kanishka's picture
End of training
39c4200 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 20.0,
"eval_steps": 500,
"global_step": 372000,
"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": 6.2221,
"step": 1000
},
{
"epoch": 0.11,
"learning_rate": 6.25e-05,
"loss": 5.01,
"step": 2000
},
{
"epoch": 0.16,
"learning_rate": 9.375e-05,
"loss": 4.6724,
"step": 3000
},
{
"epoch": 0.22,
"learning_rate": 0.000125,
"loss": 4.4552,
"step": 4000
},
{
"epoch": 0.27,
"learning_rate": 0.00015625,
"loss": 4.293,
"step": 5000
},
{
"epoch": 0.32,
"learning_rate": 0.0001875,
"loss": 4.1759,
"step": 6000
},
{
"epoch": 0.38,
"learning_rate": 0.00021875,
"loss": 4.0792,
"step": 7000
},
{
"epoch": 0.43,
"learning_rate": 0.00025,
"loss": 3.9873,
"step": 8000
},
{
"epoch": 0.48,
"learning_rate": 0.00028121875,
"loss": 3.9096,
"step": 9000
},
{
"epoch": 0.54,
"learning_rate": 0.00031246875000000003,
"loss": 3.8535,
"step": 10000
},
{
"epoch": 0.59,
"learning_rate": 0.00034371875,
"loss": 3.8057,
"step": 11000
},
{
"epoch": 0.65,
"learning_rate": 0.0003749375,
"loss": 3.7677,
"step": 12000
},
{
"epoch": 0.7,
"learning_rate": 0.0004061875,
"loss": 3.7296,
"step": 13000
},
{
"epoch": 0.75,
"learning_rate": 0.00043737500000000005,
"loss": 3.695,
"step": 14000
},
{
"epoch": 0.81,
"learning_rate": 0.000468625,
"loss": 3.6718,
"step": 15000
},
{
"epoch": 0.86,
"learning_rate": 0.00049984375,
"loss": 3.6435,
"step": 16000
},
{
"epoch": 0.91,
"learning_rate": 0.00053109375,
"loss": 3.6252,
"step": 17000
},
{
"epoch": 0.97,
"learning_rate": 0.00056234375,
"loss": 3.6018,
"step": 18000
},
{
"epoch": 1.0,
"eval_accuracy": 0.3590134207353542,
"eval_loss": 3.777891159057617,
"eval_runtime": 153.4152,
"eval_samples_per_second": 377.544,
"eval_steps_per_second": 5.906,
"step": 18600
},
{
"epoch": 1.02,
"learning_rate": 0.00059359375,
"loss": 3.5728,
"step": 19000
},
{
"epoch": 1.08,
"learning_rate": 0.0006248125,
"loss": 3.5424,
"step": 20000
},
{
"epoch": 1.13,
"learning_rate": 0.0006560312499999999,
"loss": 3.5371,
"step": 21000
},
{
"epoch": 1.18,
"learning_rate": 0.00068728125,
"loss": 3.5291,
"step": 22000
},
{
"epoch": 1.24,
"learning_rate": 0.00071853125,
"loss": 3.5138,
"step": 23000
},
{
"epoch": 1.29,
"learning_rate": 0.00074978125,
"loss": 3.5058,
"step": 24000
},
{
"epoch": 1.34,
"learning_rate": 0.0007810312499999999,
"loss": 3.4928,
"step": 25000
},
{
"epoch": 1.4,
"learning_rate": 0.00081225,
"loss": 3.4809,
"step": 26000
},
{
"epoch": 1.45,
"learning_rate": 0.00084346875,
"loss": 3.4777,
"step": 27000
},
{
"epoch": 1.51,
"learning_rate": 0.00087471875,
"loss": 3.4636,
"step": 28000
},
{
"epoch": 1.56,
"learning_rate": 0.00090596875,
"loss": 3.4606,
"step": 29000
},
{
"epoch": 1.61,
"learning_rate": 0.00093721875,
"loss": 3.4511,
"step": 30000
},
{
"epoch": 1.67,
"learning_rate": 0.0009684375,
"loss": 3.4397,
"step": 31000
},
{
"epoch": 1.72,
"learning_rate": 0.0009996875,
"loss": 3.4352,
"step": 32000
},
{
"epoch": 1.77,
"learning_rate": 0.0009970882352941177,
"loss": 3.4243,
"step": 33000
},
{
"epoch": 1.83,
"learning_rate": 0.00099415,
"loss": 3.4137,
"step": 34000
},
{
"epoch": 1.88,
"learning_rate": 0.0009912117647058825,
"loss": 3.4023,
"step": 35000
},
{
"epoch": 1.94,
"learning_rate": 0.000988270588235294,
"loss": 3.397,
"step": 36000
},
{
"epoch": 1.99,
"learning_rate": 0.000985329411764706,
"loss": 3.3799,
"step": 37000
},
{
"epoch": 2.0,
"eval_accuracy": 0.37990427992723486,
"eval_loss": 3.5989933013916016,
"eval_runtime": 154.3275,
"eval_samples_per_second": 375.312,
"eval_steps_per_second": 5.871,
"step": 37200
},
{
"epoch": 2.04,
"learning_rate": 0.0009823882352941178,
"loss": 3.3375,
"step": 38000
},
{
"epoch": 2.1,
"learning_rate": 0.00097945,
"loss": 3.3268,
"step": 39000
},
{
"epoch": 2.15,
"learning_rate": 0.0009765117647058823,
"loss": 3.3221,
"step": 40000
},
{
"epoch": 2.2,
"learning_rate": 0.0009735705882352941,
"loss": 3.3203,
"step": 41000
},
{
"epoch": 2.26,
"learning_rate": 0.0009706294117647059,
"loss": 3.3119,
"step": 42000
},
{
"epoch": 2.31,
"learning_rate": 0.0009676882352941176,
"loss": 3.3096,
"step": 43000
},
{
"epoch": 2.37,
"learning_rate": 0.0009647470588235294,
"loss": 3.3047,
"step": 44000
},
{
"epoch": 2.42,
"learning_rate": 0.0009618058823529412,
"loss": 3.2991,
"step": 45000
},
{
"epoch": 2.47,
"learning_rate": 0.0009588705882352941,
"loss": 3.293,
"step": 46000
},
{
"epoch": 2.53,
"learning_rate": 0.0009559294117647059,
"loss": 3.2854,
"step": 47000
},
{
"epoch": 2.58,
"learning_rate": 0.0009529911764705883,
"loss": 3.2802,
"step": 48000
},
{
"epoch": 2.63,
"learning_rate": 0.0009500499999999999,
"loss": 3.2805,
"step": 49000
},
{
"epoch": 2.69,
"learning_rate": 0.0009471088235294118,
"loss": 3.2725,
"step": 50000
},
{
"epoch": 2.74,
"learning_rate": 0.0009441705882352942,
"loss": 3.2704,
"step": 51000
},
{
"epoch": 2.8,
"learning_rate": 0.0009412323529411765,
"loss": 3.2664,
"step": 52000
},
{
"epoch": 2.85,
"learning_rate": 0.0009382911764705883,
"loss": 3.2581,
"step": 53000
},
{
"epoch": 2.9,
"learning_rate": 0.0009353529411764706,
"loss": 3.2615,
"step": 54000
},
{
"epoch": 2.96,
"learning_rate": 0.0009324117647058825,
"loss": 3.2535,
"step": 55000
},
{
"epoch": 3.0,
"eval_accuracy": 0.3928313870690771,
"eval_loss": 3.4629178047180176,
"eval_runtime": 154.8362,
"eval_samples_per_second": 374.079,
"eval_steps_per_second": 5.851,
"step": 55800
},
{
"epoch": 3.01,
"learning_rate": 0.0009294735294117647,
"loss": 3.2392,
"step": 56000
},
{
"epoch": 3.06,
"learning_rate": 0.000926535294117647,
"loss": 3.1885,
"step": 57000
},
{
"epoch": 3.12,
"learning_rate": 0.0009235941176470588,
"loss": 3.1928,
"step": 58000
},
{
"epoch": 3.17,
"learning_rate": 0.0009206529411764706,
"loss": 3.1915,
"step": 59000
},
{
"epoch": 3.23,
"learning_rate": 0.0009177117647058824,
"loss": 3.1906,
"step": 60000
},
{
"epoch": 3.28,
"learning_rate": 0.0009147705882352941,
"loss": 3.1953,
"step": 61000
},
{
"epoch": 3.33,
"learning_rate": 0.0009118323529411765,
"loss": 3.1935,
"step": 62000
},
{
"epoch": 3.39,
"learning_rate": 0.0009088911764705882,
"loss": 3.1937,
"step": 63000
},
{
"epoch": 3.44,
"learning_rate": 0.0009059500000000001,
"loss": 3.1911,
"step": 64000
},
{
"epoch": 3.49,
"learning_rate": 0.0009030088235294118,
"loss": 3.188,
"step": 65000
},
{
"epoch": 3.55,
"learning_rate": 0.0009000676470588236,
"loss": 3.1861,
"step": 66000
},
{
"epoch": 3.6,
"learning_rate": 0.0008971294117647059,
"loss": 3.1872,
"step": 67000
},
{
"epoch": 3.66,
"learning_rate": 0.0008941882352941176,
"loss": 3.1838,
"step": 68000
},
{
"epoch": 3.71,
"learning_rate": 0.0008912470588235294,
"loss": 3.1827,
"step": 69000
},
{
"epoch": 3.76,
"learning_rate": 0.0008883088235294117,
"loss": 3.1813,
"step": 70000
},
{
"epoch": 3.82,
"learning_rate": 0.0008853705882352941,
"loss": 3.183,
"step": 71000
},
{
"epoch": 3.87,
"learning_rate": 0.0008824294117647059,
"loss": 3.1796,
"step": 72000
},
{
"epoch": 3.92,
"learning_rate": 0.0008794882352941177,
"loss": 3.1769,
"step": 73000
},
{
"epoch": 3.98,
"learning_rate": 0.0008765470588235295,
"loss": 3.1731,
"step": 74000
},
{
"epoch": 4.0,
"eval_accuracy": 0.3979334659304163,
"eval_loss": 3.4446632862091064,
"eval_runtime": 154.4222,
"eval_samples_per_second": 375.082,
"eval_steps_per_second": 5.867,
"step": 74400
},
{
"epoch": 4.03,
"learning_rate": 0.0008736088235294118,
"loss": 3.13,
"step": 75000
},
{
"epoch": 4.09,
"learning_rate": 0.0008706676470588235,
"loss": 3.1105,
"step": 76000
},
{
"epoch": 4.14,
"learning_rate": 0.0008677294117647059,
"loss": 3.1116,
"step": 77000
},
{
"epoch": 4.19,
"learning_rate": 0.0008647911764705882,
"loss": 3.1197,
"step": 78000
},
{
"epoch": 4.25,
"learning_rate": 0.00086185,
"loss": 3.1205,
"step": 79000
},
{
"epoch": 4.3,
"learning_rate": 0.0008589088235294118,
"loss": 3.1217,
"step": 80000
},
{
"epoch": 4.35,
"learning_rate": 0.0008559676470588235,
"loss": 3.1201,
"step": 81000
},
{
"epoch": 4.41,
"learning_rate": 0.0008530294117647059,
"loss": 3.1243,
"step": 82000
},
{
"epoch": 4.46,
"learning_rate": 0.0008500882352941177,
"loss": 3.1208,
"step": 83000
},
{
"epoch": 4.52,
"learning_rate": 0.0008471470588235294,
"loss": 3.1247,
"step": 84000
},
{
"epoch": 4.57,
"learning_rate": 0.0008442088235294117,
"loss": 3.123,
"step": 85000
},
{
"epoch": 4.62,
"learning_rate": 0.0008412676470588235,
"loss": 3.1255,
"step": 86000
},
{
"epoch": 4.68,
"learning_rate": 0.0008383294117647058,
"loss": 3.1271,
"step": 87000
},
{
"epoch": 4.73,
"learning_rate": 0.0008353882352941177,
"loss": 3.1227,
"step": 88000
},
{
"epoch": 4.78,
"learning_rate": 0.00083245,
"loss": 3.1181,
"step": 89000
},
{
"epoch": 4.84,
"learning_rate": 0.0008295088235294118,
"loss": 3.1193,
"step": 90000
},
{
"epoch": 4.89,
"learning_rate": 0.0008265676470588236,
"loss": 3.1207,
"step": 91000
},
{
"epoch": 4.95,
"learning_rate": 0.0008236294117647059,
"loss": 3.1218,
"step": 92000
},
{
"epoch": 5.0,
"learning_rate": 0.0008206882352941176,
"loss": 3.1186,
"step": 93000
},
{
"epoch": 5.0,
"eval_accuracy": 0.40091726977917125,
"eval_loss": 3.4295449256896973,
"eval_runtime": 154.7977,
"eval_samples_per_second": 374.172,
"eval_steps_per_second": 5.853,
"step": 93000
},
{
"epoch": 5.05,
"learning_rate": 0.00081775,
"loss": 3.0494,
"step": 94000
},
{
"epoch": 5.11,
"learning_rate": 0.0008148088235294117,
"loss": 3.0575,
"step": 95000
},
{
"epoch": 5.16,
"learning_rate": 0.0008118705882352942,
"loss": 3.0613,
"step": 96000
},
{
"epoch": 5.22,
"learning_rate": 0.0008089294117647059,
"loss": 3.0613,
"step": 97000
},
{
"epoch": 5.27,
"learning_rate": 0.0008059911764705883,
"loss": 3.0663,
"step": 98000
},
{
"epoch": 5.32,
"learning_rate": 0.0008030529411764707,
"loss": 3.072,
"step": 99000
},
{
"epoch": 5.38,
"learning_rate": 0.0008001117647058823,
"loss": 3.0672,
"step": 100000
},
{
"epoch": 5.43,
"learning_rate": 0.0007971705882352941,
"loss": 3.0724,
"step": 101000
},
{
"epoch": 5.48,
"learning_rate": 0.0007942294117647059,
"loss": 3.0778,
"step": 102000
},
{
"epoch": 5.54,
"learning_rate": 0.0007912882352941176,
"loss": 3.0744,
"step": 103000
},
{
"epoch": 5.59,
"learning_rate": 0.0007883470588235294,
"loss": 3.0717,
"step": 104000
},
{
"epoch": 5.65,
"learning_rate": 0.0007854088235294118,
"loss": 3.074,
"step": 105000
},
{
"epoch": 5.7,
"learning_rate": 0.0007824676470588235,
"loss": 3.0752,
"step": 106000
},
{
"epoch": 5.75,
"learning_rate": 0.0007795294117647059,
"loss": 3.0816,
"step": 107000
},
{
"epoch": 5.81,
"learning_rate": 0.0007765882352941177,
"loss": 3.0765,
"step": 108000
},
{
"epoch": 5.86,
"learning_rate": 0.0007736470588235295,
"loss": 3.0727,
"step": 109000
},
{
"epoch": 5.91,
"learning_rate": 0.0007707088235294117,
"loss": 3.0803,
"step": 110000
},
{
"epoch": 5.97,
"learning_rate": 0.0007677676470588236,
"loss": 3.0776,
"step": 111000
},
{
"epoch": 6.0,
"eval_accuracy": 0.40341470070229163,
"eval_loss": 3.4003686904907227,
"eval_runtime": 154.7855,
"eval_samples_per_second": 374.202,
"eval_steps_per_second": 5.853,
"step": 111600
},
{
"epoch": 6.02,
"learning_rate": 0.0007648264705882353,
"loss": 3.047,
"step": 112000
},
{
"epoch": 6.08,
"learning_rate": 0.000761885294117647,
"loss": 3.0065,
"step": 113000
},
{
"epoch": 6.13,
"learning_rate": 0.00075895,
"loss": 3.0106,
"step": 114000
},
{
"epoch": 6.18,
"learning_rate": 0.0007560088235294118,
"loss": 3.0214,
"step": 115000
},
{
"epoch": 6.24,
"learning_rate": 0.0007530676470588236,
"loss": 3.0263,
"step": 116000
},
{
"epoch": 6.29,
"learning_rate": 0.000750129411764706,
"loss": 3.0267,
"step": 117000
},
{
"epoch": 6.34,
"learning_rate": 0.0007471882352941176,
"loss": 3.0283,
"step": 118000
},
{
"epoch": 6.4,
"learning_rate": 0.0007442470588235293,
"loss": 3.0315,
"step": 119000
},
{
"epoch": 6.45,
"learning_rate": 0.0007413088235294117,
"loss": 3.0305,
"step": 120000
},
{
"epoch": 6.51,
"learning_rate": 0.0007383676470588235,
"loss": 3.0363,
"step": 121000
},
{
"epoch": 6.56,
"learning_rate": 0.0007354294117647059,
"loss": 3.0361,
"step": 122000
},
{
"epoch": 6.61,
"learning_rate": 0.0007324882352941176,
"loss": 3.0386,
"step": 123000
},
{
"epoch": 6.67,
"learning_rate": 0.0007295470588235295,
"loss": 3.0368,
"step": 124000
},
{
"epoch": 6.72,
"learning_rate": 0.0007266058823529412,
"loss": 3.0394,
"step": 125000
},
{
"epoch": 6.77,
"learning_rate": 0.0007236705882352941,
"loss": 3.0383,
"step": 126000
},
{
"epoch": 6.83,
"learning_rate": 0.0007207294117647058,
"loss": 3.0374,
"step": 127000
},
{
"epoch": 6.88,
"learning_rate": 0.0007177882352941177,
"loss": 3.0376,
"step": 128000
},
{
"epoch": 6.94,
"learning_rate": 0.00071485,
"loss": 3.0406,
"step": 129000
},
{
"epoch": 6.99,
"learning_rate": 0.0007119088235294118,
"loss": 3.0407,
"step": 130000
},
{
"epoch": 7.0,
"eval_accuracy": 0.4053405090806296,
"eval_loss": 3.3849599361419678,
"eval_runtime": 154.433,
"eval_samples_per_second": 375.056,
"eval_steps_per_second": 5.867,
"step": 130200
},
{
"epoch": 7.04,
"learning_rate": 0.0007089705882352942,
"loss": 2.9848,
"step": 131000
},
{
"epoch": 7.1,
"learning_rate": 0.0007060294117647059,
"loss": 2.9769,
"step": 132000
},
{
"epoch": 7.15,
"learning_rate": 0.0007030882352941177,
"loss": 2.9808,
"step": 133000
},
{
"epoch": 7.2,
"learning_rate": 0.0007001500000000001,
"loss": 2.989,
"step": 134000
},
{
"epoch": 7.26,
"learning_rate": 0.0006972117647058824,
"loss": 2.9897,
"step": 135000
},
{
"epoch": 7.31,
"learning_rate": 0.0006942705882352941,
"loss": 2.9956,
"step": 136000
},
{
"epoch": 7.37,
"learning_rate": 0.0006913294117647058,
"loss": 2.9945,
"step": 137000
},
{
"epoch": 7.42,
"learning_rate": 0.0006883882352941177,
"loss": 2.9962,
"step": 138000
},
{
"epoch": 7.47,
"learning_rate": 0.00068545,
"loss": 2.9981,
"step": 139000
},
{
"epoch": 7.53,
"learning_rate": 0.0006825088235294118,
"loss": 3.0051,
"step": 140000
},
{
"epoch": 7.58,
"learning_rate": 0.0006795676470588236,
"loss": 3.0027,
"step": 141000
},
{
"epoch": 7.63,
"learning_rate": 0.000676629411764706,
"loss": 3.0015,
"step": 142000
},
{
"epoch": 7.69,
"learning_rate": 0.0006736911764705882,
"loss": 3.0058,
"step": 143000
},
{
"epoch": 7.74,
"learning_rate": 0.00067075,
"loss": 3.0009,
"step": 144000
},
{
"epoch": 7.8,
"learning_rate": 0.0006678088235294118,
"loss": 3.0072,
"step": 145000
},
{
"epoch": 7.85,
"learning_rate": 0.0006648705882352942,
"loss": 3.0086,
"step": 146000
},
{
"epoch": 7.9,
"learning_rate": 0.0006619294117647059,
"loss": 3.007,
"step": 147000
},
{
"epoch": 7.96,
"learning_rate": 0.0006589882352941176,
"loss": 3.0066,
"step": 148000
},
{
"epoch": 8.0,
"eval_accuracy": 0.40609277482942185,
"eval_loss": 3.3647897243499756,
"eval_runtime": 154.7775,
"eval_samples_per_second": 374.221,
"eval_steps_per_second": 5.854,
"step": 148800
},
{
"epoch": 8.01,
"learning_rate": 0.0006560470588235295,
"loss": 2.9921,
"step": 149000
},
{
"epoch": 8.06,
"learning_rate": 0.0006531058823529412,
"loss": 2.9407,
"step": 150000
},
{
"epoch": 8.12,
"learning_rate": 0.0006501676470588236,
"loss": 2.9487,
"step": 151000
},
{
"epoch": 8.17,
"learning_rate": 0.0006472294117647058,
"loss": 2.9502,
"step": 152000
},
{
"epoch": 8.23,
"learning_rate": 0.0006442882352941177,
"loss": 2.9528,
"step": 153000
},
{
"epoch": 8.28,
"learning_rate": 0.0006413470588235294,
"loss": 2.9604,
"step": 154000
},
{
"epoch": 8.33,
"learning_rate": 0.0006384058823529411,
"loss": 2.9659,
"step": 155000
},
{
"epoch": 8.39,
"learning_rate": 0.0006354676470588235,
"loss": 2.9664,
"step": 156000
},
{
"epoch": 8.44,
"learning_rate": 0.0006325264705882353,
"loss": 2.9699,
"step": 157000
},
{
"epoch": 8.49,
"learning_rate": 0.0006295852941176471,
"loss": 2.9722,
"step": 158000
},
{
"epoch": 8.55,
"learning_rate": 0.0006266441176470589,
"loss": 2.9748,
"step": 159000
},
{
"epoch": 8.6,
"learning_rate": 0.0006237029411764706,
"loss": 2.9758,
"step": 160000
},
{
"epoch": 8.66,
"learning_rate": 0.000620764705882353,
"loss": 2.9756,
"step": 161000
},
{
"epoch": 8.71,
"learning_rate": 0.0006178235294117646,
"loss": 2.9808,
"step": 162000
},
{
"epoch": 8.76,
"learning_rate": 0.000614885294117647,
"loss": 2.9757,
"step": 163000
},
{
"epoch": 8.82,
"learning_rate": 0.0006119441176470588,
"loss": 2.9796,
"step": 164000
},
{
"epoch": 8.87,
"learning_rate": 0.0006090058823529412,
"loss": 2.9763,
"step": 165000
},
{
"epoch": 8.92,
"learning_rate": 0.0006060647058823529,
"loss": 2.9757,
"step": 166000
},
{
"epoch": 8.98,
"learning_rate": 0.0006031235294117648,
"loss": 2.9851,
"step": 167000
},
{
"epoch": 9.0,
"eval_accuracy": 0.40737870722479436,
"eval_loss": 3.3985021114349365,
"eval_runtime": 153.5744,
"eval_samples_per_second": 377.153,
"eval_steps_per_second": 5.899,
"step": 167400
},
{
"epoch": 9.03,
"learning_rate": 0.0006001823529411765,
"loss": 2.9402,
"step": 168000
},
{
"epoch": 9.09,
"learning_rate": 0.0005972441176470589,
"loss": 2.9204,
"step": 169000
},
{
"epoch": 9.14,
"learning_rate": 0.0005943029411764705,
"loss": 2.9245,
"step": 170000
},
{
"epoch": 9.19,
"learning_rate": 0.0005913647058823529,
"loss": 2.9281,
"step": 171000
},
{
"epoch": 9.25,
"learning_rate": 0.0005884235294117647,
"loss": 2.9311,
"step": 172000
},
{
"epoch": 9.3,
"learning_rate": 0.0005854823529411765,
"loss": 2.9334,
"step": 173000
},
{
"epoch": 9.35,
"learning_rate": 0.0005825411764705883,
"loss": 2.9404,
"step": 174000
},
{
"epoch": 9.41,
"learning_rate": 0.0005796,
"loss": 2.9397,
"step": 175000
},
{
"epoch": 9.46,
"learning_rate": 0.000576664705882353,
"loss": 2.9381,
"step": 176000
},
{
"epoch": 9.52,
"learning_rate": 0.0005737264705882354,
"loss": 2.9492,
"step": 177000
},
{
"epoch": 9.57,
"learning_rate": 0.0005707852941176471,
"loss": 2.9445,
"step": 178000
},
{
"epoch": 9.62,
"learning_rate": 0.0005678441176470588,
"loss": 2.9501,
"step": 179000
},
{
"epoch": 9.68,
"learning_rate": 0.0005649029411764706,
"loss": 2.9517,
"step": 180000
},
{
"epoch": 9.73,
"learning_rate": 0.000561964705882353,
"loss": 2.9497,
"step": 181000
},
{
"epoch": 9.78,
"learning_rate": 0.0005590235294117647,
"loss": 2.9519,
"step": 182000
},
{
"epoch": 9.84,
"learning_rate": 0.0005560823529411765,
"loss": 2.9527,
"step": 183000
},
{
"epoch": 9.89,
"learning_rate": 0.0005531441176470589,
"loss": 2.9517,
"step": 184000
},
{
"epoch": 9.95,
"learning_rate": 0.0005502029411764706,
"loss": 2.9579,
"step": 185000
},
{
"epoch": 10.0,
"learning_rate": 0.0005472676470588236,
"loss": 2.953,
"step": 186000
},
{
"epoch": 10.0,
"eval_accuracy": 0.4077225943803639,
"eval_loss": 3.3963661193847656,
"eval_runtime": 153.0934,
"eval_samples_per_second": 378.338,
"eval_steps_per_second": 5.918,
"step": 186000
},
{
"epoch": 10.05,
"learning_rate": 0.0005443264705882353,
"loss": 2.8894,
"step": 187000
},
{
"epoch": 10.11,
"learning_rate": 0.000541385294117647,
"loss": 2.8973,
"step": 188000
},
{
"epoch": 10.16,
"learning_rate": 0.0005384441176470588,
"loss": 2.9017,
"step": 189000
},
{
"epoch": 10.22,
"learning_rate": 0.0005355029411764706,
"loss": 2.9084,
"step": 190000
},
{
"epoch": 10.27,
"learning_rate": 0.0005325647058823529,
"loss": 2.9047,
"step": 191000
},
{
"epoch": 10.32,
"learning_rate": 0.0005296264705882353,
"loss": 2.9129,
"step": 192000
},
{
"epoch": 10.38,
"learning_rate": 0.0005266852941176471,
"loss": 2.9153,
"step": 193000
},
{
"epoch": 10.43,
"learning_rate": 0.0005237441176470589,
"loss": 2.9163,
"step": 194000
},
{
"epoch": 10.48,
"learning_rate": 0.0005208029411764706,
"loss": 2.9232,
"step": 195000
},
{
"epoch": 10.54,
"learning_rate": 0.0005178617647058823,
"loss": 2.9191,
"step": 196000
},
{
"epoch": 10.59,
"learning_rate": 0.0005149205882352941,
"loss": 2.9213,
"step": 197000
},
{
"epoch": 10.65,
"learning_rate": 0.0005119823529411764,
"loss": 2.9242,
"step": 198000
},
{
"epoch": 10.7,
"learning_rate": 0.0005090441176470588,
"loss": 2.925,
"step": 199000
},
{
"epoch": 10.75,
"learning_rate": 0.0005061029411764706,
"loss": 2.9285,
"step": 200000
},
{
"epoch": 10.81,
"learning_rate": 0.000503164705882353,
"loss": 2.9255,
"step": 201000
},
{
"epoch": 10.86,
"learning_rate": 0.0005002235294117647,
"loss": 2.9297,
"step": 202000
},
{
"epoch": 10.91,
"learning_rate": 0.0004972852941176471,
"loss": 2.9308,
"step": 203000
},
{
"epoch": 10.97,
"learning_rate": 0.0004943441176470588,
"loss": 2.9321,
"step": 204000
},
{
"epoch": 11.0,
"eval_accuracy": 0.4087668854202796,
"eval_loss": 3.3815512657165527,
"eval_runtime": 153.3753,
"eval_samples_per_second": 377.642,
"eval_steps_per_second": 5.907,
"step": 204600
},
{
"epoch": 11.02,
"learning_rate": 0.0004914058823529412,
"loss": 2.9055,
"step": 205000
},
{
"epoch": 11.08,
"learning_rate": 0.000488464705882353,
"loss": 2.872,
"step": 206000
},
{
"epoch": 11.13,
"learning_rate": 0.0004855235294117647,
"loss": 2.8789,
"step": 207000
},
{
"epoch": 11.18,
"learning_rate": 0.00048258235294117646,
"loss": 2.8794,
"step": 208000
},
{
"epoch": 11.24,
"learning_rate": 0.00047964117647058825,
"loss": 2.8879,
"step": 209000
},
{
"epoch": 11.29,
"learning_rate": 0.0004767029411764706,
"loss": 2.8892,
"step": 210000
},
{
"epoch": 11.34,
"learning_rate": 0.00047376176470588234,
"loss": 2.8894,
"step": 211000
},
{
"epoch": 11.4,
"learning_rate": 0.0004708235294117647,
"loss": 2.8948,
"step": 212000
},
{
"epoch": 11.45,
"learning_rate": 0.0004678823529411765,
"loss": 2.896,
"step": 213000
},
{
"epoch": 11.51,
"learning_rate": 0.00046494117647058827,
"loss": 2.8989,
"step": 214000
},
{
"epoch": 11.56,
"learning_rate": 0.00046200294117647057,
"loss": 2.9017,
"step": 215000
},
{
"epoch": 11.61,
"learning_rate": 0.00045906176470588236,
"loss": 2.9009,
"step": 216000
},
{
"epoch": 11.67,
"learning_rate": 0.0004561235294117647,
"loss": 2.8993,
"step": 217000
},
{
"epoch": 11.72,
"learning_rate": 0.0004531823529411765,
"loss": 2.9067,
"step": 218000
},
{
"epoch": 11.77,
"learning_rate": 0.00045024117647058824,
"loss": 2.9063,
"step": 219000
},
{
"epoch": 11.83,
"learning_rate": 0.0004473029411764706,
"loss": 2.9044,
"step": 220000
},
{
"epoch": 11.88,
"learning_rate": 0.0004443617647058824,
"loss": 2.9051,
"step": 221000
},
{
"epoch": 11.94,
"learning_rate": 0.0004414205882352941,
"loss": 2.9095,
"step": 222000
},
{
"epoch": 11.99,
"learning_rate": 0.0004384823529411765,
"loss": 2.9082,
"step": 223000
},
{
"epoch": 12.0,
"eval_accuracy": 0.40932990910670825,
"eval_loss": 3.378004312515259,
"eval_runtime": 153.328,
"eval_samples_per_second": 377.759,
"eval_steps_per_second": 5.909,
"step": 223200
},
{
"epoch": 12.04,
"learning_rate": 0.0004355411764705882,
"loss": 2.8597,
"step": 224000
},
{
"epoch": 12.1,
"learning_rate": 0.00043260588235294117,
"loss": 2.8528,
"step": 225000
},
{
"epoch": 12.15,
"learning_rate": 0.00042966470588235296,
"loss": 2.8582,
"step": 226000
},
{
"epoch": 12.2,
"learning_rate": 0.00042672352941176475,
"loss": 2.8634,
"step": 227000
},
{
"epoch": 12.26,
"learning_rate": 0.00042378235294117644,
"loss": 2.8646,
"step": 228000
},
{
"epoch": 12.31,
"learning_rate": 0.00042084117647058823,
"loss": 2.867,
"step": 229000
},
{
"epoch": 12.37,
"learning_rate": 0.0004179029411764706,
"loss": 2.868,
"step": 230000
},
{
"epoch": 12.42,
"learning_rate": 0.00041496176470588237,
"loss": 2.875,
"step": 231000
},
{
"epoch": 12.47,
"learning_rate": 0.00041202058823529416,
"loss": 2.8725,
"step": 232000
},
{
"epoch": 12.53,
"learning_rate": 0.00040908235294117646,
"loss": 2.8762,
"step": 233000
},
{
"epoch": 12.58,
"learning_rate": 0.00040614117647058825,
"loss": 2.8777,
"step": 234000
},
{
"epoch": 12.63,
"learning_rate": 0.0004032,
"loss": 2.8833,
"step": 235000
},
{
"epoch": 12.69,
"learning_rate": 0.0004002617647058824,
"loss": 2.8854,
"step": 236000
},
{
"epoch": 12.74,
"learning_rate": 0.0003973205882352941,
"loss": 2.8843,
"step": 237000
},
{
"epoch": 12.8,
"learning_rate": 0.0003943823529411765,
"loss": 2.8856,
"step": 238000
},
{
"epoch": 12.85,
"learning_rate": 0.0003914411764705882,
"loss": 2.8873,
"step": 239000
},
{
"epoch": 12.9,
"learning_rate": 0.0003885,
"loss": 2.8843,
"step": 240000
},
{
"epoch": 12.96,
"learning_rate": 0.0003855647058823529,
"loss": 2.8881,
"step": 241000
},
{
"epoch": 13.0,
"eval_accuracy": 0.40904023506591597,
"eval_loss": 3.4020349979400635,
"eval_runtime": 153.5865,
"eval_samples_per_second": 377.123,
"eval_steps_per_second": 5.899,
"step": 241800
},
{
"epoch": 13.01,
"learning_rate": 0.0003826235294117647,
"loss": 2.8763,
"step": 242000
},
{
"epoch": 13.06,
"learning_rate": 0.0003796823529411765,
"loss": 2.8312,
"step": 243000
},
{
"epoch": 13.12,
"learning_rate": 0.00037674117647058824,
"loss": 2.8358,
"step": 244000
},
{
"epoch": 13.17,
"learning_rate": 0.00037380000000000003,
"loss": 2.8381,
"step": 245000
},
{
"epoch": 13.23,
"learning_rate": 0.00037086176470588233,
"loss": 2.8462,
"step": 246000
},
{
"epoch": 13.28,
"learning_rate": 0.0003679205882352941,
"loss": 2.849,
"step": 247000
},
{
"epoch": 13.33,
"learning_rate": 0.00036498235294117647,
"loss": 2.8528,
"step": 248000
},
{
"epoch": 13.39,
"learning_rate": 0.00036204117647058826,
"loss": 2.8491,
"step": 249000
},
{
"epoch": 13.44,
"learning_rate": 0.00035910294117647056,
"loss": 2.8541,
"step": 250000
},
{
"epoch": 13.49,
"learning_rate": 0.00035616470588235296,
"loss": 2.8578,
"step": 251000
},
{
"epoch": 13.55,
"learning_rate": 0.0003532235294117647,
"loss": 2.854,
"step": 252000
},
{
"epoch": 13.6,
"learning_rate": 0.0003502823529411765,
"loss": 2.8587,
"step": 253000
},
{
"epoch": 13.66,
"learning_rate": 0.00034734117647058823,
"loss": 2.8606,
"step": 254000
},
{
"epoch": 13.71,
"learning_rate": 0.0003444029411764706,
"loss": 2.8636,
"step": 255000
},
{
"epoch": 13.76,
"learning_rate": 0.0003414617647058824,
"loss": 2.8596,
"step": 256000
},
{
"epoch": 13.82,
"learning_rate": 0.0003385205882352941,
"loss": 2.8677,
"step": 257000
},
{
"epoch": 13.87,
"learning_rate": 0.0003355794117647059,
"loss": 2.8649,
"step": 258000
},
{
"epoch": 13.92,
"learning_rate": 0.0003326411764705882,
"loss": 2.8678,
"step": 259000
},
{
"epoch": 13.98,
"learning_rate": 0.0003297,
"loss": 2.8698,
"step": 260000
},
{
"epoch": 14.0,
"eval_accuracy": 0.4091352927578735,
"eval_loss": 3.405724287033081,
"eval_runtime": 153.5045,
"eval_samples_per_second": 377.325,
"eval_steps_per_second": 5.902,
"step": 260400
},
{
"epoch": 14.03,
"learning_rate": 0.0003267588235294118,
"loss": 2.8321,
"step": 261000
},
{
"epoch": 14.09,
"learning_rate": 0.0003238176470588236,
"loss": 2.8196,
"step": 262000
},
{
"epoch": 14.14,
"learning_rate": 0.00032088529411764704,
"loss": 2.8225,
"step": 263000
},
{
"epoch": 14.19,
"learning_rate": 0.00031794411764705883,
"loss": 2.8236,
"step": 264000
},
{
"epoch": 14.25,
"learning_rate": 0.0003150029411764706,
"loss": 2.8282,
"step": 265000
},
{
"epoch": 14.3,
"learning_rate": 0.00031206176470588236,
"loss": 2.8314,
"step": 266000
},
{
"epoch": 14.35,
"learning_rate": 0.0003091205882352941,
"loss": 2.8325,
"step": 267000
},
{
"epoch": 14.41,
"learning_rate": 0.00030618235294117645,
"loss": 2.8333,
"step": 268000
},
{
"epoch": 14.46,
"learning_rate": 0.00030324117647058824,
"loss": 2.8376,
"step": 269000
},
{
"epoch": 14.52,
"learning_rate": 0.0003003029411764706,
"loss": 2.8359,
"step": 270000
},
{
"epoch": 14.57,
"learning_rate": 0.00029736176470588233,
"loss": 2.8398,
"step": 271000
},
{
"epoch": 14.62,
"learning_rate": 0.0002944205882352941,
"loss": 2.8422,
"step": 272000
},
{
"epoch": 14.68,
"learning_rate": 0.0002914823529411765,
"loss": 2.8393,
"step": 273000
},
{
"epoch": 14.73,
"learning_rate": 0.00028854117647058827,
"loss": 2.8433,
"step": 274000
},
{
"epoch": 14.78,
"learning_rate": 0.0002856,
"loss": 2.8438,
"step": 275000
},
{
"epoch": 14.84,
"learning_rate": 0.00028265882352941174,
"loss": 2.8491,
"step": 276000
},
{
"epoch": 14.89,
"learning_rate": 0.0002797205882352941,
"loss": 2.8463,
"step": 277000
},
{
"epoch": 14.95,
"learning_rate": 0.0002767794117647059,
"loss": 2.8438,
"step": 278000
},
{
"epoch": 15.0,
"learning_rate": 0.00027384117647058823,
"loss": 2.8441,
"step": 279000
},
{
"epoch": 15.0,
"eval_accuracy": 0.40938311454277215,
"eval_loss": 3.390632390975952,
"eval_runtime": 153.3628,
"eval_samples_per_second": 377.673,
"eval_steps_per_second": 5.908,
"step": 279000
},
{
"epoch": 15.05,
"learning_rate": 0.00027089999999999997,
"loss": 2.7952,
"step": 280000
},
{
"epoch": 15.11,
"learning_rate": 0.00026796470588235293,
"loss": 2.8015,
"step": 281000
},
{
"epoch": 15.16,
"learning_rate": 0.0002650235294117647,
"loss": 2.8074,
"step": 282000
},
{
"epoch": 15.22,
"learning_rate": 0.0002620823529411765,
"loss": 2.8114,
"step": 283000
},
{
"epoch": 15.27,
"learning_rate": 0.0002591411764705882,
"loss": 2.8083,
"step": 284000
},
{
"epoch": 15.32,
"learning_rate": 0.0002562029411764706,
"loss": 2.813,
"step": 285000
},
{
"epoch": 15.38,
"learning_rate": 0.00025326176470588234,
"loss": 2.8181,
"step": 286000
},
{
"epoch": 15.43,
"learning_rate": 0.00025032058823529414,
"loss": 2.8161,
"step": 287000
},
{
"epoch": 15.48,
"learning_rate": 0.0002473823529411765,
"loss": 2.8206,
"step": 288000
},
{
"epoch": 15.54,
"learning_rate": 0.00024444411764705884,
"loss": 2.8213,
"step": 289000
},
{
"epoch": 15.59,
"learning_rate": 0.0002415029411764706,
"loss": 2.8226,
"step": 290000
},
{
"epoch": 15.65,
"learning_rate": 0.00023856176470588237,
"loss": 2.8245,
"step": 291000
},
{
"epoch": 15.7,
"learning_rate": 0.0002356205882352941,
"loss": 2.8228,
"step": 292000
},
{
"epoch": 15.75,
"learning_rate": 0.00023268235294117648,
"loss": 2.8208,
"step": 293000
},
{
"epoch": 15.81,
"learning_rate": 0.00022974117647058822,
"loss": 2.8244,
"step": 294000
},
{
"epoch": 15.86,
"learning_rate": 0.0002268,
"loss": 2.8224,
"step": 295000
},
{
"epoch": 15.91,
"learning_rate": 0.00022385882352941178,
"loss": 2.8282,
"step": 296000
},
{
"epoch": 15.97,
"learning_rate": 0.00022091764705882354,
"loss": 2.8256,
"step": 297000
},
{
"epoch": 16.0,
"eval_accuracy": 0.40941112801100277,
"eval_loss": 3.405087471008301,
"eval_runtime": 153.4984,
"eval_samples_per_second": 377.339,
"eval_steps_per_second": 5.902,
"step": 297600
},
{
"epoch": 16.02,
"learning_rate": 0.0002179764705882353,
"loss": 2.8096,
"step": 298000
},
{
"epoch": 16.08,
"learning_rate": 0.00021503823529411766,
"loss": 2.7852,
"step": 299000
},
{
"epoch": 16.13,
"learning_rate": 0.0002121,
"loss": 2.7916,
"step": 300000
},
{
"epoch": 16.18,
"learning_rate": 0.00020915882352941177,
"loss": 2.7874,
"step": 301000
},
{
"epoch": 16.24,
"learning_rate": 0.00020622058823529412,
"loss": 2.7938,
"step": 302000
},
{
"epoch": 16.29,
"learning_rate": 0.00020327941176470589,
"loss": 2.7946,
"step": 303000
},
{
"epoch": 16.34,
"learning_rate": 0.00020034117647058824,
"loss": 2.795,
"step": 304000
},
{
"epoch": 16.4,
"learning_rate": 0.0001974,
"loss": 2.7996,
"step": 305000
},
{
"epoch": 16.45,
"learning_rate": 0.00019445882352941177,
"loss": 2.798,
"step": 306000
},
{
"epoch": 16.51,
"learning_rate": 0.00019151764705882356,
"loss": 2.7995,
"step": 307000
},
{
"epoch": 16.56,
"learning_rate": 0.0001885764705882353,
"loss": 2.802,
"step": 308000
},
{
"epoch": 16.61,
"learning_rate": 0.00018563823529411767,
"loss": 2.8041,
"step": 309000
},
{
"epoch": 16.67,
"learning_rate": 0.0001827,
"loss": 2.8017,
"step": 310000
},
{
"epoch": 16.72,
"learning_rate": 0.0001797588235294118,
"loss": 2.8081,
"step": 311000
},
{
"epoch": 16.77,
"learning_rate": 0.00017681764705882353,
"loss": 2.8124,
"step": 312000
},
{
"epoch": 16.83,
"learning_rate": 0.0001738794117647059,
"loss": 2.8047,
"step": 313000
},
{
"epoch": 16.88,
"learning_rate": 0.00017094117647058823,
"loss": 2.8066,
"step": 314000
},
{
"epoch": 16.94,
"learning_rate": 0.00016800000000000002,
"loss": 2.8093,
"step": 315000
},
{
"epoch": 16.99,
"learning_rate": 0.00016505882352941176,
"loss": 2.808,
"step": 316000
},
{
"epoch": 17.0,
"eval_accuracy": 0.40925937159677506,
"eval_loss": 3.4107930660247803,
"eval_runtime": 153.936,
"eval_samples_per_second": 376.267,
"eval_steps_per_second": 5.886,
"step": 316200
},
{
"epoch": 17.04,
"learning_rate": 0.00016211764705882355,
"loss": 2.7802,
"step": 317000
},
{
"epoch": 17.1,
"learning_rate": 0.00015917647058823529,
"loss": 2.7744,
"step": 318000
},
{
"epoch": 17.15,
"learning_rate": 0.00015623529411764705,
"loss": 2.7771,
"step": 319000
},
{
"epoch": 17.2,
"learning_rate": 0.00015329705882352943,
"loss": 2.7733,
"step": 320000
},
{
"epoch": 17.26,
"learning_rate": 0.00015035588235294117,
"loss": 2.7755,
"step": 321000
},
{
"epoch": 17.31,
"learning_rate": 0.00014741764705882354,
"loss": 2.7811,
"step": 322000
},
{
"epoch": 17.37,
"learning_rate": 0.00014447647058823528,
"loss": 2.7841,
"step": 323000
},
{
"epoch": 17.42,
"learning_rate": 0.00014153529411764707,
"loss": 2.7813,
"step": 324000
},
{
"epoch": 17.47,
"learning_rate": 0.0001385941176470588,
"loss": 2.7853,
"step": 325000
},
{
"epoch": 17.53,
"learning_rate": 0.0001356529411764706,
"loss": 2.7845,
"step": 326000
},
{
"epoch": 17.58,
"learning_rate": 0.00013271470588235293,
"loss": 2.7857,
"step": 327000
},
{
"epoch": 17.63,
"learning_rate": 0.00012977352941176472,
"loss": 2.7843,
"step": 328000
},
{
"epoch": 17.69,
"learning_rate": 0.00012683529411764704,
"loss": 2.7848,
"step": 329000
},
{
"epoch": 17.74,
"learning_rate": 0.00012389411764705883,
"loss": 2.7864,
"step": 330000
},
{
"epoch": 17.8,
"learning_rate": 0.00012095588235294118,
"loss": 2.7855,
"step": 331000
},
{
"epoch": 17.85,
"learning_rate": 0.00011801470588235293,
"loss": 2.7859,
"step": 332000
},
{
"epoch": 17.9,
"learning_rate": 0.0001150735294117647,
"loss": 2.7935,
"step": 333000
},
{
"epoch": 17.96,
"learning_rate": 0.00011213529411764706,
"loss": 2.7945,
"step": 334000
},
{
"epoch": 18.0,
"eval_accuracy": 0.4093698131837562,
"eval_loss": 3.4282960891723633,
"eval_runtime": 153.2898,
"eval_samples_per_second": 377.853,
"eval_steps_per_second": 5.91,
"step": 334800
},
{
"epoch": 18.01,
"learning_rate": 0.00010919411764705883,
"loss": 2.7823,
"step": 335000
},
{
"epoch": 18.06,
"learning_rate": 0.00010625588235294118,
"loss": 2.7585,
"step": 336000
},
{
"epoch": 18.12,
"learning_rate": 0.00010331470588235294,
"loss": 2.7628,
"step": 337000
},
{
"epoch": 18.17,
"learning_rate": 0.00010037352941176471,
"loss": 2.7639,
"step": 338000
},
{
"epoch": 18.23,
"learning_rate": 9.743235294117647e-05,
"loss": 2.7637,
"step": 339000
},
{
"epoch": 18.28,
"learning_rate": 9.449705882352941e-05,
"loss": 2.7628,
"step": 340000
},
{
"epoch": 18.33,
"learning_rate": 9.155588235294119e-05,
"loss": 2.7672,
"step": 341000
},
{
"epoch": 18.39,
"learning_rate": 8.861764705882352e-05,
"loss": 2.7669,
"step": 342000
},
{
"epoch": 18.44,
"learning_rate": 8.56764705882353e-05,
"loss": 2.7623,
"step": 343000
},
{
"epoch": 18.49,
"learning_rate": 8.273529411764707e-05,
"loss": 2.7673,
"step": 344000
},
{
"epoch": 18.55,
"learning_rate": 7.979705882352942e-05,
"loss": 2.7704,
"step": 345000
},
{
"epoch": 18.6,
"learning_rate": 7.685588235294118e-05,
"loss": 2.7699,
"step": 346000
},
{
"epoch": 18.66,
"learning_rate": 7.391470588235293e-05,
"loss": 2.7699,
"step": 347000
},
{
"epoch": 18.71,
"learning_rate": 7.09764705882353e-05,
"loss": 2.7704,
"step": 348000
},
{
"epoch": 18.76,
"learning_rate": 6.803823529411765e-05,
"loss": 2.7712,
"step": 349000
},
{
"epoch": 18.82,
"learning_rate": 6.509705882352941e-05,
"loss": 2.7682,
"step": 350000
},
{
"epoch": 18.87,
"learning_rate": 6.215588235294118e-05,
"loss": 2.7717,
"step": 351000
},
{
"epoch": 18.92,
"learning_rate": 5.921470588235294e-05,
"loss": 2.7718,
"step": 352000
},
{
"epoch": 18.98,
"learning_rate": 5.62764705882353e-05,
"loss": 2.7744,
"step": 353000
},
{
"epoch": 19.0,
"eval_accuracy": 0.4094324236211445,
"eval_loss": 3.436245918273926,
"eval_runtime": 153.0633,
"eval_samples_per_second": 378.412,
"eval_steps_per_second": 5.919,
"step": 353400
},
{
"epoch": 19.03,
"learning_rate": 5.3335294117647057e-05,
"loss": 2.7619,
"step": 354000
},
{
"epoch": 19.09,
"learning_rate": 5.039411764705883e-05,
"loss": 2.7489,
"step": 355000
},
{
"epoch": 19.14,
"learning_rate": 4.745588235294118e-05,
"loss": 2.7515,
"step": 356000
},
{
"epoch": 19.19,
"learning_rate": 4.451470588235294e-05,
"loss": 2.7556,
"step": 357000
},
{
"epoch": 19.25,
"learning_rate": 4.15735294117647e-05,
"loss": 2.7508,
"step": 358000
},
{
"epoch": 19.3,
"learning_rate": 3.863235294117647e-05,
"loss": 2.7514,
"step": 359000
},
{
"epoch": 19.35,
"learning_rate": 3.569411764705882e-05,
"loss": 2.7511,
"step": 360000
},
{
"epoch": 19.41,
"learning_rate": 3.2758823529411765e-05,
"loss": 2.7546,
"step": 361000
},
{
"epoch": 19.46,
"learning_rate": 2.981764705882353e-05,
"loss": 2.7483,
"step": 362000
},
{
"epoch": 19.52,
"learning_rate": 2.6876470588235295e-05,
"loss": 2.7536,
"step": 363000
},
{
"epoch": 19.57,
"learning_rate": 2.393529411764706e-05,
"loss": 2.7534,
"step": 364000
},
{
"epoch": 19.62,
"learning_rate": 2.099705882352941e-05,
"loss": 2.7542,
"step": 365000
},
{
"epoch": 19.68,
"learning_rate": 1.8055882352941175e-05,
"loss": 2.7521,
"step": 366000
},
{
"epoch": 19.73,
"learning_rate": 1.5114705882352942e-05,
"loss": 2.7521,
"step": 367000
},
{
"epoch": 19.78,
"learning_rate": 1.2173529411764707e-05,
"loss": 2.7577,
"step": 368000
},
{
"epoch": 19.84,
"learning_rate": 9.235294117647059e-06,
"loss": 2.7544,
"step": 369000
},
{
"epoch": 19.89,
"learning_rate": 6.294117647058824e-06,
"loss": 2.7543,
"step": 370000
},
{
"epoch": 19.95,
"learning_rate": 3.3558823529411767e-06,
"loss": 2.7551,
"step": 371000
},
{
"epoch": 20.0,
"learning_rate": 4.147058823529412e-07,
"loss": 2.7567,
"step": 372000
},
{
"epoch": 20.0,
"eval_accuracy": 0.40923404527178003,
"eval_loss": 3.4371986389160156,
"eval_runtime": 153.6649,
"eval_samples_per_second": 376.931,
"eval_steps_per_second": 5.896,
"step": 372000
},
{
"epoch": 20.0,
"step": 372000,
"total_flos": 1.56735498728448e+18,
"train_loss": 3.0235826258505543,
"train_runtime": 81011.1321,
"train_samples_per_second": 146.942,
"train_steps_per_second": 4.592
}
],
"logging_steps": 1000,
"max_steps": 372000,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 5000,
"total_flos": 1.56735498728448e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}