o1_sft_all_abla_numina_oly_orca / trainer_state.json
hanbin's picture
Upload 11 files
f94dccb verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 2691,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.011148272017837236,
"grad_norm": 3.7741404834185763,
"learning_rate": 3.7037037037037036e-07,
"loss": 0.3679,
"step": 10
},
{
"epoch": 0.022296544035674472,
"grad_norm": 1.8040349271595875,
"learning_rate": 7.407407407407407e-07,
"loss": 0.3385,
"step": 20
},
{
"epoch": 0.033444816053511704,
"grad_norm": 1.1311364667628296,
"learning_rate": 1.111111111111111e-06,
"loss": 0.3008,
"step": 30
},
{
"epoch": 0.044593088071348944,
"grad_norm": 1.096280133534896,
"learning_rate": 1.4814814814814815e-06,
"loss": 0.2689,
"step": 40
},
{
"epoch": 0.055741360089186176,
"grad_norm": 0.7223895222097435,
"learning_rate": 1.8518518518518519e-06,
"loss": 0.2506,
"step": 50
},
{
"epoch": 0.06688963210702341,
"grad_norm": 0.6507379554846837,
"learning_rate": 2.222222222222222e-06,
"loss": 0.24,
"step": 60
},
{
"epoch": 0.07803790412486064,
"grad_norm": 0.5228577418159993,
"learning_rate": 2.5925925925925925e-06,
"loss": 0.2374,
"step": 70
},
{
"epoch": 0.08918617614269789,
"grad_norm": 0.5725671090349675,
"learning_rate": 2.962962962962963e-06,
"loss": 0.2343,
"step": 80
},
{
"epoch": 0.10033444816053512,
"grad_norm": 0.575895692421762,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.2279,
"step": 90
},
{
"epoch": 0.11148272017837235,
"grad_norm": 0.5723674270595959,
"learning_rate": 3.7037037037037037e-06,
"loss": 0.2263,
"step": 100
},
{
"epoch": 0.12263099219620958,
"grad_norm": 0.6264644346158856,
"learning_rate": 4.074074074074074e-06,
"loss": 0.2239,
"step": 110
},
{
"epoch": 0.13377926421404682,
"grad_norm": 0.7124762074512575,
"learning_rate": 4.444444444444444e-06,
"loss": 0.2232,
"step": 120
},
{
"epoch": 0.14492753623188406,
"grad_norm": 0.5878673377883962,
"learning_rate": 4.814814814814815e-06,
"loss": 0.2231,
"step": 130
},
{
"epoch": 0.15607580824972128,
"grad_norm": 0.821524645221957,
"learning_rate": 5.185185185185185e-06,
"loss": 0.2215,
"step": 140
},
{
"epoch": 0.16722408026755853,
"grad_norm": 0.6626631852309735,
"learning_rate": 5.555555555555557e-06,
"loss": 0.2203,
"step": 150
},
{
"epoch": 0.17837235228539577,
"grad_norm": 0.7549263527576757,
"learning_rate": 5.925925925925926e-06,
"loss": 0.2186,
"step": 160
},
{
"epoch": 0.189520624303233,
"grad_norm": 0.695178742046847,
"learning_rate": 6.296296296296297e-06,
"loss": 0.2184,
"step": 170
},
{
"epoch": 0.20066889632107024,
"grad_norm": 0.7460261830067356,
"learning_rate": 6.666666666666667e-06,
"loss": 0.22,
"step": 180
},
{
"epoch": 0.21181716833890746,
"grad_norm": 0.7489258972164311,
"learning_rate": 7.0370370370370375e-06,
"loss": 0.2181,
"step": 190
},
{
"epoch": 0.2229654403567447,
"grad_norm": 0.7563607766637321,
"learning_rate": 7.4074074074074075e-06,
"loss": 0.221,
"step": 200
},
{
"epoch": 0.23411371237458195,
"grad_norm": 0.6786891318993898,
"learning_rate": 7.77777777777778e-06,
"loss": 0.2188,
"step": 210
},
{
"epoch": 0.24526198439241917,
"grad_norm": 0.718723752297412,
"learning_rate": 8.148148148148148e-06,
"loss": 0.2191,
"step": 220
},
{
"epoch": 0.2564102564102564,
"grad_norm": 0.79515342774086,
"learning_rate": 8.518518518518519e-06,
"loss": 0.222,
"step": 230
},
{
"epoch": 0.26755852842809363,
"grad_norm": 0.6773955348554602,
"learning_rate": 8.888888888888888e-06,
"loss": 0.2189,
"step": 240
},
{
"epoch": 0.2787068004459309,
"grad_norm": 0.6019195042882624,
"learning_rate": 9.25925925925926e-06,
"loss": 0.2177,
"step": 250
},
{
"epoch": 0.2898550724637681,
"grad_norm": 0.6564030437537817,
"learning_rate": 9.62962962962963e-06,
"loss": 0.2202,
"step": 260
},
{
"epoch": 0.3010033444816054,
"grad_norm": 0.6392190084647338,
"learning_rate": 1e-05,
"loss": 0.217,
"step": 270
},
{
"epoch": 0.31215161649944256,
"grad_norm": 0.6409199129954214,
"learning_rate": 9.999579036849509e-06,
"loss": 0.2184,
"step": 280
},
{
"epoch": 0.3232998885172798,
"grad_norm": 0.6213419108722281,
"learning_rate": 9.998316218282025e-06,
"loss": 0.2165,
"step": 290
},
{
"epoch": 0.33444816053511706,
"grad_norm": 0.6837626955419673,
"learning_rate": 9.996211756937579e-06,
"loss": 0.2153,
"step": 300
},
{
"epoch": 0.3455964325529543,
"grad_norm": 0.6153911123312514,
"learning_rate": 9.993266007176446e-06,
"loss": 0.2183,
"step": 310
},
{
"epoch": 0.35674470457079155,
"grad_norm": 0.6491533849083898,
"learning_rate": 9.989479465019462e-06,
"loss": 0.2168,
"step": 320
},
{
"epoch": 0.36789297658862874,
"grad_norm": 0.6331404269468602,
"learning_rate": 9.984852768064516e-06,
"loss": 0.2164,
"step": 330
},
{
"epoch": 0.379041248606466,
"grad_norm": 0.6066866794436855,
"learning_rate": 9.979386695379179e-06,
"loss": 0.2157,
"step": 340
},
{
"epoch": 0.39018952062430323,
"grad_norm": 0.6283882559447943,
"learning_rate": 9.973082167369521e-06,
"loss": 0.2144,
"step": 350
},
{
"epoch": 0.4013377926421405,
"grad_norm": 0.620915432947406,
"learning_rate": 9.965940245625131e-06,
"loss": 0.2109,
"step": 360
},
{
"epoch": 0.4124860646599777,
"grad_norm": 0.5581175223325301,
"learning_rate": 9.95796213274036e-06,
"loss": 0.2159,
"step": 370
},
{
"epoch": 0.4236343366778149,
"grad_norm": 0.7003333387574542,
"learning_rate": 9.949149172111825e-06,
"loss": 0.2154,
"step": 380
},
{
"epoch": 0.43478260869565216,
"grad_norm": 0.6965889831011358,
"learning_rate": 9.93950284771219e-06,
"loss": 0.215,
"step": 390
},
{
"epoch": 0.4459308807134894,
"grad_norm": 0.6783561953012276,
"learning_rate": 9.929024783840302e-06,
"loss": 0.2156,
"step": 400
},
{
"epoch": 0.45707915273132665,
"grad_norm": 0.6133815858556971,
"learning_rate": 9.917716744847674e-06,
"loss": 0.2135,
"step": 410
},
{
"epoch": 0.4682274247491639,
"grad_norm": 0.5517297679124203,
"learning_rate": 9.905580634841388e-06,
"loss": 0.2127,
"step": 420
},
{
"epoch": 0.4793756967670011,
"grad_norm": 0.5672097715363509,
"learning_rate": 9.892618497363491e-06,
"loss": 0.2152,
"step": 430
},
{
"epoch": 0.49052396878483834,
"grad_norm": 0.7373862211696988,
"learning_rate": 9.878832515046873e-06,
"loss": 0.2121,
"step": 440
},
{
"epoch": 0.5016722408026756,
"grad_norm": 0.5683396526691493,
"learning_rate": 9.864225009247753e-06,
"loss": 0.2124,
"step": 450
},
{
"epoch": 0.5128205128205128,
"grad_norm": 0.5991852837643047,
"learning_rate": 9.848798439654797e-06,
"loss": 0.2116,
"step": 460
},
{
"epoch": 0.5239687848383501,
"grad_norm": 0.5945184596013314,
"learning_rate": 9.832555403874937e-06,
"loss": 0.2107,
"step": 470
},
{
"epoch": 0.5351170568561873,
"grad_norm": 0.5308451359275891,
"learning_rate": 9.815498636995982e-06,
"loss": 0.2092,
"step": 480
},
{
"epoch": 0.5462653288740246,
"grad_norm": 0.5709398866928973,
"learning_rate": 9.797631011126061e-06,
"loss": 0.2098,
"step": 490
},
{
"epoch": 0.5574136008918618,
"grad_norm": 0.570042268652287,
"learning_rate": 9.778955534910004e-06,
"loss": 0.2107,
"step": 500
},
{
"epoch": 0.5574136008918618,
"eval_accuracy": 0.9207988997610126,
"eval_loss": 0.21421977877616882,
"eval_runtime": 5.3128,
"eval_samples_per_second": 43.292,
"eval_steps_per_second": 1.506,
"step": 500
},
{
"epoch": 0.568561872909699,
"grad_norm": 0.6073438181861355,
"learning_rate": 9.759475353022735e-06,
"loss": 0.2096,
"step": 510
},
{
"epoch": 0.5797101449275363,
"grad_norm": 0.5805559872103752,
"learning_rate": 9.739193745639745e-06,
"loss": 0.2088,
"step": 520
},
{
"epoch": 0.5908584169453734,
"grad_norm": 0.569674689944485,
"learning_rate": 9.718114127884774e-06,
"loss": 0.2079,
"step": 530
},
{
"epoch": 0.6020066889632107,
"grad_norm": 0.6479348013496814,
"learning_rate": 9.696240049254744e-06,
"loss": 0.2095,
"step": 540
},
{
"epoch": 0.6131549609810479,
"grad_norm": 0.6837993590888451,
"learning_rate": 9.673575193022073e-06,
"loss": 0.2103,
"step": 550
},
{
"epoch": 0.6243032329988851,
"grad_norm": 0.5658586658235949,
"learning_rate": 9.650123375614477e-06,
"loss": 0.209,
"step": 560
},
{
"epoch": 0.6354515050167224,
"grad_norm": 0.525804021378744,
"learning_rate": 9.625888545972333e-06,
"loss": 0.2109,
"step": 570
},
{
"epoch": 0.6465997770345596,
"grad_norm": 0.5027074127285335,
"learning_rate": 9.600874784883736e-06,
"loss": 0.2072,
"step": 580
},
{
"epoch": 0.6577480490523969,
"grad_norm": 0.5503527245541562,
"learning_rate": 9.575086304297351e-06,
"loss": 0.2068,
"step": 590
},
{
"epoch": 0.6688963210702341,
"grad_norm": 0.5786235649699455,
"learning_rate": 9.548527446613198e-06,
"loss": 0.2088,
"step": 600
},
{
"epoch": 0.6800445930880713,
"grad_norm": 0.5963491322528846,
"learning_rate": 9.521202683951436e-06,
"loss": 0.2079,
"step": 610
},
{
"epoch": 0.6911928651059086,
"grad_norm": 0.5483351877002366,
"learning_rate": 9.493116617399333e-06,
"loss": 0.2086,
"step": 620
},
{
"epoch": 0.7023411371237458,
"grad_norm": 0.5231384119737766,
"learning_rate": 9.464273976236518e-06,
"loss": 0.2079,
"step": 630
},
{
"epoch": 0.7134894091415831,
"grad_norm": 0.5728914844384684,
"learning_rate": 9.434679617138624e-06,
"loss": 0.207,
"step": 640
},
{
"epoch": 0.7246376811594203,
"grad_norm": 0.6056684783613336,
"learning_rate": 9.404338523359511e-06,
"loss": 0.2072,
"step": 650
},
{
"epoch": 0.7357859531772575,
"grad_norm": 0.48483746742809725,
"learning_rate": 9.373255803892149e-06,
"loss": 0.2043,
"step": 660
},
{
"epoch": 0.7469342251950948,
"grad_norm": 0.5052853090283119,
"learning_rate": 9.341436692608341e-06,
"loss": 0.2059,
"step": 670
},
{
"epoch": 0.758082497212932,
"grad_norm": 0.5013961896834831,
"learning_rate": 9.30888654737742e-06,
"loss": 0.2045,
"step": 680
},
{
"epoch": 0.7692307692307693,
"grad_norm": 0.6136168534682885,
"learning_rate": 9.275610849164066e-06,
"loss": 0.2037,
"step": 690
},
{
"epoch": 0.7803790412486065,
"grad_norm": 0.5454932461837725,
"learning_rate": 9.241615201105372e-06,
"loss": 0.2055,
"step": 700
},
{
"epoch": 0.7915273132664437,
"grad_norm": 0.630054882059704,
"learning_rate": 9.206905327567385e-06,
"loss": 0.2084,
"step": 710
},
{
"epoch": 0.802675585284281,
"grad_norm": 0.5115790207210278,
"learning_rate": 9.171487073181198e-06,
"loss": 0.2054,
"step": 720
},
{
"epoch": 0.8138238573021181,
"grad_norm": 0.522185957547097,
"learning_rate": 9.135366401858782e-06,
"loss": 0.2073,
"step": 730
},
{
"epoch": 0.8249721293199554,
"grad_norm": 0.5370558619096407,
"learning_rate": 9.098549395788784e-06,
"loss": 0.2073,
"step": 740
},
{
"epoch": 0.8361204013377926,
"grad_norm": 0.5143052735973912,
"learning_rate": 9.061042254412346e-06,
"loss": 0.2039,
"step": 750
},
{
"epoch": 0.8472686733556298,
"grad_norm": 0.5271311850347303,
"learning_rate": 9.022851293379232e-06,
"loss": 0.2053,
"step": 760
},
{
"epoch": 0.8584169453734671,
"grad_norm": 0.54232371019945,
"learning_rate": 8.983982943484348e-06,
"loss": 0.2023,
"step": 770
},
{
"epoch": 0.8695652173913043,
"grad_norm": 0.5365286519965239,
"learning_rate": 8.944443749584908e-06,
"loss": 0.2043,
"step": 780
},
{
"epoch": 0.8807134894091416,
"grad_norm": 0.49203900940366796,
"learning_rate": 8.904240369498363e-06,
"loss": 0.2013,
"step": 790
},
{
"epoch": 0.8918617614269788,
"grad_norm": 0.6067905351468151,
"learning_rate": 8.863379572881328e-06,
"loss": 0.2038,
"step": 800
},
{
"epoch": 0.903010033444816,
"grad_norm": 0.5687299547717642,
"learning_rate": 8.821868240089676e-06,
"loss": 0.2044,
"step": 810
},
{
"epoch": 0.9141583054626533,
"grad_norm": 0.5124900416142478,
"learning_rate": 8.779713361019983e-06,
"loss": 0.2068,
"step": 820
},
{
"epoch": 0.9253065774804905,
"grad_norm": 0.48310445826631493,
"learning_rate": 8.736922033932522e-06,
"loss": 0.2022,
"step": 830
},
{
"epoch": 0.9364548494983278,
"grad_norm": 0.5272157184999492,
"learning_rate": 8.693501464256046e-06,
"loss": 0.2027,
"step": 840
},
{
"epoch": 0.947603121516165,
"grad_norm": 0.5270482869515378,
"learning_rate": 8.649458963374474e-06,
"loss": 0.2032,
"step": 850
},
{
"epoch": 0.9587513935340022,
"grad_norm": 0.5738980238907102,
"learning_rate": 8.604801947395776e-06,
"loss": 0.2,
"step": 860
},
{
"epoch": 0.9698996655518395,
"grad_norm": 0.4603491880478103,
"learning_rate": 8.55953793590321e-06,
"loss": 0.2021,
"step": 870
},
{
"epoch": 0.9810479375696767,
"grad_norm": 0.548549890115731,
"learning_rate": 8.513674550689128e-06,
"loss": 0.2002,
"step": 880
},
{
"epoch": 0.992196209587514,
"grad_norm": 0.49756226702185397,
"learning_rate": 8.46721951447158e-06,
"loss": 0.2024,
"step": 890
},
{
"epoch": 1.0033444816053512,
"grad_norm": 0.6472920299598118,
"learning_rate": 8.42018064959393e-06,
"loss": 0.1903,
"step": 900
},
{
"epoch": 1.0144927536231885,
"grad_norm": 0.5997372944954878,
"learning_rate": 8.37256587670768e-06,
"loss": 0.1599,
"step": 910
},
{
"epoch": 1.0256410256410255,
"grad_norm": 0.6661365745997817,
"learning_rate": 8.32438321343875e-06,
"loss": 0.1578,
"step": 920
},
{
"epoch": 1.0367892976588629,
"grad_norm": 0.5526716468068121,
"learning_rate": 8.275640773037432e-06,
"loss": 0.1584,
"step": 930
},
{
"epoch": 1.0479375696767002,
"grad_norm": 0.5311689895480008,
"learning_rate": 8.226346763012234e-06,
"loss": 0.1591,
"step": 940
},
{
"epoch": 1.0590858416945372,
"grad_norm": 0.6216949836162623,
"learning_rate": 8.176509483747863e-06,
"loss": 0.1587,
"step": 950
},
{
"epoch": 1.0702341137123745,
"grad_norm": 0.5643715674767006,
"learning_rate": 8.126137327107556e-06,
"loss": 0.1559,
"step": 960
},
{
"epoch": 1.0813823857302118,
"grad_norm": 0.528226825221165,
"learning_rate": 8.075238775020011e-06,
"loss": 0.1582,
"step": 970
},
{
"epoch": 1.0925306577480491,
"grad_norm": 0.5303123877825515,
"learning_rate": 8.023822398051168e-06,
"loss": 0.1588,
"step": 980
},
{
"epoch": 1.1036789297658862,
"grad_norm": 0.5437405641707471,
"learning_rate": 7.971896853961043e-06,
"loss": 0.1576,
"step": 990
},
{
"epoch": 1.1148272017837235,
"grad_norm": 0.5691619401154729,
"learning_rate": 7.919470886245886e-06,
"loss": 0.1561,
"step": 1000
},
{
"epoch": 1.1148272017837235,
"eval_accuracy": 0.9239298004895466,
"eval_loss": 0.20845723152160645,
"eval_runtime": 4.947,
"eval_samples_per_second": 46.493,
"eval_steps_per_second": 1.617,
"step": 1000
},
{
"epoch": 1.1259754738015608,
"grad_norm": 0.47581750607341355,
"learning_rate": 7.86655332266591e-06,
"loss": 0.1586,
"step": 1010
},
{
"epoch": 1.137123745819398,
"grad_norm": 0.6353712929162172,
"learning_rate": 7.813153073758833e-06,
"loss": 0.1582,
"step": 1020
},
{
"epoch": 1.1482720178372352,
"grad_norm": 0.5419698538850736,
"learning_rate": 7.759279131339455e-06,
"loss": 0.1579,
"step": 1030
},
{
"epoch": 1.1594202898550725,
"grad_norm": 0.5730038786065867,
"learning_rate": 7.704940566985592e-06,
"loss": 0.163,
"step": 1040
},
{
"epoch": 1.1705685618729098,
"grad_norm": 0.5280364919184873,
"learning_rate": 7.650146530510542e-06,
"loss": 0.1617,
"step": 1050
},
{
"epoch": 1.1817168338907469,
"grad_norm": 0.544836252028228,
"learning_rate": 7.594906248422392e-06,
"loss": 0.1596,
"step": 1060
},
{
"epoch": 1.1928651059085842,
"grad_norm": 0.5378141986574141,
"learning_rate": 7.539229022370418e-06,
"loss": 0.1603,
"step": 1070
},
{
"epoch": 1.2040133779264215,
"grad_norm": 0.5500474096682983,
"learning_rate": 7.483124227578811e-06,
"loss": 0.1579,
"step": 1080
},
{
"epoch": 1.2151616499442586,
"grad_norm": 0.5018225238945918,
"learning_rate": 7.426601311268043e-06,
"loss": 0.16,
"step": 1090
},
{
"epoch": 1.2263099219620959,
"grad_norm": 0.575725285027047,
"learning_rate": 7.369669791064085e-06,
"loss": 0.1596,
"step": 1100
},
{
"epoch": 1.2374581939799332,
"grad_norm": 0.47947733030163836,
"learning_rate": 7.312339253395778e-06,
"loss": 0.1582,
"step": 1110
},
{
"epoch": 1.2486064659977703,
"grad_norm": 0.4980356373309633,
"learning_rate": 7.254619351880625e-06,
"loss": 0.1581,
"step": 1120
},
{
"epoch": 1.2597547380156076,
"grad_norm": 0.578756138117811,
"learning_rate": 7.1965198056992615e-06,
"loss": 0.1592,
"step": 1130
},
{
"epoch": 1.2709030100334449,
"grad_norm": 0.4635139388152137,
"learning_rate": 7.138050397958889e-06,
"loss": 0.1595,
"step": 1140
},
{
"epoch": 1.282051282051282,
"grad_norm": 0.49482297151736354,
"learning_rate": 7.079220974045941e-06,
"loss": 0.1584,
"step": 1150
},
{
"epoch": 1.2931995540691192,
"grad_norm": 0.5460979920247649,
"learning_rate": 7.0200414399682745e-06,
"loss": 0.1584,
"step": 1160
},
{
"epoch": 1.3043478260869565,
"grad_norm": 0.48148732576426634,
"learning_rate": 6.96052176068713e-06,
"loss": 0.1576,
"step": 1170
},
{
"epoch": 1.3154960981047936,
"grad_norm": 0.5821793738001766,
"learning_rate": 6.900671958439192e-06,
"loss": 0.1599,
"step": 1180
},
{
"epoch": 1.326644370122631,
"grad_norm": 0.48883238562313075,
"learning_rate": 6.840502111048982e-06,
"loss": 0.158,
"step": 1190
},
{
"epoch": 1.3377926421404682,
"grad_norm": 0.5974171958012616,
"learning_rate": 6.780022350231912e-06,
"loss": 0.1604,
"step": 1200
},
{
"epoch": 1.3489409141583055,
"grad_norm": 0.5659057198248688,
"learning_rate": 6.719242859888243e-06,
"loss": 0.1576,
"step": 1210
},
{
"epoch": 1.3600891861761428,
"grad_norm": 0.6417774138796649,
"learning_rate": 6.65817387438827e-06,
"loss": 0.1588,
"step": 1220
},
{
"epoch": 1.37123745819398,
"grad_norm": 0.5414201206440971,
"learning_rate": 6.596825676849006e-06,
"loss": 0.1581,
"step": 1230
},
{
"epoch": 1.3823857302118172,
"grad_norm": 0.513016285464973,
"learning_rate": 6.535208597402658e-06,
"loss": 0.1588,
"step": 1240
},
{
"epoch": 1.3935340022296545,
"grad_norm": 0.5513989056143007,
"learning_rate": 6.473333011457181e-06,
"loss": 0.1586,
"step": 1250
},
{
"epoch": 1.4046822742474916,
"grad_norm": 0.5049419184837916,
"learning_rate": 6.411209337949214e-06,
"loss": 0.1594,
"step": 1260
},
{
"epoch": 1.415830546265329,
"grad_norm": 0.5217098963627814,
"learning_rate": 6.348848037589683e-06,
"loss": 0.159,
"step": 1270
},
{
"epoch": 1.4269788182831662,
"grad_norm": 0.5169737196009081,
"learning_rate": 6.286259611102376e-06,
"loss": 0.1589,
"step": 1280
},
{
"epoch": 1.4381270903010033,
"grad_norm": 0.5747675813836013,
"learning_rate": 6.223454597455776e-06,
"loss": 0.1587,
"step": 1290
},
{
"epoch": 1.4492753623188406,
"grad_norm": 0.5345109793646367,
"learning_rate": 6.160443572088443e-06,
"loss": 0.1595,
"step": 1300
},
{
"epoch": 1.4604236343366779,
"grad_norm": 0.49324906218088477,
"learning_rate": 6.097237145128282e-06,
"loss": 0.1584,
"step": 1310
},
{
"epoch": 1.471571906354515,
"grad_norm": 0.5093141609186953,
"learning_rate": 6.0338459596059395e-06,
"loss": 0.1554,
"step": 1320
},
{
"epoch": 1.4827201783723523,
"grad_norm": 0.509538948244533,
"learning_rate": 5.9702806896626855e-06,
"loss": 0.1585,
"step": 1330
},
{
"epoch": 1.4938684503901896,
"grad_norm": 0.6369579950149653,
"learning_rate": 5.90655203875304e-06,
"loss": 0.159,
"step": 1340
},
{
"epoch": 1.5050167224080266,
"grad_norm": 0.47625480890424343,
"learning_rate": 5.8426707378424675e-06,
"loss": 0.1579,
"step": 1350
},
{
"epoch": 1.516164994425864,
"grad_norm": 0.5438272649779904,
"learning_rate": 5.778647543600443e-06,
"loss": 0.1564,
"step": 1360
},
{
"epoch": 1.5273132664437012,
"grad_norm": 0.5161276266998152,
"learning_rate": 5.714493236589187e-06,
"loss": 0.1545,
"step": 1370
},
{
"epoch": 1.5384615384615383,
"grad_norm": 0.5315157886776762,
"learning_rate": 5.650218619448379e-06,
"loss": 0.1552,
"step": 1380
},
{
"epoch": 1.5496098104793758,
"grad_norm": 0.5390188368744134,
"learning_rate": 5.5858345150761515e-06,
"loss": 0.1556,
"step": 1390
},
{
"epoch": 1.560758082497213,
"grad_norm": 0.5357636181743068,
"learning_rate": 5.521351764806672e-06,
"loss": 0.1568,
"step": 1400
},
{
"epoch": 1.57190635451505,
"grad_norm": 0.4874443181407851,
"learning_rate": 5.456781226584621e-06,
"loss": 0.158,
"step": 1410
},
{
"epoch": 1.5830546265328875,
"grad_norm": 0.4563465609878918,
"learning_rate": 5.392133773136877e-06,
"loss": 0.1602,
"step": 1420
},
{
"epoch": 1.5942028985507246,
"grad_norm": 0.493572716982426,
"learning_rate": 5.327420290141712e-06,
"loss": 0.1571,
"step": 1430
},
{
"epoch": 1.605351170568562,
"grad_norm": 0.5297224295489479,
"learning_rate": 5.262651674395799e-06,
"loss": 0.1567,
"step": 1440
},
{
"epoch": 1.6164994425863992,
"grad_norm": 0.4990830580335757,
"learning_rate": 5.197838831979352e-06,
"loss": 0.1574,
"step": 1450
},
{
"epoch": 1.6276477146042363,
"grad_norm": 0.4877320539367503,
"learning_rate": 5.132992676419704e-06,
"loss": 0.1569,
"step": 1460
},
{
"epoch": 1.6387959866220736,
"grad_norm": 0.546166024878597,
"learning_rate": 5.068124126853633e-06,
"loss": 0.157,
"step": 1470
},
{
"epoch": 1.649944258639911,
"grad_norm": 0.4770361938111255,
"learning_rate": 5.0032441061887345e-06,
"loss": 0.1569,
"step": 1480
},
{
"epoch": 1.661092530657748,
"grad_norm": 0.44557527525358764,
"learning_rate": 4.938363539264175e-06,
"loss": 0.1571,
"step": 1490
},
{
"epoch": 1.6722408026755853,
"grad_norm": 0.48803538459110823,
"learning_rate": 4.87349335101109e-06,
"loss": 0.1547,
"step": 1500
},
{
"epoch": 1.6722408026755853,
"eval_accuracy": 0.9265380200765034,
"eval_loss": 0.19943761825561523,
"eval_runtime": 5.3541,
"eval_samples_per_second": 42.958,
"eval_steps_per_second": 1.494,
"step": 1500
},
{
"epoch": 1.6833890746934226,
"grad_norm": 0.5076348755331102,
"learning_rate": 4.808644464613015e-06,
"loss": 0.1562,
"step": 1510
},
{
"epoch": 1.6945373467112597,
"grad_norm": 0.501612702783229,
"learning_rate": 4.743827799666555e-06,
"loss": 0.1547,
"step": 1520
},
{
"epoch": 1.705685618729097,
"grad_norm": 0.4879807944214394,
"learning_rate": 4.679054270342703e-06,
"loss": 0.1563,
"step": 1530
},
{
"epoch": 1.7168338907469343,
"grad_norm": 0.46616141709578995,
"learning_rate": 4.614334783549049e-06,
"loss": 0.1538,
"step": 1540
},
{
"epoch": 1.7279821627647713,
"grad_norm": 0.5293473205445773,
"learning_rate": 4.549680237093215e-06,
"loss": 0.1571,
"step": 1550
},
{
"epoch": 1.7391304347826086,
"grad_norm": 0.4768030365700139,
"learning_rate": 4.485101517847831e-06,
"loss": 0.1561,
"step": 1560
},
{
"epoch": 1.750278706800446,
"grad_norm": 0.5395794005148075,
"learning_rate": 4.420609499917337e-06,
"loss": 0.1555,
"step": 1570
},
{
"epoch": 1.761426978818283,
"grad_norm": 0.4879668723657388,
"learning_rate": 4.3562150428069565e-06,
"loss": 0.1545,
"step": 1580
},
{
"epoch": 1.7725752508361206,
"grad_norm": 0.4939707212729394,
"learning_rate": 4.291928989594102e-06,
"loss": 0.1547,
"step": 1590
},
{
"epoch": 1.7837235228539576,
"grad_norm": 0.4807444627751433,
"learning_rate": 4.2277621651025734e-06,
"loss": 0.1548,
"step": 1600
},
{
"epoch": 1.7948717948717947,
"grad_norm": 0.49281674069166814,
"learning_rate": 4.163725374079808e-06,
"loss": 0.1559,
"step": 1610
},
{
"epoch": 1.8060200668896322,
"grad_norm": 0.52027039374057,
"learning_rate": 4.099829399377524e-06,
"loss": 0.1527,
"step": 1620
},
{
"epoch": 1.8171683389074693,
"grad_norm": 0.4693230544446258,
"learning_rate": 4.036085000136046e-06,
"loss": 0.1562,
"step": 1630
},
{
"epoch": 1.8283166109253066,
"grad_norm": 0.5063480473756882,
"learning_rate": 3.972502909972628e-06,
"loss": 0.1546,
"step": 1640
},
{
"epoch": 1.839464882943144,
"grad_norm": 0.4816499638368781,
"learning_rate": 3.909093835174066e-06,
"loss": 0.1547,
"step": 1650
},
{
"epoch": 1.850613154960981,
"grad_norm": 0.64084638347163,
"learning_rate": 3.845868452893919e-06,
"loss": 0.1546,
"step": 1660
},
{
"epoch": 1.8617614269788183,
"grad_norm": 0.5316728198827755,
"learning_rate": 3.7828374093546303e-06,
"loss": 0.1526,
"step": 1670
},
{
"epoch": 1.8729096989966556,
"grad_norm": 0.47782006941247335,
"learning_rate": 3.720011318054871e-06,
"loss": 0.1542,
"step": 1680
},
{
"epoch": 1.8840579710144927,
"grad_norm": 0.4960856794921382,
"learning_rate": 3.657400757982367e-06,
"loss": 0.1543,
"step": 1690
},
{
"epoch": 1.89520624303233,
"grad_norm": 0.5205359533145413,
"learning_rate": 3.595016271832572e-06,
"loss": 0.1531,
"step": 1700
},
{
"epoch": 1.9063545150501673,
"grad_norm": 0.5250980924250916,
"learning_rate": 3.532868364233416e-06,
"loss": 0.1541,
"step": 1710
},
{
"epoch": 1.9175027870680044,
"grad_norm": 0.4552911079184954,
"learning_rate": 3.470967499976492e-06,
"loss": 0.1552,
"step": 1720
},
{
"epoch": 1.9286510590858417,
"grad_norm": 0.47374387614448155,
"learning_rate": 3.4093241022549316e-06,
"loss": 0.1521,
"step": 1730
},
{
"epoch": 1.939799331103679,
"grad_norm": 0.4809046264515791,
"learning_rate": 3.347948550908303e-06,
"loss": 0.1536,
"step": 1740
},
{
"epoch": 1.950947603121516,
"grad_norm": 0.4611916983834658,
"learning_rate": 3.286851180674788e-06,
"loss": 0.1524,
"step": 1750
},
{
"epoch": 1.9620958751393534,
"grad_norm": 0.4496375299269273,
"learning_rate": 3.2260422794509704e-06,
"loss": 0.1527,
"step": 1760
},
{
"epoch": 1.9732441471571907,
"grad_norm": 0.49594296693835754,
"learning_rate": 3.165532086559504e-06,
"loss": 0.1523,
"step": 1770
},
{
"epoch": 1.9843924191750277,
"grad_norm": 0.4311413917530802,
"learning_rate": 3.105330791024965e-06,
"loss": 0.1528,
"step": 1780
},
{
"epoch": 1.9955406911928653,
"grad_norm": 0.46002893870709255,
"learning_rate": 3.045448529858165e-06,
"loss": 0.1523,
"step": 1790
},
{
"epoch": 2.0066889632107023,
"grad_norm": 0.6640864376769408,
"learning_rate": 2.9858953863492334e-06,
"loss": 0.1257,
"step": 1800
},
{
"epoch": 2.0178372352285394,
"grad_norm": 0.5065513503070602,
"learning_rate": 2.9266813883697342e-06,
"loss": 0.111,
"step": 1810
},
{
"epoch": 2.028985507246377,
"grad_norm": 0.5422041920978169,
"learning_rate": 2.867816506684126e-06,
"loss": 0.1125,
"step": 1820
},
{
"epoch": 2.040133779264214,
"grad_norm": 0.5207289779266513,
"learning_rate": 2.809310653270825e-06,
"loss": 0.1113,
"step": 1830
},
{
"epoch": 2.051282051282051,
"grad_norm": 0.4849357576510085,
"learning_rate": 2.751173679653184e-06,
"loss": 0.1101,
"step": 1840
},
{
"epoch": 2.0624303232998886,
"grad_norm": 0.47414870888584787,
"learning_rate": 2.693415375240629e-06,
"loss": 0.1099,
"step": 1850
},
{
"epoch": 2.0735785953177257,
"grad_norm": 0.47845407787457195,
"learning_rate": 2.636045465680282e-06,
"loss": 0.1098,
"step": 1860
},
{
"epoch": 2.084726867335563,
"grad_norm": 0.5151744285005124,
"learning_rate": 2.5790736112192893e-06,
"loss": 0.1094,
"step": 1870
},
{
"epoch": 2.0958751393534003,
"grad_norm": 0.44908780393249104,
"learning_rate": 2.522509405078187e-06,
"loss": 0.1105,
"step": 1880
},
{
"epoch": 2.1070234113712374,
"grad_norm": 0.46168216362984565,
"learning_rate": 2.4663623718355444e-06,
"loss": 0.1102,
"step": 1890
},
{
"epoch": 2.1181716833890745,
"grad_norm": 0.5003674108234805,
"learning_rate": 2.410641965824166e-06,
"loss": 0.1089,
"step": 1900
},
{
"epoch": 2.129319955406912,
"grad_norm": 0.47533896359509586,
"learning_rate": 2.355357569539114e-06,
"loss": 0.1119,
"step": 1910
},
{
"epoch": 2.140468227424749,
"grad_norm": 0.4874105883605131,
"learning_rate": 2.300518492057842e-06,
"loss": 0.114,
"step": 1920
},
{
"epoch": 2.1516164994425866,
"grad_norm": 0.5034128167417606,
"learning_rate": 2.2461339674726804e-06,
"loss": 0.1102,
"step": 1930
},
{
"epoch": 2.1627647714604237,
"grad_norm": 0.4772612798475211,
"learning_rate": 2.192213153335953e-06,
"loss": 0.1109,
"step": 1940
},
{
"epoch": 2.1739130434782608,
"grad_norm": 0.4919335114976622,
"learning_rate": 2.1387651291179775e-06,
"loss": 0.1106,
"step": 1950
},
{
"epoch": 2.1850613154960983,
"grad_norm": 0.4893367756986907,
"learning_rate": 2.085798894678217e-06,
"loss": 0.1089,
"step": 1960
},
{
"epoch": 2.1962095875139354,
"grad_norm": 0.5106660638111142,
"learning_rate": 2.0333233687498433e-06,
"loss": 0.1094,
"step": 1970
},
{
"epoch": 2.2073578595317724,
"grad_norm": 0.4698835768687655,
"learning_rate": 1.98134738743794e-06,
"loss": 0.1105,
"step": 1980
},
{
"epoch": 2.21850613154961,
"grad_norm": 0.481190970836223,
"learning_rate": 1.9298797027316474e-06,
"loss": 0.1088,
"step": 1990
},
{
"epoch": 2.229654403567447,
"grad_norm": 0.4828488140066944,
"learning_rate": 1.878928981030445e-06,
"loss": 0.1092,
"step": 2000
},
{
"epoch": 2.229654403567447,
"eval_accuracy": 0.9278099495985295,
"eval_loss": 0.20725835859775543,
"eval_runtime": 4.9127,
"eval_samples_per_second": 46.817,
"eval_steps_per_second": 1.628,
"step": 2000
},
{
"epoch": 2.240802675585284,
"grad_norm": 0.5445446664477709,
"learning_rate": 1.828503801684864e-06,
"loss": 0.1098,
"step": 2010
},
{
"epoch": 2.2519509476031216,
"grad_norm": 0.4945888022659562,
"learning_rate": 1.7786126555518484e-06,
"loss": 0.1093,
"step": 2020
},
{
"epoch": 2.2630992196209587,
"grad_norm": 0.4728964267184926,
"learning_rate": 1.729263943565022e-06,
"loss": 0.1088,
"step": 2030
},
{
"epoch": 2.274247491638796,
"grad_norm": 0.46594045141875756,
"learning_rate": 1.6804659753200925e-06,
"loss": 0.1101,
"step": 2040
},
{
"epoch": 2.2853957636566333,
"grad_norm": 0.4731193453792897,
"learning_rate": 1.6322269676756402e-06,
"loss": 0.1096,
"step": 2050
},
{
"epoch": 2.2965440356744704,
"grad_norm": 0.46507870631813036,
"learning_rate": 1.5845550433695172e-06,
"loss": 0.109,
"step": 2060
},
{
"epoch": 2.3076923076923075,
"grad_norm": 0.5041104454052192,
"learning_rate": 1.5374582296511054e-06,
"loss": 0.1097,
"step": 2070
},
{
"epoch": 2.318840579710145,
"grad_norm": 0.4506495588973169,
"learning_rate": 1.4909444569296334e-06,
"loss": 0.1084,
"step": 2080
},
{
"epoch": 2.329988851727982,
"grad_norm": 0.5081246384541027,
"learning_rate": 1.4450215574388265e-06,
"loss": 0.1094,
"step": 2090
},
{
"epoch": 2.3411371237458196,
"grad_norm": 0.449952731509518,
"learning_rate": 1.3996972639180645e-06,
"loss": 0.1086,
"step": 2100
},
{
"epoch": 2.3522853957636567,
"grad_norm": 0.5009677565037799,
"learning_rate": 1.3549792083103037e-06,
"loss": 0.1079,
"step": 2110
},
{
"epoch": 2.3634336677814938,
"grad_norm": 0.4544414237147968,
"learning_rate": 1.3108749204769732e-06,
"loss": 0.1093,
"step": 2120
},
{
"epoch": 2.374581939799331,
"grad_norm": 0.476556243372622,
"learning_rate": 1.2673918269300557e-06,
"loss": 0.109,
"step": 2130
},
{
"epoch": 2.3857302118171684,
"grad_norm": 0.48618656512125896,
"learning_rate": 1.2245372495815726e-06,
"loss": 0.1086,
"step": 2140
},
{
"epoch": 2.3968784838350055,
"grad_norm": 0.48915278085562003,
"learning_rate": 1.1823184045106816e-06,
"loss": 0.1088,
"step": 2150
},
{
"epoch": 2.408026755852843,
"grad_norm": 0.4612765468912016,
"learning_rate": 1.140742400748593e-06,
"loss": 0.1085,
"step": 2160
},
{
"epoch": 2.41917502787068,
"grad_norm": 0.49881271267886607,
"learning_rate": 1.099816239081521e-06,
"loss": 0.1084,
"step": 2170
},
{
"epoch": 2.430323299888517,
"grad_norm": 0.47510269822553847,
"learning_rate": 1.0595468108718448e-06,
"loss": 0.1089,
"step": 2180
},
{
"epoch": 2.4414715719063547,
"grad_norm": 0.47369643358112645,
"learning_rate": 1.0199408968977138e-06,
"loss": 0.1087,
"step": 2190
},
{
"epoch": 2.4526198439241917,
"grad_norm": 0.47267131195325657,
"learning_rate": 9.810051662112558e-07,
"loss": 0.108,
"step": 2200
},
{
"epoch": 2.463768115942029,
"grad_norm": 0.45402353633897424,
"learning_rate": 9.427461750156142e-07,
"loss": 0.1096,
"step": 2210
},
{
"epoch": 2.4749163879598663,
"grad_norm": 0.46683292783796754,
"learning_rate": 9.051703655609762e-07,
"loss": 0.1089,
"step": 2220
},
{
"epoch": 2.4860646599777034,
"grad_norm": 0.4654495260542525,
"learning_rate": 8.68284065059794e-07,
"loss": 0.1095,
"step": 2230
},
{
"epoch": 2.4972129319955405,
"grad_norm": 0.4819165766260374,
"learning_rate": 8.320934846213746e-07,
"loss": 0.1082,
"step": 2240
},
{
"epoch": 2.508361204013378,
"grad_norm": 0.4582207481405345,
"learning_rate": 7.966047182060226e-07,
"loss": 0.1083,
"step": 2250
},
{
"epoch": 2.519509476031215,
"grad_norm": 0.44176541619456144,
"learning_rate": 7.618237415989032e-07,
"loss": 0.1083,
"step": 2260
},
{
"epoch": 2.5306577480490526,
"grad_norm": 0.45340706173810835,
"learning_rate": 7.277564114038149e-07,
"loss": 0.1082,
"step": 2270
},
{
"epoch": 2.5418060200668897,
"grad_norm": 0.4607933599342553,
"learning_rate": 6.944084640570142e-07,
"loss": 0.1072,
"step": 2280
},
{
"epoch": 2.552954292084727,
"grad_norm": 0.4613886801061754,
"learning_rate": 6.617855148612945e-07,
"loss": 0.1071,
"step": 2290
},
{
"epoch": 2.564102564102564,
"grad_norm": 0.4490390133193121,
"learning_rate": 6.298930570404432e-07,
"loss": 0.1087,
"step": 2300
},
{
"epoch": 2.5752508361204014,
"grad_norm": 0.4561285734622287,
"learning_rate": 5.987364608142693e-07,
"loss": 0.1076,
"step": 2310
},
{
"epoch": 2.5863991081382385,
"grad_norm": 0.461278673680382,
"learning_rate": 5.683209724943344e-07,
"loss": 0.109,
"step": 2320
},
{
"epoch": 2.597547380156076,
"grad_norm": 0.45105628731245295,
"learning_rate": 5.386517136005543e-07,
"loss": 0.1081,
"step": 2330
},
{
"epoch": 2.608695652173913,
"grad_norm": 0.438445525307319,
"learning_rate": 5.097336799988067e-07,
"loss": 0.1079,
"step": 2340
},
{
"epoch": 2.61984392419175,
"grad_norm": 0.4457791687565759,
"learning_rate": 4.815717410597042e-07,
"loss": 0.1081,
"step": 2350
},
{
"epoch": 2.6309921962095872,
"grad_norm": 0.44606421814267033,
"learning_rate": 4.541706388386624e-07,
"loss": 0.1086,
"step": 2360
},
{
"epoch": 2.6421404682274248,
"grad_norm": 0.4398808910735919,
"learning_rate": 4.275349872774098e-07,
"loss": 0.1073,
"step": 2370
},
{
"epoch": 2.653288740245262,
"grad_norm": 0.4522695550459701,
"learning_rate": 4.0166927142706313e-07,
"loss": 0.1076,
"step": 2380
},
{
"epoch": 2.6644370122630994,
"grad_norm": 0.4436224867256293,
"learning_rate": 3.765778466929182e-07,
"loss": 0.1066,
"step": 2390
},
{
"epoch": 2.6755852842809364,
"grad_norm": 0.428940667218784,
"learning_rate": 3.5226493810105624e-07,
"loss": 0.1081,
"step": 2400
},
{
"epoch": 2.6867335562987735,
"grad_norm": 0.45631649150745296,
"learning_rate": 3.2873463958691675e-07,
"loss": 0.108,
"step": 2410
},
{
"epoch": 2.697881828316611,
"grad_norm": 0.4407463233212753,
"learning_rate": 3.05990913305938e-07,
"loss": 0.1077,
"step": 2420
},
{
"epoch": 2.709030100334448,
"grad_norm": 0.45899753246408376,
"learning_rate": 2.840375889663871e-07,
"loss": 0.106,
"step": 2430
},
{
"epoch": 2.7201783723522857,
"grad_norm": 0.4620830771843678,
"learning_rate": 2.62878363184495e-07,
"loss": 0.107,
"step": 2440
},
{
"epoch": 2.7313266443701227,
"grad_norm": 0.4540560827599527,
"learning_rate": 2.425167988620014e-07,
"loss": 0.1078,
"step": 2450
},
{
"epoch": 2.74247491638796,
"grad_norm": 0.4379908641559117,
"learning_rate": 2.2295632458621097e-07,
"loss": 0.1076,
"step": 2460
},
{
"epoch": 2.753623188405797,
"grad_norm": 0.44868425706880766,
"learning_rate": 2.0420023405267663e-07,
"loss": 0.1076,
"step": 2470
},
{
"epoch": 2.7647714604236344,
"grad_norm": 0.47473094706270963,
"learning_rate": 1.8625168551058115e-07,
"loss": 0.1071,
"step": 2480
},
{
"epoch": 2.7759197324414715,
"grad_norm": 0.45861955063568105,
"learning_rate": 1.6911370123094238e-07,
"loss": 0.1081,
"step": 2490
},
{
"epoch": 2.787068004459309,
"grad_norm": 0.45384731541579043,
"learning_rate": 1.5278916699770163e-07,
"loss": 0.1073,
"step": 2500
},
{
"epoch": 2.787068004459309,
"eval_accuracy": 0.9283863306145088,
"eval_loss": 0.20636913180351257,
"eval_runtime": 5.1195,
"eval_samples_per_second": 44.926,
"eval_steps_per_second": 1.563,
"step": 2500
},
{
"epoch": 2.798216276477146,
"grad_norm": 0.4773943035056343,
"learning_rate": 1.3728083162180384e-07,
"loss": 0.1075,
"step": 2510
},
{
"epoch": 2.809364548494983,
"grad_norm": 0.428735968014936,
"learning_rate": 1.2259130647833627e-07,
"loss": 0.1065,
"step": 2520
},
{
"epoch": 2.8205128205128203,
"grad_norm": 0.4385864949658292,
"learning_rate": 1.0872306506681251e-07,
"loss": 0.1079,
"step": 2530
},
{
"epoch": 2.831661092530658,
"grad_norm": 0.4743582222874101,
"learning_rate": 9.567844259467051e-08,
"loss": 0.1083,
"step": 2540
},
{
"epoch": 2.842809364548495,
"grad_norm": 0.45862196445217895,
"learning_rate": 8.345963558406001e-08,
"loss": 0.1079,
"step": 2550
},
{
"epoch": 2.8539576365663324,
"grad_norm": 0.45558482700348846,
"learning_rate": 7.206870150197831e-08,
"loss": 0.107,
"step": 2560
},
{
"epoch": 2.8651059085841695,
"grad_norm": 0.4458457370599944,
"learning_rate": 6.150755841382527e-08,
"loss": 0.1067,
"step": 2570
},
{
"epoch": 2.8762541806020065,
"grad_norm": 0.44342362029660554,
"learning_rate": 5.177798466042716e-08,
"loss": 0.1078,
"step": 2580
},
{
"epoch": 2.887402452619844,
"grad_norm": 0.4369582199234232,
"learning_rate": 4.2881618558592855e-08,
"loss": 0.109,
"step": 2590
},
{
"epoch": 2.898550724637681,
"grad_norm": 0.4355601557089695,
"learning_rate": 3.481995812524286e-08,
"loss": 0.1064,
"step": 2600
},
{
"epoch": 2.9096989966555182,
"grad_norm": 0.4377173417059771,
"learning_rate": 2.7594360825166644e-08,
"loss": 0.107,
"step": 2610
},
{
"epoch": 2.9208472686733558,
"grad_norm": 0.4495055773584237,
"learning_rate": 2.120604334244547e-08,
"loss": 0.1067,
"step": 2620
},
{
"epoch": 2.931995540691193,
"grad_norm": 0.44729903103488183,
"learning_rate": 1.565608137558128e-08,
"loss": 0.1063,
"step": 2630
},
{
"epoch": 2.94314381270903,
"grad_norm": 0.44620730977493767,
"learning_rate": 1.0945409456364353e-08,
"loss": 0.1075,
"step": 2640
},
{
"epoch": 2.9542920847268674,
"grad_norm": 0.46052917967733736,
"learning_rate": 7.0748207925103176e-09,
"loss": 0.1075,
"step": 2650
},
{
"epoch": 2.9654403567447045,
"grad_norm": 0.4450651551983087,
"learning_rate": 4.044967134099187e-09,
"loss": 0.1079,
"step": 2660
},
{
"epoch": 2.976588628762542,
"grad_norm": 0.4581682413498136,
"learning_rate": 1.8563586638281617e-09,
"loss": 0.1064,
"step": 2670
},
{
"epoch": 2.987736900780379,
"grad_norm": 0.44561324878780834,
"learning_rate": 5.093639111025673e-10,
"loss": 0.1071,
"step": 2680
},
{
"epoch": 2.998885172798216,
"grad_norm": 0.46042910026727557,
"learning_rate": 4.2096899854904285e-12,
"loss": 0.1079,
"step": 2690
},
{
"epoch": 3.0,
"step": 2691,
"total_flos": 2978959170994176.0,
"train_loss": 0.161040273276637,
"train_runtime": 34957.7462,
"train_samples_per_second": 19.698,
"train_steps_per_second": 0.077
}
],
"logging_steps": 10,
"max_steps": 2691,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"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": 2978959170994176.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}