kanishka's picture
End of training
c2189e9 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 20.0,
"eval_steps": 500,
"global_step": 371940,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"learning_rate": 3.125e-05,
"loss": 6.2401,
"step": 1000
},
{
"epoch": 0.11,
"learning_rate": 6.25e-05,
"loss": 5.0092,
"step": 2000
},
{
"epoch": 0.16,
"learning_rate": 9.375e-05,
"loss": 4.6786,
"step": 3000
},
{
"epoch": 0.22,
"learning_rate": 0.000125,
"loss": 4.4551,
"step": 4000
},
{
"epoch": 0.27,
"learning_rate": 0.00015625,
"loss": 4.2998,
"step": 5000
},
{
"epoch": 0.32,
"learning_rate": 0.0001875,
"loss": 4.1777,
"step": 6000
},
{
"epoch": 0.38,
"learning_rate": 0.00021875,
"loss": 4.0802,
"step": 7000
},
{
"epoch": 0.43,
"learning_rate": 0.00025,
"loss": 3.9801,
"step": 8000
},
{
"epoch": 0.48,
"learning_rate": 0.00028125000000000003,
"loss": 3.9145,
"step": 9000
},
{
"epoch": 0.54,
"learning_rate": 0.0003125,
"loss": 3.8495,
"step": 10000
},
{
"epoch": 0.59,
"learning_rate": 0.00034368749999999997,
"loss": 3.8054,
"step": 11000
},
{
"epoch": 0.65,
"learning_rate": 0.0003749375,
"loss": 3.7688,
"step": 12000
},
{
"epoch": 0.7,
"learning_rate": 0.00040615625,
"loss": 3.7307,
"step": 13000
},
{
"epoch": 0.75,
"learning_rate": 0.00043740625,
"loss": 3.698,
"step": 14000
},
{
"epoch": 0.81,
"learning_rate": 0.000468625,
"loss": 3.6725,
"step": 15000
},
{
"epoch": 0.86,
"learning_rate": 0.00049984375,
"loss": 3.6435,
"step": 16000
},
{
"epoch": 0.91,
"learning_rate": 0.00053109375,
"loss": 3.621,
"step": 17000
},
{
"epoch": 0.97,
"learning_rate": 0.0005623125,
"loss": 3.6004,
"step": 18000
},
{
"epoch": 1.0,
"eval_accuracy": 0.35749361468105384,
"eval_loss": 3.8218727111816406,
"eval_runtime": 152.7711,
"eval_samples_per_second": 379.116,
"eval_steps_per_second": 5.924,
"step": 18597
},
{
"epoch": 1.02,
"learning_rate": 0.00059353125,
"loss": 3.5722,
"step": 19000
},
{
"epoch": 1.08,
"learning_rate": 0.00062478125,
"loss": 3.5463,
"step": 20000
},
{
"epoch": 1.13,
"learning_rate": 0.0006560312499999999,
"loss": 3.5347,
"step": 21000
},
{
"epoch": 1.18,
"learning_rate": 0.00068725,
"loss": 3.5277,
"step": 22000
},
{
"epoch": 1.24,
"learning_rate": 0.0007185000000000001,
"loss": 3.5174,
"step": 23000
},
{
"epoch": 1.29,
"learning_rate": 0.00074971875,
"loss": 3.5043,
"step": 24000
},
{
"epoch": 1.34,
"learning_rate": 0.00078096875,
"loss": 3.5009,
"step": 25000
},
{
"epoch": 1.4,
"learning_rate": 0.0008121875,
"loss": 3.489,
"step": 26000
},
{
"epoch": 1.45,
"learning_rate": 0.0008434375,
"loss": 3.478,
"step": 27000
},
{
"epoch": 1.51,
"learning_rate": 0.0008746874999999999,
"loss": 3.4717,
"step": 28000
},
{
"epoch": 1.56,
"learning_rate": 0.0009059375,
"loss": 3.4611,
"step": 29000
},
{
"epoch": 1.61,
"learning_rate": 0.0009371562500000001,
"loss": 3.4531,
"step": 30000
},
{
"epoch": 1.67,
"learning_rate": 0.0009684062500000001,
"loss": 3.4435,
"step": 31000
},
{
"epoch": 1.72,
"learning_rate": 0.000999625,
"loss": 3.4381,
"step": 32000
},
{
"epoch": 1.77,
"learning_rate": 0.0009970936047537802,
"loss": 3.4261,
"step": 33000
},
{
"epoch": 1.83,
"learning_rate": 0.00099415190916044,
"loss": 3.4203,
"step": 34000
},
{
"epoch": 1.88,
"learning_rate": 0.0009912102135671002,
"loss": 3.406,
"step": 35000
},
{
"epoch": 1.94,
"learning_rate": 0.0009882744013649466,
"loss": 3.3939,
"step": 36000
},
{
"epoch": 1.99,
"learning_rate": 0.0009853356474672002,
"loss": 3.3852,
"step": 37000
},
{
"epoch": 2.0,
"eval_accuracy": 0.3797113267476103,
"eval_loss": 3.6091644763946533,
"eval_runtime": 153.7105,
"eval_samples_per_second": 376.799,
"eval_steps_per_second": 5.888,
"step": 37194
},
{
"epoch": 2.04,
"learning_rate": 0.0009823939518738601,
"loss": 3.3365,
"step": 38000
},
{
"epoch": 2.1,
"learning_rate": 0.0009794522562805203,
"loss": 3.3291,
"step": 39000
},
{
"epoch": 2.15,
"learning_rate": 0.0009765105606871802,
"loss": 3.3258,
"step": 40000
},
{
"epoch": 2.2,
"learning_rate": 0.0009735688650938402,
"loss": 3.3217,
"step": 41000
},
{
"epoch": 2.26,
"learning_rate": 0.0009706271695005001,
"loss": 3.3146,
"step": 42000
},
{
"epoch": 2.31,
"learning_rate": 0.0009676884156027535,
"loss": 3.3129,
"step": 43000
},
{
"epoch": 2.37,
"learning_rate": 0.0009647496617050067,
"loss": 3.3028,
"step": 44000
},
{
"epoch": 2.42,
"learning_rate": 0.0009618079661116668,
"loss": 3.3003,
"step": 45000
},
{
"epoch": 2.47,
"learning_rate": 0.00095886921221392,
"loss": 3.2966,
"step": 46000
},
{
"epoch": 2.53,
"learning_rate": 0.0009559304583161734,
"loss": 3.2909,
"step": 47000
},
{
"epoch": 2.58,
"learning_rate": 0.0009529887627228334,
"loss": 3.2885,
"step": 48000
},
{
"epoch": 2.63,
"learning_rate": 0.0009500470671294936,
"loss": 3.2834,
"step": 49000
},
{
"epoch": 2.69,
"learning_rate": 0.0009471053715361535,
"loss": 3.2797,
"step": 50000
},
{
"epoch": 2.74,
"learning_rate": 0.0009441636759428135,
"loss": 3.2697,
"step": 51000
},
{
"epoch": 2.8,
"learning_rate": 0.0009412219803494735,
"loss": 3.2692,
"step": 52000
},
{
"epoch": 2.85,
"learning_rate": 0.0009382832264517268,
"loss": 3.2633,
"step": 53000
},
{
"epoch": 2.9,
"learning_rate": 0.0009353444725539802,
"loss": 3.2586,
"step": 54000
},
{
"epoch": 2.96,
"learning_rate": 0.0009324027769606402,
"loss": 3.2597,
"step": 55000
},
{
"epoch": 3.0,
"eval_accuracy": 0.3909656655330366,
"eval_loss": 3.4836692810058594,
"eval_runtime": 153.8048,
"eval_samples_per_second": 376.568,
"eval_steps_per_second": 5.884,
"step": 55791
},
{
"epoch": 3.01,
"learning_rate": 0.0009294640230628935,
"loss": 3.2437,
"step": 56000
},
{
"epoch": 3.07,
"learning_rate": 0.0009265223274695535,
"loss": 3.1914,
"step": 57000
},
{
"epoch": 3.12,
"learning_rate": 0.0009235806318762135,
"loss": 3.1917,
"step": 58000
},
{
"epoch": 3.17,
"learning_rate": 0.0009206389362828734,
"loss": 3.1987,
"step": 59000
},
{
"epoch": 3.23,
"learning_rate": 0.0009176972406895334,
"loss": 3.1923,
"step": 60000
},
{
"epoch": 3.28,
"learning_rate": 0.0009147555450961935,
"loss": 3.193,
"step": 61000
},
{
"epoch": 3.33,
"learning_rate": 0.0009118138495028535,
"loss": 3.1958,
"step": 62000
},
{
"epoch": 3.39,
"learning_rate": 0.0009088721539095135,
"loss": 3.1988,
"step": 63000
},
{
"epoch": 3.44,
"learning_rate": 0.0009059334000117668,
"loss": 3.1987,
"step": 64000
},
{
"epoch": 3.5,
"learning_rate": 0.0009029917044184268,
"loss": 3.1947,
"step": 65000
},
{
"epoch": 3.55,
"learning_rate": 0.0009000529505206801,
"loss": 3.1923,
"step": 66000
},
{
"epoch": 3.6,
"learning_rate": 0.0008971112549273401,
"loss": 3.1845,
"step": 67000
},
{
"epoch": 3.66,
"learning_rate": 0.0008941695593340001,
"loss": 3.1875,
"step": 68000
},
{
"epoch": 3.71,
"learning_rate": 0.0008912337471318468,
"loss": 3.1892,
"step": 69000
},
{
"epoch": 3.76,
"learning_rate": 0.0008882920515385068,
"loss": 3.1828,
"step": 70000
},
{
"epoch": 3.82,
"learning_rate": 0.0008853532976407601,
"loss": 3.1899,
"step": 71000
},
{
"epoch": 3.87,
"learning_rate": 0.0008824116020474201,
"loss": 3.1775,
"step": 72000
},
{
"epoch": 3.93,
"learning_rate": 0.0008794699064540802,
"loss": 3.1812,
"step": 73000
},
{
"epoch": 3.98,
"learning_rate": 0.0008765282108607402,
"loss": 3.1758,
"step": 74000
},
{
"epoch": 4.0,
"eval_accuracy": 0.39812142834972775,
"eval_loss": 3.436387062072754,
"eval_runtime": 154.2825,
"eval_samples_per_second": 375.402,
"eval_steps_per_second": 5.866,
"step": 74388
},
{
"epoch": 4.03,
"learning_rate": 0.0008735865152674002,
"loss": 3.1337,
"step": 75000
},
{
"epoch": 4.09,
"learning_rate": 0.0008706477613696535,
"loss": 3.1139,
"step": 76000
},
{
"epoch": 4.14,
"learning_rate": 0.0008677060657763135,
"loss": 3.1202,
"step": 77000
},
{
"epoch": 4.19,
"learning_rate": 0.0008647643701829735,
"loss": 3.1162,
"step": 78000
},
{
"epoch": 4.25,
"learning_rate": 0.0008618226745896334,
"loss": 3.1242,
"step": 79000
},
{
"epoch": 4.3,
"learning_rate": 0.0008588868623874801,
"loss": 3.1254,
"step": 80000
},
{
"epoch": 4.36,
"learning_rate": 0.0008559451667941401,
"loss": 3.1248,
"step": 81000
},
{
"epoch": 4.41,
"learning_rate": 0.0008530034712008001,
"loss": 3.1254,
"step": 82000
},
{
"epoch": 4.46,
"learning_rate": 0.0008500647173030534,
"loss": 3.1243,
"step": 83000
},
{
"epoch": 4.52,
"learning_rate": 0.0008471230217097134,
"loss": 3.1294,
"step": 84000
},
{
"epoch": 4.57,
"learning_rate": 0.0008441813261163735,
"loss": 3.1251,
"step": 85000
},
{
"epoch": 4.62,
"learning_rate": 0.0008412396305230336,
"loss": 3.1259,
"step": 86000
},
{
"epoch": 4.68,
"learning_rate": 0.0008383038183208802,
"loss": 3.1267,
"step": 87000
},
{
"epoch": 4.73,
"learning_rate": 0.0008353621227275402,
"loss": 3.1244,
"step": 88000
},
{
"epoch": 4.79,
"learning_rate": 0.0008324204271342001,
"loss": 3.1275,
"step": 89000
},
{
"epoch": 4.84,
"learning_rate": 0.0008294787315408601,
"loss": 3.1259,
"step": 90000
},
{
"epoch": 4.89,
"learning_rate": 0.0008265399776431135,
"loss": 3.1271,
"step": 91000
},
{
"epoch": 4.95,
"learning_rate": 0.0008235982820497735,
"loss": 3.1197,
"step": 92000
},
{
"epoch": 5.0,
"eval_accuracy": 0.4016729407993026,
"eval_loss": 3.4115896224975586,
"eval_runtime": 154.3917,
"eval_samples_per_second": 375.137,
"eval_steps_per_second": 5.862,
"step": 92985
},
{
"epoch": 5.0,
"learning_rate": 0.0008206565864564334,
"loss": 3.1202,
"step": 93000
},
{
"epoch": 5.05,
"learning_rate": 0.0008177148908630935,
"loss": 3.0505,
"step": 94000
},
{
"epoch": 5.11,
"learning_rate": 0.0008147761369653469,
"loss": 3.0636,
"step": 95000
},
{
"epoch": 5.16,
"learning_rate": 0.0008118344413720068,
"loss": 3.0657,
"step": 96000
},
{
"epoch": 5.22,
"learning_rate": 0.0008088956874742603,
"loss": 3.065,
"step": 97000
},
{
"epoch": 5.27,
"learning_rate": 0.0008059539918809203,
"loss": 3.073,
"step": 98000
},
{
"epoch": 5.32,
"learning_rate": 0.0008030152379831736,
"loss": 3.0711,
"step": 99000
},
{
"epoch": 5.38,
"learning_rate": 0.0008000735423898336,
"loss": 3.0715,
"step": 100000
},
{
"epoch": 5.43,
"learning_rate": 0.0007971318467964936,
"loss": 3.075,
"step": 101000
},
{
"epoch": 5.48,
"learning_rate": 0.0007941901512031535,
"loss": 3.0774,
"step": 102000
},
{
"epoch": 5.54,
"learning_rate": 0.0007912543390010001,
"loss": 3.0769,
"step": 103000
},
{
"epoch": 5.59,
"learning_rate": 0.0007883126434076602,
"loss": 3.0797,
"step": 104000
},
{
"epoch": 5.65,
"learning_rate": 0.0007853709478143202,
"loss": 3.0764,
"step": 105000
},
{
"epoch": 5.7,
"learning_rate": 0.0007824292522209802,
"loss": 3.0807,
"step": 106000
},
{
"epoch": 5.75,
"learning_rate": 0.0007794904983232335,
"loss": 3.0789,
"step": 107000
},
{
"epoch": 5.81,
"learning_rate": 0.0007765488027298935,
"loss": 3.0826,
"step": 108000
},
{
"epoch": 5.86,
"learning_rate": 0.0007736071071365536,
"loss": 3.0805,
"step": 109000
},
{
"epoch": 5.91,
"learning_rate": 0.0007706654115432136,
"loss": 3.081,
"step": 110000
},
{
"epoch": 5.97,
"learning_rate": 0.0007677266576454669,
"loss": 3.08,
"step": 111000
},
{
"epoch": 6.0,
"eval_accuracy": 0.4039926701684644,
"eval_loss": 3.3782215118408203,
"eval_runtime": 154.2767,
"eval_samples_per_second": 375.416,
"eval_steps_per_second": 5.866,
"step": 111582
},
{
"epoch": 6.02,
"learning_rate": 0.0007647849620521269,
"loss": 3.0517,
"step": 112000
},
{
"epoch": 6.08,
"learning_rate": 0.0007618462081543802,
"loss": 3.0148,
"step": 113000
},
{
"epoch": 6.13,
"learning_rate": 0.0007589045125610402,
"loss": 3.0169,
"step": 114000
},
{
"epoch": 6.18,
"learning_rate": 0.0007559628169677002,
"loss": 3.0256,
"step": 115000
},
{
"epoch": 6.24,
"learning_rate": 0.0007530211213743602,
"loss": 3.0284,
"step": 116000
},
{
"epoch": 6.29,
"learning_rate": 0.0007500823674766135,
"loss": 3.03,
"step": 117000
},
{
"epoch": 6.35,
"learning_rate": 0.0007471406718832735,
"loss": 3.0312,
"step": 118000
},
{
"epoch": 6.4,
"learning_rate": 0.0007441989762899335,
"loss": 3.0341,
"step": 119000
},
{
"epoch": 6.45,
"learning_rate": 0.0007412572806965935,
"loss": 3.0385,
"step": 120000
},
{
"epoch": 6.51,
"learning_rate": 0.0007383185267988469,
"loss": 3.038,
"step": 121000
},
{
"epoch": 6.56,
"learning_rate": 0.0007353797729011003,
"loss": 3.0379,
"step": 122000
},
{
"epoch": 6.61,
"learning_rate": 0.0007324380773077602,
"loss": 3.0381,
"step": 123000
},
{
"epoch": 6.67,
"learning_rate": 0.0007294963817144202,
"loss": 3.0409,
"step": 124000
},
{
"epoch": 6.72,
"learning_rate": 0.0007265576278166736,
"loss": 3.0401,
"step": 125000
},
{
"epoch": 6.78,
"learning_rate": 0.0007236159322233335,
"loss": 3.0416,
"step": 126000
},
{
"epoch": 6.83,
"learning_rate": 0.0007206742366299935,
"loss": 3.0396,
"step": 127000
},
{
"epoch": 6.88,
"learning_rate": 0.0007177354827322469,
"loss": 3.0431,
"step": 128000
},
{
"epoch": 6.94,
"learning_rate": 0.0007147937871389068,
"loss": 3.0476,
"step": 129000
},
{
"epoch": 6.99,
"learning_rate": 0.0007118550332411602,
"loss": 3.0418,
"step": 130000
},
{
"epoch": 7.0,
"eval_accuracy": 0.40549848887391177,
"eval_loss": 3.388542413711548,
"eval_runtime": 154.3674,
"eval_samples_per_second": 375.196,
"eval_steps_per_second": 5.863,
"step": 130179
},
{
"epoch": 7.04,
"learning_rate": 0.0007089133376478202,
"loss": 2.9855,
"step": 131000
},
{
"epoch": 7.1,
"learning_rate": 0.0007059716420544801,
"loss": 2.9759,
"step": 132000
},
{
"epoch": 7.15,
"learning_rate": 0.0007030328881567336,
"loss": 2.9898,
"step": 133000
},
{
"epoch": 7.21,
"learning_rate": 0.0007000911925633936,
"loss": 2.9954,
"step": 134000
},
{
"epoch": 7.26,
"learning_rate": 0.0006971494969700536,
"loss": 2.9892,
"step": 135000
},
{
"epoch": 7.31,
"learning_rate": 0.0006942078013767135,
"loss": 2.9957,
"step": 136000
},
{
"epoch": 7.37,
"learning_rate": 0.0006912690474789669,
"loss": 3.0008,
"step": 137000
},
{
"epoch": 7.42,
"learning_rate": 0.0006883273518856269,
"loss": 3.0042,
"step": 138000
},
{
"epoch": 7.47,
"learning_rate": 0.0006853885979878802,
"loss": 3.003,
"step": 139000
},
{
"epoch": 7.53,
"learning_rate": 0.0006824469023945402,
"loss": 3.0058,
"step": 140000
},
{
"epoch": 7.58,
"learning_rate": 0.0006795081484967935,
"loss": 3.0083,
"step": 141000
},
{
"epoch": 7.64,
"learning_rate": 0.0006765664529034535,
"loss": 3.0071,
"step": 142000
},
{
"epoch": 7.69,
"learning_rate": 0.0006736276990057068,
"loss": 3.0056,
"step": 143000
},
{
"epoch": 7.74,
"learning_rate": 0.0006706860034123668,
"loss": 3.0106,
"step": 144000
},
{
"epoch": 7.8,
"learning_rate": 0.0006677472495146203,
"loss": 3.0131,
"step": 145000
},
{
"epoch": 7.85,
"learning_rate": 0.0006648055539212802,
"loss": 3.0107,
"step": 146000
},
{
"epoch": 7.9,
"learning_rate": 0.0006618668000235336,
"loss": 3.0098,
"step": 147000
},
{
"epoch": 7.96,
"learning_rate": 0.0006589251044301936,
"loss": 3.0088,
"step": 148000
},
{
"epoch": 8.0,
"eval_accuracy": 0.40621397781890217,
"eval_loss": 3.388378620147705,
"eval_runtime": 154.4388,
"eval_samples_per_second": 375.022,
"eval_steps_per_second": 5.86,
"step": 148776
},
{
"epoch": 8.01,
"learning_rate": 0.0006559834088368535,
"loss": 2.995,
"step": 149000
},
{
"epoch": 8.07,
"learning_rate": 0.0006530417132435136,
"loss": 2.9461,
"step": 150000
},
{
"epoch": 8.12,
"learning_rate": 0.0006501059010413602,
"loss": 2.9526,
"step": 151000
},
{
"epoch": 8.17,
"learning_rate": 0.0006471642054480202,
"loss": 2.9558,
"step": 152000
},
{
"epoch": 8.23,
"learning_rate": 0.0006442225098546802,
"loss": 2.9614,
"step": 153000
},
{
"epoch": 8.28,
"learning_rate": 0.0006412808142613403,
"loss": 2.9661,
"step": 154000
},
{
"epoch": 8.33,
"learning_rate": 0.0006383420603635935,
"loss": 2.9688,
"step": 155000
},
{
"epoch": 8.39,
"learning_rate": 0.0006354003647702535,
"loss": 2.9695,
"step": 156000
},
{
"epoch": 8.44,
"learning_rate": 0.0006324586691769137,
"loss": 2.9758,
"step": 157000
},
{
"epoch": 8.5,
"learning_rate": 0.0006295228569747603,
"loss": 2.9707,
"step": 158000
},
{
"epoch": 8.55,
"learning_rate": 0.0006265811613814202,
"loss": 2.9775,
"step": 159000
},
{
"epoch": 8.6,
"learning_rate": 0.0006236394657880802,
"loss": 2.9757,
"step": 160000
},
{
"epoch": 8.66,
"learning_rate": 0.0006206977701947403,
"loss": 2.9811,
"step": 161000
},
{
"epoch": 8.71,
"learning_rate": 0.0006177590162969937,
"loss": 2.9795,
"step": 162000
},
{
"epoch": 8.76,
"learning_rate": 0.0006148173207036536,
"loss": 2.9845,
"step": 163000
},
{
"epoch": 8.82,
"learning_rate": 0.000611878566805907,
"loss": 2.9813,
"step": 164000
},
{
"epoch": 8.87,
"learning_rate": 0.000608936871212567,
"loss": 2.9818,
"step": 165000
},
{
"epoch": 8.93,
"learning_rate": 0.0006059951756192269,
"loss": 2.9868,
"step": 166000
},
{
"epoch": 8.98,
"learning_rate": 0.0006030534800258869,
"loss": 2.9856,
"step": 167000
},
{
"epoch": 9.0,
"eval_accuracy": 0.4077341734853099,
"eval_loss": 3.3548429012298584,
"eval_runtime": 153.8146,
"eval_samples_per_second": 376.544,
"eval_steps_per_second": 5.884,
"step": 167373
},
{
"epoch": 9.03,
"learning_rate": 0.0006001147261281403,
"loss": 2.9389,
"step": 168000
},
{
"epoch": 9.09,
"learning_rate": 0.0005971730305348003,
"loss": 2.9204,
"step": 169000
},
{
"epoch": 9.14,
"learning_rate": 0.000594237218332647,
"loss": 2.927,
"step": 170000
},
{
"epoch": 9.2,
"learning_rate": 0.000591295522739307,
"loss": 2.9347,
"step": 171000
},
{
"epoch": 9.25,
"learning_rate": 0.000588353827145967,
"loss": 2.9347,
"step": 172000
},
{
"epoch": 9.3,
"learning_rate": 0.000585412131552627,
"loss": 2.9423,
"step": 173000
},
{
"epoch": 9.36,
"learning_rate": 0.0005824733776548803,
"loss": 2.9411,
"step": 174000
},
{
"epoch": 9.41,
"learning_rate": 0.0005795316820615403,
"loss": 2.9463,
"step": 175000
},
{
"epoch": 9.46,
"learning_rate": 0.0005765899864682003,
"loss": 2.9473,
"step": 176000
},
{
"epoch": 9.52,
"learning_rate": 0.0005736512325704536,
"loss": 2.9497,
"step": 177000
},
{
"epoch": 9.57,
"learning_rate": 0.0005707095369771136,
"loss": 2.9496,
"step": 178000
},
{
"epoch": 9.63,
"learning_rate": 0.0005677678413837736,
"loss": 2.9517,
"step": 179000
},
{
"epoch": 9.68,
"learning_rate": 0.0005648261457904336,
"loss": 2.9499,
"step": 180000
},
{
"epoch": 9.73,
"learning_rate": 0.000561887391892687,
"loss": 2.958,
"step": 181000
},
{
"epoch": 9.79,
"learning_rate": 0.0005589486379949404,
"loss": 2.9553,
"step": 182000
},
{
"epoch": 9.84,
"learning_rate": 0.0005560069424016003,
"loss": 2.9598,
"step": 183000
},
{
"epoch": 9.89,
"learning_rate": 0.0005530681885038537,
"loss": 2.9572,
"step": 184000
},
{
"epoch": 9.95,
"learning_rate": 0.0005501264929105137,
"loss": 2.9598,
"step": 185000
},
{
"epoch": 10.0,
"eval_accuracy": 0.40899585258267324,
"eval_loss": 3.3782236576080322,
"eval_runtime": 153.4575,
"eval_samples_per_second": 377.421,
"eval_steps_per_second": 5.897,
"step": 185970
},
{
"epoch": 10.0,
"learning_rate": 0.0005471847973171736,
"loss": 2.9565,
"step": 186000
},
{
"epoch": 10.06,
"learning_rate": 0.0005442431017238336,
"loss": 2.8943,
"step": 187000
},
{
"epoch": 10.11,
"learning_rate": 0.000541304347826087,
"loss": 2.9024,
"step": 188000
},
{
"epoch": 10.16,
"learning_rate": 0.0005383655939283403,
"loss": 2.9082,
"step": 189000
},
{
"epoch": 10.22,
"learning_rate": 0.0005354238983350003,
"loss": 2.9164,
"step": 190000
},
{
"epoch": 10.27,
"learning_rate": 0.0005324851444372536,
"loss": 2.9116,
"step": 191000
},
{
"epoch": 10.32,
"learning_rate": 0.0005295434488439136,
"loss": 2.9162,
"step": 192000
},
{
"epoch": 10.38,
"learning_rate": 0.000526604694946167,
"loss": 2.9193,
"step": 193000
},
{
"epoch": 10.43,
"learning_rate": 0.000523662999352827,
"loss": 2.9201,
"step": 194000
},
{
"epoch": 10.49,
"learning_rate": 0.000520721303759487,
"loss": 2.9215,
"step": 195000
},
{
"epoch": 10.54,
"learning_rate": 0.0005177825498617403,
"loss": 2.9225,
"step": 196000
},
{
"epoch": 10.59,
"learning_rate": 0.0005148408542684003,
"loss": 2.928,
"step": 197000
},
{
"epoch": 10.65,
"learning_rate": 0.0005118991586750603,
"loss": 2.9311,
"step": 198000
},
{
"epoch": 10.7,
"learning_rate": 0.0005089574630817203,
"loss": 2.9287,
"step": 199000
},
{
"epoch": 10.75,
"learning_rate": 0.0005060187091839736,
"loss": 2.9298,
"step": 200000
},
{
"epoch": 10.81,
"learning_rate": 0.0005030770135906336,
"loss": 2.9347,
"step": 201000
},
{
"epoch": 10.86,
"learning_rate": 0.000500138259692887,
"loss": 2.9303,
"step": 202000
},
{
"epoch": 10.92,
"learning_rate": 0.000497196564099547,
"loss": 2.9346,
"step": 203000
},
{
"epoch": 10.97,
"learning_rate": 0.000494254868506207,
"loss": 2.9364,
"step": 204000
},
{
"epoch": 11.0,
"eval_accuracy": 0.4093280678721547,
"eval_loss": 3.385136127471924,
"eval_runtime": 153.353,
"eval_samples_per_second": 377.678,
"eval_steps_per_second": 5.901,
"step": 204567
},
{
"epoch": 11.02,
"learning_rate": 0.0004913161146084603,
"loss": 2.9063,
"step": 205000
},
{
"epoch": 11.08,
"learning_rate": 0.0004883744190151203,
"loss": 2.8748,
"step": 206000
},
{
"epoch": 11.13,
"learning_rate": 0.00048543566511737366,
"loss": 2.8806,
"step": 207000
},
{
"epoch": 11.18,
"learning_rate": 0.00048249396952403366,
"loss": 2.8849,
"step": 208000
},
{
"epoch": 11.24,
"learning_rate": 0.000479555215626287,
"loss": 2.89,
"step": 209000
},
{
"epoch": 11.29,
"learning_rate": 0.000476613520032947,
"loss": 2.8915,
"step": 210000
},
{
"epoch": 11.35,
"learning_rate": 0.00047367182443960697,
"loss": 2.8953,
"step": 211000
},
{
"epoch": 11.4,
"learning_rate": 0.000470730128846267,
"loss": 2.8965,
"step": 212000
},
{
"epoch": 11.45,
"learning_rate": 0.0004677943166441137,
"loss": 2.8963,
"step": 213000
},
{
"epoch": 11.51,
"learning_rate": 0.0004648526210507737,
"loss": 2.9038,
"step": 214000
},
{
"epoch": 11.56,
"learning_rate": 0.00046191092545743367,
"loss": 2.9066,
"step": 215000
},
{
"epoch": 11.61,
"learning_rate": 0.0004589692298640937,
"loss": 2.9038,
"step": 216000
},
{
"epoch": 11.67,
"learning_rate": 0.000456030475966347,
"loss": 2.9145,
"step": 217000
},
{
"epoch": 11.72,
"learning_rate": 0.000453088780373007,
"loss": 2.9051,
"step": 218000
},
{
"epoch": 11.78,
"learning_rate": 0.000450147084779667,
"loss": 2.9118,
"step": 219000
},
{
"epoch": 11.83,
"learning_rate": 0.00044720538918632704,
"loss": 2.909,
"step": 220000
},
{
"epoch": 11.88,
"learning_rate": 0.000444263693592987,
"loss": 2.9115,
"step": 221000
},
{
"epoch": 11.94,
"learning_rate": 0.00044132493969524034,
"loss": 2.9129,
"step": 222000
},
{
"epoch": 11.99,
"learning_rate": 0.00043838324410190035,
"loss": 2.9156,
"step": 223000
},
{
"epoch": 12.0,
"eval_accuracy": 0.4096916571839188,
"eval_loss": 3.3802754878997803,
"eval_runtime": 153.587,
"eval_samples_per_second": 377.102,
"eval_steps_per_second": 5.892,
"step": 223164
},
{
"epoch": 12.04,
"learning_rate": 0.0004354444902041537,
"loss": 2.8619,
"step": 224000
},
{
"epoch": 12.1,
"learning_rate": 0.00043250279461081364,
"loss": 2.8604,
"step": 225000
},
{
"epoch": 12.15,
"learning_rate": 0.00042956404071306704,
"loss": 2.8593,
"step": 226000
},
{
"epoch": 12.21,
"learning_rate": 0.00042662234511972705,
"loss": 2.866,
"step": 227000
},
{
"epoch": 12.26,
"learning_rate": 0.000423680649526387,
"loss": 2.8714,
"step": 228000
},
{
"epoch": 12.31,
"learning_rate": 0.00042074189562864035,
"loss": 2.8728,
"step": 229000
},
{
"epoch": 12.37,
"learning_rate": 0.00041780020003530035,
"loss": 2.877,
"step": 230000
},
{
"epoch": 12.42,
"learning_rate": 0.0004148585044419603,
"loss": 2.8776,
"step": 231000
},
{
"epoch": 12.48,
"learning_rate": 0.00041191680884862037,
"loss": 2.8833,
"step": 232000
},
{
"epoch": 12.53,
"learning_rate": 0.0004089780549508737,
"loss": 2.8799,
"step": 233000
},
{
"epoch": 12.58,
"learning_rate": 0.00040603930105312705,
"loss": 2.8828,
"step": 234000
},
{
"epoch": 12.64,
"learning_rate": 0.000403097605459787,
"loss": 2.8836,
"step": 235000
},
{
"epoch": 12.69,
"learning_rate": 0.00040015885156204035,
"loss": 2.8862,
"step": 236000
},
{
"epoch": 12.74,
"learning_rate": 0.0003972171559687004,
"loss": 2.8845,
"step": 237000
},
{
"epoch": 12.8,
"learning_rate": 0.00039427546037536037,
"loss": 2.8895,
"step": 238000
},
{
"epoch": 12.85,
"learning_rate": 0.0003913367064776137,
"loss": 2.8926,
"step": 239000
},
{
"epoch": 12.91,
"learning_rate": 0.0003883950108842737,
"loss": 2.8876,
"step": 240000
},
{
"epoch": 12.96,
"learning_rate": 0.00038545331529093367,
"loss": 2.8949,
"step": 241000
},
{
"epoch": 13.0,
"eval_accuracy": 0.4099969324671147,
"eval_loss": 3.386857032775879,
"eval_runtime": 153.7002,
"eval_samples_per_second": 376.825,
"eval_steps_per_second": 5.888,
"step": 241761
},
{
"epoch": 13.01,
"learning_rate": 0.000382514561393187,
"loss": 2.8803,
"step": 242000
},
{
"epoch": 13.07,
"learning_rate": 0.0003795728657998471,
"loss": 2.8388,
"step": 243000
},
{
"epoch": 13.12,
"learning_rate": 0.0003766341119021004,
"loss": 2.8374,
"step": 244000
},
{
"epoch": 13.17,
"learning_rate": 0.00037369241630876037,
"loss": 2.8461,
"step": 245000
},
{
"epoch": 13.23,
"learning_rate": 0.0003707507207154204,
"loss": 2.8492,
"step": 246000
},
{
"epoch": 13.28,
"learning_rate": 0.0003678090251220804,
"loss": 2.8509,
"step": 247000
},
{
"epoch": 13.34,
"learning_rate": 0.00036486732952874034,
"loss": 2.853,
"step": 248000
},
{
"epoch": 13.39,
"learning_rate": 0.00036192563393540035,
"loss": 2.8551,
"step": 249000
},
{
"epoch": 13.44,
"learning_rate": 0.00035898688003765375,
"loss": 2.8578,
"step": 250000
},
{
"epoch": 13.5,
"learning_rate": 0.0003560451844443137,
"loss": 2.8655,
"step": 251000
},
{
"epoch": 13.55,
"learning_rate": 0.0003531093722421604,
"loss": 2.8637,
"step": 252000
},
{
"epoch": 13.6,
"learning_rate": 0.0003501676766488204,
"loss": 2.865,
"step": 253000
},
{
"epoch": 13.66,
"learning_rate": 0.00034722598105548034,
"loss": 2.8652,
"step": 254000
},
{
"epoch": 13.71,
"learning_rate": 0.0003442842854621404,
"loss": 2.8702,
"step": 255000
},
{
"epoch": 13.77,
"learning_rate": 0.0003413425898688004,
"loss": 2.8662,
"step": 256000
},
{
"epoch": 13.82,
"learning_rate": 0.00033840383597105375,
"loss": 2.8677,
"step": 257000
},
{
"epoch": 13.87,
"learning_rate": 0.0003354621403777137,
"loss": 2.8678,
"step": 258000
},
{
"epoch": 13.93,
"learning_rate": 0.0003325204447843737,
"loss": 2.8729,
"step": 259000
},
{
"epoch": 13.98,
"learning_rate": 0.0003295787491910337,
"loss": 2.8719,
"step": 260000
},
{
"epoch": 14.0,
"eval_accuracy": 0.41038846951607283,
"eval_loss": 3.381344795227051,
"eval_runtime": 153.5516,
"eval_samples_per_second": 377.189,
"eval_steps_per_second": 5.894,
"step": 260358
},
{
"epoch": 14.03,
"learning_rate": 0.00032663999529328707,
"loss": 2.8352,
"step": 261000
},
{
"epoch": 14.09,
"learning_rate": 0.0003236982996999471,
"loss": 2.8212,
"step": 262000
},
{
"epoch": 14.14,
"learning_rate": 0.0003207595458022004,
"loss": 2.8266,
"step": 263000
},
{
"epoch": 14.2,
"learning_rate": 0.00031781785020886037,
"loss": 2.8248,
"step": 264000
},
{
"epoch": 14.25,
"learning_rate": 0.0003148790963111137,
"loss": 2.8338,
"step": 265000
},
{
"epoch": 14.3,
"learning_rate": 0.0003119374007177737,
"loss": 2.8346,
"step": 266000
},
{
"epoch": 14.36,
"learning_rate": 0.0003089957051244338,
"loss": 2.8353,
"step": 267000
},
{
"epoch": 14.41,
"learning_rate": 0.0003060598929222804,
"loss": 2.8359,
"step": 268000
},
{
"epoch": 14.46,
"learning_rate": 0.0003031181973289404,
"loss": 2.8443,
"step": 269000
},
{
"epoch": 14.52,
"learning_rate": 0.0003001765017356004,
"loss": 2.843,
"step": 270000
},
{
"epoch": 14.57,
"learning_rate": 0.0002972348061422604,
"loss": 2.844,
"step": 271000
},
{
"epoch": 14.63,
"learning_rate": 0.0002942931105489204,
"loss": 2.8447,
"step": 272000
},
{
"epoch": 14.68,
"learning_rate": 0.0002913543566511738,
"loss": 2.8453,
"step": 273000
},
{
"epoch": 14.73,
"learning_rate": 0.00028841266105783373,
"loss": 2.8472,
"step": 274000
},
{
"epoch": 14.79,
"learning_rate": 0.0002854768488556804,
"loss": 2.8503,
"step": 275000
},
{
"epoch": 14.84,
"learning_rate": 0.0002825351532623404,
"loss": 2.8518,
"step": 276000
},
{
"epoch": 14.89,
"learning_rate": 0.00027959639936459376,
"loss": 2.852,
"step": 277000
},
{
"epoch": 14.95,
"learning_rate": 0.0002766547037712537,
"loss": 2.8526,
"step": 278000
},
{
"epoch": 15.0,
"eval_accuracy": 0.4108039233785912,
"eval_loss": 3.3859314918518066,
"eval_runtime": 153.7719,
"eval_samples_per_second": 376.649,
"eval_steps_per_second": 5.885,
"step": 278955
},
{
"epoch": 15.0,
"learning_rate": 0.0002737130081779138,
"loss": 2.846,
"step": 279000
},
{
"epoch": 15.06,
"learning_rate": 0.0002707713125845738,
"loss": 2.8036,
"step": 280000
},
{
"epoch": 15.11,
"learning_rate": 0.00026782961699123374,
"loss": 2.81,
"step": 281000
},
{
"epoch": 15.16,
"learning_rate": 0.0002648908630934871,
"loss": 2.8119,
"step": 282000
},
{
"epoch": 15.22,
"learning_rate": 0.0002619491675001471,
"loss": 2.8149,
"step": 283000
},
{
"epoch": 15.27,
"learning_rate": 0.00025901041360240043,
"loss": 2.816,
"step": 284000
},
{
"epoch": 15.33,
"learning_rate": 0.00025606871800906044,
"loss": 2.8168,
"step": 285000
},
{
"epoch": 15.38,
"learning_rate": 0.00025312702241572045,
"loss": 2.8184,
"step": 286000
},
{
"epoch": 15.43,
"learning_rate": 0.0002501882685179738,
"loss": 2.8235,
"step": 287000
},
{
"epoch": 15.49,
"learning_rate": 0.00024724657292463374,
"loss": 2.8206,
"step": 288000
},
{
"epoch": 15.54,
"learning_rate": 0.00024430487733129375,
"loss": 2.8246,
"step": 289000
},
{
"epoch": 15.59,
"learning_rate": 0.0002413661234335471,
"loss": 2.8273,
"step": 290000
},
{
"epoch": 15.65,
"learning_rate": 0.00023842736953580044,
"loss": 2.8266,
"step": 291000
},
{
"epoch": 15.7,
"learning_rate": 0.00023548567394246042,
"loss": 2.8287,
"step": 292000
},
{
"epoch": 15.76,
"learning_rate": 0.00023254397834912045,
"loss": 2.8297,
"step": 293000
},
{
"epoch": 15.81,
"learning_rate": 0.0002296081661469671,
"loss": 2.8276,
"step": 294000
},
{
"epoch": 15.86,
"learning_rate": 0.0002266664705536271,
"loss": 2.8313,
"step": 295000
},
{
"epoch": 15.92,
"learning_rate": 0.00022372477496028712,
"loss": 2.8306,
"step": 296000
},
{
"epoch": 15.97,
"learning_rate": 0.0002207830793669471,
"loss": 2.8289,
"step": 297000
},
{
"epoch": 16.0,
"eval_accuracy": 0.4103414420736791,
"eval_loss": 3.3979780673980713,
"eval_runtime": 153.9364,
"eval_samples_per_second": 376.246,
"eval_steps_per_second": 5.879,
"step": 297552
},
{
"epoch": 16.02,
"learning_rate": 0.0002178413837736071,
"loss": 2.8117,
"step": 298000
},
{
"epoch": 16.08,
"learning_rate": 0.00021489968818026712,
"loss": 2.7899,
"step": 299000
},
{
"epoch": 16.13,
"learning_rate": 0.00021196093428252046,
"loss": 2.7934,
"step": 300000
},
{
"epoch": 16.19,
"learning_rate": 0.00020901923868918044,
"loss": 2.7963,
"step": 301000
},
{
"epoch": 16.24,
"learning_rate": 0.00020608048479143378,
"loss": 2.7984,
"step": 302000
},
{
"epoch": 16.29,
"learning_rate": 0.0002031387891980938,
"loss": 2.7992,
"step": 303000
},
{
"epoch": 16.35,
"learning_rate": 0.00020019709360475377,
"loss": 2.8029,
"step": 304000
},
{
"epoch": 16.4,
"learning_rate": 0.00019725833970700714,
"loss": 2.8044,
"step": 305000
},
{
"epoch": 16.45,
"learning_rate": 0.00019431664411366712,
"loss": 2.8077,
"step": 306000
},
{
"epoch": 16.51,
"learning_rate": 0.00019137789021592044,
"loss": 2.8082,
"step": 307000
},
{
"epoch": 16.56,
"learning_rate": 0.00018843619462258047,
"loss": 2.8023,
"step": 308000
},
{
"epoch": 16.62,
"learning_rate": 0.00018549449902924045,
"loss": 2.8088,
"step": 309000
},
{
"epoch": 16.67,
"learning_rate": 0.00018255280343590043,
"loss": 2.8055,
"step": 310000
},
{
"epoch": 16.72,
"learning_rate": 0.0001796140495381538,
"loss": 2.8106,
"step": 311000
},
{
"epoch": 16.78,
"learning_rate": 0.00017667235394481378,
"loss": 2.8116,
"step": 312000
},
{
"epoch": 16.83,
"learning_rate": 0.00017373360004706713,
"loss": 2.8154,
"step": 313000
},
{
"epoch": 16.88,
"learning_rate": 0.00017079190445372713,
"loss": 2.813,
"step": 314000
},
{
"epoch": 16.94,
"learning_rate": 0.00016785020886038712,
"loss": 2.8118,
"step": 315000
},
{
"epoch": 16.99,
"learning_rate": 0.00016490851326704712,
"loss": 2.8104,
"step": 316000
},
{
"epoch": 17.0,
"eval_accuracy": 0.41088675885926473,
"eval_loss": 3.398145914077759,
"eval_runtime": 153.1491,
"eval_samples_per_second": 378.181,
"eval_steps_per_second": 5.909,
"step": 316149
},
{
"epoch": 17.05,
"learning_rate": 0.00016196975936930047,
"loss": 2.782,
"step": 317000
},
{
"epoch": 17.1,
"learning_rate": 0.00015902806377596047,
"loss": 2.7761,
"step": 318000
},
{
"epoch": 17.15,
"learning_rate": 0.0001560893098782138,
"loss": 2.7819,
"step": 319000
},
{
"epoch": 17.21,
"learning_rate": 0.0001531476142848738,
"loss": 2.7822,
"step": 320000
},
{
"epoch": 17.26,
"learning_rate": 0.0001502059186915338,
"loss": 2.7819,
"step": 321000
},
{
"epoch": 17.31,
"learning_rate": 0.00014726716479378712,
"loss": 2.7854,
"step": 322000
},
{
"epoch": 17.37,
"learning_rate": 0.00014432546920044716,
"loss": 2.7879,
"step": 323000
},
{
"epoch": 17.42,
"learning_rate": 0.00014138671530270047,
"loss": 2.7875,
"step": 324000
},
{
"epoch": 17.48,
"learning_rate": 0.00013844501970936045,
"loss": 2.7859,
"step": 325000
},
{
"epoch": 17.53,
"learning_rate": 0.00013550626581161382,
"loss": 2.789,
"step": 326000
},
{
"epoch": 17.58,
"learning_rate": 0.0001325645702182738,
"loss": 2.7869,
"step": 327000
},
{
"epoch": 17.64,
"learning_rate": 0.00012962581632052717,
"loss": 2.7897,
"step": 328000
},
{
"epoch": 17.69,
"learning_rate": 0.0001266870624227805,
"loss": 2.795,
"step": 329000
},
{
"epoch": 17.74,
"learning_rate": 0.0001237453668294405,
"loss": 2.7895,
"step": 330000
},
{
"epoch": 17.8,
"learning_rate": 0.00012080367123610049,
"loss": 2.7904,
"step": 331000
},
{
"epoch": 17.85,
"learning_rate": 0.0001178619756427605,
"loss": 2.7978,
"step": 332000
},
{
"epoch": 17.91,
"learning_rate": 0.00011492322174501383,
"loss": 2.7948,
"step": 333000
},
{
"epoch": 17.96,
"learning_rate": 0.00011198152615167382,
"loss": 2.7958,
"step": 334000
},
{
"epoch": 18.0,
"eval_accuracy": 0.4110488019893414,
"eval_loss": 3.4054245948791504,
"eval_runtime": 153.0764,
"eval_samples_per_second": 378.36,
"eval_steps_per_second": 5.912,
"step": 334746
},
{
"epoch": 18.01,
"learning_rate": 0.00010904277225392716,
"loss": 2.7879,
"step": 335000
},
{
"epoch": 18.07,
"learning_rate": 0.00010610107666058716,
"loss": 2.7645,
"step": 336000
},
{
"epoch": 18.12,
"learning_rate": 0.00010315938106724717,
"loss": 2.761,
"step": 337000
},
{
"epoch": 18.17,
"learning_rate": 0.00010022062716950051,
"loss": 2.7695,
"step": 338000
},
{
"epoch": 18.23,
"learning_rate": 9.727893157616049e-05,
"loss": 2.77,
"step": 339000
},
{
"epoch": 18.28,
"learning_rate": 9.434017767841383e-05,
"loss": 2.7661,
"step": 340000
},
{
"epoch": 18.34,
"learning_rate": 9.139848208507384e-05,
"loss": 2.7676,
"step": 341000
},
{
"epoch": 18.39,
"learning_rate": 8.845678649173384e-05,
"loss": 2.7748,
"step": 342000
},
{
"epoch": 18.44,
"learning_rate": 8.551509089839383e-05,
"loss": 2.7737,
"step": 343000
},
{
"epoch": 18.5,
"learning_rate": 8.257339530505384e-05,
"loss": 2.7719,
"step": 344000
},
{
"epoch": 18.55,
"learning_rate": 7.963169971171383e-05,
"loss": 2.7714,
"step": 345000
},
{
"epoch": 18.61,
"learning_rate": 7.669294581396718e-05,
"loss": 2.7772,
"step": 346000
},
{
"epoch": 18.66,
"learning_rate": 7.375125022062717e-05,
"loss": 2.7765,
"step": 347000
},
{
"epoch": 18.71,
"learning_rate": 7.08124963228805e-05,
"loss": 2.7766,
"step": 348000
},
{
"epoch": 18.77,
"learning_rate": 6.78708007295405e-05,
"loss": 2.7757,
"step": 349000
},
{
"epoch": 18.82,
"learning_rate": 6.492910513620051e-05,
"loss": 2.7752,
"step": 350000
},
{
"epoch": 18.87,
"learning_rate": 6.199035123845384e-05,
"loss": 2.7737,
"step": 351000
},
{
"epoch": 18.93,
"learning_rate": 5.9048655645113845e-05,
"loss": 2.7775,
"step": 352000
},
{
"epoch": 18.98,
"learning_rate": 5.610990174736719e-05,
"loss": 2.781,
"step": 353000
},
{
"epoch": 19.0,
"eval_accuracy": 0.4109530675530399,
"eval_loss": 3.40570068359375,
"eval_runtime": 153.2556,
"eval_samples_per_second": 377.918,
"eval_steps_per_second": 5.905,
"step": 353343
},
{
"epoch": 19.04,
"learning_rate": 5.316820615402718e-05,
"loss": 2.7637,
"step": 354000
},
{
"epoch": 19.09,
"learning_rate": 5.022945225628052e-05,
"loss": 2.7537,
"step": 355000
},
{
"epoch": 19.14,
"learning_rate": 4.7287756662940525e-05,
"loss": 2.7517,
"step": 356000
},
{
"epoch": 19.2,
"learning_rate": 4.434606106960052e-05,
"loss": 2.7559,
"step": 357000
},
{
"epoch": 19.25,
"learning_rate": 4.1404365476260514e-05,
"loss": 2.7567,
"step": 358000
},
{
"epoch": 19.3,
"learning_rate": 3.846561157851386e-05,
"loss": 2.7585,
"step": 359000
},
{
"epoch": 19.36,
"learning_rate": 3.552391598517386e-05,
"loss": 2.7578,
"step": 360000
},
{
"epoch": 19.41,
"learning_rate": 3.2585162087427194e-05,
"loss": 2.7605,
"step": 361000
},
{
"epoch": 19.47,
"learning_rate": 2.9643466494087192e-05,
"loss": 2.7599,
"step": 362000
},
{
"epoch": 19.52,
"learning_rate": 2.6707654291933872e-05,
"loss": 2.7564,
"step": 363000
},
{
"epoch": 19.57,
"learning_rate": 2.376595869859387e-05,
"loss": 2.7599,
"step": 364000
},
{
"epoch": 19.63,
"learning_rate": 2.0824263105253868e-05,
"loss": 2.7638,
"step": 365000
},
{
"epoch": 19.68,
"learning_rate": 1.788256751191387e-05,
"loss": 2.7592,
"step": 366000
},
{
"epoch": 19.73,
"learning_rate": 1.4943813614167205e-05,
"loss": 2.7596,
"step": 367000
},
{
"epoch": 19.79,
"learning_rate": 1.2002118020827206e-05,
"loss": 2.7557,
"step": 368000
},
{
"epoch": 19.84,
"learning_rate": 9.060422427487204e-06,
"loss": 2.7601,
"step": 369000
},
{
"epoch": 19.9,
"learning_rate": 6.121668529740543e-06,
"loss": 2.7576,
"step": 370000
},
{
"epoch": 19.95,
"learning_rate": 3.1799729364005417e-06,
"loss": 2.7571,
"step": 371000
},
{
"epoch": 20.0,
"eval_accuracy": 0.4109943845202858,
"eval_loss": 3.4142630100250244,
"eval_runtime": 153.3697,
"eval_samples_per_second": 377.637,
"eval_steps_per_second": 5.901,
"step": 371940
},
{
"epoch": 20.0,
"step": 371940,
"total_flos": 1.56702845389824e+18,
"train_loss": 3.027507337477987,
"train_runtime": 81074.8542,
"train_samples_per_second": 146.796,
"train_steps_per_second": 4.588
}
],
"logging_steps": 1000,
"max_steps": 371940,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 5000,
"total_flos": 1.56702845389824e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}