OpenO1-LLama-8B-v0.1 / trainer_state.json
paralym's picture
add sft model
49758f8
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 200,
"global_step": 2193,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.013679890560875513,
"grad_norm": 4.306338197381437,
"learning_rate": 4.5454545454545457e-07,
"loss": 0.9255,
"step": 10
},
{
"epoch": 0.027359781121751026,
"grad_norm": 2.40394571117872,
"learning_rate": 9.090909090909091e-07,
"loss": 0.863,
"step": 20
},
{
"epoch": 0.04103967168262654,
"grad_norm": 1.8078902930615164,
"learning_rate": 1.3636363636363636e-06,
"loss": 0.7599,
"step": 30
},
{
"epoch": 0.05471956224350205,
"grad_norm": 1.4220385901238113,
"learning_rate": 1.8181818181818183e-06,
"loss": 0.7184,
"step": 40
},
{
"epoch": 0.06839945280437756,
"grad_norm": 1.3537605726648354,
"learning_rate": 2.2727272727272728e-06,
"loss": 0.6981,
"step": 50
},
{
"epoch": 0.08207934336525308,
"grad_norm": 1.2344983469823567,
"learning_rate": 2.7272727272727272e-06,
"loss": 0.6943,
"step": 60
},
{
"epoch": 0.09575923392612859,
"grad_norm": 1.1119101183337694,
"learning_rate": 3.181818181818182e-06,
"loss": 0.6687,
"step": 70
},
{
"epoch": 0.1094391244870041,
"grad_norm": 1.2164492298241776,
"learning_rate": 3.6363636363636366e-06,
"loss": 0.6694,
"step": 80
},
{
"epoch": 0.12311901504787962,
"grad_norm": 1.3446219804481927,
"learning_rate": 4.0909090909090915e-06,
"loss": 0.6602,
"step": 90
},
{
"epoch": 0.13679890560875513,
"grad_norm": 1.2505229649313154,
"learning_rate": 4.5454545454545455e-06,
"loss": 0.6514,
"step": 100
},
{
"epoch": 0.15047879616963064,
"grad_norm": 1.1279331262953187,
"learning_rate": 5e-06,
"loss": 0.6413,
"step": 110
},
{
"epoch": 0.16415868673050615,
"grad_norm": 1.246045877795659,
"learning_rate": 5.4545454545454545e-06,
"loss": 0.6584,
"step": 120
},
{
"epoch": 0.17783857729138167,
"grad_norm": 1.1785589469428788,
"learning_rate": 5.90909090909091e-06,
"loss": 0.6292,
"step": 130
},
{
"epoch": 0.19151846785225718,
"grad_norm": 1.219408849788821,
"learning_rate": 6.363636363636364e-06,
"loss": 0.6487,
"step": 140
},
{
"epoch": 0.2051983584131327,
"grad_norm": 1.1571538455846238,
"learning_rate": 6.818181818181818e-06,
"loss": 0.6294,
"step": 150
},
{
"epoch": 0.2188782489740082,
"grad_norm": 1.3558018996617476,
"learning_rate": 7.272727272727273e-06,
"loss": 0.6443,
"step": 160
},
{
"epoch": 0.23255813953488372,
"grad_norm": 1.5617060190415994,
"learning_rate": 7.727272727272727e-06,
"loss": 0.6411,
"step": 170
},
{
"epoch": 0.24623803009575923,
"grad_norm": 1.235429384148413,
"learning_rate": 8.181818181818183e-06,
"loss": 0.6385,
"step": 180
},
{
"epoch": 0.25991792065663477,
"grad_norm": 1.2646278028286053,
"learning_rate": 8.636363636363637e-06,
"loss": 0.6466,
"step": 190
},
{
"epoch": 0.27359781121751026,
"grad_norm": 1.3092856333044969,
"learning_rate": 9.090909090909091e-06,
"loss": 0.6185,
"step": 200
},
{
"epoch": 0.27359781121751026,
"eval_loss": 0.6430878043174744,
"eval_runtime": 168.6747,
"eval_samples_per_second": 30.817,
"eval_steps_per_second": 3.854,
"step": 200
},
{
"epoch": 0.2872777017783858,
"grad_norm": 1.1897854638522036,
"learning_rate": 9.545454545454547e-06,
"loss": 0.6235,
"step": 210
},
{
"epoch": 0.3009575923392613,
"grad_norm": 1.249148897978773,
"learning_rate": 1e-05,
"loss": 0.6305,
"step": 220
},
{
"epoch": 0.3146374829001368,
"grad_norm": 1.1885808643667615,
"learning_rate": 9.99936616472178e-06,
"loss": 0.6308,
"step": 230
},
{
"epoch": 0.3283173734610123,
"grad_norm": 1.250733802249208,
"learning_rate": 9.997464819585985e-06,
"loss": 0.6331,
"step": 240
},
{
"epoch": 0.34199726402188785,
"grad_norm": 1.1199745780436605,
"learning_rate": 9.994296446648463e-06,
"loss": 0.6319,
"step": 250
},
{
"epoch": 0.35567715458276333,
"grad_norm": 1.1180646470383142,
"learning_rate": 9.989861849199833e-06,
"loss": 0.6368,
"step": 260
},
{
"epoch": 0.3693570451436389,
"grad_norm": 1.1411785528368101,
"learning_rate": 9.984162151561814e-06,
"loss": 0.628,
"step": 270
},
{
"epoch": 0.38303693570451436,
"grad_norm": 1.0640206569253472,
"learning_rate": 9.977198798802186e-06,
"loss": 0.6191,
"step": 280
},
{
"epoch": 0.3967168262653899,
"grad_norm": 1.2307131395813051,
"learning_rate": 9.9689735563684e-06,
"loss": 0.6311,
"step": 290
},
{
"epoch": 0.4103967168262654,
"grad_norm": 1.1751145352265429,
"learning_rate": 9.959488509639986e-06,
"loss": 0.6204,
"step": 300
},
{
"epoch": 0.4240766073871409,
"grad_norm": 1.1381135969095364,
"learning_rate": 9.948746063399836e-06,
"loss": 0.6185,
"step": 310
},
{
"epoch": 0.4377564979480164,
"grad_norm": 1.2008146993612843,
"learning_rate": 9.936748941224514e-06,
"loss": 0.63,
"step": 320
},
{
"epoch": 0.45143638850889195,
"grad_norm": 1.0906550245462099,
"learning_rate": 9.923500184793728e-06,
"loss": 0.6325,
"step": 330
},
{
"epoch": 0.46511627906976744,
"grad_norm": 1.2687828199372913,
"learning_rate": 9.909003153119165e-06,
"loss": 0.6202,
"step": 340
},
{
"epoch": 0.478796169630643,
"grad_norm": 1.2051295685227352,
"learning_rate": 9.893261521692865e-06,
"loss": 0.6157,
"step": 350
},
{
"epoch": 0.49247606019151846,
"grad_norm": 1.1391781521048607,
"learning_rate": 9.876279281555363e-06,
"loss": 0.6244,
"step": 360
},
{
"epoch": 0.506155950752394,
"grad_norm": 1.0911712689339368,
"learning_rate": 9.858060738283819e-06,
"loss": 0.6209,
"step": 370
},
{
"epoch": 0.5198358413132695,
"grad_norm": 1.1210123313032032,
"learning_rate": 9.838610510900412e-06,
"loss": 0.6168,
"step": 380
},
{
"epoch": 0.533515731874145,
"grad_norm": 1.2189445592306896,
"learning_rate": 9.817933530701258e-06,
"loss": 0.6273,
"step": 390
},
{
"epoch": 0.5471956224350205,
"grad_norm": 1.0483395986247415,
"learning_rate": 9.79603504000615e-06,
"loss": 0.6088,
"step": 400
},
{
"epoch": 0.5471956224350205,
"eval_loss": 0.6271325945854187,
"eval_runtime": 167.7353,
"eval_samples_per_second": 30.989,
"eval_steps_per_second": 3.875,
"step": 400
},
{
"epoch": 0.560875512995896,
"grad_norm": 1.0660522163415433,
"learning_rate": 9.772920590829471e-06,
"loss": 0.6137,
"step": 410
},
{
"epoch": 0.5745554035567716,
"grad_norm": 1.2363091668613064,
"learning_rate": 9.748596043472547e-06,
"loss": 0.6143,
"step": 420
},
{
"epoch": 0.5882352941176471,
"grad_norm": 1.0981082723618094,
"learning_rate": 9.723067565037874e-06,
"loss": 0.6068,
"step": 430
},
{
"epoch": 0.6019151846785226,
"grad_norm": 1.1847772259009521,
"learning_rate": 9.696341627865548e-06,
"loss": 0.6263,
"step": 440
},
{
"epoch": 0.615595075239398,
"grad_norm": 1.179605990278435,
"learning_rate": 9.668425007892298e-06,
"loss": 0.624,
"step": 450
},
{
"epoch": 0.6292749658002736,
"grad_norm": 1.1327982969979666,
"learning_rate": 9.639324782933556e-06,
"loss": 0.6115,
"step": 460
},
{
"epoch": 0.6429548563611491,
"grad_norm": 1.1092926273561299,
"learning_rate": 9.609048330888996e-06,
"loss": 0.6231,
"step": 470
},
{
"epoch": 0.6566347469220246,
"grad_norm": 1.2023856216186377,
"learning_rate": 9.577603327871983e-06,
"loss": 0.6087,
"step": 480
},
{
"epoch": 0.6703146374829001,
"grad_norm": 1.140079893396528,
"learning_rate": 9.544997746263409e-06,
"loss": 0.6207,
"step": 490
},
{
"epoch": 0.6839945280437757,
"grad_norm": 1.0956555350394377,
"learning_rate": 9.511239852690429e-06,
"loss": 0.5912,
"step": 500
},
{
"epoch": 0.6976744186046512,
"grad_norm": 1.0991687302711715,
"learning_rate": 9.476338205930593e-06,
"loss": 0.6084,
"step": 510
},
{
"epoch": 0.7113543091655267,
"grad_norm": 1.0969007693966542,
"learning_rate": 9.440301654741889e-06,
"loss": 0.6132,
"step": 520
},
{
"epoch": 0.7250341997264022,
"grad_norm": 1.1125573118122185,
"learning_rate": 9.403139335619302e-06,
"loss": 0.6194,
"step": 530
},
{
"epoch": 0.7387140902872777,
"grad_norm": 1.1809078700930535,
"learning_rate": 9.36486067047838e-06,
"loss": 0.5951,
"step": 540
},
{
"epoch": 0.7523939808481532,
"grad_norm": 1.1864929076204318,
"learning_rate": 9.325475364266474e-06,
"loss": 0.6007,
"step": 550
},
{
"epoch": 0.7660738714090287,
"grad_norm": 1.26583782519778,
"learning_rate": 9.28499340250219e-06,
"loss": 0.6146,
"step": 560
},
{
"epoch": 0.7797537619699042,
"grad_norm": 1.076257354571428,
"learning_rate": 9.24342504874373e-06,
"loss": 0.6077,
"step": 570
},
{
"epoch": 0.7934336525307798,
"grad_norm": 1.0622000039574993,
"learning_rate": 9.200780841986717e-06,
"loss": 0.6001,
"step": 580
},
{
"epoch": 0.8071135430916553,
"grad_norm": 1.188474169737116,
"learning_rate": 9.157071593992217e-06,
"loss": 0.6289,
"step": 590
},
{
"epoch": 0.8207934336525308,
"grad_norm": 1.1602582650897655,
"learning_rate": 9.112308386545574e-06,
"loss": 0.6083,
"step": 600
},
{
"epoch": 0.8207934336525308,
"eval_loss": 0.6138648390769958,
"eval_runtime": 169.037,
"eval_samples_per_second": 30.751,
"eval_steps_per_second": 3.845,
"step": 600
},
{
"epoch": 0.8344733242134063,
"grad_norm": 1.1422692672202526,
"learning_rate": 9.066502568646805e-06,
"loss": 0.6028,
"step": 610
},
{
"epoch": 0.8481532147742818,
"grad_norm": 1.1671002032829694,
"learning_rate": 9.019665753633247e-06,
"loss": 0.6025,
"step": 620
},
{
"epoch": 0.8618331053351573,
"grad_norm": 1.1006027336870072,
"learning_rate": 8.971809816235164e-06,
"loss": 0.6072,
"step": 630
},
{
"epoch": 0.8755129958960328,
"grad_norm": 1.14419096908491,
"learning_rate": 8.922946889565119e-06,
"loss": 0.6014,
"step": 640
},
{
"epoch": 0.8891928864569083,
"grad_norm": 1.1499692013955807,
"learning_rate": 8.873089362041797e-06,
"loss": 0.584,
"step": 650
},
{
"epoch": 0.9028727770177839,
"grad_norm": 1.0137383119426882,
"learning_rate": 8.822249874249131e-06,
"loss": 0.5929,
"step": 660
},
{
"epoch": 0.9165526675786594,
"grad_norm": 1.0739771517494396,
"learning_rate": 8.770441315731477e-06,
"loss": 0.5954,
"step": 670
},
{
"epoch": 0.9302325581395349,
"grad_norm": 1.066521535106681,
"learning_rate": 8.717676821725679e-06,
"loss": 0.5876,
"step": 680
},
{
"epoch": 0.9439124487004104,
"grad_norm": 1.1050537240337106,
"learning_rate": 8.663969769830824e-06,
"loss": 0.6006,
"step": 690
},
{
"epoch": 0.957592339261286,
"grad_norm": 1.0834708122941283,
"learning_rate": 8.609333776616592e-06,
"loss": 0.5974,
"step": 700
},
{
"epoch": 0.9712722298221614,
"grad_norm": 1.2256953791044325,
"learning_rate": 8.553782694170963e-06,
"loss": 0.6043,
"step": 710
},
{
"epoch": 0.9849521203830369,
"grad_norm": 1.0369686902126203,
"learning_rate": 8.497330606588257e-06,
"loss": 0.5779,
"step": 720
},
{
"epoch": 0.9986320109439124,
"grad_norm": 1.0669076543876443,
"learning_rate": 8.439991826398331e-06,
"loss": 0.6108,
"step": 730
},
{
"epoch": 1.012311901504788,
"grad_norm": 1.38284058642527,
"learning_rate": 8.38178089093786e-06,
"loss": 0.4662,
"step": 740
},
{
"epoch": 1.0259917920656634,
"grad_norm": 1.36236398170643,
"learning_rate": 8.322712558664635e-06,
"loss": 0.4296,
"step": 750
},
{
"epoch": 1.039671682626539,
"grad_norm": 1.2100527044799618,
"learning_rate": 8.262801805415783e-06,
"loss": 0.4295,
"step": 760
},
{
"epoch": 1.0533515731874146,
"grad_norm": 1.1588309328691597,
"learning_rate": 8.202063820610887e-06,
"loss": 0.4244,
"step": 770
},
{
"epoch": 1.06703146374829,
"grad_norm": 1.1590821782151508,
"learning_rate": 8.140514003400945e-06,
"loss": 0.4257,
"step": 780
},
{
"epoch": 1.0807113543091655,
"grad_norm": 1.1515662295268354,
"learning_rate": 8.078167958764162e-06,
"loss": 0.4325,
"step": 790
},
{
"epoch": 1.094391244870041,
"grad_norm": 1.1972938041742296,
"learning_rate": 8.015041493549562e-06,
"loss": 0.4314,
"step": 800
},
{
"epoch": 1.094391244870041,
"eval_loss": 0.6188511848449707,
"eval_runtime": 168.4076,
"eval_samples_per_second": 30.866,
"eval_steps_per_second": 3.86,
"step": 800
},
{
"epoch": 1.1080711354309165,
"grad_norm": 1.2310788704520765,
"learning_rate": 7.951150612469396e-06,
"loss": 0.4335,
"step": 810
},
{
"epoch": 1.121751025991792,
"grad_norm": 1.2391064990548075,
"learning_rate": 7.886511514041422e-06,
"loss": 0.4271,
"step": 820
},
{
"epoch": 1.1354309165526675,
"grad_norm": 1.1957485922230724,
"learning_rate": 7.821140586482013e-06,
"loss": 0.4367,
"step": 830
},
{
"epoch": 1.1491108071135432,
"grad_norm": 1.2204287898575044,
"learning_rate": 7.755054403551191e-06,
"loss": 0.4359,
"step": 840
},
{
"epoch": 1.1627906976744187,
"grad_norm": 1.2398786083740503,
"learning_rate": 7.688269720350616e-06,
"loss": 0.433,
"step": 850
},
{
"epoch": 1.1764705882352942,
"grad_norm": 1.1869051225812688,
"learning_rate": 7.620803469075588e-06,
"loss": 0.4387,
"step": 860
},
{
"epoch": 1.1901504787961696,
"grad_norm": 1.1936581208347872,
"learning_rate": 7.552672754722169e-06,
"loss": 0.4198,
"step": 870
},
{
"epoch": 1.2038303693570451,
"grad_norm": 1.153245583578447,
"learning_rate": 7.483894850750473e-06,
"loss": 0.4356,
"step": 880
},
{
"epoch": 1.2175102599179206,
"grad_norm": 1.2121786191778063,
"learning_rate": 7.414487194705258e-06,
"loss": 0.4334,
"step": 890
},
{
"epoch": 1.231190150478796,
"grad_norm": 1.1865361362733824,
"learning_rate": 7.344467383794917e-06,
"loss": 0.4335,
"step": 900
},
{
"epoch": 1.2448700410396718,
"grad_norm": 1.2578535517364966,
"learning_rate": 7.2738531704299845e-06,
"loss": 0.4338,
"step": 910
},
{
"epoch": 1.2585499316005473,
"grad_norm": 1.208342728404754,
"learning_rate": 7.2026624577222855e-06,
"loss": 0.4506,
"step": 920
},
{
"epoch": 1.2722298221614228,
"grad_norm": 1.2293841459842745,
"learning_rate": 7.1309132949459e-06,
"loss": 0.4356,
"step": 930
},
{
"epoch": 1.2859097127222983,
"grad_norm": 1.1995897552412818,
"learning_rate": 7.058623872961051e-06,
"loss": 0.4293,
"step": 940
},
{
"epoch": 1.2995896032831737,
"grad_norm": 1.3990655945335009,
"learning_rate": 6.985812519602094e-06,
"loss": 0.4268,
"step": 950
},
{
"epoch": 1.3132694938440492,
"grad_norm": 1.2252593246536305,
"learning_rate": 6.912497695030796e-06,
"loss": 0.4286,
"step": 960
},
{
"epoch": 1.3269493844049247,
"grad_norm": 1.1833071160287538,
"learning_rate": 6.838697987056046e-06,
"loss": 0.4313,
"step": 970
},
{
"epoch": 1.3406292749658002,
"grad_norm": 1.1637401442465505,
"learning_rate": 6.764432106421223e-06,
"loss": 0.4279,
"step": 980
},
{
"epoch": 1.3543091655266757,
"grad_norm": 1.2180322469819673,
"learning_rate": 6.68971888206037e-06,
"loss": 0.4305,
"step": 990
},
{
"epoch": 1.3679890560875512,
"grad_norm": 1.2229086904650939,
"learning_rate": 6.614577256324426e-06,
"loss": 0.4322,
"step": 1000
},
{
"epoch": 1.3679890560875512,
"eval_loss": 0.617301881313324,
"eval_runtime": 168.0881,
"eval_samples_per_second": 30.924,
"eval_steps_per_second": 3.867,
"step": 1000
},
{
"epoch": 1.3816689466484269,
"grad_norm": 1.1947852132614785,
"learning_rate": 6.539026280178695e-06,
"loss": 0.4279,
"step": 1010
},
{
"epoch": 1.3953488372093024,
"grad_norm": 1.247192723450866,
"learning_rate": 6.46308510837277e-06,
"loss": 0.4309,
"step": 1020
},
{
"epoch": 1.4090287277701778,
"grad_norm": 1.185576441473233,
"learning_rate": 6.386772994584156e-06,
"loss": 0.4349,
"step": 1030
},
{
"epoch": 1.4227086183310533,
"grad_norm": 1.129361008569495,
"learning_rate": 6.310109286536801e-06,
"loss": 0.4232,
"step": 1040
},
{
"epoch": 1.4363885088919288,
"grad_norm": 1.1431132187777882,
"learning_rate": 6.233113421095794e-06,
"loss": 0.4355,
"step": 1050
},
{
"epoch": 1.4500683994528043,
"grad_norm": 1.1747059710212222,
"learning_rate": 6.155804919339453e-06,
"loss": 0.4286,
"step": 1060
},
{
"epoch": 1.46374829001368,
"grad_norm": 1.151624752794334,
"learning_rate": 6.078203381610064e-06,
"loss": 0.4314,
"step": 1070
},
{
"epoch": 1.4774281805745555,
"grad_norm": 1.2141807720724713,
"learning_rate": 6.000328482544532e-06,
"loss": 0.4292,
"step": 1080
},
{
"epoch": 1.491108071135431,
"grad_norm": 1.2395901378501935,
"learning_rate": 5.92219996608618e-06,
"loss": 0.4295,
"step": 1090
},
{
"epoch": 1.5047879616963065,
"grad_norm": 1.3241796863033826,
"learning_rate": 5.843837640479e-06,
"loss": 0.4399,
"step": 1100
},
{
"epoch": 1.518467852257182,
"grad_norm": 1.1677488632090005,
"learning_rate": 5.765261373245567e-06,
"loss": 0.424,
"step": 1110
},
{
"epoch": 1.5321477428180574,
"grad_norm": 1.2084792827131878,
"learning_rate": 5.686491086149965e-06,
"loss": 0.4285,
"step": 1120
},
{
"epoch": 1.545827633378933,
"grad_norm": 1.2057361531766408,
"learning_rate": 5.60754675014693e-06,
"loss": 0.4397,
"step": 1130
},
{
"epoch": 1.5595075239398084,
"grad_norm": 1.0459853203349079,
"learning_rate": 5.528448380318532e-06,
"loss": 0.4184,
"step": 1140
},
{
"epoch": 1.573187414500684,
"grad_norm": 1.1565826426536214,
"learning_rate": 5.449216030799667e-06,
"loss": 0.4199,
"step": 1150
},
{
"epoch": 1.5868673050615594,
"grad_norm": 1.337080485300612,
"learning_rate": 5.369869789693656e-06,
"loss": 0.4392,
"step": 1160
},
{
"epoch": 1.6005471956224349,
"grad_norm": 1.2455984858098026,
"learning_rate": 5.290429773979224e-06,
"loss": 0.4337,
"step": 1170
},
{
"epoch": 1.6142270861833106,
"grad_norm": 1.218208739067707,
"learning_rate": 5.2109161244101545e-06,
"loss": 0.4448,
"step": 1180
},
{
"epoch": 1.627906976744186,
"grad_norm": 1.1285789923935337,
"learning_rate": 5.1313490004089265e-06,
"loss": 0.4286,
"step": 1190
},
{
"epoch": 1.6415868673050615,
"grad_norm": 1.2264365996543631,
"learning_rate": 5.05174857495561e-06,
"loss": 0.4368,
"step": 1200
},
{
"epoch": 1.6415868673050615,
"eval_loss": 0.608180046081543,
"eval_runtime": 168.5858,
"eval_samples_per_second": 30.833,
"eval_steps_per_second": 3.856,
"step": 1200
},
{
"epoch": 1.655266757865937,
"grad_norm": 1.243465209947156,
"learning_rate": 4.972135029473333e-06,
"loss": 0.4276,
"step": 1210
},
{
"epoch": 1.6689466484268127,
"grad_norm": 1.1806815681626925,
"learning_rate": 4.892528548711594e-06,
"loss": 0.4334,
"step": 1220
},
{
"epoch": 1.6826265389876882,
"grad_norm": 1.1380113337270097,
"learning_rate": 4.812949315628746e-06,
"loss": 0.4272,
"step": 1230
},
{
"epoch": 1.6963064295485637,
"grad_norm": 1.1677234792880147,
"learning_rate": 4.7334175062749275e-06,
"loss": 0.4164,
"step": 1240
},
{
"epoch": 1.7099863201094392,
"grad_norm": 1.1552072694954474,
"learning_rate": 4.653953284676737e-06,
"loss": 0.4415,
"step": 1250
},
{
"epoch": 1.7236662106703147,
"grad_norm": 1.2021058595481928,
"learning_rate": 4.574576797724984e-06,
"loss": 0.4161,
"step": 1260
},
{
"epoch": 1.7373461012311902,
"grad_norm": 1.3115148564903278,
"learning_rate": 4.495308170066739e-06,
"loss": 0.4326,
"step": 1270
},
{
"epoch": 1.7510259917920656,
"grad_norm": 1.2648526427702413,
"learning_rate": 4.4161674990030715e-06,
"loss": 0.4426,
"step": 1280
},
{
"epoch": 1.7647058823529411,
"grad_norm": 1.137931826008329,
"learning_rate": 4.3371748493936834e-06,
"loss": 0.4263,
"step": 1290
},
{
"epoch": 1.7783857729138166,
"grad_norm": 1.203799366153618,
"learning_rate": 4.2583502485697945e-06,
"loss": 0.422,
"step": 1300
},
{
"epoch": 1.792065663474692,
"grad_norm": 1.1215051415533788,
"learning_rate": 4.179713681256523e-06,
"loss": 0.4278,
"step": 1310
},
{
"epoch": 1.8057455540355676,
"grad_norm": 1.1919550892429331,
"learning_rate": 4.101285084506075e-06,
"loss": 0.4126,
"step": 1320
},
{
"epoch": 1.819425444596443,
"grad_norm": 1.2602652902830995,
"learning_rate": 4.023084342643026e-06,
"loss": 0.4302,
"step": 1330
},
{
"epoch": 1.8331053351573188,
"grad_norm": 1.1856280985422172,
"learning_rate": 3.94513128222297e-06,
"loss": 0.4308,
"step": 1340
},
{
"epoch": 1.8467852257181943,
"grad_norm": 1.1188116484677932,
"learning_rate": 3.867445667005801e-06,
"loss": 0.4343,
"step": 1350
},
{
"epoch": 1.8604651162790697,
"grad_norm": 1.1588258315903728,
"learning_rate": 3.790047192944931e-06,
"loss": 0.4301,
"step": 1360
},
{
"epoch": 1.8741450068399452,
"grad_norm": 1.1459817706332724,
"learning_rate": 3.7129554831936998e-06,
"loss": 0.4175,
"step": 1370
},
{
"epoch": 1.887824897400821,
"grad_norm": 1.104555953617468,
"learning_rate": 3.6361900831302264e-06,
"loss": 0.4175,
"step": 1380
},
{
"epoch": 1.9015047879616964,
"grad_norm": 1.1942887584277926,
"learning_rate": 3.5597704554019907e-06,
"loss": 0.4211,
"step": 1390
},
{
"epoch": 1.915184678522572,
"grad_norm": 1.0668265437929074,
"learning_rate": 3.4837159749913944e-06,
"loss": 0.4195,
"step": 1400
},
{
"epoch": 1.915184678522572,
"eval_loss": 0.6006618142127991,
"eval_runtime": 167.6616,
"eval_samples_per_second": 31.003,
"eval_steps_per_second": 3.877,
"step": 1400
},
{
"epoch": 1.9288645690834474,
"grad_norm": 1.088376315283852,
"learning_rate": 3.40804592430354e-06,
"loss": 0.4205,
"step": 1410
},
{
"epoch": 1.9425444596443229,
"grad_norm": 1.1227517978501422,
"learning_rate": 3.3327794882774785e-06,
"loss": 0.4295,
"step": 1420
},
{
"epoch": 1.9562243502051984,
"grad_norm": 1.1732094471787033,
"learning_rate": 3.2579357495221786e-06,
"loss": 0.4104,
"step": 1430
},
{
"epoch": 1.9699042407660738,
"grad_norm": 1.2162059196521249,
"learning_rate": 3.183533683478427e-06,
"loss": 0.4091,
"step": 1440
},
{
"epoch": 1.9835841313269493,
"grad_norm": 1.108696878534993,
"learning_rate": 3.1095921536079215e-06,
"loss": 0.43,
"step": 1450
},
{
"epoch": 1.9972640218878248,
"grad_norm": 1.148281960391559,
"learning_rate": 3.0361299066107225e-06,
"loss": 0.4146,
"step": 1460
},
{
"epoch": 2.0109439124487003,
"grad_norm": 1.853284154156804,
"learning_rate": 2.9631655676723358e-06,
"loss": 0.3057,
"step": 1470
},
{
"epoch": 2.024623803009576,
"grad_norm": 1.2698401561195134,
"learning_rate": 2.890717635741589e-06,
"loss": 0.2671,
"step": 1480
},
{
"epoch": 2.0383036935704513,
"grad_norm": 1.1953162644205662,
"learning_rate": 2.81880447884052e-06,
"loss": 0.2676,
"step": 1490
},
{
"epoch": 2.0519835841313268,
"grad_norm": 1.120927117498952,
"learning_rate": 2.7474443294074537e-06,
"loss": 0.2593,
"step": 1500
},
{
"epoch": 2.0656634746922027,
"grad_norm": 1.324110443129841,
"learning_rate": 2.6766552796744583e-06,
"loss": 0.2703,
"step": 1510
},
{
"epoch": 2.079343365253078,
"grad_norm": 1.3413599300031922,
"learning_rate": 2.606455277080346e-06,
"loss": 0.2708,
"step": 1520
},
{
"epoch": 2.0930232558139537,
"grad_norm": 1.1788852376372527,
"learning_rate": 2.536862119720387e-06,
"loss": 0.2664,
"step": 1530
},
{
"epoch": 2.106703146374829,
"grad_norm": 1.1789132415176646,
"learning_rate": 2.467893451833884e-06,
"loss": 0.2673,
"step": 1540
},
{
"epoch": 2.1203830369357046,
"grad_norm": 1.2137095943380327,
"learning_rate": 2.3995667593307563e-06,
"loss": 0.2622,
"step": 1550
},
{
"epoch": 2.13406292749658,
"grad_norm": 1.2537108019590875,
"learning_rate": 2.331899365358266e-06,
"loss": 0.2695,
"step": 1560
},
{
"epoch": 2.1477428180574556,
"grad_norm": 1.2719182887175577,
"learning_rate": 2.2649084259090058e-06,
"loss": 0.2593,
"step": 1570
},
{
"epoch": 2.161422708618331,
"grad_norm": 1.164630355430247,
"learning_rate": 2.1986109254712745e-06,
"loss": 0.2708,
"step": 1580
},
{
"epoch": 2.1751025991792066,
"grad_norm": 1.1779155039946436,
"learning_rate": 2.133023672722923e-06,
"loss": 0.2697,
"step": 1590
},
{
"epoch": 2.188782489740082,
"grad_norm": 1.1294417346333414,
"learning_rate": 2.0681632962697955e-06,
"loss": 0.2528,
"step": 1600
},
{
"epoch": 2.188782489740082,
"eval_loss": 0.6378765106201172,
"eval_runtime": 171.4166,
"eval_samples_per_second": 30.324,
"eval_steps_per_second": 3.792,
"step": 1600
},
{
"epoch": 2.2024623803009575,
"grad_norm": 1.2183820949308963,
"learning_rate": 2.0040462404297873e-06,
"loss": 0.2719,
"step": 1610
},
{
"epoch": 2.216142270861833,
"grad_norm": 1.1698549285231044,
"learning_rate": 1.9406887610636716e-06,
"loss": 0.2659,
"step": 1620
},
{
"epoch": 2.2298221614227085,
"grad_norm": 1.2697486173216241,
"learning_rate": 1.8781069214536711e-06,
"loss": 0.2677,
"step": 1630
},
{
"epoch": 2.243502051983584,
"grad_norm": 1.214820822899955,
"learning_rate": 1.8163165882308776e-06,
"loss": 0.2662,
"step": 1640
},
{
"epoch": 2.2571819425444595,
"grad_norm": 1.3319711950469666,
"learning_rate": 1.7553334273525107e-06,
"loss": 0.2648,
"step": 1650
},
{
"epoch": 2.270861833105335,
"grad_norm": 1.245211504895954,
"learning_rate": 1.6951729001300626e-06,
"loss": 0.2659,
"step": 1660
},
{
"epoch": 2.2845417236662104,
"grad_norm": 1.2357536814491976,
"learning_rate": 1.6358502593093417e-06,
"loss": 0.2638,
"step": 1670
},
{
"epoch": 2.2982216142270864,
"grad_norm": 1.2488899759446517,
"learning_rate": 1.5773805452033664e-06,
"loss": 0.2647,
"step": 1680
},
{
"epoch": 2.311901504787962,
"grad_norm": 1.1504334028748302,
"learning_rate": 1.5197785818791395e-06,
"loss": 0.2611,
"step": 1690
},
{
"epoch": 2.3255813953488373,
"grad_norm": 1.2488245466965713,
"learning_rate": 1.463058973399241e-06,
"loss": 0.2597,
"step": 1700
},
{
"epoch": 2.339261285909713,
"grad_norm": 1.2329978043228385,
"learning_rate": 1.4072361001191997e-06,
"loss": 0.2664,
"step": 1710
},
{
"epoch": 2.3529411764705883,
"grad_norm": 1.200114038379274,
"learning_rate": 1.3523241150415822e-06,
"loss": 0.2579,
"step": 1720
},
{
"epoch": 2.366621067031464,
"grad_norm": 1.1866913191947843,
"learning_rate": 1.298336940227724e-06,
"loss": 0.2627,
"step": 1730
},
{
"epoch": 2.3803009575923393,
"grad_norm": 1.215070259088293,
"learning_rate": 1.2452882632680125e-06,
"loss": 0.2629,
"step": 1740
},
{
"epoch": 2.3939808481532148,
"grad_norm": 1.2848393192245922,
"learning_rate": 1.1931915338116157e-06,
"loss": 0.2579,
"step": 1750
},
{
"epoch": 2.4076607387140903,
"grad_norm": 1.178082536045285,
"learning_rate": 1.1420599601565375e-06,
"loss": 0.2655,
"step": 1760
},
{
"epoch": 2.4213406292749657,
"grad_norm": 1.1599349607603797,
"learning_rate": 1.0919065059008626e-06,
"loss": 0.2614,
"step": 1770
},
{
"epoch": 2.4350205198358412,
"grad_norm": 1.218569747775109,
"learning_rate": 1.042743886656043e-06,
"loss": 0.2633,
"step": 1780
},
{
"epoch": 2.4487004103967167,
"grad_norm": 1.1984843383370676,
"learning_rate": 9.94584566823062e-07,
"loss": 0.2626,
"step": 1790
},
{
"epoch": 2.462380300957592,
"grad_norm": 1.1483398440958268,
"learning_rate": 9.474407564322685e-07,
"loss": 0.2648,
"step": 1800
},
{
"epoch": 2.462380300957592,
"eval_loss": 0.6390905976295471,
"eval_runtime": 167.5515,
"eval_samples_per_second": 31.023,
"eval_steps_per_second": 3.879,
"step": 1800
},
{
"epoch": 2.4760601915184677,
"grad_norm": 1.2020289928203793,
"learning_rate": 9.01324408047734e-07,
"loss": 0.2649,
"step": 1810
},
{
"epoch": 2.4897400820793436,
"grad_norm": 1.1514025291135426,
"learning_rate": 8.562472137368638e-07,
"loss": 0.2598,
"step": 1820
},
{
"epoch": 2.503419972640219,
"grad_norm": 1.1443562664564093,
"learning_rate": 8.122206021060552e-07,
"loss": 0.2627,
"step": 1830
},
{
"epoch": 2.5170998632010946,
"grad_norm": 1.2145886027335204,
"learning_rate": 7.692557354031633e-07,
"loss": 0.2554,
"step": 1840
},
{
"epoch": 2.53077975376197,
"grad_norm": 1.2003000862318016,
"learning_rate": 7.273635066874796e-07,
"loss": 0.2595,
"step": 1850
},
{
"epoch": 2.5444596443228455,
"grad_norm": 1.2102887918479095,
"learning_rate": 6.86554537067986e-07,
"loss": 0.2562,
"step": 1860
},
{
"epoch": 2.558139534883721,
"grad_norm": 1.1976060832616344,
"learning_rate": 6.46839173010525e-07,
"loss": 0.2589,
"step": 1870
},
{
"epoch": 2.5718194254445965,
"grad_norm": 1.1509734934067806,
"learning_rate": 6.08227483714629e-07,
"loss": 0.2565,
"step": 1880
},
{
"epoch": 2.585499316005472,
"grad_norm": 1.237509833168683,
"learning_rate": 5.707292585606277e-07,
"loss": 0.2646,
"step": 1890
},
{
"epoch": 2.5991792065663475,
"grad_norm": 1.2849404614912345,
"learning_rate": 5.343540046277113e-07,
"loss": 0.2571,
"step": 1900
},
{
"epoch": 2.612859097127223,
"grad_norm": 1.2020133789054372,
"learning_rate": 4.991109442835578e-07,
"loss": 0.2605,
"step": 1910
},
{
"epoch": 2.6265389876880985,
"grad_norm": 1.2567141981825356,
"learning_rate": 4.6500901284615094e-07,
"loss": 0.2621,
"step": 1920
},
{
"epoch": 2.640218878248974,
"grad_norm": 1.1549261891969707,
"learning_rate": 4.320568563183708e-07,
"loss": 0.2646,
"step": 1930
},
{
"epoch": 2.6538987688098494,
"grad_norm": 1.227541734725619,
"learning_rate": 4.0026282919593815e-07,
"loss": 0.2653,
"step": 1940
},
{
"epoch": 2.667578659370725,
"grad_norm": 1.1729506883600467,
"learning_rate": 3.6963499234926314e-07,
"loss": 0.2675,
"step": 1950
},
{
"epoch": 2.6812585499316004,
"grad_norm": 1.2392479528804268,
"learning_rate": 3.401811109797404e-07,
"loss": 0.2805,
"step": 1960
},
{
"epoch": 2.694938440492476,
"grad_norm": 1.1834886023051063,
"learning_rate": 3.1190865265100954e-07,
"loss": 0.2594,
"step": 1970
},
{
"epoch": 2.7086183310533514,
"grad_norm": 1.1342198262790109,
"learning_rate": 2.8482478539566595e-07,
"loss": 0.2662,
"step": 1980
},
{
"epoch": 2.722298221614227,
"grad_norm": 1.2035486656539252,
"learning_rate": 2.58936375897923e-07,
"loss": 0.2688,
"step": 1990
},
{
"epoch": 2.7359781121751023,
"grad_norm": 1.181326670816563,
"learning_rate": 2.342499877526755e-07,
"loss": 0.2586,
"step": 2000
},
{
"epoch": 2.7359781121751023,
"eval_loss": 0.6391583681106567,
"eval_runtime": 167.8571,
"eval_samples_per_second": 30.967,
"eval_steps_per_second": 3.872,
"step": 2000
},
{
"epoch": 2.7496580027359783,
"grad_norm": 1.1676009303739256,
"learning_rate": 2.1077187980140467e-07,
"loss": 0.2616,
"step": 2010
},
{
"epoch": 2.7633378932968538,
"grad_norm": 1.219603623890368,
"learning_rate": 1.8850800454534358e-07,
"loss": 0.2591,
"step": 2020
},
{
"epoch": 2.7770177838577292,
"grad_norm": 1.2643792382182215,
"learning_rate": 1.6746400663631923e-07,
"loss": 0.2608,
"step": 2030
},
{
"epoch": 2.7906976744186047,
"grad_norm": 1.2780674547136892,
"learning_rate": 1.4764522144563865e-07,
"loss": 0.2625,
"step": 2040
},
{
"epoch": 2.80437756497948,
"grad_norm": 1.2502610941549783,
"learning_rate": 1.2905667371139298e-07,
"loss": 0.259,
"step": 2050
},
{
"epoch": 2.8180574555403557,
"grad_norm": 1.246880298259785,
"learning_rate": 1.117030762645116e-07,
"loss": 0.2613,
"step": 2060
},
{
"epoch": 2.831737346101231,
"grad_norm": 1.2060582120323238,
"learning_rate": 9.558882883390075e-08,
"loss": 0.2607,
"step": 2070
},
{
"epoch": 2.8454172366621067,
"grad_norm": 1.1847791928318971,
"learning_rate": 8.071801693096237e-08,
"loss": 0.2575,
"step": 2080
},
{
"epoch": 2.859097127222982,
"grad_norm": 1.2366555323688222,
"learning_rate": 6.70944108137761e-08,
"loss": 0.2601,
"step": 2090
},
{
"epoch": 2.8727770177838576,
"grad_norm": 1.1353358354521152,
"learning_rate": 5.472146453121163e-08,
"loss": 0.2565,
"step": 2100
},
{
"epoch": 2.886456908344733,
"grad_norm": 1.2500960828119085,
"learning_rate": 4.360231504720813e-08,
"loss": 0.2726,
"step": 2110
},
{
"epoch": 2.9001367989056086,
"grad_norm": 1.234389078825029,
"learning_rate": 3.3739781445449316e-08,
"loss": 0.2544,
"step": 2120
},
{
"epoch": 2.9138166894664845,
"grad_norm": 1.1537121999523,
"learning_rate": 2.513636421462573e-08,
"loss": 0.2587,
"step": 2130
},
{
"epoch": 2.92749658002736,
"grad_norm": 1.1582097586418938,
"learning_rate": 1.779424461447965e-08,
"loss": 0.2622,
"step": 2140
},
{
"epoch": 2.9411764705882355,
"grad_norm": 1.1715551395170911,
"learning_rate": 1.1715284122779114e-08,
"loss": 0.261,
"step": 2150
},
{
"epoch": 2.954856361149111,
"grad_norm": 1.14493606362213,
"learning_rate": 6.901023963369336e-09,
"loss": 0.2581,
"step": 2160
},
{
"epoch": 2.9685362517099865,
"grad_norm": 1.2230232820768532,
"learning_rate": 3.3526847154219387e-09,
"loss": 0.2577,
"step": 2170
},
{
"epoch": 2.982216142270862,
"grad_norm": 1.2703211507866685,
"learning_rate": 1.0711660039747219e-09,
"loss": 0.2691,
"step": 2180
},
{
"epoch": 2.9958960328317374,
"grad_norm": 1.1271004957793302,
"learning_rate": 5.704627184577316e-11,
"loss": 0.2604,
"step": 2190
},
{
"epoch": 3.0,
"step": 2193,
"total_flos": 391576352194560.0,
"train_loss": 0.44177393167463547,
"train_runtime": 22000.6442,
"train_samples_per_second": 6.378,
"train_steps_per_second": 0.1
}
],
"logging_steps": 10,
"max_steps": 2193,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 391576352194560.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}