zephyr-llama-3-8b-sft / trainer_state.json
tanliboy's picture
Model save
5ab63b1 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 1624,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0006157635467980296,
"grad_norm": 39.11833832570066,
"learning_rate": 1.8404907975460124e-08,
"loss": 1.0171,
"step": 1
},
{
"epoch": 0.006157635467980296,
"grad_norm": 97.51191996179428,
"learning_rate": 1.8404907975460122e-07,
"loss": 1.0435,
"step": 10
},
{
"epoch": 0.012315270935960592,
"grad_norm": 33.104108639558106,
"learning_rate": 3.6809815950920245e-07,
"loss": 1.0398,
"step": 20
},
{
"epoch": 0.01847290640394089,
"grad_norm": 25.392368172808187,
"learning_rate": 5.521472392638037e-07,
"loss": 1.0188,
"step": 30
},
{
"epoch": 0.024630541871921183,
"grad_norm": 2.5252784333814122,
"learning_rate": 7.361963190184049e-07,
"loss": 0.9958,
"step": 40
},
{
"epoch": 0.03078817733990148,
"grad_norm": 1.4934120741666812,
"learning_rate": 9.202453987730062e-07,
"loss": 0.9789,
"step": 50
},
{
"epoch": 0.03694581280788178,
"grad_norm": 10.681983660964159,
"learning_rate": 1.1042944785276073e-06,
"loss": 0.9309,
"step": 60
},
{
"epoch": 0.04310344827586207,
"grad_norm": 1.1792371247881757,
"learning_rate": 1.2883435582822088e-06,
"loss": 0.9487,
"step": 70
},
{
"epoch": 0.04926108374384237,
"grad_norm": 3.5967827346653807,
"learning_rate": 1.4723926380368098e-06,
"loss": 0.9392,
"step": 80
},
{
"epoch": 0.05541871921182266,
"grad_norm": 1.3027564465976336,
"learning_rate": 1.6564417177914112e-06,
"loss": 0.9278,
"step": 90
},
{
"epoch": 0.06157635467980296,
"grad_norm": 1.868580833085548,
"learning_rate": 1.8404907975460124e-06,
"loss": 0.9329,
"step": 100
},
{
"epoch": 0.06773399014778325,
"grad_norm": 8.034382889984798,
"learning_rate": 2.0245398773006137e-06,
"loss": 0.9353,
"step": 110
},
{
"epoch": 0.07389162561576355,
"grad_norm": 2.971665740965309,
"learning_rate": 2.2085889570552147e-06,
"loss": 0.9192,
"step": 120
},
{
"epoch": 0.08004926108374384,
"grad_norm": 3.681564626469473,
"learning_rate": 2.3926380368098157e-06,
"loss": 0.9046,
"step": 130
},
{
"epoch": 0.08620689655172414,
"grad_norm": 4.346804226811641,
"learning_rate": 2.5766871165644175e-06,
"loss": 0.9297,
"step": 140
},
{
"epoch": 0.09236453201970443,
"grad_norm": 1.8357015303994761,
"learning_rate": 2.7607361963190186e-06,
"loss": 0.906,
"step": 150
},
{
"epoch": 0.09852216748768473,
"grad_norm": 1.1670278503111207,
"learning_rate": 2.9447852760736196e-06,
"loss": 0.9123,
"step": 160
},
{
"epoch": 0.10467980295566502,
"grad_norm": 1.7533219735333505,
"learning_rate": 2.9998300784514776e-06,
"loss": 0.9075,
"step": 170
},
{
"epoch": 0.11083743842364532,
"grad_norm": 1.9794621815752764,
"learning_rate": 2.9989979023179235e-06,
"loss": 0.9104,
"step": 180
},
{
"epoch": 0.11699507389162561,
"grad_norm": 0.9849983349999529,
"learning_rate": 2.9974726458049776e-06,
"loss": 0.8953,
"step": 190
},
{
"epoch": 0.12315270935960591,
"grad_norm": 1.0588037960942727,
"learning_rate": 2.9952550141340154e-06,
"loss": 0.8935,
"step": 200
},
{
"epoch": 0.12931034482758622,
"grad_norm": 1.2914329429720137,
"learning_rate": 2.9923460326547038e-06,
"loss": 0.898,
"step": 210
},
{
"epoch": 0.1354679802955665,
"grad_norm": 1.462714450533501,
"learning_rate": 2.988747046370918e-06,
"loss": 0.8964,
"step": 220
},
{
"epoch": 0.1416256157635468,
"grad_norm": 1.5479755472032886,
"learning_rate": 2.984459719318862e-06,
"loss": 0.901,
"step": 230
},
{
"epoch": 0.1477832512315271,
"grad_norm": 1.0726588199016611,
"learning_rate": 2.9794860337976802e-06,
"loss": 0.8898,
"step": 240
},
{
"epoch": 0.1539408866995074,
"grad_norm": 1.2064222174518724,
"learning_rate": 2.9738282894529177e-06,
"loss": 0.8906,
"step": 250
},
{
"epoch": 0.16009852216748768,
"grad_norm": 1.0779645832395361,
"learning_rate": 2.96748910221325e-06,
"loss": 0.8956,
"step": 260
},
{
"epoch": 0.16625615763546797,
"grad_norm": 1.350106799679849,
"learning_rate": 2.9604714030809755e-06,
"loss": 0.8908,
"step": 270
},
{
"epoch": 0.1724137931034483,
"grad_norm": 1.4528864294030424,
"learning_rate": 2.9527784367768305e-06,
"loss": 0.8918,
"step": 280
},
{
"epoch": 0.17857142857142858,
"grad_norm": 1.093974116558019,
"learning_rate": 2.9444137602397515e-06,
"loss": 0.9045,
"step": 290
},
{
"epoch": 0.18472906403940886,
"grad_norm": 1.1335514527922086,
"learning_rate": 2.935381240982281e-06,
"loss": 0.8919,
"step": 300
},
{
"epoch": 0.19088669950738915,
"grad_norm": 1.1578136051328392,
"learning_rate": 2.9256850553023724e-06,
"loss": 0.8932,
"step": 310
},
{
"epoch": 0.19704433497536947,
"grad_norm": 1.2524948494126782,
"learning_rate": 2.9153296863524315e-06,
"loss": 0.8779,
"step": 320
},
{
"epoch": 0.20320197044334976,
"grad_norm": 0.9562337520138312,
"learning_rate": 2.9043199220664704e-06,
"loss": 0.8848,
"step": 330
},
{
"epoch": 0.20935960591133004,
"grad_norm": 1.014920165575569,
"learning_rate": 2.8926608529463473e-06,
"loss": 0.891,
"step": 340
},
{
"epoch": 0.21551724137931033,
"grad_norm": 1.2764221494335906,
"learning_rate": 2.880357869708111e-06,
"loss": 0.9045,
"step": 350
},
{
"epoch": 0.22167487684729065,
"grad_norm": 0.9543614414538709,
"learning_rate": 2.8674166607895357e-06,
"loss": 0.886,
"step": 360
},
{
"epoch": 0.22783251231527094,
"grad_norm": 1.0200791854817908,
"learning_rate": 2.85384320972e-06,
"loss": 0.8798,
"step": 370
},
{
"epoch": 0.23399014778325122,
"grad_norm": 0.9423219737843089,
"learning_rate": 2.839643792353928e-06,
"loss": 0.888,
"step": 380
},
{
"epoch": 0.24014778325123154,
"grad_norm": 1.807370938563018,
"learning_rate": 2.824824973969069e-06,
"loss": 0.9058,
"step": 390
},
{
"epoch": 0.24630541871921183,
"grad_norm": 0.9624203166584838,
"learning_rate": 2.8093936062309614e-06,
"loss": 0.88,
"step": 400
},
{
"epoch": 0.2524630541871921,
"grad_norm": 1.0446763066450997,
"learning_rate": 2.7933568240249776e-06,
"loss": 0.8916,
"step": 410
},
{
"epoch": 0.25862068965517243,
"grad_norm": 1.031965401372882,
"learning_rate": 2.776722042157421e-06,
"loss": 0.8937,
"step": 420
},
{
"epoch": 0.2647783251231527,
"grad_norm": 1.0397880014843512,
"learning_rate": 2.7594969519271988e-06,
"loss": 0.8802,
"step": 430
},
{
"epoch": 0.270935960591133,
"grad_norm": 0.9409991870080029,
"learning_rate": 2.7416895175696533e-06,
"loss": 0.8952,
"step": 440
},
{
"epoch": 0.2770935960591133,
"grad_norm": 0.9068233549186512,
"learning_rate": 2.723307972574199e-06,
"loss": 0.8951,
"step": 450
},
{
"epoch": 0.2832512315270936,
"grad_norm": 1.0260952139368371,
"learning_rate": 2.7043608158774645e-06,
"loss": 0.8848,
"step": 460
},
{
"epoch": 0.2894088669950739,
"grad_norm": 0.8928699960345075,
"learning_rate": 2.684856807933706e-06,
"loss": 0.8844,
"step": 470
},
{
"epoch": 0.2955665024630542,
"grad_norm": 0.9088682466856005,
"learning_rate": 2.664804966664298e-06,
"loss": 0.8935,
"step": 480
},
{
"epoch": 0.3017241379310345,
"grad_norm": 0.94527180620086,
"learning_rate": 2.6442145632881894e-06,
"loss": 0.8908,
"step": 490
},
{
"epoch": 0.3078817733990148,
"grad_norm": 0.9455656466636171,
"learning_rate": 2.623095118035235e-06,
"loss": 0.8972,
"step": 500
},
{
"epoch": 0.31403940886699505,
"grad_norm": 0.9278270508038465,
"learning_rate": 2.601456395744403e-06,
"loss": 0.8963,
"step": 510
},
{
"epoch": 0.32019704433497537,
"grad_norm": 1.0267218503341182,
"learning_rate": 2.579308401348876e-06,
"loss": 0.8982,
"step": 520
},
{
"epoch": 0.3263546798029557,
"grad_norm": 0.8668867739209858,
"learning_rate": 2.556661375250149e-06,
"loss": 0.8826,
"step": 530
},
{
"epoch": 0.33251231527093594,
"grad_norm": 0.9466156644717508,
"learning_rate": 2.533525788583248e-06,
"loss": 0.8824,
"step": 540
},
{
"epoch": 0.33866995073891626,
"grad_norm": 1.079466437044084,
"learning_rate": 2.509912338375275e-06,
"loss": 0.8756,
"step": 550
},
{
"epoch": 0.3448275862068966,
"grad_norm": 1.5912074264549028,
"learning_rate": 2.4858319425994978e-06,
"loss": 0.894,
"step": 560
},
{
"epoch": 0.35098522167487683,
"grad_norm": 0.9313999461191238,
"learning_rate": 2.4612957351272963e-06,
"loss": 0.8895,
"step": 570
},
{
"epoch": 0.35714285714285715,
"grad_norm": 1.1801770634441218,
"learning_rate": 2.4363150605802704e-06,
"loss": 0.8665,
"step": 580
},
{
"epoch": 0.3633004926108374,
"grad_norm": 1.0068892948188721,
"learning_rate": 2.41090146908492e-06,
"loss": 0.876,
"step": 590
},
{
"epoch": 0.3694581280788177,
"grad_norm": 1.0826722320337905,
"learning_rate": 2.385066710932294e-06,
"loss": 0.8673,
"step": 600
},
{
"epoch": 0.37561576354679804,
"grad_norm": 0.8776518626640517,
"learning_rate": 2.3588227311451007e-06,
"loss": 0.871,
"step": 610
},
{
"epoch": 0.3817733990147783,
"grad_norm": 0.8970209724037471,
"learning_rate": 2.3321816639547747e-06,
"loss": 0.8754,
"step": 620
},
{
"epoch": 0.3879310344827586,
"grad_norm": 0.865408238823646,
"learning_rate": 2.305155827191066e-06,
"loss": 0.8828,
"step": 630
},
{
"epoch": 0.39408866995073893,
"grad_norm": 0.9191133091733539,
"learning_rate": 2.2777577165867354e-06,
"loss": 0.893,
"step": 640
},
{
"epoch": 0.4002463054187192,
"grad_norm": 1.0084087069366179,
"learning_rate": 2.25e-06,
"loss": 0.8707,
"step": 650
},
{
"epoch": 0.4064039408866995,
"grad_norm": 0.9641835901570172,
"learning_rate": 2.2218955115573864e-06,
"loss": 0.8941,
"step": 660
},
{
"epoch": 0.4125615763546798,
"grad_norm": 1.0960250718373252,
"learning_rate": 2.1934572457197163e-06,
"loss": 0.8801,
"step": 670
},
{
"epoch": 0.4187192118226601,
"grad_norm": 0.9620003659319739,
"learning_rate": 2.164698351273952e-06,
"loss": 0.8838,
"step": 680
},
{
"epoch": 0.4248768472906404,
"grad_norm": 0.945391462529571,
"learning_rate": 2.1356321252536947e-06,
"loss": 0.8773,
"step": 690
},
{
"epoch": 0.43103448275862066,
"grad_norm": 0.9952181170766936,
"learning_rate": 2.10627200679113e-06,
"loss": 0.8809,
"step": 700
},
{
"epoch": 0.437192118226601,
"grad_norm": 1.2866159652764533,
"learning_rate": 2.0766315709032837e-06,
"loss": 0.8778,
"step": 710
},
{
"epoch": 0.4433497536945813,
"grad_norm": 0.9256729280886802,
"learning_rate": 2.046724522215437e-06,
"loss": 0.8711,
"step": 720
},
{
"epoch": 0.44950738916256155,
"grad_norm": 0.830245767183929,
"learning_rate": 2.016564688624627e-06,
"loss": 0.8848,
"step": 730
},
{
"epoch": 0.45566502463054187,
"grad_norm": 0.9400170628704059,
"learning_rate": 1.9861660149061435e-06,
"loss": 0.8746,
"step": 740
},
{
"epoch": 0.4618226600985222,
"grad_norm": 0.9114604227226025,
"learning_rate": 1.9555425562659878e-06,
"loss": 0.8839,
"step": 750
},
{
"epoch": 0.46798029556650245,
"grad_norm": 0.9636472157159227,
"learning_rate": 1.924708471842276e-06,
"loss": 0.8644,
"step": 760
},
{
"epoch": 0.47413793103448276,
"grad_norm": 0.8916431000744512,
"learning_rate": 1.8936780181585799e-06,
"loss": 0.8749,
"step": 770
},
{
"epoch": 0.4802955665024631,
"grad_norm": 0.9584025191544591,
"learning_rate": 1.86246554253225e-06,
"loss": 0.886,
"step": 780
},
{
"epoch": 0.48645320197044334,
"grad_norm": 1.1713799337704054,
"learning_rate": 1.831085476440753e-06,
"loss": 0.8738,
"step": 790
},
{
"epoch": 0.49261083743842365,
"grad_norm": 1.4368869246034384,
"learning_rate": 1.7995523288490959e-06,
"loss": 0.8848,
"step": 800
},
{
"epoch": 0.4987684729064039,
"grad_norm": 0.9684556270804614,
"learning_rate": 1.7678806795014293e-06,
"loss": 0.8806,
"step": 810
},
{
"epoch": 0.5049261083743842,
"grad_norm": 0.931294683241216,
"learning_rate": 1.7360851721799163e-06,
"loss": 0.8609,
"step": 820
},
{
"epoch": 0.5110837438423645,
"grad_norm": 0.9532178001849707,
"learning_rate": 1.7041805079340006e-06,
"loss": 0.871,
"step": 830
},
{
"epoch": 0.5172413793103449,
"grad_norm": 0.91289729076867,
"learning_rate": 1.6721814382831911e-06,
"loss": 0.8706,
"step": 840
},
{
"epoch": 0.5233990147783252,
"grad_norm": 0.8701261925434093,
"learning_rate": 1.6401027583965135e-06,
"loss": 0.874,
"step": 850
},
{
"epoch": 0.5295566502463054,
"grad_norm": 1.3157172918251052,
"learning_rate": 1.6079593002517785e-06,
"loss": 0.8972,
"step": 860
},
{
"epoch": 0.5357142857142857,
"grad_norm": 0.9460694314665482,
"learning_rate": 1.575765925777834e-06,
"loss": 0.8705,
"step": 870
},
{
"epoch": 0.541871921182266,
"grad_norm": 0.8663250399265648,
"learning_rate": 1.543537519982963e-06,
"loss": 0.8757,
"step": 880
},
{
"epoch": 0.5480295566502463,
"grad_norm": 0.8476656469038841,
"learning_rate": 1.5112889840726194e-06,
"loss": 0.8821,
"step": 890
},
{
"epoch": 0.5541871921182266,
"grad_norm": 0.9793899276197807,
"learning_rate": 1.4790352285596656e-06,
"loss": 0.867,
"step": 900
},
{
"epoch": 0.5603448275862069,
"grad_norm": 0.9471504761102012,
"learning_rate": 1.4467911663703118e-06,
"loss": 0.8669,
"step": 910
},
{
"epoch": 0.5665024630541872,
"grad_norm": 1.0080484253850497,
"learning_rate": 1.4145717059489405e-06,
"loss": 0.8782,
"step": 920
},
{
"epoch": 0.5726600985221675,
"grad_norm": 1.1941676655278535,
"learning_rate": 1.3823917443649994e-06,
"loss": 0.8543,
"step": 930
},
{
"epoch": 0.5788177339901478,
"grad_norm": 1.155321605934009,
"learning_rate": 1.3502661604251562e-06,
"loss": 0.8751,
"step": 940
},
{
"epoch": 0.5849753694581281,
"grad_norm": 1.0349420537193152,
"learning_rate": 1.3182098077938954e-06,
"loss": 0.863,
"step": 950
},
{
"epoch": 0.5911330049261084,
"grad_norm": 0.8668489498799227,
"learning_rate": 1.286237508125744e-06,
"loss": 0.8889,
"step": 960
},
{
"epoch": 0.5972906403940886,
"grad_norm": 1.1105014519516443,
"learning_rate": 1.254364044212291e-06,
"loss": 0.8817,
"step": 970
},
{
"epoch": 0.603448275862069,
"grad_norm": 1.1510310259609142,
"learning_rate": 1.2226041531471835e-06,
"loss": 0.8686,
"step": 980
},
{
"epoch": 0.6096059113300493,
"grad_norm": 0.9362548110546389,
"learning_rate": 1.1909725195122443e-06,
"loss": 0.8676,
"step": 990
},
{
"epoch": 0.6157635467980296,
"grad_norm": 0.8788738119989281,
"learning_rate": 1.1594837685878725e-06,
"loss": 0.8523,
"step": 1000
},
{
"epoch": 0.6219211822660099,
"grad_norm": 0.887194098681397,
"learning_rate": 1.1281524595908653e-06,
"loss": 0.8762,
"step": 1010
},
{
"epoch": 0.6280788177339901,
"grad_norm": 0.8612235803063835,
"learning_rate": 1.0969930789427798e-06,
"loss": 0.877,
"step": 1020
},
{
"epoch": 0.6342364532019704,
"grad_norm": 0.8734309546117548,
"learning_rate": 1.0660200335719569e-06,
"loss": 0.8761,
"step": 1030
},
{
"epoch": 0.6403940886699507,
"grad_norm": 0.8944912206450915,
"learning_rate": 1.0352476442522963e-06,
"loss": 0.8847,
"step": 1040
},
{
"epoch": 0.646551724137931,
"grad_norm": 0.8553821783678947,
"learning_rate": 1.004690138981871e-06,
"loss": 0.8731,
"step": 1050
},
{
"epoch": 0.6527093596059114,
"grad_norm": 0.8628572848264044,
"learning_rate": 9.74361646404432e-07,
"loss": 0.8606,
"step": 1060
},
{
"epoch": 0.6588669950738916,
"grad_norm": 0.8555513910364403,
"learning_rate": 9.442761892768561e-07,
"loss": 0.8656,
"step": 1070
},
{
"epoch": 0.6650246305418719,
"grad_norm": 0.9062019842589142,
"learning_rate": 9.144476779855462e-07,
"loss": 0.8892,
"step": 1080
},
{
"epoch": 0.6711822660098522,
"grad_norm": 0.8327590617691879,
"learning_rate": 8.848899041147947e-07,
"loss": 0.8746,
"step": 1090
},
{
"epoch": 0.6773399014778325,
"grad_norm": 0.8338019269330827,
"learning_rate": 8.556165340700687e-07,
"loss": 0.8615,
"step": 1100
},
{
"epoch": 0.6834975369458128,
"grad_norm": 0.8295491614862206,
"learning_rate": 8.266411027591801e-07,
"loss": 0.8732,
"step": 1110
},
{
"epoch": 0.6896551724137931,
"grad_norm": 0.9395669949159903,
"learning_rate": 7.979770073342484e-07,
"loss": 0.8729,
"step": 1120
},
{
"epoch": 0.6958128078817734,
"grad_norm": 0.867733164384384,
"learning_rate": 7.696375009973643e-07,
"loss": 0.8608,
"step": 1130
},
{
"epoch": 0.7019704433497537,
"grad_norm": 0.9064838208046303,
"learning_rate": 7.41635686872804e-07,
"loss": 0.8765,
"step": 1140
},
{
"epoch": 0.708128078817734,
"grad_norm": 0.9195681549910459,
"learning_rate": 7.139845119486371e-07,
"loss": 0.8473,
"step": 1150
},
{
"epoch": 0.7142857142857143,
"grad_norm": 0.924047168481818,
"learning_rate": 6.866967610905234e-07,
"loss": 0.8793,
"step": 1160
},
{
"epoch": 0.7204433497536946,
"grad_norm": 0.8440847823992839,
"learning_rate": 6.597850511304739e-07,
"loss": 0.8602,
"step": 1170
},
{
"epoch": 0.7266009852216748,
"grad_norm": 0.9108312916499829,
"learning_rate": 6.332618250332988e-07,
"loss": 0.854,
"step": 1180
},
{
"epoch": 0.7327586206896551,
"grad_norm": 0.8778675133667185,
"learning_rate": 6.071393461434488e-07,
"loss": 0.8797,
"step": 1190
},
{
"epoch": 0.7389162561576355,
"grad_norm": 0.862025362499502,
"learning_rate": 5.814296925149026e-07,
"loss": 0.876,
"step": 1200
},
{
"epoch": 0.7450738916256158,
"grad_norm": 0.9153055267247704,
"learning_rate": 5.561447513267311e-07,
"loss": 0.8714,
"step": 1210
},
{
"epoch": 0.7512315270935961,
"grad_norm": 0.8294104219179396,
"learning_rate": 5.312962133869093e-07,
"loss": 0.8659,
"step": 1220
},
{
"epoch": 0.7573891625615764,
"grad_norm": 0.8628035692028492,
"learning_rate": 5.068955677269281e-07,
"loss": 0.8653,
"step": 1230
},
{
"epoch": 0.7635467980295566,
"grad_norm": 0.8789751770218951,
"learning_rate": 4.829540962896927e-07,
"loss": 0.8787,
"step": 1240
},
{
"epoch": 0.7697044334975369,
"grad_norm": 0.880811469553608,
"learning_rate": 4.594828687131814e-07,
"loss": 0.8719,
"step": 1250
},
{
"epoch": 0.7758620689655172,
"grad_norm": 0.8589600352283664,
"learning_rate": 4.36492737212255e-07,
"loss": 0.8649,
"step": 1260
},
{
"epoch": 0.7820197044334976,
"grad_norm": 1.97188374324455,
"learning_rate": 4.13994331561004e-07,
"loss": 0.8559,
"step": 1270
},
{
"epoch": 0.7881773399014779,
"grad_norm": 1.2314345966965359,
"learning_rate": 3.9199805417793833e-07,
"loss": 0.866,
"step": 1280
},
{
"epoch": 0.7943349753694581,
"grad_norm": 0.8564385884815968,
"learning_rate": 3.705140753162973e-07,
"loss": 0.8767,
"step": 1290
},
{
"epoch": 0.8004926108374384,
"grad_norm": 0.8935220357617403,
"learning_rate": 3.495523283617106e-07,
"loss": 0.8713,
"step": 1300
},
{
"epoch": 0.8066502463054187,
"grad_norm": 1.0237618875404009,
"learning_rate": 3.2912250523937e-07,
"loss": 0.8718,
"step": 1310
},
{
"epoch": 0.812807881773399,
"grad_norm": 0.8442304790394727,
"learning_rate": 3.092340519328474e-07,
"loss": 0.8864,
"step": 1320
},
{
"epoch": 0.8189655172413793,
"grad_norm": 1.4691847919498742,
"learning_rate": 2.8989616411662826e-07,
"loss": 0.8597,
"step": 1330
},
{
"epoch": 0.8251231527093597,
"grad_norm": 0.8880158508210901,
"learning_rate": 2.7111778290437465e-07,
"loss": 0.8675,
"step": 1340
},
{
"epoch": 0.8312807881773399,
"grad_norm": 0.8439570654706303,
"learning_rate": 2.529075907148916e-07,
"loss": 0.8701,
"step": 1350
},
{
"epoch": 0.8374384236453202,
"grad_norm": 0.9041155686907367,
"learning_rate": 2.352740072577002e-07,
"loss": 0.8668,
"step": 1360
},
{
"epoch": 0.8435960591133005,
"grad_norm": 0.9150682000805254,
"learning_rate": 2.182251856400826e-07,
"loss": 0.8594,
"step": 1370
},
{
"epoch": 0.8497536945812808,
"grad_norm": 0.8626796470648748,
"learning_rate": 2.0176900859738906e-07,
"loss": 0.8839,
"step": 1380
},
{
"epoch": 0.8559113300492611,
"grad_norm": 0.8349539026741156,
"learning_rate": 1.8591308484835833e-07,
"loss": 0.8733,
"step": 1390
},
{
"epoch": 0.8620689655172413,
"grad_norm": 0.9059703603517487,
"learning_rate": 1.706647455771302e-07,
"loss": 0.8739,
"step": 1400
},
{
"epoch": 0.8682266009852216,
"grad_norm": 0.8855388011148819,
"learning_rate": 1.56031041043582e-07,
"loss": 0.879,
"step": 1410
},
{
"epoch": 0.874384236453202,
"grad_norm": 0.8551754428690189,
"learning_rate": 1.4201873732355343e-07,
"loss": 0.8794,
"step": 1420
},
{
"epoch": 0.8805418719211823,
"grad_norm": 0.9468709892180046,
"learning_rate": 1.2863431318046615e-07,
"loss": 0.8931,
"step": 1430
},
{
"epoch": 0.8866995073891626,
"grad_norm": 0.8556767525548026,
"learning_rate": 1.158839570697861e-07,
"loss": 0.8777,
"step": 1440
},
{
"epoch": 0.8928571428571429,
"grad_norm": 0.864133807831794,
"learning_rate": 1.0377356427771567e-07,
"loss": 0.8627,
"step": 1450
},
{
"epoch": 0.8990147783251231,
"grad_norm": 0.8989986768130477,
"learning_rate": 9.230873419543373e-08,
"loss": 0.8616,
"step": 1460
},
{
"epoch": 0.9051724137931034,
"grad_norm": 0.8711608226459443,
"learning_rate": 8.14947677301468e-08,
"loss": 0.8665,
"step": 1470
},
{
"epoch": 0.9113300492610837,
"grad_norm": 0.8224276565204746,
"learning_rate": 7.133666485414858e-08,
"loss": 0.8642,
"step": 1480
},
{
"epoch": 0.9174876847290641,
"grad_norm": 0.841920633489749,
"learning_rate": 6.183912229302135e-08,
"loss": 0.8646,
"step": 1490
},
{
"epoch": 0.9236453201970444,
"grad_norm": 0.8581528480622548,
"learning_rate": 5.3006531354045596e-08,
"loss": 0.8671,
"step": 1500
},
{
"epoch": 0.9298029556650246,
"grad_norm": 0.808232107109127,
"learning_rate": 4.4842975895823926e-08,
"loss": 0.8749,
"step": 1510
},
{
"epoch": 0.9359605911330049,
"grad_norm": 0.8903690610551493,
"learning_rate": 3.7352230440058534e-08,
"loss": 0.8747,
"step": 1520
},
{
"epoch": 0.9421182266009852,
"grad_norm": 0.8304916684961995,
"learning_rate": 3.053775842635453e-08,
"loss": 0.8766,
"step": 1530
},
{
"epoch": 0.9482758620689655,
"grad_norm": 0.8415141188924126,
"learning_rate": 2.4402710610854582e-08,
"loss": 0.8622,
"step": 1540
},
{
"epoch": 0.9544334975369458,
"grad_norm": 0.9097293475369945,
"learning_rate": 1.894992360944786e-08,
"loss": 0.8555,
"step": 1550
},
{
"epoch": 0.9605911330049262,
"grad_norm": 0.8268939856081935,
"learning_rate": 1.4181918586225029e-08,
"loss": 0.8669,
"step": 1560
},
{
"epoch": 0.9667487684729064,
"grad_norm": 0.8721871446970374,
"learning_rate": 1.0100900087787357e-08,
"loss": 0.8705,
"step": 1570
},
{
"epoch": 0.9729064039408867,
"grad_norm": 0.8696350718282767,
"learning_rate": 6.708755023946245e-09,
"loss": 0.8761,
"step": 1580
},
{
"epoch": 0.979064039408867,
"grad_norm": 0.8251276019284721,
"learning_rate": 4.007051795287098e-09,
"loss": 0.8688,
"step": 1590
},
{
"epoch": 0.9852216748768473,
"grad_norm": 0.8414267048570288,
"learning_rate": 1.997039568000403e-09,
"loss": 0.8574,
"step": 1600
},
{
"epoch": 0.9913793103448276,
"grad_norm": 0.8837062633704493,
"learning_rate": 6.796476963130683e-10,
"loss": 0.8721,
"step": 1610
},
{
"epoch": 0.9975369458128078,
"grad_norm": 1.0761674680711684,
"learning_rate": 5.548529279081338e-11,
"loss": 0.8681,
"step": 1620
},
{
"epoch": 1.0,
"eval_loss": 0.8712338209152222,
"eval_runtime": 531.6277,
"eval_samples_per_second": 43.468,
"eval_steps_per_second": 1.36,
"step": 1624
},
{
"epoch": 1.0,
"step": 1624,
"total_flos": 555967626641408.0,
"train_loss": 0.8854433513186836,
"train_runtime": 19158.9512,
"train_samples_per_second": 10.849,
"train_steps_per_second": 0.085
}
],
"logging_steps": 10,
"max_steps": 1624,
"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": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 555967626641408.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}