Files
qwen3-8b-deita-sft-teacher/trainer_state.json

1661 lines
40 KiB
JSON
Raw Normal View History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 1156,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.004325259515570935,
"grad_norm": 4.35123215481323,
"learning_rate": 8.620689655172415e-07,
"loss": 1.1904,
"step": 5
},
{
"epoch": 0.00865051903114187,
"grad_norm": 4.403748317626318,
"learning_rate": 1.9396551724137932e-06,
"loss": 1.1054,
"step": 10
},
{
"epoch": 0.012975778546712802,
"grad_norm": 1.4606257815501247,
"learning_rate": 3.017241379310345e-06,
"loss": 1.067,
"step": 15
},
{
"epoch": 0.01730103806228374,
"grad_norm": 1.2597796533288468,
"learning_rate": 4.094827586206896e-06,
"loss": 1.0359,
"step": 20
},
{
"epoch": 0.02162629757785467,
"grad_norm": 1.1637436559096648,
"learning_rate": 5.172413793103448e-06,
"loss": 0.982,
"step": 25
},
{
"epoch": 0.025951557093425604,
"grad_norm": 1.129170232112561,
"learning_rate": 6.25e-06,
"loss": 0.99,
"step": 30
},
{
"epoch": 0.03027681660899654,
"grad_norm": 1.4809412766751884,
"learning_rate": 7.3275862068965514e-06,
"loss": 1.0141,
"step": 35
},
{
"epoch": 0.03460207612456748,
"grad_norm": 1.6602790320663552,
"learning_rate": 8.405172413793103e-06,
"loss": 1.0115,
"step": 40
},
{
"epoch": 0.03892733564013841,
"grad_norm": 1.693504275805329,
"learning_rate": 9.482758620689655e-06,
"loss": 0.9672,
"step": 45
},
{
"epoch": 0.04325259515570934,
"grad_norm": 1.867767030271666,
"learning_rate": 1.0560344827586207e-05,
"loss": 0.9495,
"step": 50
},
{
"epoch": 0.04757785467128028,
"grad_norm": 1.6855487950579542,
"learning_rate": 1.163793103448276e-05,
"loss": 0.9299,
"step": 55
},
{
"epoch": 0.05190311418685121,
"grad_norm": 1.532082300541662,
"learning_rate": 1.271551724137931e-05,
"loss": 0.9432,
"step": 60
},
{
"epoch": 0.056228373702422146,
"grad_norm": 1.5444864937497627,
"learning_rate": 1.3793103448275863e-05,
"loss": 0.9188,
"step": 65
},
{
"epoch": 0.06055363321799308,
"grad_norm": 1.1522756727876342,
"learning_rate": 1.4870689655172415e-05,
"loss": 0.8527,
"step": 70
},
{
"epoch": 0.06487889273356401,
"grad_norm": 1.0202505047323622,
"learning_rate": 1.5948275862068967e-05,
"loss": 0.8731,
"step": 75
},
{
"epoch": 0.06920415224913495,
"grad_norm": 0.9717090561320276,
"learning_rate": 1.7025862068965516e-05,
"loss": 0.9088,
"step": 80
},
{
"epoch": 0.07352941176470588,
"grad_norm": 0.9554882216281484,
"learning_rate": 1.810344827586207e-05,
"loss": 0.904,
"step": 85
},
{
"epoch": 0.07785467128027682,
"grad_norm": 1.1405179407279404,
"learning_rate": 1.9181034482758623e-05,
"loss": 0.8859,
"step": 90
},
{
"epoch": 0.08217993079584775,
"grad_norm": 0.9183014201649258,
"learning_rate": 2.0258620689655172e-05,
"loss": 0.8127,
"step": 95
},
{
"epoch": 0.08650519031141868,
"grad_norm": 0.8498041251214743,
"learning_rate": 2.1336206896551724e-05,
"loss": 0.8867,
"step": 100
},
{
"epoch": 0.09083044982698962,
"grad_norm": 0.8507455158196404,
"learning_rate": 2.2413793103448276e-05,
"loss": 0.8432,
"step": 105
},
{
"epoch": 0.09515570934256055,
"grad_norm": 0.8198012289472929,
"learning_rate": 2.349137931034483e-05,
"loss": 0.891,
"step": 110
},
{
"epoch": 0.09948096885813149,
"grad_norm": 0.922337984622785,
"learning_rate": 2.456896551724138e-05,
"loss": 0.9028,
"step": 115
},
{
"epoch": 0.10380622837370242,
"grad_norm": 0.7209214584903633,
"learning_rate": 2.4999486722033878e-05,
"loss": 0.8598,
"step": 120
},
{
"epoch": 0.10813148788927336,
"grad_norm": 0.7542513167702001,
"learning_rate": 2.4996350176002415e-05,
"loss": 0.8971,
"step": 125
},
{
"epoch": 0.11245674740484429,
"grad_norm": 0.7078673089301097,
"learning_rate": 2.4990362953009037e-05,
"loss": 0.8883,
"step": 130
},
{
"epoch": 0.11678200692041522,
"grad_norm": 1.8867646071760962,
"learning_rate": 2.4981526418863643e-05,
"loss": 0.8598,
"step": 135
},
{
"epoch": 0.12110726643598616,
"grad_norm": 0.6374023611261029,
"learning_rate": 2.4969842589363157e-05,
"loss": 0.9004,
"step": 140
},
{
"epoch": 0.1254325259515571,
"grad_norm": 0.7009722074552326,
"learning_rate": 2.4955314129831702e-05,
"loss": 0.9058,
"step": 145
},
{
"epoch": 0.12975778546712802,
"grad_norm": 0.6122532527973501,
"learning_rate": 2.4937944354512583e-05,
"loss": 0.8695,
"step": 150
},
{
"epoch": 0.13408304498269896,
"grad_norm": 0.6755550221895136,
"learning_rate": 2.4917737225812236e-05,
"loss": 0.912,
"step": 155
},
{
"epoch": 0.1384083044982699,
"grad_norm": 0.5958534813871882,
"learning_rate": 2.4894697353396314e-05,
"loss": 0.92,
"step": 160
},
{
"epoch": 0.14273356401384082,
"grad_norm": 0.5754584150821744,
"learning_rate": 2.486882999313814e-05,
"loss": 0.8976,
"step": 165
},
{
"epoch": 0.14705882352941177,
"grad_norm": 0.6296808029048555,
"learning_rate": 2.4840141045919725e-05,
"loss": 0.8834,
"step": 170
},
{
"epoch": 0.1513840830449827,
"grad_norm": 0.5709916645310777,
"learning_rate": 2.4808637056285654e-05,
"loss": 0.8744,
"step": 175
},
{
"epoch": 0.15570934256055363,
"grad_norm": 0.6195204752553578,
"learning_rate": 2.4774325210950146e-05,
"loss": 0.8285,
"step": 180
},
{
"epoch": 0.16003460207612458,
"grad_norm": 0.5981886603079031,
"learning_rate": 2.473721333715761e-05,
"loss": 0.8795,
"step": 185
},
{
"epoch": 0.1643598615916955,
"grad_norm": 0.5873981893364288,
"learning_rate": 2.4697309900897076e-05,
"loss": 0.8477,
"step": 190
},
{
"epoch": 0.16868512110726644,
"grad_norm": 0.6154967555267769,
"learning_rate": 2.4654624004970957e-05,
"loss": 0.8901,
"step": 195
},
{
"epoch": 0.17301038062283736,
"grad_norm": 0.6545395385572848,
"learning_rate": 2.460916538691849e-05,
"loss": 0.8874,
"step": 200
},
{
"epoch": 0.1773356401384083,
"grad_norm": 0.5701589240432223,
"learning_rate": 2.4560944416794397e-05,
"loss": 0.9025,
"step": 205
},
{
"epoch": 0.18166089965397925,
"grad_norm": 0.7287328700260651,
"learning_rate": 2.450997209480327e-05,
"loss": 0.9533,
"step": 210
},
{
"epoch": 0.18598615916955016,
"grad_norm": 0.6318744664922232,
"learning_rate": 2.4456260048790198e-05,
"loss": 0.9212,
"step": 215
},
{
"epoch": 0.1903114186851211,
"grad_norm": 0.593526964948498,
"learning_rate": 2.4399820531588202e-05,
"loss": 0.8467,
"step": 220
},
{
"epoch": 0.19463667820069205,
"grad_norm": 0.5527811855267959,
"learning_rate": 2.4340666418223134e-05,
"loss": 0.8611,
"step": 225
},
{
"epoch": 0.19896193771626297,
"grad_norm": 0.5752044594613943,
"learning_rate": 2.427881120297658e-05,
"loss": 0.8744,
"step": 230
},
{
"epoch": 0.20328719723183392,
"grad_norm": 0.5806611958413757,
"learning_rate": 2.4214268996307565e-05,
"loss": 0.8259,
"step": 235
},
{
"epoch": 0.20761245674740483,
"grad_norm": 0.5545392774453544,
"learning_rate": 2.414705452163365e-05,
"loss": 0.8293,
"step": 240
},
{
"epoch": 0.21193771626297578,
"grad_norm": 0.5330970193480532,
"learning_rate": 2.4077183111972207e-05,
"loss": 0.8474,
"step": 245
},
{
"epoch": 0.21626297577854672,
"grad_norm": 0.5677979399377041,
"learning_rate": 2.400467070644267e-05,
"loss": 0.8796,
"step": 250
},
{
"epoch": 0.22058823529411764,
"grad_norm": 0.6016601269861623,
"learning_rate": 2.3929533846630463e-05,
"loss": 0.9105,
"step": 255
},
{
"epoch": 0.22491349480968859,
"grad_norm": 0.5479556885398574,
"learning_rate": 2.3851789672813517e-05,
"loss": 0.8645,
"step": 260
},
{
"epoch": 0.2292387543252595,
"grad_norm": 0.6095223065882377,
"learning_rate": 2.3771455920052244e-05,
"loss": 0.9416,
"step": 265
},
{
"epoch": 0.23356401384083045,
"grad_norm": 0.5290930780655559,
"learning_rate": 2.368855091414378e-05,
"loss": 0.8738,
"step": 270
},
{
"epoch": 0.2378892733564014,
"grad_norm": 0.5697800907847043,
"learning_rate": 2.3603093567441488e-05,
"loss": 0.8568,
"step": 275
},
{
"epoch": 0.2422145328719723,
"grad_norm": 0.6447957140556879,
"learning_rate": 2.3515103374540682e-05,
"loss": 0.8701,
"step": 280
},
{
"epoch": 0.24653979238754326,
"grad_norm": 0.5860229140661292,
"learning_rate": 2.3424600407831476e-05,
"loss": 0.8618,
"step": 285
},
{
"epoch": 0.2508650519031142,
"grad_norm": 0.7669002724780247,
"learning_rate": 2.333160531291987e-05,
"loss": 0.8899,
"step": 290
},
{
"epoch": 0.25519031141868515,
"grad_norm": 0.5620842120948024,
"learning_rate": 2.323613930391806e-05,
"loss": 0.8848,
"step": 295
},
{
"epoch": 0.25951557093425603,
"grad_norm": 0.54716558076972,
"learning_rate": 2.3138224158605043e-05,
"loss": 0.8277,
"step": 300
},
{
"epoch": 0.263840830449827,
"grad_norm": 1.0465118120487207,
"learning_rate": 2.3037882213458685e-05,
"loss": 0.8863,
"step": 305
},
{
"epoch": 0.2681660899653979,
"grad_norm": 0.5106226720494563,
"learning_rate": 2.2935136358560263e-05,
"loss": 0.8399,
"step": 310
},
{
"epoch": 0.27249134948096887,
"grad_norm": 0.5239424882259534,
"learning_rate": 2.283001003237281e-05,
"loss": 0.8574,
"step": 315
},
{
"epoch": 0.2768166089965398,
"grad_norm": 0.5937613416602285,
"learning_rate": 2.2722527216394267e-05,
"loss": 0.8668,
"step": 320
},
{
"epoch": 0.2811418685121107,
"grad_norm": 0.5688236948434819,
"learning_rate": 2.2612712429686845e-05,
"loss": 0.8724,
"step": 325
},
{
"epoch": 0.28546712802768165,
"grad_norm": 0.6349052588438815,
"learning_rate": 2.2500590723283696e-05,
"loss": 0.8661,
"step": 330
},
{
"epoch": 0.2897923875432526,
"grad_norm": 0.5891502805591111,
"learning_rate": 2.2386187674474257e-05,
"loss": 0.8449,
"step": 335
},
{
"epoch": 0.29411764705882354,
"grad_norm": 0.5294750103265548,
"learning_rate": 2.2269529380969545e-05,
"loss": 0.8445,
"step": 340
},
{
"epoch": 0.2984429065743945,
"grad_norm": 0.5339518191163046,
"learning_rate": 2.215064245494871e-05,
"loss": 0.8462,
"step": 345
},
{
"epoch": 0.3027681660899654,
"grad_norm": 0.6053170526605012,
"learning_rate": 2.2029554016988273e-05,
"loss": 0.8995,
"step": 350
},
{
"epoch": 0.3070934256055363,
"grad_norm": 0.5274943916525573,
"learning_rate": 2.1906291689875334e-05,
"loss": 0.8975,
"step": 355
},
{
"epoch": 0.31141868512110726,
"grad_norm": 0.5490639498780314,
"learning_rate": 2.178088359230625e-05,
"loss": 0.8723,
"step": 360
},
{
"epoch": 0.3157439446366782,
"grad_norm": 0.5546347111911519,
"learning_rate": 2.1653358332472172e-05,
"loss": 0.8442,
"step": 365
},
{
"epoch": 0.32006920415224915,
"grad_norm": 0.5231958215763314,
"learning_rate": 2.152374500153293e-05,
"loss": 0.8414,
"step": 370
},
{
"epoch": 0.32439446366782004,
"grad_norm": 0.5373746721490149,
"learning_rate": 2.1392073166980742e-05,
"loss": 0.8547,
"step": 375
},
{
"epoch": 0.328719723183391,
"grad_norm": 0.5857136188797172,
"learning_rate": 2.1258372865895222e-05,
"loss": 0.8204,
"step": 380
},
{
"epoch": 0.33304498269896193,
"grad_norm": 0.5707682901836567,
"learning_rate": 2.1122674598091345e-05,
"loss": 0.8859,
"step": 385
},
{
"epoch": 0.3373702422145329,
"grad_norm": 0.5586518838132467,
"learning_rate": 2.0985009319161775e-05,
"loss": 0.8525,
"step": 390
},
{
"epoch": 0.3416955017301038,
"grad_norm": 0.568756673783547,
"learning_rate": 2.0845408433415243e-05,
"loss": 0.8515,
"step": 395
},
{
"epoch": 0.3460207612456747,
"grad_norm": 0.5766380643843076,
"learning_rate": 2.0703903786712616e-05,
"loss": 0.8765,
"step": 400
},
{
"epoch": 0.35034602076124566,
"grad_norm": 0.533905355892338,
"learning_rate": 2.0560527659202132e-05,
"loss": 0.8304,
"step": 405
},
{
"epoch": 0.3546712802768166,
"grad_norm": 0.5457677323719116,
"learning_rate": 2.04153127579557e-05,
"loss": 0.9004,
"step": 410
},
{
"epoch": 0.35899653979238755,
"grad_norm": 0.5285914104827892,
"learning_rate": 2.026829220950768e-05,
"loss": 0.8463,
"step": 415
},
{
"epoch": 0.3633217993079585,
"grad_norm": 0.5102402946377723,
"learning_rate": 2.0119499552298045e-05,
"loss": 0.8065,
"step": 420
},
{
"epoch": 0.36764705882352944,
"grad_norm": 0.5431748377731394,
"learning_rate": 1.9968968729021594e-05,
"loss": 0.8356,
"step": 425
},
{
"epoch": 0.3719723183391003,
"grad_norm": 0.5045996820024975,
"learning_rate": 1.9816734078884886e-05,
"loss": 0.8606,
"step": 430
},
{
"epoch": 0.3762975778546713,
"grad_norm": 0.607916853539143,
"learning_rate": 1.9662830329772765e-05,
"loss": 0.844,
"step": 435
},
{
"epoch": 0.3806228373702422,
"grad_norm": 0.4982162623452115,
"learning_rate": 1.950729259032623e-05,
"loss": 0.841,
"step": 440
},
{
"epoch": 0.38494809688581316,
"grad_norm": 0.6951767690853428,
"learning_rate": 1.935015634193338e-05,
"loss": 0.8458,
"step": 445
},
{
"epoch": 0.3892733564013841,
"grad_norm": 0.5280798256175327,
"learning_rate": 1.9191457430635406e-05,
"loss": 0.8761,
"step": 450
},
{
"epoch": 0.393598615916955,
"grad_norm": 0.5469434558515678,
"learning_rate": 1.9031232058949362e-05,
"loss": 0.8428,
"step": 455
},
{
"epoch": 0.39792387543252594,
"grad_norm": 0.5566658489241726,
"learning_rate": 1.8869516777609615e-05,
"loss": 0.8721,
"step": 460
},
{
"epoch": 0.4022491349480969,
"grad_norm": 0.5147959979260325,
"learning_rate": 1.870634847722985e-05,
"loss": 0.8098,
"step": 465
},
{
"epoch": 0.40657439446366783,
"grad_norm": 0.5026942020350078,
"learning_rate": 1.8541764379887576e-05,
"loss": 0.8642,
"step": 470
},
{
"epoch": 0.4108996539792388,
"grad_norm": 0.5276811854210969,
"learning_rate": 1.837580203063299e-05,
"loss": 0.9132,
"step": 475
},
{
"epoch": 0.41522491349480967,
"grad_norm": 0.5098393290890468,
"learning_rate": 1.8208499288924162e-05,
"loss": 0.8359,
"step": 480
},
{
"epoch": 0.4195501730103806,
"grad_norm": 0.49770354206186807,
"learning_rate": 1.8039894319990518e-05,
"loss": 0.8286,
"step": 485
},
{
"epoch": 0.42387543252595156,
"grad_norm": 0.5068062704589165,
"learning_rate": 1.7870025586126557e-05,
"loss": 0.8409,
"step": 490
},
{
"epoch": 0.4282006920415225,
"grad_norm": 0.5392259286489979,
"learning_rate": 1.7698931837917794e-05,
"loss": 0.803,
"step": 495
},
{
"epoch": 0.43252595155709345,
"grad_norm": 0.54905695464169,
"learning_rate": 1.7526652105400963e-05,
"loss": 0.8716,
"step": 500
},
{
"epoch": 0.43685121107266434,
"grad_norm": 0.5264824920979494,
"learning_rate": 1.7353225689160436e-05,
"loss": 0.8454,
"step": 505
},
{
"epoch": 0.4411764705882353,
"grad_norm": 0.5127421501556995,
"learning_rate": 1.7178692151362965e-05,
"loss": 0.8543,
"step": 510
},
{
"epoch": 0.4455017301038062,
"grad_norm": 0.5053456726100279,
"learning_rate": 1.7003091306732706e-05,
"loss": 0.8574,
"step": 515
},
{
"epoch": 0.44982698961937717,
"grad_norm": 0.5389098963163003,
"learning_rate": 1.682646321346866e-05,
"loss": 0.8378,
"step": 520
},
{
"epoch": 0.4541522491349481,
"grad_norm": 0.47163595528466123,
"learning_rate": 1.6648848164106606e-05,
"loss": 0.8166,
"step": 525
},
{
"epoch": 0.458477508650519,
"grad_norm": 0.501177183638914,
"learning_rate": 1.647028667632747e-05,
"loss": 0.839,
"step": 530
},
{
"epoch": 0.46280276816608995,
"grad_norm": 0.516911388563668,
"learning_rate": 1.629081948371444e-05,
"loss": 0.8318,
"step": 535
},
{
"epoch": 0.4671280276816609,
"grad_norm": 0.5324144463482655,
"learning_rate": 1.6110487526460794e-05,
"loss": 0.8474,
"step": 540
},
{
"epoch": 0.47145328719723184,
"grad_norm": 0.5372490848083202,
"learning_rate": 1.5929331942030574e-05,
"loss": 0.8681,
"step": 545
},
{
"epoch": 0.4757785467128028,
"grad_norm": 0.5460547567563712,
"learning_rate": 1.5747394055774285e-05,
"loss": 0.8781,
"step": 550
},
{
"epoch": 0.4801038062283737,
"grad_norm": 0.5384550374861695,
"learning_rate": 1.556471537150172e-05,
"loss": 0.8155,
"step": 555
},
{
"epoch": 0.4844290657439446,
"grad_norm": 0.49799019916537807,
"learning_rate": 1.538133756201409e-05,
"loss": 0.828,
"step": 560
},
{
"epoch": 0.48875432525951557,
"grad_norm": 0.511737356494711,
"learning_rate": 1.5197302459597578e-05,
"loss": 0.8495,
"step": 565
},
{
"epoch": 0.4930795847750865,
"grad_norm": 0.610006369399749,
"learning_rate": 1.5012652046480535e-05,
"loss": 0.8787,
"step": 570
},
{
"epoch": 0.49740484429065746,
"grad_norm": 0.4828688382612867,
"learning_rate": 1.4827428445256462e-05,
"loss": 0.8353,
"step": 575
},
{
"epoch": 0.5017301038062284,
"grad_norm": 0.49851171619747253,
"learning_rate": 1.4641673909274961e-05,
"loss": 0.8439,
"step": 580
},
{
"epoch": 0.5060553633217993,
"grad_norm": 0.5188684614142125,
"learning_rate": 1.4455430813002888e-05,
"loss": 0.8224,
"step": 585
},
{
"epoch": 0.5103806228373703,
"grad_norm": 0.5235163481718514,
"learning_rate": 1.426874164235784e-05,
"loss": 0.8184,
"step": 590
},
{
"epoch": 0.5147058823529411,
"grad_norm": 0.5125427555965906,
"learning_rate": 1.4081648985016243e-05,
"loss": 0.8603,
"step": 595
},
{
"epoch": 0.5190311418685121,
"grad_norm": 0.5095437020058842,
"learning_rate": 1.3894195520698213e-05,
"loss": 0.8005,
"step": 600
},
{
"epoch": 0.523356401384083,
"grad_norm": 0.5105112205987176,
"learning_rate": 1.370642401143144e-05,
"loss": 0.8046,
"step": 605
},
{
"epoch": 0.527681660899654,
"grad_norm": 0.5323826305507945,
"learning_rate": 1.3518377291796262e-05,
"loss": 0.7734,
"step": 610
},
{
"epoch": 0.5320069204152249,
"grad_norm": 0.5062582757468418,
"learning_rate": 1.3330098259154208e-05,
"loss": 0.8444,
"step": 615
},
{
"epoch": 0.5363321799307958,
"grad_norm": 0.5052320608565245,
"learning_rate": 1.3141629863862254e-05,
"loss": 0.866,
"step": 620
},
{
"epoch": 0.5406574394463668,
"grad_norm": 0.5158556498655326,
"learning_rate": 1.2953015099474904e-05,
"loss": 0.8262,
"step": 625
},
{
"epoch": 0.5449826989619377,
"grad_norm": 0.5324471489741694,
"learning_rate": 1.2764296992936503e-05,
"loss": 0.8192,
"step": 630
},
{
"epoch": 0.5493079584775087,
"grad_norm": 0.788357636352125,
"learning_rate": 1.2575518594765886e-05,
"loss": 0.8448,
"step": 635
},
{
"epoch": 0.5536332179930796,
"grad_norm": 0.5408366880159132,
"learning_rate": 1.238672296923566e-05,
"loss": 0.9008,
"step": 640
},
{
"epoch": 0.5579584775086506,
"grad_norm": 0.5068320183557856,
"learning_rate": 1.219795318454835e-05,
"loss": 0.8568,
"step": 645
},
{
"epoch": 0.5622837370242214,
"grad_norm": 0.5089724974628004,
"learning_rate": 1.2009252303011647e-05,
"loss": 0.8267,
"step": 650
},
{
"epoch": 0.5666089965397924,
"grad_norm": 0.5071763485055888,
"learning_rate": 1.1820663371214999e-05,
"loss": 0.8411,
"step": 655
},
{
"epoch": 0.5709342560553633,
"grad_norm": 0.5187138851073148,
"learning_rate": 1.1632229410209811e-05,
"loss": 0.8017,
"step": 660
},
{
"epoch": 0.5752595155709342,
"grad_norm": 0.5141145359916918,
"learning_rate": 1.1443993405695414e-05,
"loss": 0.8116,
"step": 665
},
{
"epoch": 0.5795847750865052,
"grad_norm": 0.5567070506875661,
"learning_rate": 1.1255998298213158e-05,
"loss": 0.8794,
"step": 670
},
{
"epoch": 0.5839100346020761,
"grad_norm": 0.49741438511126207,
"learning_rate": 1.106828697335078e-05,
"loss": 0.8131,
"step": 675
},
{
"epoch": 0.5882352941176471,
"grad_norm": 0.5066122940429205,
"learning_rate": 1.088090225195931e-05,
"loss": 0.8317,
"step": 680
},
{
"epoch": 0.592560553633218,
"grad_norm": 0.5363235218723706,
"learning_rate": 1.0693886880384737e-05,
"loss": 0.7839,
"step": 685
},
{
"epoch": 0.596885813148789,
"grad_norm": 0.53560471840449,
"learning_rate": 1.0507283520716691e-05,
"loss": 0.7977,
"step": 690
},
{
"epoch": 0.6012110726643599,
"grad_norm": 0.4861268370712016,
"learning_rate": 1.032113474105634e-05,
"loss": 0.7808,
"step": 695
},
{
"epoch": 0.6055363321799307,
"grad_norm": 0.5301669979705685,
"learning_rate": 1.0135483005805701e-05,
"loss": 0.8274,
"step": 700
},
{
"epoch": 0.6098615916955017,
"grad_norm": 0.530439935809521,
"learning_rate": 9.950370665980644e-06,
"loss": 0.8172,
"step": 705
},
{
"epoch": 0.6141868512110726,
"grad_norm": 0.5385001374970331,
"learning_rate": 9.765839949549738e-06,
"loss": 0.8677,
"step": 710
},
{
"epoch": 0.6185121107266436,
"grad_norm": 0.4993627239475041,
"learning_rate": 9.581932951801187e-06,
"loss": 0.8514,
"step": 715
},
{
"epoch": 0.6228373702422145,
"grad_norm": 0.5284418571933815,
"learning_rate": 9.398691625739997e-06,
"loss": 0.8334,
"step": 720
},
{
"epoch": 0.6271626297577855,
"grad_norm": 0.5392676440821361,
"learning_rate": 9.216157772517664e-06,
"loss": 0.8377,
"step": 725
},
{
"epoch": 0.6314878892733564,
"grad_norm": 0.5136126084601406,
"learning_rate": 9.034373031896417e-06,
"loss": 0.8313,
"step": 730
},
{
"epoch": 0.6358131487889274,
"grad_norm": 0.5324014111860309,
"learning_rate": 8.853378872750352e-06,
"loss": 0.8779,
"step": 735
},
{
"epoch": 0.6401384083044983,
"grad_norm": 0.749263003013231,
"learning_rate": 8.673216583605517e-06,
"loss": 0.8637,
"step": 740
},
{
"epoch": 0.6444636678200693,
"grad_norm": 0.5526744196093801,
"learning_rate": 8.493927263221132e-06,
"loss": 0.8736,
"step": 745
},
{
"epoch": 0.6487889273356401,
"grad_norm": 0.5063072263291449,
"learning_rate": 8.315551811214128e-06,
"loss": 0.8084,
"step": 750
},
{
"epoch": 0.653114186851211,
"grad_norm": 0.5473154955121547,
"learning_rate": 8.138130918729061e-06,
"loss": 0.8723,
"step": 755
},
{
"epoch": 0.657439446366782,
"grad_norm": 0.5523057815889402,
"learning_rate": 7.961705059155669e-06,
"loss": 0.8576,
"step": 760
},
{
"epoch": 0.6617647058823529,
"grad_norm": 0.5073374122462716,
"learning_rate": 7.786314478895995e-06,
"loss": 0.8686,
"step": 765
},
{
"epoch": 0.6660899653979239,
"grad_norm": 0.5061687468551056,
"learning_rate": 7.611999188183386e-06,
"loss": 0.7825,
"step": 770
},
{
"epoch": 0.6704152249134948,
"grad_norm": 0.6316502152940354,
"learning_rate": 7.438798951955308e-06,
"loss": 0.8363,
"step": 775
},
{
"epoch": 0.6747404844290658,
"grad_norm": 0.5067998953618004,
"learning_rate": 7.266753280782153e-06,
"loss": 0.8381,
"step": 780
},
{
"epoch": 0.6790657439446367,
"grad_norm": 0.5139250194422382,
"learning_rate": 7.095901421854031e-06,
"loss": 0.8425,
"step": 785
},
{
"epoch": 0.6833910034602076,
"grad_norm": 0.6296340797901989,
"learning_rate": 6.926282350027726e-06,
"loss": 0.8414,
"step": 790
},
{
"epoch": 0.6877162629757786,
"grad_norm": 0.5047756045208669,
"learning_rate": 6.757934758935644e-06,
"loss": 0.7665,
"step": 795
},
{
"epoch": 0.6920415224913494,
"grad_norm": 0.572738019347974,
"learning_rate": 6.5908970521590615e-06,
"loss": 0.8161,
"step": 800
},
{
"epoch": 0.6963667820069204,
"grad_norm": 0.5069382413264995,
"learning_rate": 6.425207334467425e-06,
"loss": 0.8278,
"step": 805
},
{
"epoch": 0.7006920415224913,
"grad_norm": 0.4898090135258845,
"learning_rate": 6.260903403125895e-06,
"loss": 0.8034,
"step": 810
},
{
"epoch": 0.7050173010380623,
"grad_norm": 0.5081634917932453,
"learning_rate": 6.098022739272975e-06,
"loss": 0.8131,
"step": 815
},
{
"epoch": 0.7093425605536332,
"grad_norm": 0.5175833945088627,
"learning_rate": 5.936602499370311e-06,
"loss": 0.8265,
"step": 820
},
{
"epoch": 0.7136678200692042,
"grad_norm": 0.5386658076448018,
"learning_rate": 5.776679506726518e-06,
"loss": 0.8364,
"step": 825
},
{
"epoch": 0.7179930795847751,
"grad_norm": 0.5216927460030648,
"learning_rate": 5.618290243097023e-06,
"loss": 0.7991,
"step": 830
},
{
"epoch": 0.722318339100346,
"grad_norm": 0.5110983811093638,
"learning_rate": 5.4614708403618044e-06,
"loss": 0.861,
"step": 835
},
{
"epoch": 0.726643598615917,
"grad_norm": 0.5053372906071909,
"learning_rate": 5.30625707228299e-06,
"loss": 0.8299,
"step": 840
},
{
"epoch": 0.7309688581314879,
"grad_norm": 0.5004085082572955,
"learning_rate": 5.152684346344087e-06,
"loss": 0.7934,
"step": 845
},
{
"epoch": 0.7352941176470589,
"grad_norm": 0.49425689688033725,
"learning_rate": 5.0007876956728215e-06,
"loss": 0.8283,
"step": 850
},
{
"epoch": 0.7396193771626297,
"grad_norm": 0.5272884015689866,
"learning_rate": 4.85060177104935e-06,
"loss": 0.8162,
"step": 855
},
{
"epoch": 0.7439446366782007,
"grad_norm": 0.521595928251936,
"learning_rate": 4.702160833001691e-06,
"loss": 0.8657,
"step": 860
},
{
"epoch": 0.7482698961937716,
"grad_norm": 0.5137260796340287,
"learning_rate": 4.555498743990184e-06,
"loss": 0.8308,
"step": 865
},
{
"epoch": 0.7525951557093425,
"grad_norm": 0.5107813329761777,
"learning_rate": 4.410648960682778e-06,
"loss": 0.7963,
"step": 870
},
{
"epoch": 0.7569204152249135,
"grad_norm": 0.5099358774599103,
"learning_rate": 4.2676445263228445e-06,
"loss": 0.7941,
"step": 875
},
{
"epoch": 0.7612456747404844,
"grad_norm": 0.5009824820130271,
"learning_rate": 4.126518063191356e-06,
"loss": 0.8048,
"step": 880
},
{
"epoch": 0.7655709342560554,
"grad_norm": 0.4939027961806981,
"learning_rate": 3.9873017651650345e-06,
"loss": 0.8394,
"step": 885
},
{
"epoch": 0.7698961937716263,
"grad_norm": 0.5176545471119196,
"learning_rate": 3.850027390372299e-06,
"loss": 0.8265,
"step": 890
},
{
"epoch": 0.7742214532871973,
"grad_norm": 0.528833669832955,
"learning_rate": 3.7147262539485346e-06,
"loss": 0.8594,
"step": 895
},
{
"epoch": 0.7785467128027682,
"grad_norm": 0.5012299670914705,
"learning_rate": 3.581429220892493e-06,
"loss": 0.8628,
"step": 900
},
{
"epoch": 0.782871972318339,
"grad_norm": 0.4872299919889065,
"learning_rate": 3.450166699025323e-06,
"loss": 0.8257,
"step": 905
},
{
"epoch": 0.78719723183391,
"grad_norm": 0.528370450443096,
"learning_rate": 3.3209686320539333e-06,
"loss": 0.8136,
"step": 910
},
{
"epoch": 0.7915224913494809,
"grad_norm": 0.5792281420730306,
"learning_rate": 3.1938644927401907e-06,
"loss": 0.8379,
"step": 915
},
{
"epoch": 0.7958477508650519,
"grad_norm": 0.5215022441313045,
"learning_rate": 3.068883276177603e-06,
"loss": 0.8442,
"step": 920
},
{
"epoch": 0.8001730103806228,
"grad_norm": 0.5302991353359664,
"learning_rate": 2.946053493176895e-06,
"loss": 0.7983,
"step": 925
},
{
"epoch": 0.8044982698961938,
"grad_norm": 0.5052616638889765,
"learning_rate": 2.825403163762136e-06,
"loss": 0.8289,
"step": 930
},
{
"epoch": 0.8088235294117647,
"grad_norm": 0.5079415891014708,
"learning_rate": 2.7069598107787614e-06,
"loss": 0.8117,
"step": 935
},
{
"epoch": 0.8131487889273357,
"grad_norm": 0.5106410010796052,
"learning_rate": 2.5907504536150734e-06,
"loss": 0.8411,
"step": 940
},
{
"epoch": 0.8174740484429066,
"grad_norm": 0.5507935287130962,
"learning_rate": 2.47680160203852e-06,
"loss": 0.849,
"step": 945
},
{
"epoch": 0.8217993079584776,
"grad_norm": 0.49137356817379463,
"learning_rate": 2.365139250148299e-06,
"loss": 0.8179,
"step": 950
},
{
"epoch": 0.8261245674740484,
"grad_norm": 0.4639212947765931,
"learning_rate": 2.2557888704455437e-06,
"loss": 0.7913,
"step": 955
},
{
"epoch": 0.8304498269896193,
"grad_norm": 0.5043002160977934,
"learning_rate": 2.1487754080225296e-06,
"loss": 0.7796,
"step": 960
},
{
"epoch": 0.8347750865051903,
"grad_norm": 0.4944119145187766,
"learning_rate": 2.0441232748721657e-06,
"loss": 0.8185,
"step": 965
},
{
"epoch": 0.8391003460207612,
"grad_norm": 0.5290366850290401,
"learning_rate": 1.941856344319129e-06,
"loss": 0.8552,
"step": 970
},
{
"epoch": 0.8434256055363322,
"grad_norm": 0.4999874951847485,
"learning_rate": 1.8419979455738478e-06,
"loss": 0.8656,
"step": 975
},
{
"epoch": 0.8477508650519031,
"grad_norm": 0.49955855199934796,
"learning_rate": 1.744570858410638e-06,
"loss": 0.8039,
"step": 980
},
{
"epoch": 0.8520761245674741,
"grad_norm": 0.5135220149854418,
"learning_rate": 1.6495973079711569e-06,
"loss": 0.8149,
"step": 985
},
{
"epoch": 0.856401384083045,
"grad_norm": 0.5046972531572713,
"learning_rate": 1.5570989596943864e-06,
"loss": 0.7999,
"step": 990
},
{
"epoch": 0.860726643598616,
"grad_norm": 0.5179440222287706,
"learning_rate": 1.4670969143742928e-06,
"loss": 0.8104,
"step": 995
},
{
"epoch": 0.8650519031141869,
"grad_norm": 0.5228153223587654,
"learning_rate": 1.3796117033463118e-06,
"loss": 0.8566,
"step": 1000
},
{
"epoch": 0.8693771626297578,
"grad_norm": 0.5391211673663446,
"learning_rate": 1.294663283803703e-06,
"loss": 0.8518,
"step": 1005
},
{
"epoch": 0.8737024221453287,
"grad_norm": 0.504384382386959,
"learning_rate": 1.212271034244926e-06,
"loss": 0.801,
"step": 1010
},
{
"epoch": 0.8780276816608996,
"grad_norm": 0.4785088098756923,
"learning_rate": 1.1324537500529725e-06,
"loss": 0.8131,
"step": 1015
},
{
"epoch": 0.8823529411764706,
"grad_norm": 0.4963438651591217,
"learning_rate": 1.0552296392077805e-06,
"loss": 0.7988,
"step": 1020
},
{
"epoch": 0.8866782006920415,
"grad_norm": 0.5339083354032369,
"learning_rate": 9.806163181325879e-07,
"loss": 0.774,
"step": 1025
},
{
"epoch": 0.8910034602076125,
"grad_norm": 0.5715619749859293,
"learning_rate": 9.086308076752814e-07,
"loss": 0.8132,
"step": 1030
},
{
"epoch": 0.8953287197231834,
"grad_norm": 0.5371813287339363,
"learning_rate": 8.392895292255717e-07,
"loss": 0.839,
"step": 1035
},
{
"epoch": 0.8996539792387543,
"grad_norm": 0.49428906553101953,
"learning_rate": 7.726083009689494e-07,
"loss": 0.779,
"step": 1040
},
{
"epoch": 0.9039792387543253,
"grad_norm": 0.5249338454718393,
"learning_rate": 7.086023342782106e-07,
"loss": 0.8145,
"step": 1045
},
{
"epoch": 0.9083044982698962,
"grad_norm": 0.5519654935007201,
"learning_rate": 6.472862302434476e-07,
"loss": 0.8668,
"step": 1050
},
{
"epoch": 0.9126297577854672,
"grad_norm": 0.5116299810045446,
"learning_rate": 5.886739763412141e-07,
"loss": 0.7868,
"step": 1055
},
{
"epoch": 0.916955017301038,
"grad_norm": 0.4974593798556511,
"learning_rate": 5.327789432437141e-07,
"loss": 0.8086,
"step": 1060
},
{
"epoch": 0.921280276816609,
"grad_norm": 0.5265647244032939,
"learning_rate": 4.796138817686549e-07,
"loss": 0.8559,
"step": 1065
},
{
"epoch": 0.9256055363321799,
"grad_norm": 0.49809187311204983,
"learning_rate": 4.2919091997053325e-07,
"loss": 0.7893,
"step": 1070
},
{
"epoch": 0.9299307958477508,
"grad_norm": 0.5022109153114724,
"learning_rate": 3.8152156037396387e-07,
"loss": 0.8019,
"step": 1075
},
{
"epoch": 0.9342560553633218,
"grad_norm": 0.5928320960488397,
"learning_rate": 3.366166773497312e-07,
"loss": 0.8326,
"step": 1080
},
{
"epoch": 0.9385813148788927,
"grad_norm": 0.5130802515210321,
"learning_rate": 2.944865146340986e-07,
"loss": 0.899,
"step": 1085
},
{
"epoch": 0.9429065743944637,
"grad_norm": 0.4719362769911749,
"learning_rate": 2.55140682992018e-07,
"loss": 0.8444,
"step": 1090
},
{
"epoch": 0.9472318339100346,
"grad_norm": 0.528933199844152,
"learning_rate": 2.185881580247004e-07,
"loss": 0.8169,
"step": 1095
},
{
"epoch": 0.9515570934256056,
"grad_norm": 0.5063093744503498,
"learning_rate": 1.8483727812210748e-07,
"loss": 0.7663,
"step": 1100
},
{
"epoch": 0.9558823529411765,
"grad_norm": 0.5290010051023691,
"learning_rate": 1.5389574256077926e-07,
"loss": 0.809,
"step": 1105
},
{
"epoch": 0.9602076124567474,
"grad_norm": 0.4872560757289188,
"learning_rate": 1.2577060974747757e-07,
"loss": 0.7593,
"step": 1110
},
{
"epoch": 0.9645328719723183,
"grad_norm": 0.5050163209919502,
"learning_rate": 1.004682956090186e-07,
"loss": 0.8483,
"step": 1115
},
{
"epoch": 0.9688581314878892,
"grad_norm": 7.776625466704461,
"learning_rate": 7.79945721286604e-08,
"loss": 0.8139,
"step": 1120
},
{
"epoch": 0.9731833910034602,
"grad_norm": 0.4887738375903042,
"learning_rate": 5.835456602939215e-08,
"loss": 0.7895,
"step": 1125
},
{
"epoch": 0.9775086505190311,
"grad_norm": 0.5108808230832508,
"learning_rate": 4.155275760443222e-08,
"loss": 0.847,
"step": 1130
},
{
"epoch": 0.9818339100346021,
"grad_norm": 0.5072678680238547,
"learning_rate": 2.7592979695172107e-08,
"loss": 0.7916,
"step": 1135
},
{
"epoch": 0.986159169550173,
"grad_norm": 0.5044352077176663,
"learning_rate": 1.6478416816827214e-08,
"loss": 0.8123,
"step": 1140
},
{
"epoch": 0.990484429065744,
"grad_norm": 0.5228080963566056,
"learning_rate": 8.211604431984655e-09,
"loss": 0.8227,
"step": 1145
},
{
"epoch": 0.9948096885813149,
"grad_norm": 0.4884395803733054,
"learning_rate": 2.7944283722075583e-09,
"loss": 0.7721,
"step": 1150
},
{
"epoch": 0.9991349480968859,
"grad_norm": 0.5950204077170628,
"learning_rate": 2.2812440784586443e-10,
"loss": 0.8012,
"step": 1155
},
{
"epoch": 1.0,
"step": 1156,
"total_flos": 140105591291904.0,
"train_loss": 0.8524971550723673,
"train_runtime": 3492.0427,
"train_samples_per_second": 10.585,
"train_steps_per_second": 0.331
}
],
"logging_steps": 5,
"max_steps": 1156,
"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": 140105591291904.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}