namhokaist's picture
Upload Qwen3 continue-SFT checkpoint
22f2854 verified
Raw
History Blame Contribute Delete
55.4 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 100.0,
"global_step": 2420,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0004133512452206262,
"grad_norm": 15.187292098999023,
"learning_rate": 4.132231404958678e-09,
"loss": 0.41774559020996094,
"step": 1,
"token_acc": 0.8747855917667239
},
{
"epoch": 0.004133512452206263,
"grad_norm": 19.65194320678711,
"learning_rate": 4.1322314049586776e-08,
"loss": 0.5101552539401584,
"step": 10,
"token_acc": 0.8672614712308813
},
{
"epoch": 0.008267024904412525,
"grad_norm": 17.633342742919922,
"learning_rate": 8.264462809917355e-08,
"loss": 0.500081729888916,
"step": 20,
"token_acc": 0.8690631099544567
},
{
"epoch": 0.012400537356618786,
"grad_norm": 14.556735038757324,
"learning_rate": 1.2396694214876034e-07,
"loss": 0.47970046997070315,
"step": 30,
"token_acc": 0.8761444980855668
},
{
"epoch": 0.01653404980882505,
"grad_norm": 13.404815673828125,
"learning_rate": 1.652892561983471e-07,
"loss": 0.4663290023803711,
"step": 40,
"token_acc": 0.8768464730290456
},
{
"epoch": 0.02066756226103131,
"grad_norm": 16.647714614868164,
"learning_rate": 2.0661157024793388e-07,
"loss": 0.47812185287475584,
"step": 50,
"token_acc": 0.8676254289916653
},
{
"epoch": 0.024801074713237573,
"grad_norm": 12.500774383544922,
"learning_rate": 2.4793388429752067e-07,
"loss": 0.4136521339416504,
"step": 60,
"token_acc": 0.8737117618190741
},
{
"epoch": 0.028934587165443837,
"grad_norm": 6.377337455749512,
"learning_rate": 2.892561983471074e-07,
"loss": 0.36531882286071776,
"step": 70,
"token_acc": 0.876664447403462
},
{
"epoch": 0.0330680996176501,
"grad_norm": 5.292947292327881,
"learning_rate": 3.305785123966942e-07,
"loss": 0.32203586101531984,
"step": 80,
"token_acc": 0.8768644348690753
},
{
"epoch": 0.03720161206985636,
"grad_norm": 4.2747650146484375,
"learning_rate": 3.71900826446281e-07,
"loss": 0.3041213035583496,
"step": 90,
"token_acc": 0.8846544715447154
},
{
"epoch": 0.04133512452206262,
"grad_norm": 4.908409595489502,
"learning_rate": 4.1322314049586775e-07,
"loss": 0.3379059791564941,
"step": 100,
"token_acc": 0.8750799744081894
},
{
"epoch": 0.04546863697426889,
"grad_norm": 3.656118392944336,
"learning_rate": 4.545454545454545e-07,
"loss": 0.31875288486480713,
"step": 110,
"token_acc": 0.8804008542796123
},
{
"epoch": 0.049602149426475145,
"grad_norm": 3.3246145248413086,
"learning_rate": 4.958677685950413e-07,
"loss": 0.29662423133850097,
"step": 120,
"token_acc": 0.8869823461475004
},
{
"epoch": 0.05373566187868141,
"grad_norm": 3.7789199352264404,
"learning_rate": 4.999810934903514e-07,
"loss": 0.29268321990966795,
"step": 130,
"token_acc": 0.8861788617886179
},
{
"epoch": 0.057869174330887674,
"grad_norm": 4.397291660308838,
"learning_rate": 4.999157413258781e-07,
"loss": 0.29063005447387696,
"step": 140,
"token_acc": 0.8904428904428905
},
{
"epoch": 0.06200268678309393,
"grad_norm": 3.78377366065979,
"learning_rate": 4.998037222932112e-07,
"loss": 0.29862408638000487,
"step": 150,
"token_acc": 0.8850214450676345
},
{
"epoch": 0.0661361992353002,
"grad_norm": 2.576719284057617,
"learning_rate": 4.996450573097103e-07,
"loss": 0.29778199195861815,
"step": 160,
"token_acc": 0.8898598516075845
},
{
"epoch": 0.07026971168750645,
"grad_norm": 2.56404709815979,
"learning_rate": 4.994397760029525e-07,
"loss": 0.29409146308898926,
"step": 170,
"token_acc": 0.8850082372322899
},
{
"epoch": 0.07440322413971272,
"grad_norm": 4.336145401000977,
"learning_rate": 4.991879167051996e-07,
"loss": 0.26966080665588377,
"step": 180,
"token_acc": 0.8925689307330195
},
{
"epoch": 0.07853673659191898,
"grad_norm": 3.390794515609741,
"learning_rate": 4.988895264462412e-07,
"loss": 0.29469354152679444,
"step": 190,
"token_acc": 0.8883736336535276
},
{
"epoch": 0.08267024904412525,
"grad_norm": 3.1859688758850098,
"learning_rate": 4.985446609446118e-07,
"loss": 0.2899317741394043,
"step": 200,
"token_acc": 0.8896125584502338
},
{
"epoch": 0.08680376149633151,
"grad_norm": 3.3690783977508545,
"learning_rate": 4.98153384597187e-07,
"loss": 0.28981895446777345,
"step": 210,
"token_acc": 0.8872266973532796
},
{
"epoch": 0.09093727394853778,
"grad_norm": 4.324132442474365,
"learning_rate": 4.977157704671584e-07,
"loss": 0.28389716148376465,
"step": 220,
"token_acc": 0.8907058234607041
},
{
"epoch": 0.09507078640074403,
"grad_norm": 3.3958513736724854,
"learning_rate": 4.97231900270391e-07,
"loss": 0.28700425624847414,
"step": 230,
"token_acc": 0.8925086819910699
},
{
"epoch": 0.09920429885295029,
"grad_norm": 4.404426574707031,
"learning_rate": 4.967018643601637e-07,
"loss": 0.27868449687957764,
"step": 240,
"token_acc": 0.8929693961952027
},
{
"epoch": 0.10333781130515655,
"grad_norm": 3.217381715774536,
"learning_rate": 4.961257617102981e-07,
"loss": 0.2906649112701416,
"step": 250,
"token_acc": 0.889583676155998
},
{
"epoch": 0.10747132375736282,
"grad_norm": 4.56388521194458,
"learning_rate": 4.955036998966764e-07,
"loss": 0.29251632690429685,
"step": 260,
"token_acc": 0.8891044293566117
},
{
"epoch": 0.11160483620956908,
"grad_norm": 3.9011070728302,
"learning_rate": 4.948357950771547e-07,
"loss": 0.2851822853088379,
"step": 270,
"token_acc": 0.8942994847930863
},
{
"epoch": 0.11573834866177535,
"grad_norm": 3.7518205642700195,
"learning_rate": 4.94122171969872e-07,
"loss": 0.276279616355896,
"step": 280,
"token_acc": 0.8945077026121903
},
{
"epoch": 0.11987186111398161,
"grad_norm": 3.027475357055664,
"learning_rate": 4.933629638299619e-07,
"loss": 0.276011323928833,
"step": 290,
"token_acc": 0.896625868342706
},
{
"epoch": 0.12400537356618786,
"grad_norm": 22.47526741027832,
"learning_rate": 4.925583124246696e-07,
"loss": 0.27693781852722166,
"step": 300,
"token_acc": 0.8915203709837695
},
{
"epoch": 0.12813888601839413,
"grad_norm": 2.701873302459717,
"learning_rate": 4.917083680068799e-07,
"loss": 0.2832486629486084,
"step": 310,
"token_acc": 0.8900897904888593
},
{
"epoch": 0.1322723984706004,
"grad_norm": 3.029818534851074,
"learning_rate": 4.908132892870602e-07,
"loss": 0.2849259614944458,
"step": 320,
"token_acc": 0.8893092105263158
},
{
"epoch": 0.13640591092280666,
"grad_norm": 3.6257729530334473,
"learning_rate": 4.898732434036243e-07,
"loss": 0.30432281494140623,
"step": 330,
"token_acc": 0.8885283141327276
},
{
"epoch": 0.1405394233750129,
"grad_norm": 3.2731194496154785,
"learning_rate": 4.888884058917233e-07,
"loss": 0.2886804580688477,
"step": 340,
"token_acc": 0.8868455931080186
},
{
"epoch": 0.14467293582721918,
"grad_norm": 2.3444454669952393,
"learning_rate": 4.878589606504668e-07,
"loss": 0.2758631229400635,
"step": 350,
"token_acc": 0.8895339193896169
},
{
"epoch": 0.14880644827942544,
"grad_norm": 2.296010732650757,
"learning_rate": 4.867850999085843e-07,
"loss": 0.28838496208190917,
"step": 360,
"token_acc": 0.8871601703242712
},
{
"epoch": 0.1529399607316317,
"grad_norm": 3.069838523864746,
"learning_rate": 4.856670241885293e-07,
"loss": 0.283137583732605,
"step": 370,
"token_acc": 0.8913007997388608
},
{
"epoch": 0.15707347318383796,
"grad_norm": 2.7073512077331543,
"learning_rate": 4.845049422690368e-07,
"loss": 0.26442747116088866,
"step": 380,
"token_acc": 0.8952525764487245
},
{
"epoch": 0.16120698563604424,
"grad_norm": 3.427346706390381,
"learning_rate": 4.832990711461371e-07,
"loss": 0.2623051643371582,
"step": 390,
"token_acc": 0.8944868532654792
},
{
"epoch": 0.1653404980882505,
"grad_norm": 3.874199390411377,
"learning_rate": 4.82049635992636e-07,
"loss": 0.27241015434265137,
"step": 400,
"token_acc": 0.8945147679324894
},
{
"epoch": 0.16947401054045674,
"grad_norm": 3.385901927947998,
"learning_rate": 4.807568701160688e-07,
"loss": 0.2917393684387207,
"step": 410,
"token_acc": 0.8884169253389969
},
{
"epoch": 0.17360752299266302,
"grad_norm": 2.8726325035095215,
"learning_rate": 4.794210149151344e-07,
"loss": 0.29250597953796387,
"step": 420,
"token_acc": 0.8883935679032001
},
{
"epoch": 0.17774103544486927,
"grad_norm": 2.249457836151123,
"learning_rate": 4.780423198346187e-07,
"loss": 0.2777788877487183,
"step": 430,
"token_acc": 0.8955669931927611
},
{
"epoch": 0.18187454789707555,
"grad_norm": 2.388742685317993,
"learning_rate": 4.7662104231881575e-07,
"loss": 0.2727825164794922,
"step": 440,
"token_acc": 0.8921021478099104
},
{
"epoch": 0.1860080603492818,
"grad_norm": 11.276811599731445,
"learning_rate": 4.75157447763455e-07,
"loss": 0.2678534507751465,
"step": 450,
"token_acc": 0.8970366649924661
},
{
"epoch": 0.19014157280148805,
"grad_norm": 3.2645626068115234,
"learning_rate": 4.7365180946614415e-07,
"loss": 0.28694725036621094,
"step": 460,
"token_acc": 0.8915203709837695
},
{
"epoch": 0.19427508525369433,
"grad_norm": 4.024374485015869,
"learning_rate": 4.721044085753358e-07,
"loss": 0.2869134426116943,
"step": 470,
"token_acc": 0.8864149725717974
},
{
"epoch": 0.19840859770590058,
"grad_norm": 2.646841287612915,
"learning_rate": 4.705155340378287e-07,
"loss": 0.27790093421936035,
"step": 480,
"token_acc": 0.8927627230667547
},
{
"epoch": 0.20254211015810686,
"grad_norm": 4.402639389038086,
"learning_rate": 4.6888548254481263e-07,
"loss": 0.2458963394165039,
"step": 490,
"token_acc": 0.9047208821502413
},
{
"epoch": 0.2066756226103131,
"grad_norm": 4.239990711212158,
"learning_rate": 4.672145584764669e-07,
"loss": 0.2684264421463013,
"step": 500,
"token_acc": 0.8942339926248742
},
{
"epoch": 0.2108091350625194,
"grad_norm": 2.78176212310791,
"learning_rate": 4.6550307384512353e-07,
"loss": 0.279697847366333,
"step": 510,
"token_acc": 0.8904379202886665
},
{
"epoch": 0.21494264751472564,
"grad_norm": 2.863884925842285,
"learning_rate": 4.6375134823700503e-07,
"loss": 0.2946455001831055,
"step": 520,
"token_acc": 0.8851153721156971
},
{
"epoch": 0.2190761599669319,
"grad_norm": 2.9876773357391357,
"learning_rate": 4.6195970875254773e-07,
"loss": 0.27865424156188967,
"step": 530,
"token_acc": 0.892393625759816
},
{
"epoch": 0.22320967241913817,
"grad_norm": 7.1545562744140625,
"learning_rate": 4.601284899453226e-07,
"loss": 0.2536644458770752,
"step": 540,
"token_acc": 0.9022222222222223
},
{
"epoch": 0.22734318487134442,
"grad_norm": 3.1952993869781494,
"learning_rate": 4.5825803375956355e-07,
"loss": 0.2653358459472656,
"step": 550,
"token_acc": 0.8977581521739131
},
{
"epoch": 0.2314766973235507,
"grad_norm": 3.3041276931762695,
"learning_rate": 4.5634868946631614e-07,
"loss": 0.29339802265167236,
"step": 560,
"token_acc": 0.8858834827144686
},
{
"epoch": 0.23561020977575695,
"grad_norm": 2.3186426162719727,
"learning_rate": 4.5440081359821816e-07,
"loss": 0.26668615341186525,
"step": 570,
"token_acc": 0.8962453905464298
},
{
"epoch": 0.23974372222796322,
"grad_norm": 10.634851455688477,
"learning_rate": 4.524147698829239e-07,
"loss": 0.3000948905944824,
"step": 580,
"token_acc": 0.8845410628019323
},
{
"epoch": 0.24387723468016947,
"grad_norm": 3.039048671722412,
"learning_rate": 4.503909291751856e-07,
"loss": 0.2688342571258545,
"step": 590,
"token_acc": 0.8956870611835507
},
{
"epoch": 0.24801074713237573,
"grad_norm": 5.39232873916626,
"learning_rate": 4.4832966938760303e-07,
"loss": 0.2932869911193848,
"step": 600,
"token_acc": 0.8858075040783034
},
{
"epoch": 0.252144259584582,
"grad_norm": 4.740464210510254,
"learning_rate": 4.462313754200561e-07,
"loss": 0.2607602596282959,
"step": 610,
"token_acc": 0.9009314140558848
},
{
"epoch": 0.25627777203678825,
"grad_norm": 4.451042175292969,
"learning_rate": 4.4409643908783225e-07,
"loss": 0.29047460556030275,
"step": 620,
"token_acc": 0.8890701468189234
},
{
"epoch": 0.26041128448899453,
"grad_norm": 40.26980972290039,
"learning_rate": 4.4192525904846246e-07,
"loss": 0.27602272033691405,
"step": 630,
"token_acc": 0.8922743922743923
},
{
"epoch": 0.2645447969412008,
"grad_norm": 2.3861398696899414,
"learning_rate": 4.3971824072728016e-07,
"loss": 0.26196789741516113,
"step": 640,
"token_acc": 0.896882898062342
},
{
"epoch": 0.26867830939340703,
"grad_norm": 3.5515103340148926,
"learning_rate": 4.374757962417155e-07,
"loss": 0.2640524864196777,
"step": 650,
"token_acc": 0.8970958536175927
},
{
"epoch": 0.2728118218456133,
"grad_norm": 2.9944169521331787,
"learning_rate": 4.351983443243409e-07,
"loss": 0.26778354644775393,
"step": 660,
"token_acc": 0.8965807646959744
},
{
"epoch": 0.2769453342978196,
"grad_norm": 2.6837551593780518,
"learning_rate": 4.328863102446807e-07,
"loss": 0.26420319080352783,
"step": 670,
"token_acc": 0.9015952980688497
},
{
"epoch": 0.2810788467500258,
"grad_norm": 3.8792166709899902,
"learning_rate": 4.305401257298006e-07,
"loss": 0.2633915901184082,
"step": 680,
"token_acc": 0.8977387782652717
},
{
"epoch": 0.2852123592022321,
"grad_norm": 3.0519802570343018,
"learning_rate": 4.2816022888369076e-07,
"loss": 0.2695890426635742,
"step": 690,
"token_acc": 0.8971804828634138
},
{
"epoch": 0.28934587165443837,
"grad_norm": 5.02327299118042,
"learning_rate": 4.257470641054589e-07,
"loss": 0.27457616329193113,
"step": 700,
"token_acc": 0.8917720479553293
},
{
"epoch": 0.29347938410664465,
"grad_norm": 3.205557107925415,
"learning_rate": 4.2330108200634723e-07,
"loss": 0.2886713981628418,
"step": 710,
"token_acc": 0.8879117364886473
},
{
"epoch": 0.29761289655885087,
"grad_norm": 3.19230580329895,
"learning_rate": 4.2082273932558953e-07,
"loss": 0.2535089492797852,
"step": 720,
"token_acc": 0.8990020646937371
},
{
"epoch": 0.30174640901105715,
"grad_norm": 3.3932271003723145,
"learning_rate": 4.183124988451243e-07,
"loss": 0.2608576059341431,
"step": 730,
"token_acc": 0.8972371967654986
},
{
"epoch": 0.3058799214632634,
"grad_norm": 3.8836634159088135,
"learning_rate": 4.157708293031792e-07,
"loss": 0.2700679302215576,
"step": 740,
"token_acc": 0.8951531470885223
},
{
"epoch": 0.31001343391546965,
"grad_norm": 2.8291547298431396,
"learning_rate": 4.131982053067431e-07,
"loss": 0.26695971488952636,
"step": 750,
"token_acc": 0.8984375
},
{
"epoch": 0.3141469463676759,
"grad_norm": 4.157771587371826,
"learning_rate": 4.10595107242943e-07,
"loss": 0.26310505867004397,
"step": 760,
"token_acc": 0.8973281801377919
},
{
"epoch": 0.3182804588198822,
"grad_norm": 2.9965980052948,
"learning_rate": 4.0796202118934096e-07,
"loss": 0.27557735443115233,
"step": 770,
"token_acc": 0.8943264764432648
},
{
"epoch": 0.3224139712720885,
"grad_norm": 2.7542531490325928,
"learning_rate": 4.052994388231684e-07,
"loss": 0.281093692779541,
"step": 780,
"token_acc": 0.8885085574572127
},
{
"epoch": 0.3265474837242947,
"grad_norm": 3.779541015625,
"learning_rate": 4.026078573295154e-07,
"loss": 0.2782632350921631,
"step": 790,
"token_acc": 0.8945878434637802
},
{
"epoch": 0.330680996176501,
"grad_norm": 5.1954474449157715,
"learning_rate": 3.998877793084906e-07,
"loss": 0.27410087585449217,
"step": 800,
"token_acc": 0.8901967378510173
},
{
"epoch": 0.33481450862870726,
"grad_norm": 7.226117134094238,
"learning_rate": 3.9713971268137123e-07,
"loss": 0.2809471607208252,
"step": 810,
"token_acc": 0.8907059585492227
},
{
"epoch": 0.3389480210809135,
"grad_norm": 2.590132474899292,
"learning_rate": 3.943641705957581e-07,
"loss": 0.2636889934539795,
"step": 820,
"token_acc": 0.8986338336987688
},
{
"epoch": 0.34308153353311976,
"grad_norm": 2.9958176612854004,
"learning_rate": 3.915616713297556e-07,
"loss": 0.2696376800537109,
"step": 830,
"token_acc": 0.8965865142088448
},
{
"epoch": 0.34721504598532604,
"grad_norm": 7.742995738983154,
"learning_rate": 3.887327381951934e-07,
"loss": 0.281767463684082,
"step": 840,
"token_acc": 0.8927512355848435
},
{
"epoch": 0.35134855843753227,
"grad_norm": 2.001983404159546,
"learning_rate": 3.858778994399081e-07,
"loss": 0.2946297645568848,
"step": 850,
"token_acc": 0.8862207896857373
},
{
"epoch": 0.35548207088973854,
"grad_norm": 2.323462963104248,
"learning_rate": 3.8299768814910374e-07,
"loss": 0.2778719425201416,
"step": 860,
"token_acc": 0.8916474843801381
},
{
"epoch": 0.3596155833419448,
"grad_norm": 2.8264565467834473,
"learning_rate": 3.8009264214580805e-07,
"loss": 0.2745103359222412,
"step": 870,
"token_acc": 0.8917322834645669
},
{
"epoch": 0.3637490957941511,
"grad_norm": 6.078236103057861,
"learning_rate": 3.771633038904446e-07,
"loss": 0.294478702545166,
"step": 880,
"token_acc": 0.8898250162022034
},
{
"epoch": 0.3678826082463573,
"grad_norm": 2.497274398803711,
"learning_rate": 3.742102203795392e-07,
"loss": 0.27193236351013184,
"step": 890,
"token_acc": 0.8928221859706362
},
{
"epoch": 0.3720161206985636,
"grad_norm": 3.3112785816192627,
"learning_rate": 3.712339430435792e-07,
"loss": 0.28939318656921387,
"step": 900,
"token_acc": 0.8878336335352873
},
{
"epoch": 0.3761496331507699,
"grad_norm": 2.605825424194336,
"learning_rate": 3.6823502764404376e-07,
"loss": 0.2712101936340332,
"step": 910,
"token_acc": 0.894501490559788
},
{
"epoch": 0.3802831456029761,
"grad_norm": 3.187836170196533,
"learning_rate": 3.652140341696274e-07,
"loss": 0.271730899810791,
"step": 920,
"token_acc": 0.8975674333940096
},
{
"epoch": 0.3844166580551824,
"grad_norm": 7.310910701751709,
"learning_rate": 3.621715267316723e-07,
"loss": 0.2949850559234619,
"step": 930,
"token_acc": 0.8846963562753036
},
{
"epoch": 0.38855017050738866,
"grad_norm": 2.690484046936035,
"learning_rate": 3.5910807345883175e-07,
"loss": 0.2591448545455933,
"step": 940,
"token_acc": 0.898834656308056
},
{
"epoch": 0.39268368295959494,
"grad_norm": 3.423795461654663,
"learning_rate": 3.560242463909836e-07,
"loss": 0.27901949882507326,
"step": 950,
"token_acc": 0.8928336620644313
},
{
"epoch": 0.39681719541180116,
"grad_norm": 3.1476311683654785,
"learning_rate": 3.5292062137241264e-07,
"loss": 0.2721529722213745,
"step": 960,
"token_acc": 0.8943731490621916
},
{
"epoch": 0.40095070786400744,
"grad_norm": 3.0869557857513428,
"learning_rate": 3.497977779442827e-07,
"loss": 0.27245345115661623,
"step": 970,
"token_acc": 0.8961685188256759
},
{
"epoch": 0.4050842203162137,
"grad_norm": 2.2117760181427,
"learning_rate": 3.466562992364197e-07,
"loss": 0.28414316177368165,
"step": 980,
"token_acc": 0.8968448585908125
},
{
"epoch": 0.40921773276841994,
"grad_norm": 2.3948702812194824,
"learning_rate": 3.4349677185842243e-07,
"loss": 0.2586636781692505,
"step": 990,
"token_acc": 0.9002820640451302
},
{
"epoch": 0.4133512452206262,
"grad_norm": 2.862246036529541,
"learning_rate": 3.4031978579012534e-07,
"loss": 0.266216516494751,
"step": 1000,
"token_acc": 0.8937520882058135
},
{
"epoch": 0.4174847576728325,
"grad_norm": 2.563786029815674,
"learning_rate": 3.3712593427143136e-07,
"loss": 0.27124173641204835,
"step": 1010,
"token_acc": 0.8961279461279461
},
{
"epoch": 0.4216182701250388,
"grad_norm": 2.9208412170410156,
"learning_rate": 3.3391581369153577e-07,
"loss": 0.25082371234893797,
"step": 1020,
"token_acc": 0.900375939849624
},
{
"epoch": 0.425751782577245,
"grad_norm": 3.598951578140259,
"learning_rate": 3.306900234775622e-07,
"loss": 0.27564148902893065,
"step": 1030,
"token_acc": 0.893046357615894
},
{
"epoch": 0.4298852950294513,
"grad_norm": 2.1928725242614746,
"learning_rate": 3.274491659826315e-07,
"loss": 0.26628134250640867,
"step": 1040,
"token_acc": 0.8938008974572046
},
{
"epoch": 0.43401880748165755,
"grad_norm": 2.6296000480651855,
"learning_rate": 3.24193846373384e-07,
"loss": 0.27588448524475095,
"step": 1050,
"token_acc": 0.8928809048569527
},
{
"epoch": 0.4381523199338638,
"grad_norm": 2.9763195514678955,
"learning_rate": 3.2092467251697666e-07,
"loss": 0.25474910736083983,
"step": 1060,
"token_acc": 0.899932042133877
},
{
"epoch": 0.44228583238607005,
"grad_norm": 3.3110697269439697,
"learning_rate": 3.176422548675754e-07,
"loss": 0.27116897106170657,
"step": 1070,
"token_acc": 0.8963963963963963
},
{
"epoch": 0.44641934483827633,
"grad_norm": 2.6853387355804443,
"learning_rate": 3.143472063523652e-07,
"loss": 0.2755572319030762,
"step": 1080,
"token_acc": 0.892444295526889
},
{
"epoch": 0.4505528572904826,
"grad_norm": 2.4194726943969727,
"learning_rate": 3.1104014225709784e-07,
"loss": 0.2763054370880127,
"step": 1090,
"token_acc": 0.8904599382415082
},
{
"epoch": 0.45468636974268883,
"grad_norm": 3.427236557006836,
"learning_rate": 3.077216801111989e-07,
"loss": 0.26727967262268065,
"step": 1100,
"token_acc": 0.8977576974564926
},
{
"epoch": 0.4588198821948951,
"grad_norm": 2.6082870960235596,
"learning_rate": 3.043924395724571e-07,
"loss": 0.29102773666381837,
"step": 1110,
"token_acc": 0.8888158975201182
},
{
"epoch": 0.4629533946471014,
"grad_norm": 2.626399517059326,
"learning_rate": 3.010530423113144e-07,
"loss": 0.27240285873413084,
"step": 1120,
"token_acc": 0.8930702917771883
},
{
"epoch": 0.4670869070993076,
"grad_norm": 2.829056978225708,
"learning_rate": 2.977041118947818e-07,
"loss": 0.27853941917419434,
"step": 1130,
"token_acc": 0.8952905811623246
},
{
"epoch": 0.4712204195515139,
"grad_norm": 3.656513214111328,
"learning_rate": 2.9434627367000043e-07,
"loss": 0.26388509273529054,
"step": 1140,
"token_acc": 0.8956347215253387
},
{
"epoch": 0.47535393200372017,
"grad_norm": 3.179504632949829,
"learning_rate": 2.9098015464747003e-07,
"loss": 0.2650350570678711,
"step": 1150,
"token_acc": 0.9002514668901928
},
{
"epoch": 0.47948744445592645,
"grad_norm": 1.9940325021743774,
"learning_rate": 2.876063833839669e-07,
"loss": 0.27263484001159666,
"step": 1160,
"token_acc": 0.8941737985516788
},
{
"epoch": 0.48362095690813267,
"grad_norm": 2.556058645248413,
"learning_rate": 2.842255898651736e-07,
"loss": 0.27724528312683105,
"step": 1170,
"token_acc": 0.8932891466445734
},
{
"epoch": 0.48775446936033895,
"grad_norm": 2.78503680229187,
"learning_rate": 2.808384053880413e-07,
"loss": 0.2747943639755249,
"step": 1180,
"token_acc": 0.8945209973753281
},
{
"epoch": 0.4918879818125452,
"grad_norm": 3.950742244720459,
"learning_rate": 2.774454624429073e-07,
"loss": 0.25716724395751955,
"step": 1190,
"token_acc": 0.9012387578482945
},
{
"epoch": 0.49602149426475145,
"grad_norm": 2.3381898403167725,
"learning_rate": 2.7404739459539007e-07,
"loss": 0.27159938812255857,
"step": 1200,
"token_acc": 0.8916296661733267
},
{
"epoch": 0.5001550067169578,
"grad_norm": 2.3968288898468018,
"learning_rate": 2.706448363680831e-07,
"loss": 0.2611609697341919,
"step": 1210,
"token_acc": 0.8995780590717299
},
{
"epoch": 0.504288519169164,
"grad_norm": 2.8506569862365723,
"learning_rate": 2.672384231220704e-07,
"loss": 0.27340211868286135,
"step": 1220,
"token_acc": 0.8947804473902237
},
{
"epoch": 0.5084220316213702,
"grad_norm": 3.3233721256256104,
"learning_rate": 2.6382879093828514e-07,
"loss": 0.2683706760406494,
"step": 1230,
"token_acc": 0.8963037297206891
},
{
"epoch": 0.5125555440735765,
"grad_norm": 2.4866981506347656,
"learning_rate": 2.6041657649873413e-07,
"loss": 0.26338682174682615,
"step": 1240,
"token_acc": 0.8985190171659374
},
{
"epoch": 0.5166890565257828,
"grad_norm": 3.7649965286254883,
"learning_rate": 2.5700241696760985e-07,
"loss": 0.28077752590179444,
"step": 1250,
"token_acc": 0.8936486933509759
},
{
"epoch": 0.5208225689779891,
"grad_norm": 2.989053964614868,
"learning_rate": 2.5358694987231236e-07,
"loss": 0.2896185636520386,
"step": 1260,
"token_acc": 0.8880983182406209
},
{
"epoch": 0.5249560814301953,
"grad_norm": 2.023987293243408,
"learning_rate": 2.5017081298440385e-07,
"loss": 0.28451263904571533,
"step": 1270,
"token_acc": 0.891644908616188
},
{
"epoch": 0.5290895938824016,
"grad_norm": 3.300025224685669,
"learning_rate": 2.467546442005173e-07,
"loss": 0.2620598793029785,
"step": 1280,
"token_acc": 0.900350525788683
},
{
"epoch": 0.5332231063346078,
"grad_norm": 3.0800909996032715,
"learning_rate": 2.433390814232413e-07,
"loss": 0.2617885828018188,
"step": 1290,
"token_acc": 0.8993594066082266
},
{
"epoch": 0.5373566187868141,
"grad_norm": 2.6922402381896973,
"learning_rate": 2.399247624420047e-07,
"loss": 0.26834275722503664,
"step": 1300,
"token_acc": 0.8978016445712368
},
{
"epoch": 0.5414901312390203,
"grad_norm": 3.132843017578125,
"learning_rate": 2.365123248139821e-07,
"loss": 0.2786297559738159,
"step": 1310,
"token_acc": 0.8910678971654581
},
{
"epoch": 0.5456236436912266,
"grad_norm": 2.4884276390075684,
"learning_rate": 2.3310240574504182e-07,
"loss": 0.2636242151260376,
"step": 1320,
"token_acc": 0.8977688977688978
},
{
"epoch": 0.5497571561434329,
"grad_norm": 2.82171368598938,
"learning_rate": 2.2969564197076079e-07,
"loss": 0.2741330623626709,
"step": 1330,
"token_acc": 0.8915287563493364
},
{
"epoch": 0.5538906685956392,
"grad_norm": 4.288270473480225,
"learning_rate": 2.2629266963752667e-07,
"loss": 0.2554323196411133,
"step": 1340,
"token_acc": 0.9015640938456307
},
{
"epoch": 0.5580241810478455,
"grad_norm": 3.2320291996002197,
"learning_rate": 2.2289412418374936e-07,
"loss": 0.2432697296142578,
"step": 1350,
"token_acc": 0.907323055360897
},
{
"epoch": 0.5621576935000516,
"grad_norm": 2.923374652862549,
"learning_rate": 2.1950064022120633e-07,
"loss": 0.2673605680465698,
"step": 1360,
"token_acc": 0.8946932006633499
},
{
"epoch": 0.5662912059522579,
"grad_norm": 2.6756703853607178,
"learning_rate": 2.1611285141654016e-07,
"loss": 0.25931453704833984,
"step": 1370,
"token_acc": 0.9023279352226721
},
{
"epoch": 0.5704247184044642,
"grad_norm": 2.919431686401367,
"learning_rate": 2.1273139037293464e-07,
"loss": 0.2924022674560547,
"step": 1380,
"token_acc": 0.8861129136924075
},
{
"epoch": 0.5745582308566705,
"grad_norm": 2.697990894317627,
"learning_rate": 2.0935688851198806e-07,
"loss": 0.2696035861968994,
"step": 1390,
"token_acc": 0.8975341552815728
},
{
"epoch": 0.5786917433088767,
"grad_norm": 2.2708208560943604,
"learning_rate": 2.0598997595580767e-07,
"loss": 0.26399407386779783,
"step": 1400,
"token_acc": 0.8985267339844396
},
{
"epoch": 0.582825255761083,
"grad_norm": 2.6366543769836426,
"learning_rate": 2.026312814093467e-07,
"loss": 0.27448673248291017,
"step": 1410,
"token_acc": 0.8944939004286185
},
{
"epoch": 0.5869587682132893,
"grad_norm": 3.1638495922088623,
"learning_rate": 1.992814320430063e-07,
"loss": 0.25971574783325196,
"step": 1420,
"token_acc": 0.8989193682460516
},
{
"epoch": 0.5910922806654955,
"grad_norm": 2.5996718406677246,
"learning_rate": 1.959410533755232e-07,
"loss": 0.2614893436431885,
"step": 1430,
"token_acc": 0.8977443609022556
},
{
"epoch": 0.5952257931177017,
"grad_norm": 3.4245429039001465,
"learning_rate": 1.926107691571669e-07,
"loss": 0.25876526832580565,
"step": 1440,
"token_acc": 0.8990978950885399
},
{
"epoch": 0.599359305569908,
"grad_norm": 2.5762760639190674,
"learning_rate": 1.8929120125326638e-07,
"loss": 0.2643924951553345,
"step": 1450,
"token_acc": 0.8950617283950617
},
{
"epoch": 0.6034928180221143,
"grad_norm": 5.55498743057251,
"learning_rate": 1.8598296952808862e-07,
"loss": 0.25945119857788085,
"step": 1460,
"token_acc": 0.9014960497562615
},
{
"epoch": 0.6076263304743206,
"grad_norm": 2.8441519737243652,
"learning_rate": 1.8268669172909136e-07,
"loss": 0.27311277389526367,
"step": 1470,
"token_acc": 0.8933951332560834
},
{
"epoch": 0.6117598429265269,
"grad_norm": 3.4796242713928223,
"learning_rate": 1.7940298337157112e-07,
"loss": 0.2703615427017212,
"step": 1480,
"token_acc": 0.8952948972829688
},
{
"epoch": 0.6158933553787331,
"grad_norm": 2.238431930541992,
"learning_rate": 1.7613245762372708e-07,
"loss": 0.2600128173828125,
"step": 1490,
"token_acc": 0.8996632996632996
},
{
"epoch": 0.6200268678309393,
"grad_norm": 2.2030811309814453,
"learning_rate": 1.7287572519216464e-07,
"loss": 0.25577716827392577,
"step": 1500,
"token_acc": 0.904350770272558
},
{
"epoch": 0.6241603802831456,
"grad_norm": 3.3314075469970703,
"learning_rate": 1.6963339420785754e-07,
"loss": 0.26538925170898436,
"step": 1510,
"token_acc": 0.8960908787170063
},
{
"epoch": 0.6282938927353519,
"grad_norm": 2.1006789207458496,
"learning_rate": 1.6640607011259151e-07,
"loss": 0.2399357795715332,
"step": 1520,
"token_acc": 0.902493938344302
},
{
"epoch": 0.6324274051875581,
"grad_norm": 2.626713752746582,
"learning_rate": 1.6319435554590944e-07,
"loss": 0.2515601396560669,
"step": 1530,
"token_acc": 0.9024597116200169
},
{
"epoch": 0.6365609176397644,
"grad_norm": 3.986639976501465,
"learning_rate": 1.5999885023258097e-07,
"loss": 0.25067887306213377,
"step": 1540,
"token_acc": 0.9026355548094679
},
{
"epoch": 0.6406944300919707,
"grad_norm": 3.3434393405914307,
"learning_rate": 1.5682015087061518e-07,
"loss": 0.2590707540512085,
"step": 1550,
"token_acc": 0.9005226774574271
},
{
"epoch": 0.644827942544177,
"grad_norm": 3.366339683532715,
"learning_rate": 1.5365885101983873e-07,
"loss": 0.2801948070526123,
"step": 1560,
"token_acc": 0.8898828887443071
},
{
"epoch": 0.6489614549963831,
"grad_norm": 2.367339611053467,
"learning_rate": 1.5051554099106078e-07,
"loss": 0.27551071643829345,
"step": 1570,
"token_acc": 0.8944099378881988
},
{
"epoch": 0.6530949674485894,
"grad_norm": 2.5332202911376953,
"learning_rate": 1.473908077358437e-07,
"loss": 0.24955625534057618,
"step": 1580,
"token_acc": 0.8998467041389883
},
{
"epoch": 0.6572284799007957,
"grad_norm": 3.8126466274261475,
"learning_rate": 1.4428523473690096e-07,
"loss": 0.26283018589019774,
"step": 1590,
"token_acc": 0.8978897208985704
},
{
"epoch": 0.661361992353002,
"grad_norm": 3.0068535804748535,
"learning_rate": 1.4119940189914423e-07,
"loss": 0.2719302177429199,
"step": 1600,
"token_acc": 0.895405179615706
},
{
"epoch": 0.6654955048052082,
"grad_norm": 2.681375026702881,
"learning_rate": 1.3813388544139627e-07,
"loss": 0.2649834632873535,
"step": 1610,
"token_acc": 0.8958127266732608
},
{
"epoch": 0.6696290172574145,
"grad_norm": 2.4036672115325928,
"learning_rate": 1.3508925778879404e-07,
"loss": 0.2687398910522461,
"step": 1620,
"token_acc": 0.8934707903780069
},
{
"epoch": 0.6737625297096208,
"grad_norm": 2.5061728954315186,
"learning_rate": 1.3206608746589966e-07,
"loss": 0.26619772911071776,
"step": 1630,
"token_acc": 0.8967507834405409
},
{
"epoch": 0.677896042161827,
"grad_norm": 2.5941014289855957,
"learning_rate": 1.2906493899053916e-07,
"loss": 0.26439173221588136,
"step": 1640,
"token_acc": 0.8992673992673993
},
{
"epoch": 0.6820295546140333,
"grad_norm": 2.642897844314575,
"learning_rate": 1.2608637276838984e-07,
"loss": 0.29090008735656736,
"step": 1650,
"token_acc": 0.8849372384937239
},
{
"epoch": 0.6861630670662395,
"grad_norm": 2.4777562618255615,
"learning_rate": 1.231309449883361e-07,
"loss": 0.2851018667221069,
"step": 1660,
"token_acc": 0.887654917109287
},
{
"epoch": 0.6902965795184458,
"grad_norm": 2.3737990856170654,
"learning_rate": 1.201992075186112e-07,
"loss": 0.260064959526062,
"step": 1670,
"token_acc": 0.8986190636577972
},
{
"epoch": 0.6944300919706521,
"grad_norm": 3.5159294605255127,
"learning_rate": 1.1729170780374717e-07,
"loss": 0.28465135097503663,
"step": 1680,
"token_acc": 0.8883774453394706
},
{
"epoch": 0.6985636044228584,
"grad_norm": 2.8983967304229736,
"learning_rate": 1.1440898876235036e-07,
"loss": 0.2606903314590454,
"step": 1690,
"token_acc": 0.8997312730937185
},
{
"epoch": 0.7026971168750645,
"grad_norm": 3.8028135299682617,
"learning_rate": 1.115515886857219e-07,
"loss": 0.27002012729644775,
"step": 1700,
"token_acc": 0.8957915831663327
},
{
"epoch": 0.7068306293272708,
"grad_norm": 2.745276927947998,
"learning_rate": 1.0872004113734201e-07,
"loss": 0.26640658378601073,
"step": 1710,
"token_acc": 0.8981512605042017
},
{
"epoch": 0.7109641417794771,
"grad_norm": 2.575197696685791,
"learning_rate": 1.0591487485323816e-07,
"loss": 0.2605179786682129,
"step": 1720,
"token_acc": 0.8981295925183701
},
{
"epoch": 0.7150976542316834,
"grad_norm": 2.4712765216827393,
"learning_rate": 1.0313661364325336e-07,
"loss": 0.2793430805206299,
"step": 1730,
"token_acc": 0.8935962196512954
},
{
"epoch": 0.7192311666838896,
"grad_norm": 3.8916172981262207,
"learning_rate": 1.0038577629323506e-07,
"loss": 0.256488037109375,
"step": 1740,
"token_acc": 0.8990278014668258
},
{
"epoch": 0.7233646791360959,
"grad_norm": 2.930061101913452,
"learning_rate": 9.766287646816241e-08,
"loss": 0.27172183990478516,
"step": 1750,
"token_acc": 0.8951014349332014
},
{
"epoch": 0.7274981915883022,
"grad_norm": 3.304394483566284,
"learning_rate": 9.49684226162292e-08,
"loss": 0.27311196327209475,
"step": 1760,
"token_acc": 0.8935924888815681
},
{
"epoch": 0.7316317040405084,
"grad_norm": 2.3063862323760986,
"learning_rate": 9.23029178739007e-08,
"loss": 0.2707731008529663,
"step": 1770,
"token_acc": 0.8950394218134035
},
{
"epoch": 0.7357652164927146,
"grad_norm": 2.8351504802703857,
"learning_rate": 8.96668599719636e-08,
"loss": 0.26172828674316406,
"step": 1780,
"token_acc": 0.8987975951903807
},
{
"epoch": 0.7398987289449209,
"grad_norm": 2.895158290863037,
"learning_rate": 8.706074114258364e-08,
"loss": 0.28212413787841795,
"step": 1790,
"token_acc": 0.8900865588763678
},
{
"epoch": 0.7440322413971272,
"grad_norm": 2.090304374694824,
"learning_rate": 8.448504802739165e-08,
"loss": 0.26543664932250977,
"step": 1800,
"token_acc": 0.8973930481283422
},
{
"epoch": 0.7481657538493335,
"grad_norm": 3.581449508666992,
"learning_rate": 8.19402615866125e-08,
"loss": 0.2609621524810791,
"step": 1810,
"token_acc": 0.897105571356868
},
{
"epoch": 0.7522992663015398,
"grad_norm": 3.7827906608581543,
"learning_rate": 7.942685700925477e-08,
"loss": 0.2822258710861206,
"step": 1820,
"token_acc": 0.8955717118307998
},
{
"epoch": 0.756432778753746,
"grad_norm": 4.347923278808594,
"learning_rate": 7.694530362437942e-08,
"loss": 0.27174232006072996,
"step": 1830,
"token_acc": 0.8972891059100859
},
{
"epoch": 0.7605662912059522,
"grad_norm": 2.4850101470947266,
"learning_rate": 7.449606481346085e-08,
"loss": 0.27895298004150393,
"step": 1840,
"token_acc": 0.8953719008264462
},
{
"epoch": 0.7646998036581585,
"grad_norm": 2.907944679260254,
"learning_rate": 7.207959792385998e-08,
"loss": 0.2639348268508911,
"step": 1850,
"token_acc": 0.8955551799898597
},
{
"epoch": 0.7688333161103648,
"grad_norm": 2.6346871852874756,
"learning_rate": 6.969635418342365e-08,
"loss": 0.280044412612915,
"step": 1860,
"token_acc": 0.8924205378973105
},
{
"epoch": 0.772966828562571,
"grad_norm": 2.931610107421875,
"learning_rate": 6.73467786162265e-08,
"loss": 0.2777987003326416,
"step": 1870,
"token_acc": 0.8917720479553293
},
{
"epoch": 0.7771003410147773,
"grad_norm": 2.1244542598724365,
"learning_rate": 6.503130995947131e-08,
"loss": 0.26816506385803224,
"step": 1880,
"token_acc": 0.8972116812407194
},
{
"epoch": 0.7812338534669836,
"grad_norm": 3.9334847927093506,
"learning_rate": 6.275038058156379e-08,
"loss": 0.29413180351257323,
"step": 1890,
"token_acc": 0.8877047876570702
},
{
"epoch": 0.7853673659191899,
"grad_norm": 2.048691749572754,
"learning_rate": 6.050441640137582e-08,
"loss": 0.2679485321044922,
"step": 1900,
"token_acc": 0.8930240320427236
},
{
"epoch": 0.789500878371396,
"grad_norm": 2.922384738922119,
"learning_rate": 5.8293836808713805e-08,
"loss": 0.25333313941955565,
"step": 1910,
"token_acc": 0.8993645886999828
},
{
"epoch": 0.7936343908236023,
"grad_norm": 4.081994533538818,
"learning_rate": 5.611905458600574e-08,
"loss": 0.2619889736175537,
"step": 1920,
"token_acc": 0.9003006012024048
},
{
"epoch": 0.7977679032758086,
"grad_norm": 2.331836700439453,
"learning_rate": 5.398047583122206e-08,
"loss": 0.2741227388381958,
"step": 1930,
"token_acc": 0.8928689336414658
},
{
"epoch": 0.8019014157280149,
"grad_norm": 2.6420035362243652,
"learning_rate": 5.187849988204479e-08,
"loss": 0.2765916109085083,
"step": 1940,
"token_acc": 0.8925619834710744
},
{
"epoch": 0.8060349281802212,
"grad_norm": 2.482302665710449,
"learning_rate": 4.9813519241299425e-08,
"loss": 0.29370822906494143,
"step": 1950,
"token_acc": 0.8826374334999194
},
{
"epoch": 0.8101684406324274,
"grad_norm": 3.364745855331421,
"learning_rate": 4.778591950366212e-08,
"loss": 0.2546257734298706,
"step": 1960,
"token_acc": 0.9023733378219155
},
{
"epoch": 0.8143019530846337,
"grad_norm": 3.09533429145813,
"learning_rate": 4.579607928365803e-08,
"loss": 0.26116080284118653,
"step": 1970,
"token_acc": 0.895497311827957
},
{
"epoch": 0.8184354655368399,
"grad_norm": 2.486457586288452,
"learning_rate": 4.384437014496214e-08,
"loss": 0.26124980449676516,
"step": 1980,
"token_acc": 0.9021720828422294
},
{
"epoch": 0.8225689779890462,
"grad_norm": 2.3879621028900146,
"learning_rate": 4.1931156531017e-08,
"loss": 0.2727025032043457,
"step": 1990,
"token_acc": 0.8939494680851063
},
{
"epoch": 0.8267024904412524,
"grad_norm": 2.0487353801727295,
"learning_rate": 4.0056795696980144e-08,
"loss": 0.2716191291809082,
"step": 2000,
"token_acc": 0.8951921477291632
},
{
"epoch": 0.8308360028934587,
"grad_norm": 4.274483680725098,
"learning_rate": 3.822163764301378e-08,
"loss": 0.2700816631317139,
"step": 2010,
"token_acc": 0.8948497854077253
},
{
"epoch": 0.834969515345665,
"grad_norm": 2.1714582443237305,
"learning_rate": 3.6426025048928914e-08,
"loss": 0.2807120084762573,
"step": 2020,
"token_acc": 0.8919180512985966
},
{
"epoch": 0.8391030277978713,
"grad_norm": 3.727752208709717,
"learning_rate": 3.4670293210196794e-08,
"loss": 0.2586544036865234,
"step": 2030,
"token_acc": 0.8995808885163453
},
{
"epoch": 0.8432365402500775,
"grad_norm": 2.8159375190734863,
"learning_rate": 3.295476997533905e-08,
"loss": 0.28347413539886473,
"step": 2040,
"token_acc": 0.8905654383735705
},
{
"epoch": 0.8473700527022837,
"grad_norm": 2.0777461528778076,
"learning_rate": 3.127977568470813e-08,
"loss": 0.25567169189453126,
"step": 2050,
"token_acc": 0.9038004750593824
},
{
"epoch": 0.85150356515449,
"grad_norm": 2.6102845668792725,
"learning_rate": 2.9645623110670286e-08,
"loss": 0.2771745204925537,
"step": 2060,
"token_acc": 0.8919854881266491
},
{
"epoch": 0.8556370776066963,
"grad_norm": 5.132630348205566,
"learning_rate": 2.8052617399201476e-08,
"loss": 0.2776770114898682,
"step": 2070,
"token_acc": 0.8938922547895857
},
{
"epoch": 0.8597705900589026,
"grad_norm": 4.320744037628174,
"learning_rate": 2.6501056012907252e-08,
"loss": 0.2832627773284912,
"step": 2080,
"token_acc": 0.8891581838746162
},
{
"epoch": 0.8639041025111088,
"grad_norm": 3.808979034423828,
"learning_rate": 2.4991228675477288e-08,
"loss": 0.26563119888305664,
"step": 2090,
"token_acc": 0.8978029294274301
},
{
"epoch": 0.8680376149633151,
"grad_norm": 2.5682902336120605,
"learning_rate": 2.3523417317585477e-08,
"loss": 0.25737996101379396,
"step": 2100,
"token_acc": 0.9021499915354664
},
{
"epoch": 0.8721711274155214,
"grad_norm": 2.6730620861053467,
"learning_rate": 2.209789602424461e-08,
"loss": 0.28185770511627195,
"step": 2110,
"token_acc": 0.888924975641442
},
{
"epoch": 0.8763046398677276,
"grad_norm": 2.259251594543457,
"learning_rate": 2.0714930983626534e-08,
"loss": 0.2834534406661987,
"step": 2120,
"token_acc": 0.8912690684842584
},
{
"epoch": 0.8804381523199338,
"grad_norm": 2.275806427001953,
"learning_rate": 1.937478043735677e-08,
"loss": 0.27589869499206543,
"step": 2130,
"token_acc": 0.8949448378066853
},
{
"epoch": 0.8845716647721401,
"grad_norm": 3.5575265884399414,
"learning_rate": 1.8077694632292777e-08,
"loss": 0.2688805103302002,
"step": 2140,
"token_acc": 0.897500844309355
},
{
"epoch": 0.8887051772243464,
"grad_norm": 2.558871030807495,
"learning_rate": 1.6823915773795354e-08,
"loss": 0.26161561012268064,
"step": 2150,
"token_acc": 0.9005506424161521
},
{
"epoch": 0.8928386896765527,
"grad_norm": 2.106569528579712,
"learning_rate": 1.561367798050159e-08,
"loss": 0.2594000816345215,
"step": 2160,
"token_acc": 0.9025995948683322
},
{
"epoch": 0.8969722021287589,
"grad_norm": 2.074631690979004,
"learning_rate": 1.4447207240607579e-08,
"loss": 0.2612841367721558,
"step": 2170,
"token_acc": 0.8996145466733702
},
{
"epoch": 0.9011057145809652,
"grad_norm": 3.111022472381592,
"learning_rate": 1.3324721369669862e-08,
"loss": 0.28264079093933103,
"step": 2180,
"token_acc": 0.89177131526768
},
{
"epoch": 0.9052392270331714,
"grad_norm": 2.248342275619507,
"learning_rate": 1.2246429969932442e-08,
"loss": 0.2776078224182129,
"step": 2190,
"token_acc": 0.8949741192185674
},
{
"epoch": 0.9093727394853777,
"grad_norm": 3.3940088748931885,
"learning_rate": 1.1212534391187689e-08,
"loss": 0.2858419895172119,
"step": 2200,
"token_acc": 0.8873626373626373
},
{
"epoch": 0.913506251937584,
"grad_norm": 2.3772406578063965,
"learning_rate": 1.0223227693178155e-08,
"loss": 0.27225029468536377,
"step": 2210,
"token_acc": 0.8944840152393573
},
{
"epoch": 0.9176397643897902,
"grad_norm": 6.223030090332031,
"learning_rate": 9.278694609546722e-09,
"loss": 0.27013497352600097,
"step": 2220,
"token_acc": 0.8913370055082624
},
{
"epoch": 0.9217732768419965,
"grad_norm": 2.280205488204956,
"learning_rate": 8.379111513340753e-09,
"loss": 0.2775579452514648,
"step": 2230,
"token_acc": 0.8897431702928186
},
{
"epoch": 0.9259067892942028,
"grad_norm": 2.102734088897705,
"learning_rate": 7.524646384078404e-09,
"loss": 0.2551942110061646,
"step": 2240,
"token_acc": 0.902751666381815
},
{
"epoch": 0.9300403017464091,
"grad_norm": 2.731247901916504,
"learning_rate": 6.715458776381383e-09,
"loss": 0.27170391082763673,
"step": 2250,
"token_acc": 0.8929798723613157
},
{
"epoch": 0.9341738141986152,
"grad_norm": 2.7572975158691406,
"learning_rate": 5.951699790181258e-09,
"loss": 0.2469815731048584,
"step": 2260,
"token_acc": 0.9041237113402062
},
{
"epoch": 0.9383073266508215,
"grad_norm": 3.6743345260620117,
"learning_rate": 5.233512042504718e-09,
"loss": 0.267983078956604,
"step": 2270,
"token_acc": 0.8989370676564873
},
{
"epoch": 0.9424408391030278,
"grad_norm": 2.9522664546966553,
"learning_rate": 4.561029640842484e-09,
"loss": 0.28539676666259767,
"step": 2280,
"token_acc": 0.8891777705557361
},
{
"epoch": 0.9465743515552341,
"grad_norm": 2.377556324005127,
"learning_rate": 3.934378158107349e-09,
"loss": 0.2863355875015259,
"step": 2290,
"token_acc": 0.8862681744749596
},
{
"epoch": 0.9507078640074403,
"grad_norm": 2.461761951446533,
"learning_rate": 3.3536746091860635e-09,
"loss": 0.27661709785461425,
"step": 2300,
"token_acc": 0.8950495049504951
},
{
"epoch": 0.9548413764596466,
"grad_norm": 7.1646504402160645,
"learning_rate": 2.8190274290888215e-09,
"loss": 0.2638131856918335,
"step": 2310,
"token_acc": 0.8973592426507224
},
{
"epoch": 0.9589748889118529,
"grad_norm": 4.375582218170166,
"learning_rate": 2.3305364527013984e-09,
"loss": 0.27904419898986815,
"step": 2320,
"token_acc": 0.8944037308461026
},
{
"epoch": 0.9631084013640591,
"grad_norm": 2.156973123550415,
"learning_rate": 1.888292896142868e-09,
"loss": 0.2675792217254639,
"step": 2330,
"token_acc": 0.8935467898993233
},
{
"epoch": 0.9672419138162653,
"grad_norm": 3.002598285675049,
"learning_rate": 1.4923793397326467e-09,
"loss": 0.2722255945205688,
"step": 2340,
"token_acc": 0.8944089194949992
},
{
"epoch": 0.9713754262684716,
"grad_norm": 2.815256118774414,
"learning_rate": 1.1428697125704379e-09,
"loss": 0.2843726634979248,
"step": 2350,
"token_acc": 0.8908282562160382
},
{
"epoch": 0.9755089387206779,
"grad_norm": 2.310183525085449,
"learning_rate": 8.398292787313576e-10,
"loss": 0.2593583583831787,
"step": 2360,
"token_acc": 0.8995342648037259
},
{
"epoch": 0.9796424511728842,
"grad_norm": 2.1561472415924072,
"learning_rate": 5.833146250790177e-10,
"loss": 0.2610877275466919,
"step": 2370,
"token_acc": 0.8955997992303831
},
{
"epoch": 0.9837759636250905,
"grad_norm": 3.440030097961426,
"learning_rate": 3.733736506992269e-10,
"loss": 0.2442765235900879,
"step": 2380,
"token_acc": 0.9032476319350473
},
{
"epoch": 0.9879094760772967,
"grad_norm": 2.758301019668579,
"learning_rate": 2.100455579556737e-10,
"loss": 0.2496877670288086,
"step": 2390,
"token_acc": 0.9015686950525772
},
{
"epoch": 0.9920429885295029,
"grad_norm": 1.9135822057724,
"learning_rate": 9.336084516964327e-11,
"loss": 0.2654644727706909,
"step": 2400,
"token_acc": 0.8951626217599472
},
{
"epoch": 0.9961765009817092,
"grad_norm": 2.8369832038879395,
"learning_rate": 2.3341300925072693e-11,
"loss": 0.2826450109481812,
"step": 2410,
"token_acc": 0.8910778248127645
},
{
"epoch": 1.0,
"grad_norm": 3.27158784866333,
"learning_rate": 0.0,
"loss": 0.27386786937713625,
"step": 2420,
"token_acc": 0.8922451863628335
}
],
"logging_steps": 10,
"max_steps": 2420,
"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": 2.928351858378408e+18,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}