loss": 0.4518,
"step": 710
},
{
"epoch": 0.73,
"grad_norm": 0.33971870382433816,
"learning_rate": 4.0124445638384366e-05,
"loss": 0.6093,
"step": 715
},
{
"epoch": 0.74,
"grad_norm": 0.4086041810801181,
"learning_rate": 3.869796035586625e-05,
"loss": 0.6063,
"step": 720
},
{
"epoch": 0.74,
"grad_norm": 0.3648240136823399,
"learning_rate": 3.7291185445736444e-05,
"loss": 0.4986,
"step": 725
},
{
"epoch": 0.75,
"grad_norm": 0.37137184349965324,
"learning_rate": 3.590457322666997e-05,
"loss": 0.5459,
"step": 730
},
{
"epoch": 0.75,
"grad_norm": 0.3202982297697742,
"learning_rate": 3.453856953445557e-05,
"loss": 0.5487,
"step": 735
},
{
"epoch": 0.76,
"grad_norm": 0.2743331793572931,
"learning_rate": 3.319361357864663e-05,
"loss": 0.5254,
"step": 740
},
{
"epoch": 0.76,
"grad_norm": 0.3002561379742104,
"learning_rate": 3.187013780134291e-05,
"loss": 0.5548,
"step": 745
},
{
"epoch": 0.77,
"grad_norm": 0.35602828538254794,
"learning_rate": 3.05685677381475e-05,
"loss": 0.5441,
"step": 750
},
{
"epoch": 0.77,
"grad_norm": 0.34925281382951245,
"learning_rate": 2.9289321881345254e-05,
"loss": 0.5641,
"step": 755
},
{
"epoch": 0.78,
"grad_norm": 0.3365150010367925,
"learning_rate": 2.8032811545345294e-05,
"loss": 0.6116,
"step": 760
},
{
"epoch": 0.79,
"grad_norm": 0.3475156862552588,
"learning_rate": 2.679944073443158e-05,
"loss": 0.6135,
"step": 765
},
{
"epoch": 0.79,
"grad_norm": 0.3146717649143792,
"learning_rate": 2.5589606012863963e-05,
"loss": 0.5197,
"step": 770
},
{
"epoch": 0.8,
"grad_norm": 0.3211819564312899,
"learning_rate": 2.4403696377371142e-05,
"loss": 0.5392,
"step": 775
},
{
"epoch": 0.8,
"grad_norm": 0.3409828677067338,
"learning_rate": 2.324209313207736e-05,
"loss": 0.5052,
"step": 780
},
{
"epoch": 0.81,
"grad_norm": 0.2557678515935505,
"learning_rate": 2.210516976590179e-05,
"loss": 0.4917,
"step": 785
},
{
"epoch": 0.81,
"grad_norm": 0.24519712151032516,
"learning_rate": 2.099329183247126e-05,
"loss": 0.3827,
"step": 790
},
{
"epoch": 0.82,
"grad_norm": 0.32798523064161217,
"learning_rate": 1.9906816832584253e-05,
"loss": 0.4955,
"step": 795
},
{
"epoch": 0.82,
"grad_norm": 0.21681035987591052,
"learning_rate": 1.8846094099263912e-05,
"loss": 0.4815,
"step": 800
},
{
"epoch": 0.83,
"grad_norm": 0.3625179077510228,
"learning_rate": 1.781146468543765e-05,
"loss": 0.5803,
"step": 805
},
{
"epoch": 0.83,
"grad_norm": 0.4443723352307208,
"learning_rate": 1.6803261254278636e-05,
"loss": 0.5739,
"step": 810
},
{
"epoch": 0.84,
"grad_norm": 0.42475678943437195,
"learning_rate": 1.582180797224507e-05,
"loss": 0.5068,
"step": 815
},
{
"epoch": 0.84,
"grad_norm": 0.46644455691617004,
"learning_rate": 1.4867420404851307e-05,
"loss": 0.5777,
"step": 820
},
{
"epoch": 0.85,
"grad_norm": 0.2850490204012201,
"learning_rate": 1.3940405415204416e-05,
"loss": 0.6615,
"step": 825
},
{
"epoch": 0.85,
"grad_norm": 0.38102633045200973,
"learning_rate": 1.30410610653389e-05,
"loss": 0.5049,
"step": 830
},
{
"epoch": 0.86,
"grad_norm": 0.2792649404054081,
"learning_rate": 1.2169676520381168e-05,
"loss": 0.4754,
"step": 835
},
{
"epoch": 0.86,
"grad_norm": 0.2655446144792418,
"learning_rate": 1.1326531955574526e-05,
"loss": 0.5295,
"step": 840
},
{
"epoch": 0.87,
"grad_norm": 0.2775130936028629,
"learning_rate": 1.0511898466194903e-05,
"loss": 0.4933,
"step": 845
},
{
"epoch": 0.87,
"grad_norm": 0.24711621838841755,
"learning_rate": 9.726037980385738e-06,
"loss": 0.5456,
"step": 850
},
{
"epoch": 0.88,
"grad_norm": 0.393271652854802,
"learning_rate": 8.969203174940654e-06,
"loss": 0.5453,
"step": 855
},
{
"epoch": 0.88,
"grad_norm": 0.2694238731763909,
"learning_rate": 8.24163739406062e-06,
"loss": 0.5382,
"step": 860
},
{
"epoch": 0.89,
"grad_norm": 0.29367966818919566,
"learning_rate": 7.543574571111655e-06,
"loss": 0.4812,
"step": 865
},
{
"epoch": 0.89,
"grad_norm": 0.3400136578447068,
"learning_rate": 6.875239153408542e-06,
"loss": 0.5184,
"step": 870
},
{
"epoch": 0.9,
"grad_norm": 0.4285681368195809,
"learning_rate": 6.236846030048604e-06,
"loss": 0.611,
"step": 875
},
{
"epoch": 0.9,
"grad_norm": 0.46776985117421765,
"learning_rate": 5.6286004628186675e-06,
"loss": 0.4348,
"step": 880
},
{
"epoch": 0.91,
"grad_norm": 0.3617441270450328,
"learning_rate": 5.0506980201973974e-06,
"loss": 0.5767,
"step": 885
},
{
"epoch": 0.91,
"grad_norm": 0.33384517109285833,
"learning_rate": 4.503324514474483e-06,
"loss": 0.5594,
"step": 890
},
{
"epoch": 0.92,
"grad_norm": 0.4731448498076414,
"learning_rate": 3.986655942006579e-06,
"loss": 0.4884,
"step": 895
},
{
"epoch": 0.92,
"grad_norm": 0.3078236949280879,
"learning_rate": 3.5008584266294386e-06,
"loss": 0.5628,
"step": 900
},
{
"epoch": 0.93,
"grad_norm": 0.32888917087743735,
"learning_rate": 3.0460881662442763e-06,
"loss": 0.4988,
"step": 905
},
{
"epoch": 0.93,
"grad_norm": 0.29639144474137563,
"learning_rate": 2.622491382595693e-06,
"loss": 0.5666,
"step": 910
},
{
"epoch": 0.94,
"grad_norm": 0.37502576344896726,
"learning_rate": 2.2302042742571193e-06,
"loss": 0.4814,
"step": 915
},
{
"epoch": 0.94,
"grad_norm": 0.3978025873196548,
"learning_rate": 1.869352972839067e-06,
"loss": 0.4309,
"step": 920
},
{
"epoch": 0.95,
"grad_norm": 0.37364204566135195,
"learning_rate": 1.5400535024342022e-06,
"loss": 0.4488,
"step": 925
},
{
"epoch": 0.95,
"grad_norm": 0.2906768898262274,
"learning_rate": 1.2424117423122328e-06,
"loss": 0.5843,
"step": 930
},
{
"epoch": 0.96,
"grad_norm": 0.2940342755596854,
"learning_rate": 9.765233928766493e-07,
"loss": 0.6249,
"step": 935
},
{
"epoch": 0.96,
"grad_norm": 0.40595136224277073,
"learning_rate": 7.42473944894384e-07,
"loss": 0.5441,
"step": 940
},
{
"epoch": 0.97,
"grad_norm": 0.3682171569394288,
"learning_rate": 5.403386520079323e-07,
"loss": 0.4915,
"step": 945
},
{
"epoch": 0.97,
"grad_norm": 0.3409683342269183,
"learning_rate": 3.701825065392184e-07,
"loss": 0.4854,
"step": 950
},
{
"epoch": 0.98,
"grad_norm": 0.3753876090426702,
"learning_rate": 2.320602185927001e-07,
"loss": 0.4656,
"step": 955
},
{
"epoch": 0.99,
"grad_norm": 0.3824583155896019,
"learning_rate": 1.2601619846444035e-07,
"loss": 0.5535,
"step": 960
},
{
"epoch": 0.99,
"grad_norm": 0.33353700957958915,
"learning_rate": 5.208454236296234e-08,
"loss": 0.5956,
"step": 965
},
{
"epoch": 1.0,
"grad_norm": 0.460947189729548,
"learning_rate": 1.0289021446308056e-08,
"loss": 0.5259,
"step": 970
},
{
"epoch": 1.0,
"eval_loss": NaN,
"eval_runtime": 1480.6147,
"eval_samples_per_second": 1.561,
"eval_steps_per_second": 0.39,
"step": 974
},
{
"epoch": 1.0,
"step": 974,
"total_flos": 4299354448986112.0,
"train_loss": 0.2056859987472362,
"train_runtime": 8762.8151,
"train_samples_per_second": 3.558,
"train_steps_per_second": 0.111
}
],
"logging_steps": 5,
"max_steps": 974,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 40,
"total_flos": 4299354448986112.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}