uploaded / best_eval_4 /train_state.json
reneeice's picture
Upload folder using huggingface_hub
b7a489e verified
Raw
History Blame Contribute Delete
9.88 kB
{
"step": 401,
"losses": [
2.1343601644039154,
2.0724628567695618,
1.8829140663146973,
1.5485129058361053,
2.0023295283317566,
1.9346746504306793,
2.039268136024475,
2.092041403055191,
1.945346713066101,
1.7569688260555267,
1.8735843896865845,
1.9090708792209625,
1.6151848435401917,
1.7625981569290161,
2.068992555141449,
1.7361219227313995,
2.0706606209278107,
1.9792323410511017,
1.9772961139678955,
1.8365838527679443,
1.6523617506027222,
1.9725398421287537,
1.9878025949001312,
1.7896310240030289,
2.046401262283325,
1.8547812402248383,
2.1061335802078247,
1.9161535799503326,
2.1761361956596375,
2.1191433668136597,
1.5494413673877716,
2.041987180709839,
1.799335092306137,
1.8639665395021439,
1.8064149618148804,
1.9582627415657043,
2.3036161065101624,
2.061281830072403,
1.7895398437976837,
2.1972492933273315,
1.8500069081783295,
2.01457941532135,
2.311652898788452,
2.0377022325992584,
2.1203809678554535,
1.9050646126270294,
2.1400992274284363,
2.0162739157676697,
2.0760363936424255,
1.9967831671237946,
2.199576199054718,
1.903381735086441,
1.9348146319389343,
1.565977357327938,
2.216809928417206,
2.279655873775482,
2.146350234746933,
1.8592103123664856,
2.0420128405094147,
2.202657461166382,
2.0355475544929504,
1.9122264087200165,
1.7364317774772644,
1.8870865404605865,
1.9341412782669067,
2.0909554958343506,
2.204548269510269,
2.0142270624637604,
1.6295936107635498,
1.788179874420166,
2.022219479084015,
2.15500944852829,
1.7791253924369812,
1.9474472403526306,
1.8155687749385834,
1.8396632373332977,
2.2242822349071503,
2.0792854130268097,
1.6167261078953743,
1.8821334838867188,
1.8196536004543304,
2.087564706802368,
1.9993097186088562,
2.1168552935123444,
2.0352262556552887,
2.2895873188972473,
2.1865938305854797,
1.8330432772636414,
2.10781130194664,
1.875508338212967,
1.6532813906669617,
2.096515417098999,
1.939268171787262,
1.852915108203888,
2.4114430844783783,
2.036993145942688,
2.056860029697418,
1.9688847362995148,
1.729723185300827,
1.9320075511932373,
2.1951059699058533,
1.4285495057702065,
1.8822387158870697,
1.893274337053299,
1.7797814011573792,
1.9108840823173523,
1.8049851953983307,
1.8350046575069427,
1.991377979516983,
2.1966324746608734,
2.154123991727829,
1.9668664038181305,
1.9106942415237427,
2.020862102508545,
1.996761977672577,
1.6555583775043488,
1.9698050320148468,
1.8792638778686523,
1.8288727700710297,
2.0352847278118134,
1.8880273401737213,
2.014271706342697,
1.7977636754512787,
2.267457276582718,
1.9233829975128174,
1.7882795333862305,
2.111464351415634,
2.046808570623398,
2.127010613679886,
2.096220314502716,
1.8619424104690552,
1.9859545826911926,
2.2351702451705933,
2.0095019936561584,
1.8124884068965912,
2.0566355288028717,
1.8829358518123627,
2.228476822376251,
2.169349581003189,
1.9356842339038849,
2.197207033634186,
1.9552882313728333,
2.005706787109375,
2.0790003836154938,
2.1652873158454895,
2.2506291270256042,
2.1488280296325684,
1.666898250579834,
1.9886826276779175,
1.8917436003684998,
1.917124480009079,
1.932131290435791,
2.257670372724533,
1.8668333888053894,
2.0241182148456573,
2.155973106622696,
1.832424134016037,
2.0300442576408386,
1.869031399488449,
1.9497641324996948,
2.218937933444977,
1.7895354628562927,
1.9639611840248108,
2.0145443379879,
1.7729362547397614,
1.8493213057518005,
2.1741292774677277,
2.0540651679039,
1.9173554182052612,
1.9990263879299164,
2.1634564995765686,
1.9584728181362152,
1.4916782528162003,
2.0375813841819763,
1.9370076954364777,
1.8460400998592377,
2.141355186700821,
2.048035830259323,
2.0943254232406616,
1.3870432674884796,
1.9739985466003418,
2.079814076423645,
1.960821509361267,
1.724249616265297,
1.6594904959201813,
2.0725796818733215,
2.0049106776714325,
1.7692183256149292,
2.252512514591217,
1.8543781340122223,
2.7010736763477325,
2.0170003473758698,
1.9308967888355255,
1.9603045284748077,
2.166411817073822,
2.075620085000992,
2.125515103340149,
2.0684770047664642,
1.9685010015964508,
2.057872623205185,
2.0869134664535522,
1.8150206804275513,
1.6579519212245941,
1.9104999899864197,
1.9154045581817627,
1.9604035019874573,
2.056108742952347,
2.097057729959488,
2.1572895646095276,
1.9608762860298157,
1.4785484373569489,
1.9157750606536865,
1.752137005329132,
2.0132217705249786,
2.244666576385498,
2.1824870109558105,
1.9102648496627808,
1.8477981090545654,
1.8056239187717438,
2.080883741378784,
1.8958399593830109,
1.9592851400375366,
2.1995291113853455,
2.0795817971229553,
1.8578779101371765,
1.9151493608951569,
1.9029805362224579,
1.8901777863502502,
1.7989002466201782,
1.775819331407547,
1.4867726117372513,
1.8664362132549286,
2.1231315433979034,
1.9697103202342987,
1.849808692932129,
1.8471928536891937,
1.992421269416809,
1.873524695634842,
1.9098478853702545,
1.7680681347846985,
1.870894968509674,
1.9761443138122559,
1.8797105252742767,
1.5107271373271942,
2.033596396446228,
2.0826768279075623,
1.6832334995269775,
1.675403743982315,
2.1581598222255707,
1.8126455545425415,
1.9812375009059906,
2.0296229124069214,
2.163834869861603,
1.9061608910560608,
2.349096179008484,
1.7541834115982056,
2.172113597393036,
1.9886090159416199,
1.8932957351207733,
2.2767199277877808,
1.886280745267868,
1.890844613313675,
2.071047067642212,
1.9397878348827362,
2.0238566398620605,
2.1563323736190796,
1.76157745718956,
2.215013861656189,
1.6411037147045135,
1.7615033239126205,
1.8542159497737885,
1.795795127749443,
2.225602298974991,
1.9437134563922882,
1.986325353384018,
2.0043095648288727,
1.9501238763332367,
1.7597680985927582,
2.275161862373352,
2.080092966556549,
1.4906146079301834,
1.8959785997867584,
1.974496841430664,
2.0430229008197784,
2.2613201439380646,
1.914484053850174,
2.0284277200698853,
1.901948243379593,
1.7956669330596924,
1.9412107467651367,
1.9384097754955292,
2.1359619796276093,
2.008501648902893,
2.0292378962039948,
1.9525749683380127,
1.9676333367824554,
1.596329241991043,
2.1911933422088623,
2.0755282640457153,
2.125846654176712,
2.273774266242981,
1.897540658712387,
2.124108701944351,
1.9193576574325562,
2.3038374185562134,
1.9048315286636353,
1.6724864840507507,
2.047564446926117,
1.934387519955635,
2.1303321719169617,
2.233566999435425,
1.9496371150016785,
1.8122275173664093,
1.814949780702591,
1.7847777605056763,
1.8882033824920654,
2.0106955468654633,
1.916115641593933,
2.0246517658233643,
2.1212706565856934,
2.4152793884277344,
1.8375939726829529,
1.850978285074234,
2.053068608045578,
1.9256162643432617,
1.7829122245311737,
1.9772546291351318,
2.022214114665985,
2.019051045179367,
1.8135776221752167,
1.9114308059215546,
2.14043927192688,
2.239055573940277,
1.6795585751533508,
2.276009351015091,
2.087787240743637,
1.876219779253006,
1.6404992938041687,
1.9292358756065369,
2.205354690551758,
1.9994370341300964,
2.0519250631332397,
2.075588583946228,
2.2462772130966187,
2.0940224528312683,
1.8732309937477112,
1.7715218365192413,
1.5410477817058563,
1.7525133341550827,
2.080644190311432,
1.6389579102396965,
1.8843775987625122,
2.1427654027938843,
2.111940711736679,
1.9778867065906525,
1.9787790179252625,
1.7654073536396027,
1.8432354927062988,
2.234671473503113,
2.278365671634674,
2.0803157687187195,
2.053450584411621,
2.063992917537689,
1.7682743966579437,
2.0207310020923615,
1.9868452847003937,
2.0650334656238556,
1.9444032907485962,
1.7403165996074677,
2.008295029401779,
1.987241953611374,
2.078886032104492,
2.2060952186584473,
1.9917446970939636,
2.253327250480652,
1.9419922828674316,
1.8198893070220947,
1.9853444695472717,
1.9338945746421814,
1.9119682610034943,
2.07899010181427,
2.1189143657684326,
1.8148274719715118,
2.2248027622699738,
2.0726819932460785,
2.0100126564502716,
2.15337797999382,
2.0229094326496124,
1.7959956228733063,
1.9197712242603302,
2.0216313004493713,
1.6409029960632324,
1.9614338874816895,
1.9463849663734436,
2.2429850101470947,
2.031428813934326,
1.5765005052089691,
1.9564637243747711,
2.084881007671356,
2.040048509836197,
2.1490495800971985
],
"dataset_offset": 0,
"config": {
"king_dir": "/root/king_165_hf",
"optimizer": "muon_adamw",
"lr": 5e-05,
"muon_lr": 0.0001,
"lora_r": 128,
"steps": 5000,
"grad_accum": 4,
"seq_len": 2048,
"unfreeze_head": false,
"source_dirs": [
"/root/datasets/finewebedu"
],
"source_weights": [
1.0
]
}
}