jflotz's picture
Training in progress, step 500000
78e9441
raw
history blame
226 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 11.152748037116346,
"global_step": 500000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 5.999999999999999e-06,
"loss": 0.9165,
"step": 500
},
{
"epoch": 0.02,
"learning_rate": 1.1999999999999999e-05,
"loss": 0.7051,
"step": 1000
},
{
"epoch": 0.02,
"eval_loss": 0.6818633079528809,
"eval_runtime": 1.9324,
"eval_samples_per_second": 1188.697,
"eval_steps_per_second": 18.63,
"step": 1000
},
{
"epoch": 0.03,
"learning_rate": 1.7999999999999997e-05,
"loss": 0.6818,
"step": 1500
},
{
"epoch": 0.04,
"learning_rate": 2.3999999999999997e-05,
"loss": 0.6808,
"step": 2000
},
{
"epoch": 0.04,
"eval_loss": 0.6788448691368103,
"eval_runtime": 1.762,
"eval_samples_per_second": 1303.62,
"eval_steps_per_second": 20.431,
"step": 2000
},
{
"epoch": 0.06,
"learning_rate": 2.9999999999999997e-05,
"loss": 0.6804,
"step": 2500
},
{
"epoch": 0.07,
"learning_rate": 3.5999999999999994e-05,
"loss": 0.6803,
"step": 3000
},
{
"epoch": 0.07,
"eval_loss": 0.678871750831604,
"eval_runtime": 1.9035,
"eval_samples_per_second": 1206.735,
"eval_steps_per_second": 18.913,
"step": 3000
},
{
"epoch": 0.08,
"learning_rate": 4.2e-05,
"loss": 0.6801,
"step": 3500
},
{
"epoch": 0.09,
"learning_rate": 4.7999999999999994e-05,
"loss": 0.68,
"step": 4000
},
{
"epoch": 0.09,
"eval_loss": 0.6781899333000183,
"eval_runtime": 1.9139,
"eval_samples_per_second": 1200.176,
"eval_steps_per_second": 18.81,
"step": 4000
},
{
"epoch": 0.1,
"learning_rate": 5.399999999999999e-05,
"loss": 0.6798,
"step": 4500
},
{
"epoch": 0.11,
"learning_rate": 5.9999999999999995e-05,
"loss": 0.6796,
"step": 5000
},
{
"epoch": 0.11,
"eval_loss": 0.6784412860870361,
"eval_runtime": 1.8311,
"eval_samples_per_second": 1254.405,
"eval_steps_per_second": 19.66,
"step": 5000
},
{
"epoch": 0.12,
"learning_rate": 6.599999999999999e-05,
"loss": 0.6792,
"step": 5500
},
{
"epoch": 0.13,
"learning_rate": 7.199999999999999e-05,
"loss": 0.6788,
"step": 6000
},
{
"epoch": 0.13,
"eval_loss": 0.67804354429245,
"eval_runtime": 1.8509,
"eval_samples_per_second": 1240.986,
"eval_steps_per_second": 19.449,
"step": 6000
},
{
"epoch": 0.14,
"learning_rate": 7.8e-05,
"loss": 0.6783,
"step": 6500
},
{
"epoch": 0.16,
"learning_rate": 8.4e-05,
"loss": 0.678,
"step": 7000
},
{
"epoch": 0.16,
"eval_loss": 0.6776260733604431,
"eval_runtime": 1.8893,
"eval_samples_per_second": 1215.825,
"eval_steps_per_second": 19.055,
"step": 7000
},
{
"epoch": 0.17,
"learning_rate": 8.999999999999999e-05,
"loss": 0.6778,
"step": 7500
},
{
"epoch": 0.18,
"learning_rate": 9.599999999999999e-05,
"loss": 0.678,
"step": 8000
},
{
"epoch": 0.18,
"eval_loss": 0.6782737374305725,
"eval_runtime": 1.8707,
"eval_samples_per_second": 1227.889,
"eval_steps_per_second": 19.244,
"step": 8000
},
{
"epoch": 0.19,
"learning_rate": 0.000102,
"loss": 0.6783,
"step": 8500
},
{
"epoch": 0.2,
"learning_rate": 0.00010799999999999998,
"loss": 0.6672,
"step": 9000
},
{
"epoch": 0.2,
"eval_loss": 0.6526092290878296,
"eval_runtime": 1.8141,
"eval_samples_per_second": 1266.219,
"eval_steps_per_second": 19.845,
"step": 9000
},
{
"epoch": 0.21,
"learning_rate": 0.00011399999999999999,
"loss": 0.6453,
"step": 9500
},
{
"epoch": 0.22,
"learning_rate": 0.00011999999999999999,
"loss": 0.632,
"step": 10000
},
{
"epoch": 0.22,
"eval_loss": 0.6212196350097656,
"eval_runtime": 1.9366,
"eval_samples_per_second": 1186.109,
"eval_steps_per_second": 18.589,
"step": 10000
},
{
"epoch": 0.23,
"learning_rate": 0.00012599999999999997,
"loss": 0.6235,
"step": 10500
},
{
"epoch": 0.25,
"learning_rate": 0.00013199999999999998,
"loss": 0.6146,
"step": 11000
},
{
"epoch": 0.25,
"eval_loss": 0.6032013893127441,
"eval_runtime": 1.9441,
"eval_samples_per_second": 1181.504,
"eval_steps_per_second": 18.517,
"step": 11000
},
{
"epoch": 0.26,
"learning_rate": 0.000138,
"loss": 0.604,
"step": 11500
},
{
"epoch": 0.27,
"learning_rate": 0.00014399999999999998,
"loss": 0.5936,
"step": 12000
},
{
"epoch": 0.27,
"eval_loss": 0.5801298022270203,
"eval_runtime": 1.9051,
"eval_samples_per_second": 1205.71,
"eval_steps_per_second": 18.897,
"step": 12000
},
{
"epoch": 0.28,
"learning_rate": 0.00015,
"loss": 0.583,
"step": 12500
},
{
"epoch": 0.29,
"learning_rate": 0.000156,
"loss": 0.573,
"step": 13000
},
{
"epoch": 0.29,
"eval_loss": 0.5577847361564636,
"eval_runtime": 1.8487,
"eval_samples_per_second": 1242.481,
"eval_steps_per_second": 19.473,
"step": 13000
},
{
"epoch": 0.3,
"learning_rate": 0.000162,
"loss": 0.5636,
"step": 13500
},
{
"epoch": 0.31,
"learning_rate": 0.000168,
"loss": 0.5544,
"step": 14000
},
{
"epoch": 0.31,
"eval_loss": 0.5371876955032349,
"eval_runtime": 1.956,
"eval_samples_per_second": 1174.317,
"eval_steps_per_second": 18.405,
"step": 14000
},
{
"epoch": 0.32,
"learning_rate": 0.00017399999999999997,
"loss": 0.546,
"step": 14500
},
{
"epoch": 0.33,
"learning_rate": 0.00017999999999999998,
"loss": 0.5382,
"step": 15000
},
{
"epoch": 0.33,
"eval_loss": 0.5219064354896545,
"eval_runtime": 1.9007,
"eval_samples_per_second": 1208.524,
"eval_steps_per_second": 18.941,
"step": 15000
},
{
"epoch": 0.35,
"learning_rate": 0.000186,
"loss": 0.5291,
"step": 15500
},
{
"epoch": 0.36,
"learning_rate": 0.00019199999999999998,
"loss": 0.5214,
"step": 16000
},
{
"epoch": 0.36,
"eval_loss": 0.5052428841590881,
"eval_runtime": 1.9463,
"eval_samples_per_second": 1180.202,
"eval_steps_per_second": 18.497,
"step": 16000
},
{
"epoch": 0.37,
"learning_rate": 0.000198,
"loss": 0.5141,
"step": 16500
},
{
"epoch": 0.38,
"learning_rate": 0.000204,
"loss": 0.5074,
"step": 17000
},
{
"epoch": 0.38,
"eval_loss": 0.4891608655452728,
"eval_runtime": 1.9578,
"eval_samples_per_second": 1173.284,
"eval_steps_per_second": 18.388,
"step": 17000
},
{
"epoch": 0.39,
"learning_rate": 0.00020999999999999998,
"loss": 0.5006,
"step": 17500
},
{
"epoch": 0.4,
"learning_rate": 0.00021599999999999996,
"loss": 0.4943,
"step": 18000
},
{
"epoch": 0.4,
"eval_loss": 0.4745289087295532,
"eval_runtime": 1.9643,
"eval_samples_per_second": 1169.354,
"eval_steps_per_second": 18.327,
"step": 18000
},
{
"epoch": 0.41,
"learning_rate": 0.00022199999999999998,
"loss": 0.4886,
"step": 18500
},
{
"epoch": 0.42,
"learning_rate": 0.00022799999999999999,
"loss": 0.4834,
"step": 19000
},
{
"epoch": 0.42,
"eval_loss": 0.4653691351413727,
"eval_runtime": 1.8946,
"eval_samples_per_second": 1212.382,
"eval_steps_per_second": 19.001,
"step": 19000
},
{
"epoch": 0.43,
"learning_rate": 0.000234,
"loss": 0.4781,
"step": 19500
},
{
"epoch": 0.45,
"learning_rate": 0.00023999999999999998,
"loss": 0.4727,
"step": 20000
},
{
"epoch": 0.45,
"eval_loss": 0.4524119198322296,
"eval_runtime": 1.8731,
"eval_samples_per_second": 1226.293,
"eval_steps_per_second": 19.219,
"step": 20000
},
{
"epoch": 0.46,
"learning_rate": 0.00024599999999999996,
"loss": 0.4662,
"step": 20500
},
{
"epoch": 0.47,
"learning_rate": 0.00025199999999999995,
"loss": 0.4596,
"step": 21000
},
{
"epoch": 0.47,
"eval_loss": 0.4388866424560547,
"eval_runtime": 1.9513,
"eval_samples_per_second": 1177.152,
"eval_steps_per_second": 18.449,
"step": 21000
},
{
"epoch": 0.48,
"learning_rate": 0.000258,
"loss": 0.4536,
"step": 21500
},
{
"epoch": 0.49,
"learning_rate": 0.00026399999999999997,
"loss": 0.448,
"step": 22000
},
{
"epoch": 0.49,
"eval_loss": 0.42909005284309387,
"eval_runtime": 1.8841,
"eval_samples_per_second": 1219.142,
"eval_steps_per_second": 19.107,
"step": 22000
},
{
"epoch": 0.5,
"learning_rate": 0.00027,
"loss": 0.4428,
"step": 22500
},
{
"epoch": 0.51,
"learning_rate": 0.000276,
"loss": 0.4378,
"step": 23000
},
{
"epoch": 0.51,
"eval_loss": 0.41723570227622986,
"eval_runtime": 1.8673,
"eval_samples_per_second": 1230.107,
"eval_steps_per_second": 19.279,
"step": 23000
},
{
"epoch": 0.52,
"learning_rate": 0.00028199999999999997,
"loss": 0.4336,
"step": 23500
},
{
"epoch": 0.54,
"learning_rate": 0.00028799999999999995,
"loss": 0.4294,
"step": 24000
},
{
"epoch": 0.54,
"eval_loss": 0.4092707931995392,
"eval_runtime": 1.8847,
"eval_samples_per_second": 1218.745,
"eval_steps_per_second": 19.101,
"step": 24000
},
{
"epoch": 0.55,
"learning_rate": 0.000294,
"loss": 0.4253,
"step": 24500
},
{
"epoch": 0.56,
"learning_rate": 0.0003,
"loss": 0.4218,
"step": 25000
},
{
"epoch": 0.56,
"eval_loss": 0.40236151218414307,
"eval_runtime": 1.9002,
"eval_samples_per_second": 1208.849,
"eval_steps_per_second": 18.946,
"step": 25000
},
{
"epoch": 0.57,
"learning_rate": 0.00029999920715161553,
"loss": 0.4181,
"step": 25500
},
{
"epoch": 0.58,
"learning_rate": 0.0002999968286151326,
"loss": 0.4146,
"step": 26000
},
{
"epoch": 0.58,
"eval_loss": 0.3960426449775696,
"eval_runtime": 1.8399,
"eval_samples_per_second": 1248.469,
"eval_steps_per_second": 19.567,
"step": 26000
},
{
"epoch": 0.59,
"learning_rate": 0.0002999928644165624,
"loss": 0.4113,
"step": 26500
},
{
"epoch": 0.6,
"learning_rate": 0.0002999873145992569,
"loss": 0.408,
"step": 27000
},
{
"epoch": 0.6,
"eval_loss": 0.38567298650741577,
"eval_runtime": 1.8013,
"eval_samples_per_second": 1275.179,
"eval_steps_per_second": 19.985,
"step": 27000
},
{
"epoch": 0.61,
"learning_rate": 0.000299980179223908,
"loss": 0.4048,
"step": 27500
},
{
"epoch": 0.62,
"learning_rate": 0.0002999714583685469,
"loss": 0.4018,
"step": 28000
},
{
"epoch": 0.62,
"eval_loss": 0.3801218569278717,
"eval_runtime": 1.9159,
"eval_samples_per_second": 1198.935,
"eval_steps_per_second": 18.79,
"step": 28000
},
{
"epoch": 0.64,
"learning_rate": 0.00029996115212854366,
"loss": 0.3991,
"step": 28500
},
{
"epoch": 0.65,
"learning_rate": 0.00029994926061660554,
"loss": 0.3963,
"step": 29000
},
{
"epoch": 0.65,
"eval_loss": 0.37746793031692505,
"eval_runtime": 1.9561,
"eval_samples_per_second": 1174.286,
"eval_steps_per_second": 18.404,
"step": 29000
},
{
"epoch": 0.66,
"learning_rate": 0.0002999357839627762,
"loss": 0.394,
"step": 29500
},
{
"epoch": 0.67,
"learning_rate": 0.00029992072231443425,
"loss": 0.3917,
"step": 30000
},
{
"epoch": 0.67,
"eval_loss": 0.3722357749938965,
"eval_runtime": 1.841,
"eval_samples_per_second": 1247.702,
"eval_steps_per_second": 19.555,
"step": 30000
},
{
"epoch": 0.68,
"learning_rate": 0.0002999040758362914,
"loss": 0.3894,
"step": 30500
},
{
"epoch": 0.69,
"learning_rate": 0.00029988584471039094,
"loss": 0.3869,
"step": 31000
},
{
"epoch": 0.69,
"eval_loss": 0.3689954876899719,
"eval_runtime": 1.9129,
"eval_samples_per_second": 1200.807,
"eval_steps_per_second": 18.82,
"step": 31000
},
{
"epoch": 0.7,
"learning_rate": 0.0002998660291361054,
"loss": 0.3851,
"step": 31500
},
{
"epoch": 0.71,
"learning_rate": 0.0002998446293301349,
"loss": 0.3831,
"step": 32000
},
{
"epoch": 0.71,
"eval_loss": 0.3645932674407959,
"eval_runtime": 1.8679,
"eval_samples_per_second": 1229.739,
"eval_steps_per_second": 19.273,
"step": 32000
},
{
"epoch": 0.72,
"learning_rate": 0.0002998216455265042,
"loss": 0.3813,
"step": 32500
},
{
"epoch": 0.74,
"learning_rate": 0.00029979707797656046,
"loss": 0.3794,
"step": 33000
},
{
"epoch": 0.74,
"eval_loss": 0.35992059111595154,
"eval_runtime": 1.8813,
"eval_samples_per_second": 1220.942,
"eval_steps_per_second": 19.135,
"step": 33000
},
{
"epoch": 0.75,
"learning_rate": 0.00029977092694897053,
"loss": 0.3773,
"step": 33500
},
{
"epoch": 0.76,
"learning_rate": 0.0002997431927297178,
"loss": 0.3756,
"step": 34000
},
{
"epoch": 0.76,
"eval_loss": 0.3545241057872772,
"eval_runtime": 1.8945,
"eval_samples_per_second": 1212.449,
"eval_steps_per_second": 19.002,
"step": 34000
},
{
"epoch": 0.77,
"learning_rate": 0.00029971387562209936,
"loss": 0.3739,
"step": 34500
},
{
"epoch": 0.78,
"learning_rate": 0.00029968297594672226,
"loss": 0.3723,
"step": 35000
},
{
"epoch": 0.78,
"eval_loss": 0.3539046347141266,
"eval_runtime": 1.8855,
"eval_samples_per_second": 1218.214,
"eval_steps_per_second": 19.093,
"step": 35000
},
{
"epoch": 0.79,
"learning_rate": 0.0002996504940415005,
"loss": 0.3712,
"step": 35500
},
{
"epoch": 0.8,
"learning_rate": 0.00029961643026165096,
"loss": 0.3693,
"step": 36000
},
{
"epoch": 0.8,
"eval_loss": 0.352187842130661,
"eval_runtime": 1.9629,
"eval_samples_per_second": 1170.187,
"eval_steps_per_second": 18.34,
"step": 36000
},
{
"epoch": 0.81,
"learning_rate": 0.00029958078497968973,
"loss": 0.3681,
"step": 36500
},
{
"epoch": 0.83,
"learning_rate": 0.0002995435585854278,
"loss": 0.3668,
"step": 37000
},
{
"epoch": 0.83,
"eval_loss": 0.3482549488544464,
"eval_runtime": 1.9671,
"eval_samples_per_second": 1167.724,
"eval_steps_per_second": 18.301,
"step": 37000
},
{
"epoch": 0.84,
"learning_rate": 0.0002995047514859671,
"loss": 0.365,
"step": 37500
},
{
"epoch": 0.85,
"learning_rate": 0.0002994643641056959,
"loss": 0.3635,
"step": 38000
},
{
"epoch": 0.85,
"eval_loss": 0.3442147970199585,
"eval_runtime": 1.9101,
"eval_samples_per_second": 1202.583,
"eval_steps_per_second": 18.848,
"step": 38000
},
{
"epoch": 0.86,
"learning_rate": 0.000299422396886284,
"loss": 0.3627,
"step": 38500
},
{
"epoch": 0.87,
"learning_rate": 0.0002993788502866783,
"loss": 0.3613,
"step": 39000
},
{
"epoch": 0.87,
"eval_loss": 0.3413570523262024,
"eval_runtime": 1.8505,
"eval_samples_per_second": 1241.299,
"eval_steps_per_second": 19.454,
"step": 39000
},
{
"epoch": 0.88,
"learning_rate": 0.00029933372478309746,
"loss": 0.3602,
"step": 39500
},
{
"epoch": 0.89,
"learning_rate": 0.00029928702086902664,
"loss": 0.3589,
"step": 40000
},
{
"epoch": 0.89,
"eval_loss": 0.3388674557209015,
"eval_runtime": 1.8714,
"eval_samples_per_second": 1227.432,
"eval_steps_per_second": 19.237,
"step": 40000
},
{
"epoch": 0.9,
"learning_rate": 0.00029923873905521244,
"loss": 0.3579,
"step": 40500
},
{
"epoch": 0.91,
"learning_rate": 0.000299188879869657,
"loss": 0.3568,
"step": 41000
},
{
"epoch": 0.91,
"eval_loss": 0.33952972292900085,
"eval_runtime": 2.0164,
"eval_samples_per_second": 1139.183,
"eval_steps_per_second": 17.854,
"step": 41000
},
{
"epoch": 0.93,
"learning_rate": 0.00029913744385761244,
"loss": 0.3554,
"step": 41500
},
{
"epoch": 0.94,
"learning_rate": 0.00029908443158157465,
"loss": 0.3549,
"step": 42000
},
{
"epoch": 0.94,
"eval_loss": 0.334180623292923,
"eval_runtime": 1.9257,
"eval_samples_per_second": 1192.785,
"eval_steps_per_second": 18.694,
"step": 42000
},
{
"epoch": 0.95,
"learning_rate": 0.0002990298436212775,
"loss": 0.3538,
"step": 42500
},
{
"epoch": 0.96,
"learning_rate": 0.0002989736805736861,
"loss": 0.3524,
"step": 43000
},
{
"epoch": 0.96,
"eval_loss": 0.3344336450099945,
"eval_runtime": 1.9242,
"eval_samples_per_second": 1193.754,
"eval_steps_per_second": 18.709,
"step": 43000
},
{
"epoch": 0.97,
"learning_rate": 0.00029891594305299065,
"loss": 0.3519,
"step": 43500
},
{
"epoch": 0.98,
"learning_rate": 0.00029885663169059926,
"loss": 0.3504,
"step": 44000
},
{
"epoch": 0.98,
"eval_loss": 0.3332378566265106,
"eval_runtime": 1.9034,
"eval_samples_per_second": 1206.779,
"eval_steps_per_second": 18.913,
"step": 44000
},
{
"epoch": 0.99,
"learning_rate": 0.0002987957471351316,
"loss": 0.3499,
"step": 44500
},
{
"epoch": 1.0,
"learning_rate": 0.00029873329005241137,
"loss": 0.349,
"step": 45000
},
{
"epoch": 1.0,
"eval_loss": 0.3305933475494385,
"eval_runtime": 2.0165,
"eval_samples_per_second": 1139.102,
"eval_steps_per_second": 17.853,
"step": 45000
},
{
"epoch": 1.01,
"learning_rate": 0.00029866926112545925,
"loss": 0.3482,
"step": 45500
},
{
"epoch": 1.03,
"learning_rate": 0.00029860366105448534,
"loss": 0.3473,
"step": 46000
},
{
"epoch": 1.03,
"eval_loss": 0.32961174845695496,
"eval_runtime": 2.0634,
"eval_samples_per_second": 1113.205,
"eval_steps_per_second": 17.447,
"step": 46000
},
{
"epoch": 1.04,
"learning_rate": 0.00029853649055688143,
"loss": 0.3464,
"step": 46500
},
{
"epoch": 1.05,
"learning_rate": 0.00029846775036721337,
"loss": 0.3459,
"step": 47000
},
{
"epoch": 1.05,
"eval_loss": 0.32834696769714355,
"eval_runtime": 1.9893,
"eval_samples_per_second": 1154.665,
"eval_steps_per_second": 18.097,
"step": 47000
},
{
"epoch": 1.06,
"learning_rate": 0.0002983974412372129,
"loss": 0.3446,
"step": 47500
},
{
"epoch": 1.07,
"learning_rate": 0.00029832556393576934,
"loss": 0.3441,
"step": 48000
},
{
"epoch": 1.07,
"eval_loss": 0.3241877555847168,
"eval_runtime": 2.051,
"eval_samples_per_second": 1119.916,
"eval_steps_per_second": 17.552,
"step": 48000
},
{
"epoch": 1.08,
"learning_rate": 0.0002982521192489214,
"loss": 0.3434,
"step": 48500
},
{
"epoch": 1.09,
"learning_rate": 0.0002981771079798483,
"loss": 0.3427,
"step": 49000
},
{
"epoch": 1.09,
"eval_loss": 0.3232521414756775,
"eval_runtime": 1.9619,
"eval_samples_per_second": 1170.819,
"eval_steps_per_second": 18.35,
"step": 49000
},
{
"epoch": 1.1,
"learning_rate": 0.00029810053094886136,
"loss": 0.3417,
"step": 49500
},
{
"epoch": 1.12,
"learning_rate": 0.00029802238899339473,
"loss": 0.3411,
"step": 50000
},
{
"epoch": 1.12,
"eval_loss": 0.322670042514801,
"eval_runtime": 2.0139,
"eval_samples_per_second": 1140.566,
"eval_steps_per_second": 17.876,
"step": 50000
},
{
"epoch": 1.13,
"learning_rate": 0.0002979426829679962,
"loss": 0.3406,
"step": 50500
},
{
"epoch": 1.14,
"learning_rate": 0.0002978614137443183,
"loss": 0.3398,
"step": 51000
},
{
"epoch": 1.14,
"eval_loss": 0.32256901264190674,
"eval_runtime": 1.9699,
"eval_samples_per_second": 1166.043,
"eval_steps_per_second": 18.275,
"step": 51000
},
{
"epoch": 1.15,
"learning_rate": 0.000297778582211108,
"loss": 0.3391,
"step": 51500
},
{
"epoch": 1.16,
"learning_rate": 0.00029769418927419786,
"loss": 0.3385,
"step": 52000
},
{
"epoch": 1.16,
"eval_loss": 0.3191610276699066,
"eval_runtime": 2.0164,
"eval_samples_per_second": 1139.174,
"eval_steps_per_second": 17.854,
"step": 52000
},
{
"epoch": 1.17,
"learning_rate": 0.0002976082358564954,
"loss": 0.3381,
"step": 52500
},
{
"epoch": 1.18,
"learning_rate": 0.00029752072289797353,
"loss": 0.3372,
"step": 53000
},
{
"epoch": 1.18,
"eval_loss": 0.31954050064086914,
"eval_runtime": 1.913,
"eval_samples_per_second": 1200.727,
"eval_steps_per_second": 18.819,
"step": 53000
},
{
"epoch": 1.19,
"learning_rate": 0.00029743165135565986,
"loss": 0.3368,
"step": 53500
},
{
"epoch": 1.2,
"learning_rate": 0.00029734102220362654,
"loss": 0.3359,
"step": 54000
},
{
"epoch": 1.2,
"eval_loss": 0.31627157330513,
"eval_runtime": 2.0014,
"eval_samples_per_second": 1147.684,
"eval_steps_per_second": 17.987,
"step": 54000
},
{
"epoch": 1.22,
"learning_rate": 0.00029724883643297937,
"loss": 0.3356,
"step": 54500
},
{
"epoch": 1.23,
"learning_rate": 0.0002971550950518473,
"loss": 0.3348,
"step": 55000
},
{
"epoch": 1.23,
"eval_loss": 0.3175150454044342,
"eval_runtime": 2.0666,
"eval_samples_per_second": 1111.49,
"eval_steps_per_second": 17.42,
"step": 55000
},
{
"epoch": 1.24,
"learning_rate": 0.000297059799085371,
"loss": 0.3343,
"step": 55500
},
{
"epoch": 1.25,
"learning_rate": 0.00029696294957569196,
"loss": 0.3338,
"step": 56000
},
{
"epoch": 1.25,
"eval_loss": 0.3165690302848816,
"eval_runtime": 2.0136,
"eval_samples_per_second": 1140.732,
"eval_steps_per_second": 17.878,
"step": 56000
},
{
"epoch": 1.26,
"learning_rate": 0.00029686454758194076,
"loss": 0.3331,
"step": 56500
},
{
"epoch": 1.27,
"learning_rate": 0.00029676459418022594,
"loss": 0.3327,
"step": 57000
},
{
"epoch": 1.27,
"eval_loss": 0.316009521484375,
"eval_runtime": 1.9665,
"eval_samples_per_second": 1168.059,
"eval_steps_per_second": 18.307,
"step": 57000
},
{
"epoch": 1.28,
"learning_rate": 0.0002966630904636219,
"loss": 0.3321,
"step": 57500
},
{
"epoch": 1.29,
"learning_rate": 0.0002965600375421569,
"loss": 0.3315,
"step": 58000
},
{
"epoch": 1.29,
"eval_loss": 0.31569892168045044,
"eval_runtime": 2.026,
"eval_samples_per_second": 1133.75,
"eval_steps_per_second": 17.769,
"step": 58000
},
{
"epoch": 1.3,
"learning_rate": 0.0002964554365428013,
"loss": 0.3313,
"step": 58500
},
{
"epoch": 1.32,
"learning_rate": 0.00029634928860945486,
"loss": 0.3301,
"step": 59000
},
{
"epoch": 1.32,
"eval_loss": 0.310556024312973,
"eval_runtime": 2.0156,
"eval_samples_per_second": 1139.586,
"eval_steps_per_second": 17.86,
"step": 59000
},
{
"epoch": 1.33,
"learning_rate": 0.0002962415949029343,
"loss": 0.33,
"step": 59500
},
{
"epoch": 1.34,
"learning_rate": 0.00029613235660096084,
"loss": 0.3294,
"step": 60000
},
{
"epoch": 1.34,
"eval_loss": 0.3107939064502716,
"eval_runtime": 1.9893,
"eval_samples_per_second": 1154.672,
"eval_steps_per_second": 18.097,
"step": 60000
},
{
"epoch": 1.35,
"learning_rate": 0.00029602157489814693,
"loss": 0.3291,
"step": 60500
},
{
"epoch": 1.36,
"learning_rate": 0.00029590925100598365,
"loss": 0.3285,
"step": 61000
},
{
"epoch": 1.36,
"eval_loss": 0.31057360768318176,
"eval_runtime": 2.0555,
"eval_samples_per_second": 1117.488,
"eval_steps_per_second": 17.514,
"step": 61000
},
{
"epoch": 1.37,
"learning_rate": 0.000295795386152827,
"loss": 0.3284,
"step": 61500
},
{
"epoch": 1.38,
"learning_rate": 0.0002956799815838848,
"loss": 0.3285,
"step": 62000
},
{
"epoch": 1.38,
"eval_loss": 0.3091903328895569,
"eval_runtime": 2.0051,
"eval_samples_per_second": 1145.598,
"eval_steps_per_second": 17.955,
"step": 62000
},
{
"epoch": 1.39,
"learning_rate": 0.0002955630385612029,
"loss": 0.3268,
"step": 62500
},
{
"epoch": 1.41,
"learning_rate": 0.0002954445583636515,
"loss": 0.3266,
"step": 63000
},
{
"epoch": 1.41,
"eval_loss": 0.309052050113678,
"eval_runtime": 1.9663,
"eval_samples_per_second": 1168.204,
"eval_steps_per_second": 18.309,
"step": 63000
},
{
"epoch": 1.42,
"learning_rate": 0.00029532454228691103,
"loss": 0.3263,
"step": 63500
},
{
"epoch": 1.43,
"learning_rate": 0.0002952029916434581,
"loss": 0.3259,
"step": 64000
},
{
"epoch": 1.43,
"eval_loss": 0.3066043555736542,
"eval_runtime": 2.0287,
"eval_samples_per_second": 1132.238,
"eval_steps_per_second": 17.745,
"step": 64000
},
{
"epoch": 1.44,
"learning_rate": 0.00029507990776255107,
"loss": 0.3257,
"step": 64500
},
{
"epoch": 1.45,
"learning_rate": 0.00029495529199021555,
"loss": 0.3251,
"step": 65000
},
{
"epoch": 1.45,
"eval_loss": 0.3062935769557953,
"eval_runtime": 2.0366,
"eval_samples_per_second": 1127.874,
"eval_steps_per_second": 17.677,
"step": 65000
},
{
"epoch": 1.46,
"learning_rate": 0.0002948291456892296,
"loss": 0.325,
"step": 65500
},
{
"epoch": 1.47,
"learning_rate": 0.00029470147023910907,
"loss": 0.3245,
"step": 66000
},
{
"epoch": 1.47,
"eval_loss": 0.3044787049293518,
"eval_runtime": 1.9965,
"eval_samples_per_second": 1150.53,
"eval_steps_per_second": 18.032,
"step": 66000
},
{
"epoch": 1.48,
"learning_rate": 0.0002945722670360921,
"loss": 0.3261,
"step": 66500
},
{
"epoch": 1.49,
"learning_rate": 0.0002944415374931243,
"loss": 0.3233,
"step": 67000
},
{
"epoch": 1.49,
"eval_loss": 0.3051415681838989,
"eval_runtime": 2.0098,
"eval_samples_per_second": 1142.892,
"eval_steps_per_second": 17.912,
"step": 67000
},
{
"epoch": 1.51,
"learning_rate": 0.00029430928303984295,
"loss": 0.3231,
"step": 67500
},
{
"epoch": 1.52,
"learning_rate": 0.0002941755051225616,
"loss": 0.3227,
"step": 68000
},
{
"epoch": 1.52,
"eval_loss": 0.303621768951416,
"eval_runtime": 2.0068,
"eval_samples_per_second": 1144.62,
"eval_steps_per_second": 17.939,
"step": 68000
},
{
"epoch": 1.53,
"learning_rate": 0.00029404020520425417,
"loss": 0.3226,
"step": 68500
},
{
"epoch": 1.54,
"learning_rate": 0.0002939033847645388,
"loss": 0.3221,
"step": 69000
},
{
"epoch": 1.54,
"eval_loss": 0.30241796374320984,
"eval_runtime": 1.989,
"eval_samples_per_second": 1154.871,
"eval_steps_per_second": 18.1,
"step": 69000
},
{
"epoch": 1.55,
"learning_rate": 0.00029376504529966195,
"loss": 0.3216,
"step": 69500
},
{
"epoch": 1.56,
"learning_rate": 0.00029362518832248184,
"loss": 0.3216,
"step": 70000
},
{
"epoch": 1.56,
"eval_loss": 0.3033766746520996,
"eval_runtime": 1.9913,
"eval_samples_per_second": 1153.526,
"eval_steps_per_second": 18.079,
"step": 70000
},
{
"epoch": 1.57,
"learning_rate": 0.0002934838153624519,
"loss": 0.3212,
"step": 70500
},
{
"epoch": 1.58,
"learning_rate": 0.00029334092796560427,
"loss": 0.3206,
"step": 71000
},
{
"epoch": 1.58,
"eval_loss": 0.3010749816894531,
"eval_runtime": 1.9979,
"eval_samples_per_second": 1149.696,
"eval_steps_per_second": 18.019,
"step": 71000
},
{
"epoch": 1.59,
"learning_rate": 0.0002931965276945326,
"loss": 0.3203,
"step": 71500
},
{
"epoch": 1.61,
"learning_rate": 0.0002930506161283751,
"loss": 0.3202,
"step": 72000
},
{
"epoch": 1.61,
"eval_loss": 0.30220872163772583,
"eval_runtime": 2.0424,
"eval_samples_per_second": 1124.632,
"eval_steps_per_second": 17.626,
"step": 72000
},
{
"epoch": 1.62,
"learning_rate": 0.00029290319486279724,
"loss": 0.3195,
"step": 72500
},
{
"epoch": 1.63,
"learning_rate": 0.0002927542655099744,
"loss": 0.3194,
"step": 73000
},
{
"epoch": 1.63,
"eval_loss": 0.30245041847229004,
"eval_runtime": 2.0617,
"eval_samples_per_second": 1114.154,
"eval_steps_per_second": 17.462,
"step": 73000
},
{
"epoch": 1.64,
"learning_rate": 0.00029260382969857417,
"loss": 0.319,
"step": 73500
},
{
"epoch": 1.65,
"learning_rate": 0.00029245188907373845,
"loss": 0.3188,
"step": 74000
},
{
"epoch": 1.65,
"eval_loss": 0.2999315857887268,
"eval_runtime": 2.0046,
"eval_samples_per_second": 1145.863,
"eval_steps_per_second": 17.959,
"step": 74000
},
{
"epoch": 1.66,
"learning_rate": 0.0002922984452970655,
"loss": 0.3185,
"step": 74500
},
{
"epoch": 1.67,
"learning_rate": 0.000292143500046592,
"loss": 0.3184,
"step": 75000
},
{
"epoch": 1.67,
"eval_loss": 0.2986513078212738,
"eval_runtime": 1.898,
"eval_samples_per_second": 1210.225,
"eval_steps_per_second": 18.967,
"step": 75000
},
{
"epoch": 1.68,
"learning_rate": 0.0002919870550167743,
"loss": 0.3178,
"step": 75500
},
{
"epoch": 1.7,
"learning_rate": 0.0002918291119184702,
"loss": 0.3184,
"step": 76000
},
{
"epoch": 1.7,
"eval_loss": 0.30084192752838135,
"eval_runtime": 2.0127,
"eval_samples_per_second": 1141.227,
"eval_steps_per_second": 17.886,
"step": 76000
},
{
"epoch": 1.71,
"learning_rate": 0.0002916696724789201,
"loss": 0.3171,
"step": 76500
},
{
"epoch": 1.72,
"learning_rate": 0.00029150873844172823,
"loss": 0.3168,
"step": 77000
},
{
"epoch": 1.72,
"eval_loss": 0.2981663942337036,
"eval_runtime": 2.1151,
"eval_samples_per_second": 1085.979,
"eval_steps_per_second": 17.02,
"step": 77000
},
{
"epoch": 1.73,
"learning_rate": 0.00029134631156684334,
"loss": 0.3167,
"step": 77500
},
{
"epoch": 1.74,
"learning_rate": 0.0002911823936305398,
"loss": 0.3162,
"step": 78000
},
{
"epoch": 1.74,
"eval_loss": 0.2964055836200714,
"eval_runtime": 2.046,
"eval_samples_per_second": 1122.697,
"eval_steps_per_second": 17.596,
"step": 78000
},
{
"epoch": 1.75,
"learning_rate": 0.0002910169864253979,
"loss": 0.3162,
"step": 78500
},
{
"epoch": 1.76,
"learning_rate": 0.0002908500917602842,
"loss": 0.3157,
"step": 79000
},
{
"epoch": 1.76,
"eval_loss": 0.2958185076713562,
"eval_runtime": 1.9809,
"eval_samples_per_second": 1159.565,
"eval_steps_per_second": 18.173,
"step": 79000
},
{
"epoch": 1.77,
"learning_rate": 0.00029068171146033226,
"loss": 0.3154,
"step": 79500
},
{
"epoch": 1.78,
"learning_rate": 0.0002905118473669218,
"loss": 0.3149,
"step": 80000
},
{
"epoch": 1.78,
"eval_loss": 0.2947300374507904,
"eval_runtime": 1.9379,
"eval_samples_per_second": 1185.314,
"eval_steps_per_second": 18.577,
"step": 80000
},
{
"epoch": 1.8,
"learning_rate": 0.00029034050133765947,
"loss": 0.3149,
"step": 80500
},
{
"epoch": 1.81,
"learning_rate": 0.00029016767524635804,
"loss": 0.3145,
"step": 81000
},
{
"epoch": 1.81,
"eval_loss": 0.2963857054710388,
"eval_runtime": 1.9198,
"eval_samples_per_second": 1196.488,
"eval_steps_per_second": 18.752,
"step": 81000
},
{
"epoch": 1.82,
"learning_rate": 0.00028999337098301585,
"loss": 0.3144,
"step": 81500
},
{
"epoch": 1.83,
"learning_rate": 0.0002898175904537964,
"loss": 0.314,
"step": 82000
},
{
"epoch": 1.83,
"eval_loss": 0.295415997505188,
"eval_runtime": 1.9531,
"eval_samples_per_second": 1176.079,
"eval_steps_per_second": 18.432,
"step": 82000
},
{
"epoch": 1.84,
"learning_rate": 0.0002896403355810075,
"loss": 0.3135,
"step": 82500
},
{
"epoch": 1.85,
"learning_rate": 0.00028946160830307997,
"loss": 0.3135,
"step": 83000
},
{
"epoch": 1.85,
"eval_loss": 0.2971150577068329,
"eval_runtime": 2.0357,
"eval_samples_per_second": 1128.362,
"eval_steps_per_second": 17.684,
"step": 83000
},
{
"epoch": 1.86,
"learning_rate": 0.00028928141057454665,
"loss": 0.3136,
"step": 83500
},
{
"epoch": 1.87,
"learning_rate": 0.0002890997443660211,
"loss": 0.3127,
"step": 84000
},
{
"epoch": 1.87,
"eval_loss": 0.2954028248786926,
"eval_runtime": 2.0203,
"eval_samples_per_second": 1136.969,
"eval_steps_per_second": 17.819,
"step": 84000
},
{
"epoch": 1.88,
"learning_rate": 0.00028891661166417586,
"loss": 0.3126,
"step": 84500
},
{
"epoch": 1.9,
"learning_rate": 0.00028873201447172074,
"loss": 0.3125,
"step": 85000
},
{
"epoch": 1.9,
"eval_loss": 0.2963060438632965,
"eval_runtime": 1.9991,
"eval_samples_per_second": 1149.017,
"eval_steps_per_second": 18.008,
"step": 85000
},
{
"epoch": 1.91,
"learning_rate": 0.0002885459548073812,
"loss": 0.3124,
"step": 85500
},
{
"epoch": 1.92,
"learning_rate": 0.0002883584347058758,
"loss": 0.3122,
"step": 86000
},
{
"epoch": 1.92,
"eval_loss": 0.2932322025299072,
"eval_runtime": 1.9426,
"eval_samples_per_second": 1182.434,
"eval_steps_per_second": 18.532,
"step": 86000
},
{
"epoch": 1.93,
"learning_rate": 0.00028816945621789437,
"loss": 0.3121,
"step": 86500
},
{
"epoch": 1.94,
"learning_rate": 0.0002879790214100753,
"loss": 0.3119,
"step": 87000
},
{
"epoch": 1.94,
"eval_loss": 0.29384109377861023,
"eval_runtime": 1.9978,
"eval_samples_per_second": 1149.741,
"eval_steps_per_second": 18.019,
"step": 87000
},
{
"epoch": 1.95,
"learning_rate": 0.0002877871323649833,
"loss": 0.3115,
"step": 87500
},
{
"epoch": 1.96,
"learning_rate": 0.0002875937911810861,
"loss": 0.3111,
"step": 88000
},
{
"epoch": 1.96,
"eval_loss": 0.29092082381248474,
"eval_runtime": 1.9165,
"eval_samples_per_second": 1198.512,
"eval_steps_per_second": 18.784,
"step": 88000
},
{
"epoch": 1.97,
"learning_rate": 0.000287398999972732,
"loss": 0.3107,
"step": 88500
},
{
"epoch": 1.99,
"learning_rate": 0.00028720276087012636,
"loss": 0.3107,
"step": 89000
},
{
"epoch": 1.99,
"eval_loss": 0.2925170063972473,
"eval_runtime": 1.9863,
"eval_samples_per_second": 1156.436,
"eval_steps_per_second": 18.124,
"step": 89000
},
{
"epoch": 2.0,
"learning_rate": 0.0002870050760193086,
"loss": 0.3103,
"step": 89500
},
{
"epoch": 2.01,
"learning_rate": 0.00028680594758212854,
"loss": 0.3105,
"step": 90000
},
{
"epoch": 2.01,
"eval_loss": 0.2921919524669647,
"eval_runtime": 2.0098,
"eval_samples_per_second": 1142.913,
"eval_steps_per_second": 17.912,
"step": 90000
},
{
"epoch": 2.02,
"learning_rate": 0.00028660537773622294,
"loss": 0.3098,
"step": 90500
},
{
"epoch": 2.03,
"learning_rate": 0.00028640336867499143,
"loss": 0.3098,
"step": 91000
},
{
"epoch": 2.03,
"eval_loss": 0.2905424237251282,
"eval_runtime": 1.9993,
"eval_samples_per_second": 1148.881,
"eval_steps_per_second": 18.006,
"step": 91000
},
{
"epoch": 2.04,
"learning_rate": 0.0002861999226075728,
"loss": 0.3095,
"step": 91500
},
{
"epoch": 2.05,
"learning_rate": 0.0002859950417588206,
"loss": 0.309,
"step": 92000
},
{
"epoch": 2.05,
"eval_loss": 0.2886371612548828,
"eval_runtime": 1.9517,
"eval_samples_per_second": 1176.934,
"eval_steps_per_second": 18.446,
"step": 92000
},
{
"epoch": 2.06,
"learning_rate": 0.00028578872836927904,
"loss": 0.4773,
"step": 92500
},
{
"epoch": 2.07,
"learning_rate": 0.0002855809846951582,
"loss": 0.6769,
"step": 93000
},
{
"epoch": 2.07,
"eval_loss": 0.6724876165390015,
"eval_runtime": 1.9377,
"eval_samples_per_second": 1185.409,
"eval_steps_per_second": 18.578,
"step": 93000
},
{
"epoch": 2.09,
"learning_rate": 0.00028537181300830963,
"loss": 0.4512,
"step": 93500
},
{
"epoch": 2.1,
"learning_rate": 0.0002851612155962014,
"loss": 0.3124,
"step": 94000
},
{
"epoch": 2.1,
"eval_loss": 0.29317304491996765,
"eval_runtime": 2.3779,
"eval_samples_per_second": 965.986,
"eval_steps_per_second": 15.14,
"step": 94000
},
{
"epoch": 2.11,
"learning_rate": 0.0002849491947618932,
"loss": 0.3108,
"step": 94500
},
{
"epoch": 2.12,
"learning_rate": 0.0002847357528240107,
"loss": 0.3099,
"step": 95000
},
{
"epoch": 2.12,
"eval_loss": 0.2914526164531708,
"eval_runtime": 2.0271,
"eval_samples_per_second": 1133.13,
"eval_steps_per_second": 17.759,
"step": 95000
},
{
"epoch": 2.13,
"learning_rate": 0.0002845208921167208,
"loss": 0.3093,
"step": 95500
},
{
"epoch": 2.14,
"learning_rate": 0.00028430461498970584,
"loss": 0.3088,
"step": 96000
},
{
"epoch": 2.14,
"eval_loss": 0.2922168970108032,
"eval_runtime": 2.0567,
"eval_samples_per_second": 1116.857,
"eval_steps_per_second": 17.504,
"step": 96000
},
{
"epoch": 2.15,
"learning_rate": 0.00028408692380813775,
"loss": 0.3082,
"step": 96500
},
{
"epoch": 2.16,
"learning_rate": 0.00028386782095265247,
"loss": 0.3081,
"step": 97000
},
{
"epoch": 2.16,
"eval_loss": 0.2883111238479614,
"eval_runtime": 2.0553,
"eval_samples_per_second": 1117.577,
"eval_steps_per_second": 17.515,
"step": 97000
},
{
"epoch": 2.17,
"learning_rate": 0.0002836473088193237,
"loss": 0.308,
"step": 97500
},
{
"epoch": 2.19,
"learning_rate": 0.00028342538981963677,
"loss": 0.3074,
"step": 98000
},
{
"epoch": 2.19,
"eval_loss": 0.28875553607940674,
"eval_runtime": 2.0294,
"eval_samples_per_second": 1131.888,
"eval_steps_per_second": 17.74,
"step": 98000
},
{
"epoch": 2.2,
"learning_rate": 0.0002832020663804624,
"loss": 0.3074,
"step": 98500
},
{
"epoch": 2.21,
"learning_rate": 0.00028297734094402986,
"loss": 0.3067,
"step": 99000
},
{
"epoch": 2.21,
"eval_loss": 0.2898538112640381,
"eval_runtime": 1.9557,
"eval_samples_per_second": 1174.502,
"eval_steps_per_second": 18.408,
"step": 99000
},
{
"epoch": 2.22,
"learning_rate": 0.0002827512159679005,
"loss": 0.3072,
"step": 99500
},
{
"epoch": 2.23,
"learning_rate": 0.00028252369392494086,
"loss": 0.3069,
"step": 100000
},
{
"epoch": 2.23,
"eval_loss": 0.2872167229652405,
"eval_runtime": 2.0341,
"eval_samples_per_second": 1129.255,
"eval_steps_per_second": 17.698,
"step": 100000
},
{
"epoch": 2.24,
"learning_rate": 0.0002822947773032956,
"loss": 0.3066,
"step": 100500
},
{
"epoch": 2.25,
"learning_rate": 0.0002820644686063602,
"loss": 0.3061,
"step": 101000
},
{
"epoch": 2.25,
"eval_loss": 0.28897979855537415,
"eval_runtime": 1.9983,
"eval_samples_per_second": 1149.501,
"eval_steps_per_second": 18.016,
"step": 101000
},
{
"epoch": 2.26,
"learning_rate": 0.00028183277035275363,
"loss": 0.3058,
"step": 101500
},
{
"epoch": 2.28,
"learning_rate": 0.0002815996850762909,
"loss": 0.3059,
"step": 102000
},
{
"epoch": 2.28,
"eval_loss": 0.2893044054508209,
"eval_runtime": 2.0103,
"eval_samples_per_second": 1142.629,
"eval_steps_per_second": 17.908,
"step": 102000
},
{
"epoch": 2.29,
"learning_rate": 0.00028136521532595515,
"loss": 0.3059,
"step": 102500
},
{
"epoch": 2.3,
"learning_rate": 0.00028112936366587023,
"loss": 0.3053,
"step": 103000
},
{
"epoch": 2.3,
"eval_loss": 0.2896111011505127,
"eval_runtime": 2.0298,
"eval_samples_per_second": 1131.621,
"eval_steps_per_second": 17.735,
"step": 103000
},
{
"epoch": 2.31,
"learning_rate": 0.00028089213267527184,
"loss": 0.305,
"step": 103500
},
{
"epoch": 2.32,
"learning_rate": 0.0002806535249484803,
"loss": 0.3052,
"step": 104000
},
{
"epoch": 2.32,
"eval_loss": 0.28629741072654724,
"eval_runtime": 1.9678,
"eval_samples_per_second": 1167.305,
"eval_steps_per_second": 18.295,
"step": 104000
},
{
"epoch": 2.33,
"learning_rate": 0.00028041354309487135,
"loss": 0.3049,
"step": 104500
},
{
"epoch": 2.34,
"learning_rate": 0.0002801721897388482,
"loss": 0.3046,
"step": 105000
},
{
"epoch": 2.34,
"eval_loss": 0.288398802280426,
"eval_runtime": 2.0315,
"eval_samples_per_second": 1130.7,
"eval_steps_per_second": 17.721,
"step": 105000
},
{
"epoch": 2.35,
"learning_rate": 0.0002799294675198124,
"loss": 0.3062,
"step": 105500
},
{
"epoch": 2.36,
"learning_rate": 0.00027968537909213524,
"loss": 0.3041,
"step": 106000
},
{
"epoch": 2.36,
"eval_loss": 0.2860032618045807,
"eval_runtime": 1.9922,
"eval_samples_per_second": 1152.987,
"eval_steps_per_second": 18.07,
"step": 106000
},
{
"epoch": 2.38,
"learning_rate": 0.0002794399271251287,
"loss": 0.3038,
"step": 106500
},
{
"epoch": 2.39,
"learning_rate": 0.0002791931143030162,
"loss": 0.3046,
"step": 107000
},
{
"epoch": 2.39,
"eval_loss": 0.28703776001930237,
"eval_runtime": 1.8567,
"eval_samples_per_second": 1237.118,
"eval_steps_per_second": 19.389,
"step": 107000
},
{
"epoch": 2.4,
"learning_rate": 0.00027894494332490315,
"loss": 0.3034,
"step": 107500
},
{
"epoch": 2.41,
"learning_rate": 0.0002786954169047476,
"loss": 0.3034,
"step": 108000
},
{
"epoch": 2.41,
"eval_loss": 0.2861557602882385,
"eval_runtime": 1.9458,
"eval_samples_per_second": 1180.495,
"eval_steps_per_second": 18.501,
"step": 108000
},
{
"epoch": 2.42,
"learning_rate": 0.0002784445377713306,
"loss": 0.3037,
"step": 108500
},
{
"epoch": 2.43,
"learning_rate": 0.0002781923086682261,
"loss": 0.3034,
"step": 109000
},
{
"epoch": 2.43,
"eval_loss": 0.2837345004081726,
"eval_runtime": 2.0145,
"eval_samples_per_second": 1140.255,
"eval_steps_per_second": 17.871,
"step": 109000
},
{
"epoch": 2.44,
"learning_rate": 0.0002779387323537711,
"loss": 0.3033,
"step": 109500
},
{
"epoch": 2.45,
"learning_rate": 0.0002776838116010356,
"loss": 0.3029,
"step": 110000
},
{
"epoch": 2.45,
"eval_loss": 0.2857857942581177,
"eval_runtime": 1.9616,
"eval_samples_per_second": 1170.988,
"eval_steps_per_second": 18.352,
"step": 110000
},
{
"epoch": 2.46,
"learning_rate": 0.0002774275491977922,
"loss": 0.3028,
"step": 110500
},
{
"epoch": 2.48,
"learning_rate": 0.0002771699479464853,
"loss": 0.3026,
"step": 111000
},
{
"epoch": 2.48,
"eval_loss": 0.284712016582489,
"eval_runtime": 2.0468,
"eval_samples_per_second": 1122.254,
"eval_steps_per_second": 17.589,
"step": 111000
},
{
"epoch": 2.49,
"learning_rate": 0.00027691101066420104,
"loss": 0.3024,
"step": 111500
},
{
"epoch": 2.5,
"learning_rate": 0.0002766507401826361,
"loss": 0.3022,
"step": 112000
},
{
"epoch": 2.5,
"eval_loss": 0.2855146527290344,
"eval_runtime": 1.9866,
"eval_samples_per_second": 1156.261,
"eval_steps_per_second": 18.122,
"step": 112000
},
{
"epoch": 2.51,
"learning_rate": 0.0002763891393480666,
"loss": 0.3019,
"step": 112500
},
{
"epoch": 2.52,
"learning_rate": 0.0002761262110213175,
"loss": 0.3021,
"step": 113000
},
{
"epoch": 2.52,
"eval_loss": 0.2834793031215668,
"eval_runtime": 1.9281,
"eval_samples_per_second": 1191.346,
"eval_steps_per_second": 18.672,
"step": 113000
},
{
"epoch": 2.53,
"learning_rate": 0.00027586195807773083,
"loss": 0.3017,
"step": 113500
},
{
"epoch": 2.54,
"learning_rate": 0.00027559638340713435,
"loss": 0.3014,
"step": 114000
},
{
"epoch": 2.54,
"eval_loss": 0.28450024127960205,
"eval_runtime": 2.0139,
"eval_samples_per_second": 1140.591,
"eval_steps_per_second": 17.876,
"step": 114000
},
{
"epoch": 2.55,
"learning_rate": 0.00027532948991381025,
"loss": 0.3016,
"step": 114500
},
{
"epoch": 2.57,
"learning_rate": 0.00027506128051646287,
"loss": 0.3012,
"step": 115000
},
{
"epoch": 2.57,
"eval_loss": 0.2832774817943573,
"eval_runtime": 1.9735,
"eval_samples_per_second": 1163.926,
"eval_steps_per_second": 18.242,
"step": 115000
},
{
"epoch": 2.58,
"learning_rate": 0.00027479175814818733,
"loss": 0.3041,
"step": 115500
},
{
"epoch": 2.59,
"learning_rate": 0.000274520925756437,
"loss": 0.306,
"step": 116000
},
{
"epoch": 2.59,
"eval_loss": 0.2821986675262451,
"eval_runtime": 1.9791,
"eval_samples_per_second": 1160.624,
"eval_steps_per_second": 18.19,
"step": 116000
},
{
"epoch": 2.6,
"learning_rate": 0.00027424878630299157,
"loss": 0.301,
"step": 116500
},
{
"epoch": 2.61,
"learning_rate": 0.0002739753427639244,
"loss": 0.3004,
"step": 117000
},
{
"epoch": 2.61,
"eval_loss": 0.28125685453414917,
"eval_runtime": 2.0653,
"eval_samples_per_second": 1112.189,
"eval_steps_per_second": 17.431,
"step": 117000
},
{
"epoch": 2.62,
"learning_rate": 0.0002737005981295704,
"loss": 0.3005,
"step": 117500
},
{
"epoch": 2.63,
"learning_rate": 0.0002734245554044927,
"loss": 0.3,
"step": 118000
},
{
"epoch": 2.63,
"eval_loss": 0.28397423028945923,
"eval_runtime": 2.0248,
"eval_samples_per_second": 1134.431,
"eval_steps_per_second": 17.78,
"step": 118000
},
{
"epoch": 2.64,
"learning_rate": 0.0002731472176074504,
"loss": 0.3002,
"step": 118500
},
{
"epoch": 2.65,
"learning_rate": 0.0002728685877713653,
"loss": 0.3,
"step": 119000
},
{
"epoch": 2.65,
"eval_loss": 0.28364112973213196,
"eval_runtime": 1.9818,
"eval_samples_per_second": 1159.029,
"eval_steps_per_second": 18.165,
"step": 119000
},
{
"epoch": 2.67,
"learning_rate": 0.0002725886689432884,
"loss": 0.3007,
"step": 119500
},
{
"epoch": 2.68,
"learning_rate": 0.0002723074641843674,
"loss": 0.2996,
"step": 120000
},
{
"epoch": 2.68,
"eval_loss": 0.280828058719635,
"eval_runtime": 2.0656,
"eval_samples_per_second": 1112.04,
"eval_steps_per_second": 17.429,
"step": 120000
},
{
"epoch": 2.69,
"learning_rate": 0.0002720249765698123,
"loss": 0.2999,
"step": 120500
},
{
"epoch": 2.7,
"learning_rate": 0.0002717412091888626,
"loss": 0.3004,
"step": 121000
},
{
"epoch": 2.7,
"eval_loss": 0.28047671914100647,
"eval_runtime": 1.9731,
"eval_samples_per_second": 1164.181,
"eval_steps_per_second": 18.246,
"step": 121000
},
{
"epoch": 2.71,
"learning_rate": 0.00027145616514475274,
"loss": 0.2996,
"step": 121500
},
{
"epoch": 2.72,
"learning_rate": 0.0002711698475546788,
"loss": 0.299,
"step": 122000
},
{
"epoch": 2.72,
"eval_loss": 0.2800908386707306,
"eval_runtime": 2.05,
"eval_samples_per_second": 1120.501,
"eval_steps_per_second": 17.561,
"step": 122000
},
{
"epoch": 2.73,
"learning_rate": 0.00027088225954976407,
"loss": 0.2995,
"step": 122500
},
{
"epoch": 2.74,
"learning_rate": 0.0002705934042750249,
"loss": 0.2991,
"step": 123000
},
{
"epoch": 2.74,
"eval_loss": 0.27821099758148193,
"eval_runtime": 2.0199,
"eval_samples_per_second": 1137.17,
"eval_steps_per_second": 17.822,
"step": 123000
},
{
"epoch": 2.75,
"learning_rate": 0.00027030328488933625,
"loss": 0.2989,
"step": 123500
},
{
"epoch": 2.77,
"learning_rate": 0.00027001190456539726,
"loss": 0.2987,
"step": 124000
},
{
"epoch": 2.77,
"eval_loss": 0.2801131308078766,
"eval_runtime": 1.9909,
"eval_samples_per_second": 1153.743,
"eval_steps_per_second": 18.082,
"step": 124000
},
{
"epoch": 2.78,
"learning_rate": 0.0002697192664896965,
"loss": 0.2987,
"step": 124500
},
{
"epoch": 2.79,
"learning_rate": 0.00026942537386247706,
"loss": 0.2987,
"step": 125000
},
{
"epoch": 2.79,
"eval_loss": 0.2810814380645752,
"eval_runtime": 1.998,
"eval_samples_per_second": 1149.64,
"eval_steps_per_second": 18.018,
"step": 125000
},
{
"epoch": 2.8,
"learning_rate": 0.0002691302298977016,
"loss": 0.2985,
"step": 125500
},
{
"epoch": 2.81,
"learning_rate": 0.0002688338378230173,
"loss": 0.298,
"step": 126000
},
{
"epoch": 2.81,
"eval_loss": 0.27943727374076843,
"eval_runtime": 2.0072,
"eval_samples_per_second": 1144.392,
"eval_steps_per_second": 17.936,
"step": 126000
},
{
"epoch": 2.82,
"learning_rate": 0.00026853620087972035,
"loss": 0.2983,
"step": 126500
},
{
"epoch": 2.83,
"learning_rate": 0.00026823732232272065,
"loss": 0.2981,
"step": 127000
},
{
"epoch": 2.83,
"eval_loss": 0.27999699115753174,
"eval_runtime": 2.0113,
"eval_samples_per_second": 1142.034,
"eval_steps_per_second": 17.899,
"step": 127000
},
{
"epoch": 2.84,
"learning_rate": 0.0002679372054205063,
"loss": 0.2985,
"step": 127500
},
{
"epoch": 2.86,
"learning_rate": 0.0002676358534551076,
"loss": 0.2976,
"step": 128000
},
{
"epoch": 2.86,
"eval_loss": 0.2797393500804901,
"eval_runtime": 2.072,
"eval_samples_per_second": 1108.589,
"eval_steps_per_second": 17.374,
"step": 128000
},
{
"epoch": 2.87,
"learning_rate": 0.00026733326972206133,
"loss": 0.2981,
"step": 128500
},
{
"epoch": 2.88,
"learning_rate": 0.0002670294575303748,
"loss": 0.2977,
"step": 129000
},
{
"epoch": 2.88,
"eval_loss": 0.2784021496772766,
"eval_runtime": 2.0212,
"eval_samples_per_second": 1136.428,
"eval_steps_per_second": 17.811,
"step": 129000
},
{
"epoch": 2.89,
"learning_rate": 0.0002667244202024894,
"loss": 0.2975,
"step": 129500
},
{
"epoch": 2.9,
"learning_rate": 0.00026641816107424453,
"loss": 0.297,
"step": 130000
},
{
"epoch": 2.9,
"eval_loss": 0.2790050208568573,
"eval_runtime": 1.9904,
"eval_samples_per_second": 1154.041,
"eval_steps_per_second": 18.087,
"step": 130000
},
{
"epoch": 2.91,
"learning_rate": 0.0002661106834948409,
"loss": 0.2971,
"step": 130500
},
{
"epoch": 2.92,
"learning_rate": 0.0002658019908268041,
"loss": 0.2968,
"step": 131000
},
{
"epoch": 2.92,
"eval_loss": 0.27784037590026855,
"eval_runtime": 1.9858,
"eval_samples_per_second": 1156.736,
"eval_steps_per_second": 18.129,
"step": 131000
},
{
"epoch": 2.93,
"learning_rate": 0.00026549208644594766,
"loss": 0.2965,
"step": 131500
},
{
"epoch": 2.94,
"learning_rate": 0.00026518097374133627,
"loss": 0.2965,
"step": 132000
},
{
"epoch": 2.94,
"eval_loss": 0.27897170186042786,
"eval_runtime": 1.9786,
"eval_samples_per_second": 1160.938,
"eval_steps_per_second": 18.195,
"step": 132000
},
{
"epoch": 2.96,
"learning_rate": 0.00026486865611524853,
"loss": 0.2964,
"step": 132500
},
{
"epoch": 2.97,
"learning_rate": 0.00026455513698314003,
"loss": 0.2968,
"step": 133000
},
{
"epoch": 2.97,
"eval_loss": 0.27647751569747925,
"eval_runtime": 1.9935,
"eval_samples_per_second": 1152.23,
"eval_steps_per_second": 18.058,
"step": 133000
},
{
"epoch": 2.98,
"learning_rate": 0.0002642404197736058,
"loss": 0.2959,
"step": 133500
},
{
"epoch": 2.99,
"learning_rate": 0.0002639245079283428,
"loss": 0.2962,
"step": 134000
},
{
"epoch": 2.99,
"eval_loss": 0.2772797644138336,
"eval_runtime": 1.997,
"eval_samples_per_second": 1150.213,
"eval_steps_per_second": 18.027,
"step": 134000
},
{
"epoch": 3.0,
"learning_rate": 0.00026360740490211234,
"loss": 0.2957,
"step": 134500
},
{
"epoch": 3.01,
"learning_rate": 0.0002632891141627023,
"loss": 0.2961,
"step": 135000
},
{
"epoch": 3.01,
"eval_loss": 0.27765029668807983,
"eval_runtime": 1.9854,
"eval_samples_per_second": 1156.944,
"eval_steps_per_second": 18.132,
"step": 135000
},
{
"epoch": 3.02,
"learning_rate": 0.00026296963919088923,
"loss": 0.2956,
"step": 135500
},
{
"epoch": 3.03,
"learning_rate": 0.00026264898348040024,
"loss": 0.2955,
"step": 136000
},
{
"epoch": 3.03,
"eval_loss": 0.27664855122566223,
"eval_runtime": 1.9862,
"eval_samples_per_second": 1156.498,
"eval_steps_per_second": 18.125,
"step": 136000
},
{
"epoch": 3.04,
"learning_rate": 0.0002623271505378748,
"loss": 0.2952,
"step": 136500
},
{
"epoch": 3.06,
"learning_rate": 0.00026200414388282637,
"loss": 0.2951,
"step": 137000
},
{
"epoch": 3.06,
"eval_loss": 0.2762826979160309,
"eval_runtime": 1.9933,
"eval_samples_per_second": 1152.379,
"eval_steps_per_second": 18.061,
"step": 137000
},
{
"epoch": 3.07,
"learning_rate": 0.00026167996704760406,
"loss": 0.2952,
"step": 137500
},
{
"epoch": 3.08,
"learning_rate": 0.00026135462357735375,
"loss": 0.295,
"step": 138000
},
{
"epoch": 3.08,
"eval_loss": 0.27805039286613464,
"eval_runtime": 2.0209,
"eval_samples_per_second": 1136.598,
"eval_steps_per_second": 17.813,
"step": 138000
},
{
"epoch": 3.09,
"learning_rate": 0.0002610281170299795,
"loss": 0.2948,
"step": 138500
},
{
"epoch": 3.1,
"learning_rate": 0.00026070045097610465,
"loss": 0.2948,
"step": 139000
},
{
"epoch": 3.1,
"eval_loss": 0.27585625648498535,
"eval_runtime": 2.0475,
"eval_samples_per_second": 1121.859,
"eval_steps_per_second": 17.582,
"step": 139000
},
{
"epoch": 3.11,
"learning_rate": 0.0002603716289990326,
"loss": 0.2944,
"step": 139500
},
{
"epoch": 3.12,
"learning_rate": 0.00026004165469470787,
"loss": 0.2945,
"step": 140000
},
{
"epoch": 3.12,
"eval_loss": 0.28066232800483704,
"eval_runtime": 1.9112,
"eval_samples_per_second": 1201.849,
"eval_steps_per_second": 18.836,
"step": 140000
},
{
"epoch": 3.13,
"learning_rate": 0.0002597105316716766,
"loss": 0.2948,
"step": 140500
},
{
"epoch": 3.15,
"learning_rate": 0.000259378263551047,
"loss": 0.2945,
"step": 141000
},
{
"epoch": 3.15,
"eval_loss": 0.27713122963905334,
"eval_runtime": 1.9955,
"eval_samples_per_second": 1151.101,
"eval_steps_per_second": 18.041,
"step": 141000
},
{
"epoch": 3.16,
"learning_rate": 0.0002590448539664501,
"loss": 0.3036,
"step": 141500
},
{
"epoch": 3.17,
"learning_rate": 0.00025871030656399966,
"loss": 0.2997,
"step": 142000
},
{
"epoch": 3.17,
"eval_loss": 0.2775452733039856,
"eval_runtime": 1.9855,
"eval_samples_per_second": 1156.879,
"eval_steps_per_second": 18.131,
"step": 142000
},
{
"epoch": 3.18,
"learning_rate": 0.00025837462500225255,
"loss": 0.297,
"step": 142500
},
{
"epoch": 3.19,
"learning_rate": 0.0002580378129521685,
"loss": 0.2998,
"step": 143000
},
{
"epoch": 3.19,
"eval_loss": 0.2764818072319031,
"eval_runtime": 1.9519,
"eval_samples_per_second": 1176.822,
"eval_steps_per_second": 18.444,
"step": 143000
},
{
"epoch": 3.2,
"learning_rate": 0.0002576998740970701,
"loss": 0.2949,
"step": 143500
},
{
"epoch": 3.21,
"learning_rate": 0.00025736081213260253,
"loss": 0.2945,
"step": 144000
},
{
"epoch": 3.21,
"eval_loss": 0.27885541319847107,
"eval_runtime": 1.9999,
"eval_samples_per_second": 1148.549,
"eval_steps_per_second": 18.001,
"step": 144000
},
{
"epoch": 3.22,
"learning_rate": 0.0002570206307666931,
"loss": 0.2937,
"step": 144500
},
{
"epoch": 3.23,
"learning_rate": 0.0002566793337195108,
"loss": 0.2935,
"step": 145000
},
{
"epoch": 3.23,
"eval_loss": 0.2775752544403076,
"eval_runtime": 1.9798,
"eval_samples_per_second": 1160.196,
"eval_steps_per_second": 18.183,
"step": 145000
},
{
"epoch": 3.25,
"learning_rate": 0.0002563369247234254,
"loss": 0.2935,
"step": 145500
},
{
"epoch": 3.26,
"learning_rate": 0.0002559934075229669,
"loss": 0.2941,
"step": 146000
},
{
"epoch": 3.26,
"eval_loss": 0.2758236229419708,
"eval_runtime": 1.9855,
"eval_samples_per_second": 1156.897,
"eval_steps_per_second": 18.132,
"step": 146000
},
{
"epoch": 3.27,
"learning_rate": 0.0002556487858747843,
"loss": 0.2931,
"step": 146500
},
{
"epoch": 3.28,
"learning_rate": 0.00025530306354760464,
"loss": 0.2934,
"step": 147000
},
{
"epoch": 3.28,
"eval_loss": 0.2756287157535553,
"eval_runtime": 1.9891,
"eval_samples_per_second": 1154.8,
"eval_steps_per_second": 18.099,
"step": 147000
},
{
"epoch": 3.29,
"learning_rate": 0.000254956244322192,
"loss": 0.2929,
"step": 147500
},
{
"epoch": 3.3,
"learning_rate": 0.00025460833199130595,
"loss": 0.2931,
"step": 148000
},
{
"epoch": 3.3,
"eval_loss": 0.2749306857585907,
"eval_runtime": 1.9885,
"eval_samples_per_second": 1155.115,
"eval_steps_per_second": 18.104,
"step": 148000
},
{
"epoch": 3.31,
"learning_rate": 0.00025425933035965983,
"loss": 0.293,
"step": 148500
},
{
"epoch": 3.32,
"learning_rate": 0.00025390924324387965,
"loss": 0.2929,
"step": 149000
},
{
"epoch": 3.32,
"eval_loss": 0.2756015658378601,
"eval_runtime": 2.0342,
"eval_samples_per_second": 1129.181,
"eval_steps_per_second": 17.697,
"step": 149000
},
{
"epoch": 3.33,
"learning_rate": 0.0002535580744724621,
"loss": 0.2928,
"step": 149500
},
{
"epoch": 3.35,
"learning_rate": 0.00025320582788573246,
"loss": 0.2927,
"step": 150000
},
{
"epoch": 3.35,
"eval_loss": 0.27484118938446045,
"eval_runtime": 1.9877,
"eval_samples_per_second": 1155.618,
"eval_steps_per_second": 18.112,
"step": 150000
},
{
"epoch": 3.36,
"learning_rate": 0.000252852507335803,
"loss": 0.2923,
"step": 150500
},
{
"epoch": 3.37,
"learning_rate": 0.0002524981166865307,
"loss": 0.2924,
"step": 151000
},
{
"epoch": 3.37,
"eval_loss": 0.2738962769508362,
"eval_runtime": 1.9816,
"eval_samples_per_second": 1159.146,
"eval_steps_per_second": 18.167,
"step": 151000
},
{
"epoch": 3.38,
"learning_rate": 0.00025214265981347487,
"loss": 0.2921,
"step": 151500
},
{
"epoch": 3.39,
"learning_rate": 0.000251786140603855,
"loss": 0.292,
"step": 152000
},
{
"epoch": 3.39,
"eval_loss": 0.27476462721824646,
"eval_runtime": 1.9539,
"eval_samples_per_second": 1175.601,
"eval_steps_per_second": 18.425,
"step": 152000
},
{
"epoch": 3.4,
"learning_rate": 0.00025142856295650795,
"loss": 0.2919,
"step": 152500
},
{
"epoch": 3.41,
"learning_rate": 0.0002510699307818457,
"loss": 0.2917,
"step": 153000
},
{
"epoch": 3.41,
"eval_loss": 0.2740522623062134,
"eval_runtime": 1.9887,
"eval_samples_per_second": 1155.036,
"eval_steps_per_second": 18.102,
"step": 153000
},
{
"epoch": 3.42,
"learning_rate": 0.00025071024800181214,
"loss": 0.2913,
"step": 153500
},
{
"epoch": 3.44,
"learning_rate": 0.0002503495185498405,
"loss": 0.2913,
"step": 154000
},
{
"epoch": 3.44,
"eval_loss": 0.27570095658302307,
"eval_runtime": 1.9561,
"eval_samples_per_second": 1174.293,
"eval_steps_per_second": 18.404,
"step": 154000
},
{
"epoch": 3.45,
"learning_rate": 0.00024998774637081044,
"loss": 0.2914,
"step": 154500
},
{
"epoch": 3.46,
"learning_rate": 0.00024962493542100443,
"loss": 0.2911,
"step": 155000
},
{
"epoch": 3.46,
"eval_loss": 0.27467048168182373,
"eval_runtime": 1.9751,
"eval_samples_per_second": 1163.003,
"eval_steps_per_second": 18.227,
"step": 155000
},
{
"epoch": 3.47,
"learning_rate": 0.0002492610896680649,
"loss": 0.2911,
"step": 155500
},
{
"epoch": 3.48,
"learning_rate": 0.00024889621309095067,
"loss": 0.2911,
"step": 156000
},
{
"epoch": 3.48,
"eval_loss": 0.27558836340904236,
"eval_runtime": 1.9879,
"eval_samples_per_second": 1155.504,
"eval_steps_per_second": 18.11,
"step": 156000
},
{
"epoch": 3.49,
"learning_rate": 0.00024853030967989366,
"loss": 0.2913,
"step": 156500
},
{
"epoch": 3.5,
"learning_rate": 0.00024816338343635485,
"loss": 0.2909,
"step": 157000
},
{
"epoch": 3.5,
"eval_loss": 0.27131345868110657,
"eval_runtime": 1.9424,
"eval_samples_per_second": 1182.575,
"eval_steps_per_second": 18.534,
"step": 157000
},
{
"epoch": 3.51,
"learning_rate": 0.0002477954383729809,
"loss": 0.2908,
"step": 157500
},
{
"epoch": 3.52,
"learning_rate": 0.00024742647851355997,
"loss": 0.2908,
"step": 158000
},
{
"epoch": 3.52,
"eval_loss": 0.2729162871837616,
"eval_runtime": 2.0222,
"eval_samples_per_second": 1135.884,
"eval_steps_per_second": 17.802,
"step": 158000
},
{
"epoch": 3.54,
"learning_rate": 0.0002470565078929781,
"loss": 0.2909,
"step": 158500
},
{
"epoch": 3.55,
"learning_rate": 0.00024668553055717465,
"loss": 0.2905,
"step": 159000
},
{
"epoch": 3.55,
"eval_loss": 0.2731362581253052,
"eval_runtime": 1.928,
"eval_samples_per_second": 1191.362,
"eval_steps_per_second": 18.672,
"step": 159000
},
{
"epoch": 3.56,
"learning_rate": 0.0002463135505630984,
"loss": 0.2905,
"step": 159500
},
{
"epoch": 3.57,
"learning_rate": 0.00024594057197866283,
"loss": 0.2902,
"step": 160000
},
{
"epoch": 3.57,
"eval_loss": 0.27327215671539307,
"eval_runtime": 1.9424,
"eval_samples_per_second": 1182.586,
"eval_steps_per_second": 18.534,
"step": 160000
},
{
"epoch": 3.58,
"learning_rate": 0.0002455665988827021,
"loss": 0.2899,
"step": 160500
},
{
"epoch": 3.59,
"learning_rate": 0.0002451916353649261,
"loss": 0.2903,
"step": 161000
},
{
"epoch": 3.59,
"eval_loss": 0.2730531692504883,
"eval_runtime": 1.9636,
"eval_samples_per_second": 1169.786,
"eval_steps_per_second": 18.334,
"step": 161000
},
{
"epoch": 3.6,
"learning_rate": 0.00024481568552587566,
"loss": 0.2901,
"step": 161500
},
{
"epoch": 3.61,
"learning_rate": 0.0002444387534768781,
"loss": 0.2895,
"step": 162000
},
{
"epoch": 3.61,
"eval_loss": 0.2728096544742584,
"eval_runtime": 1.9828,
"eval_samples_per_second": 1158.488,
"eval_steps_per_second": 18.157,
"step": 162000
},
{
"epoch": 3.62,
"learning_rate": 0.0002440608433400018,
"loss": 0.2938,
"step": 162500
},
{
"epoch": 3.64,
"learning_rate": 0.00024368195924801158,
"loss": 0.2904,
"step": 163000
},
{
"epoch": 3.64,
"eval_loss": 0.27427318692207336,
"eval_runtime": 1.9845,
"eval_samples_per_second": 1157.485,
"eval_steps_per_second": 18.141,
"step": 163000
},
{
"epoch": 3.65,
"learning_rate": 0.00024330210534432314,
"loss": 0.29,
"step": 163500
},
{
"epoch": 3.66,
"learning_rate": 0.0002429212857829579,
"loss": 0.2891,
"step": 164000
},
{
"epoch": 3.66,
"eval_loss": 0.2683263421058655,
"eval_runtime": 1.9558,
"eval_samples_per_second": 1174.46,
"eval_steps_per_second": 18.407,
"step": 164000
},
{
"epoch": 3.67,
"learning_rate": 0.00024253950472849758,
"loss": 0.2893,
"step": 164500
},
{
"epoch": 3.68,
"learning_rate": 0.0002421567663560386,
"loss": 0.2888,
"step": 165000
},
{
"epoch": 3.68,
"eval_loss": 0.27206218242645264,
"eval_runtime": 1.9344,
"eval_samples_per_second": 1187.465,
"eval_steps_per_second": 18.611,
"step": 165000
},
{
"epoch": 3.69,
"learning_rate": 0.00024177307485114653,
"loss": 0.2891,
"step": 165500
},
{
"epoch": 3.7,
"learning_rate": 0.0002413884344098101,
"loss": 0.2889,
"step": 166000
},
{
"epoch": 3.7,
"eval_loss": 0.2694759666919708,
"eval_runtime": 1.9374,
"eval_samples_per_second": 1185.599,
"eval_steps_per_second": 18.581,
"step": 166000
},
{
"epoch": 3.71,
"learning_rate": 0.00024100284923839568,
"loss": 0.2892,
"step": 166500
},
{
"epoch": 3.73,
"learning_rate": 0.0002406163235536008,
"loss": 0.289,
"step": 167000
},
{
"epoch": 3.73,
"eval_loss": 0.27140599489212036,
"eval_runtime": 2.0107,
"eval_samples_per_second": 1142.373,
"eval_steps_per_second": 17.904,
"step": 167000
},
{
"epoch": 3.74,
"learning_rate": 0.00024022886158240857,
"loss": 0.2887,
"step": 167500
},
{
"epoch": 3.75,
"learning_rate": 0.0002398404675620409,
"loss": 0.6252,
"step": 168000
},
{
"epoch": 3.75,
"eval_loss": 0.6785263419151306,
"eval_runtime": 1.9867,
"eval_samples_per_second": 1156.175,
"eval_steps_per_second": 18.12,
"step": 168000
},
{
"epoch": 3.76,
"learning_rate": 0.0002394511457399126,
"loss": 0.6774,
"step": 168500
},
{
"epoch": 3.77,
"learning_rate": 0.00023906090037358478,
"loss": 0.6773,
"step": 169000
},
{
"epoch": 3.77,
"eval_loss": 0.6770340800285339,
"eval_runtime": 1.9319,
"eval_samples_per_second": 1188.977,
"eval_steps_per_second": 18.634,
"step": 169000
},
{
"epoch": 3.78,
"learning_rate": 0.0002386697357307182,
"loss": 0.6769,
"step": 169500
},
{
"epoch": 3.79,
"learning_rate": 0.00023827765608902676,
"loss": 0.6547,
"step": 170000
},
{
"epoch": 3.79,
"eval_loss": 0.6202901601791382,
"eval_runtime": 1.9551,
"eval_samples_per_second": 1174.884,
"eval_steps_per_second": 18.414,
"step": 170000
},
{
"epoch": 3.8,
"learning_rate": 0.0002378846657362306,
"loss": 0.3248,
"step": 170500
},
{
"epoch": 3.81,
"learning_rate": 0.00023749076897000928,
"loss": 0.2956,
"step": 171000
},
{
"epoch": 3.81,
"eval_loss": 0.27639877796173096,
"eval_runtime": 1.9923,
"eval_samples_per_second": 1152.936,
"eval_steps_per_second": 18.07,
"step": 171000
},
{
"epoch": 3.83,
"learning_rate": 0.00023709597009795465,
"loss": 0.2929,
"step": 171500
},
{
"epoch": 3.84,
"learning_rate": 0.000236700273437524,
"loss": 0.2914,
"step": 172000
},
{
"epoch": 3.84,
"eval_loss": 0.27524837851524353,
"eval_runtime": 1.9856,
"eval_samples_per_second": 1156.853,
"eval_steps_per_second": 18.131,
"step": 172000
},
{
"epoch": 3.85,
"learning_rate": 0.0002363036833159925,
"loss": 0.2913,
"step": 172500
},
{
"epoch": 3.86,
"learning_rate": 0.00023590620407040633,
"loss": 0.2901,
"step": 173000
},
{
"epoch": 3.86,
"eval_loss": 0.27291467785835266,
"eval_runtime": 2.0289,
"eval_samples_per_second": 1132.125,
"eval_steps_per_second": 17.743,
"step": 173000
},
{
"epoch": 3.87,
"learning_rate": 0.00023550784004753471,
"loss": 0.2896,
"step": 173500
},
{
"epoch": 3.88,
"learning_rate": 0.0002351085956038229,
"loss": 0.2891,
"step": 174000
},
{
"epoch": 3.88,
"eval_loss": 0.2731077969074249,
"eval_runtime": 1.9499,
"eval_samples_per_second": 1177.996,
"eval_steps_per_second": 18.462,
"step": 174000
},
{
"epoch": 3.89,
"learning_rate": 0.00023470847510534407,
"loss": 0.2891,
"step": 174500
},
{
"epoch": 3.9,
"learning_rate": 0.00023430748292775188,
"loss": 0.2886,
"step": 175000
},
{
"epoch": 3.9,
"eval_loss": 0.2711014151573181,
"eval_runtime": 1.958,
"eval_samples_per_second": 1173.126,
"eval_steps_per_second": 18.386,
"step": 175000
},
{
"epoch": 3.91,
"learning_rate": 0.0002339056234562326,
"loss": 0.2887,
"step": 175500
},
{
"epoch": 3.93,
"learning_rate": 0.00023350290108545694,
"loss": 0.2882,
"step": 176000
},
{
"epoch": 3.93,
"eval_loss": 0.27429357171058655,
"eval_runtime": 1.9855,
"eval_samples_per_second": 1156.89,
"eval_steps_per_second": 18.131,
"step": 176000
},
{
"epoch": 3.94,
"learning_rate": 0.00023309932021953238,
"loss": 0.2883,
"step": 176500
},
{
"epoch": 3.95,
"learning_rate": 0.00023269488527195446,
"loss": 0.2881,
"step": 177000
},
{
"epoch": 3.95,
"eval_loss": 0.26987290382385254,
"eval_runtime": 1.9002,
"eval_samples_per_second": 1208.791,
"eval_steps_per_second": 18.945,
"step": 177000
},
{
"epoch": 3.96,
"learning_rate": 0.00023228960066555907,
"loss": 0.2878,
"step": 177500
},
{
"epoch": 3.97,
"learning_rate": 0.00023188347083247365,
"loss": 0.2875,
"step": 178000
},
{
"epoch": 3.97,
"eval_loss": 0.2710895836353302,
"eval_runtime": 1.9126,
"eval_samples_per_second": 1200.973,
"eval_steps_per_second": 18.822,
"step": 178000
},
{
"epoch": 3.98,
"learning_rate": 0.00023147650021406905,
"loss": 0.2875,
"step": 178500
},
{
"epoch": 3.99,
"learning_rate": 0.00023106869326091075,
"loss": 0.2873,
"step": 179000
},
{
"epoch": 3.99,
"eval_loss": 0.27097710967063904,
"eval_runtime": 2.0543,
"eval_samples_per_second": 1118.139,
"eval_steps_per_second": 17.524,
"step": 179000
},
{
"epoch": 4.0,
"learning_rate": 0.00023066005443271017,
"loss": 0.2877,
"step": 179500
},
{
"epoch": 4.01,
"learning_rate": 0.00023025058819827618,
"loss": 0.287,
"step": 180000
},
{
"epoch": 4.01,
"eval_loss": 0.2711479663848877,
"eval_runtime": 2.0281,
"eval_samples_per_second": 1132.606,
"eval_steps_per_second": 17.751,
"step": 180000
},
{
"epoch": 4.03,
"learning_rate": 0.00022984029903546578,
"loss": 0.2873,
"step": 180500
},
{
"epoch": 4.04,
"learning_rate": 0.00022942919143113572,
"loss": 0.2873,
"step": 181000
},
{
"epoch": 4.04,
"eval_loss": 0.2718214690685272,
"eval_runtime": 1.9128,
"eval_samples_per_second": 1200.871,
"eval_steps_per_second": 18.821,
"step": 181000
},
{
"epoch": 4.05,
"learning_rate": 0.0002290172698810927,
"loss": 0.525,
"step": 181500
},
{
"epoch": 4.06,
"learning_rate": 0.00022860453889004493,
"loss": 0.6785,
"step": 182000
},
{
"epoch": 4.06,
"eval_loss": 0.6782656311988831,
"eval_runtime": 1.9047,
"eval_samples_per_second": 1205.991,
"eval_steps_per_second": 18.901,
"step": 182000
},
{
"epoch": 4.07,
"learning_rate": 0.00022819100297155235,
"loss": 0.6774,
"step": 182500
},
{
"epoch": 4.08,
"learning_rate": 0.0002277766666479774,
"loss": 0.6773,
"step": 183000
},
{
"epoch": 4.08,
"eval_loss": 0.6771561503410339,
"eval_runtime": 2.0046,
"eval_samples_per_second": 1145.885,
"eval_steps_per_second": 17.959,
"step": 183000
},
{
"epoch": 4.09,
"learning_rate": 0.00022736153445043595,
"loss": 0.6595,
"step": 183500
},
{
"epoch": 4.1,
"learning_rate": 0.00022694561091874706,
"loss": 0.3007,
"step": 184000
},
{
"epoch": 4.1,
"eval_loss": 0.2753528952598572,
"eval_runtime": 2.0143,
"eval_samples_per_second": 1140.359,
"eval_steps_per_second": 17.872,
"step": 184000
},
{
"epoch": 4.12,
"learning_rate": 0.00022652890060138387,
"loss": 0.2927,
"step": 184500
},
{
"epoch": 4.13,
"learning_rate": 0.00022611140805542366,
"loss": 0.2905,
"step": 185000
},
{
"epoch": 4.13,
"eval_loss": 0.27167341113090515,
"eval_runtime": 1.9686,
"eval_samples_per_second": 1166.834,
"eval_steps_per_second": 18.287,
"step": 185000
},
{
"epoch": 4.14,
"learning_rate": 0.00022569313784649798,
"loss": 0.2891,
"step": 185500
},
{
"epoch": 4.15,
"learning_rate": 0.0002252740945487429,
"loss": 0.288,
"step": 186000
},
{
"epoch": 4.15,
"eval_loss": 0.272190660238266,
"eval_runtime": 1.9619,
"eval_samples_per_second": 1170.775,
"eval_steps_per_second": 18.349,
"step": 186000
},
{
"epoch": 4.16,
"learning_rate": 0.00022485428274474867,
"loss": 0.2881,
"step": 186500
},
{
"epoch": 4.17,
"learning_rate": 0.00022443370702551,
"loss": 0.2873,
"step": 187000
},
{
"epoch": 4.17,
"eval_loss": 0.27140477299690247,
"eval_runtime": 1.9557,
"eval_samples_per_second": 1174.514,
"eval_steps_per_second": 18.408,
"step": 187000
},
{
"epoch": 4.18,
"learning_rate": 0.00022401237199037565,
"loss": 0.287,
"step": 187500
},
{
"epoch": 4.19,
"learning_rate": 0.0002235902822469979,
"loss": 0.2866,
"step": 188000
},
{
"epoch": 4.19,
"eval_loss": 0.2686367928981781,
"eval_runtime": 1.9954,
"eval_samples_per_second": 1151.13,
"eval_steps_per_second": 18.041,
"step": 188000
},
{
"epoch": 4.2,
"learning_rate": 0.00022316744241128268,
"loss": 0.2864,
"step": 188500
},
{
"epoch": 4.22,
"learning_rate": 0.00022274385710733855,
"loss": 0.2862,
"step": 189000
},
{
"epoch": 4.22,
"eval_loss": 0.2684060335159302,
"eval_runtime": 2.0107,
"eval_samples_per_second": 1142.364,
"eval_steps_per_second": 17.904,
"step": 189000
},
{
"epoch": 4.23,
"learning_rate": 0.00022231953096742672,
"loss": 0.286,
"step": 189500
},
{
"epoch": 4.24,
"learning_rate": 0.00022189446863190974,
"loss": 0.2862,
"step": 190000
},
{
"epoch": 4.24,
"eval_loss": 0.2695789337158203,
"eval_runtime": 1.9916,
"eval_samples_per_second": 1153.341,
"eval_steps_per_second": 18.076,
"step": 190000
},
{
"epoch": 4.25,
"learning_rate": 0.00022146867474920118,
"loss": 0.2859,
"step": 190500
},
{
"epoch": 4.26,
"learning_rate": 0.00022104215397571484,
"loss": 0.2858,
"step": 191000
},
{
"epoch": 4.26,
"eval_loss": 0.270698606967926,
"eval_runtime": 2.0195,
"eval_samples_per_second": 1137.436,
"eval_steps_per_second": 17.827,
"step": 191000
},
{
"epoch": 4.27,
"learning_rate": 0.0002206149109758135,
"loss": 0.2855,
"step": 191500
},
{
"epoch": 4.28,
"learning_rate": 0.00022018695042175818,
"loss": 0.2865,
"step": 192000
},
{
"epoch": 4.28,
"eval_loss": 0.2688451111316681,
"eval_runtime": 2.0,
"eval_samples_per_second": 1148.488,
"eval_steps_per_second": 18.0,
"step": 192000
},
{
"epoch": 4.29,
"learning_rate": 0.00021975827699365693,
"loss": 0.2849,
"step": 192500
},
{
"epoch": 4.3,
"learning_rate": 0.00021932889537941365,
"loss": 0.2851,
"step": 193000
},
{
"epoch": 4.3,
"eval_loss": 0.26711875200271606,
"eval_runtime": 1.9779,
"eval_samples_per_second": 1161.305,
"eval_steps_per_second": 18.201,
"step": 193000
},
{
"epoch": 4.32,
"learning_rate": 0.0002188988102746769,
"loss": 0.2871,
"step": 193500
},
{
"epoch": 4.33,
"learning_rate": 0.0002184680263827885,
"loss": 0.2853,
"step": 194000
},
{
"epoch": 4.33,
"eval_loss": 0.26893720030784607,
"eval_runtime": 1.9488,
"eval_samples_per_second": 1178.699,
"eval_steps_per_second": 18.473,
"step": 194000
},
{
"epoch": 4.34,
"learning_rate": 0.00021803654841473204,
"loss": 0.286,
"step": 194500
},
{
"epoch": 4.35,
"learning_rate": 0.00021760438108908142,
"loss": 0.2853,
"step": 195000
},
{
"epoch": 4.35,
"eval_loss": 0.2694932520389557,
"eval_runtime": 1.9477,
"eval_samples_per_second": 1179.331,
"eval_steps_per_second": 18.483,
"step": 195000
},
{
"epoch": 4.36,
"learning_rate": 0.0002171715291319494,
"loss": 0.2845,
"step": 195500
},
{
"epoch": 4.37,
"learning_rate": 0.0002167379972769355,
"loss": 0.2853,
"step": 196000
},
{
"epoch": 4.37,
"eval_loss": 0.2688761353492737,
"eval_runtime": 1.976,
"eval_samples_per_second": 1162.467,
"eval_steps_per_second": 18.219,
"step": 196000
},
{
"epoch": 4.38,
"learning_rate": 0.0002163037902650747,
"loss": 0.2845,
"step": 196500
},
{
"epoch": 4.39,
"learning_rate": 0.0002158689128447853,
"loss": 0.2844,
"step": 197000
},
{
"epoch": 4.39,
"eval_loss": 0.2665814757347107,
"eval_runtime": 1.9647,
"eval_samples_per_second": 1169.149,
"eval_steps_per_second": 18.324,
"step": 197000
},
{
"epoch": 4.41,
"learning_rate": 0.00021543336977181704,
"loss": 0.2841,
"step": 197500
},
{
"epoch": 4.42,
"learning_rate": 0.00021499716580919933,
"loss": 0.2837,
"step": 198000
},
{
"epoch": 4.42,
"eval_loss": 0.2675575613975525,
"eval_runtime": 2.0149,
"eval_samples_per_second": 1140.003,
"eval_steps_per_second": 17.867,
"step": 198000
},
{
"epoch": 4.43,
"learning_rate": 0.00021456030572718866,
"loss": 0.2841,
"step": 198500
},
{
"epoch": 4.44,
"learning_rate": 0.000214122794303217,
"loss": 0.2837,
"step": 199000
},
{
"epoch": 4.44,
"eval_loss": 0.2669568657875061,
"eval_runtime": 1.9768,
"eval_samples_per_second": 1161.955,
"eval_steps_per_second": 18.211,
"step": 199000
},
{
"epoch": 4.45,
"learning_rate": 0.00021368463632183912,
"loss": 0.2838,
"step": 199500
},
{
"epoch": 4.46,
"learning_rate": 0.00021324583657468055,
"loss": 0.2838,
"step": 200000
},
{
"epoch": 4.46,
"eval_loss": 0.2675861120223999,
"eval_runtime": 2.0013,
"eval_samples_per_second": 1147.731,
"eval_steps_per_second": 17.988,
"step": 200000
},
{
"epoch": 4.47,
"learning_rate": 0.000212806399860385,
"loss": 0.2837,
"step": 200500
},
{
"epoch": 4.48,
"learning_rate": 0.00021236633098456196,
"loss": 0.2835,
"step": 201000
},
{
"epoch": 4.48,
"eval_loss": 0.26529645919799805,
"eval_runtime": 2.0487,
"eval_samples_per_second": 1121.175,
"eval_steps_per_second": 17.572,
"step": 201000
},
{
"epoch": 4.49,
"learning_rate": 0.0002119256347597342,
"loss": 0.2834,
"step": 201500
},
{
"epoch": 4.51,
"learning_rate": 0.000211484316005285,
"loss": 0.2835,
"step": 202000
},
{
"epoch": 4.51,
"eval_loss": 0.2684980034828186,
"eval_runtime": 2.049,
"eval_samples_per_second": 1121.029,
"eval_steps_per_second": 17.569,
"step": 202000
},
{
"epoch": 4.52,
"learning_rate": 0.00021104237954740554,
"loss": 0.2833,
"step": 202500
},
{
"epoch": 4.53,
"learning_rate": 0.00021059983021904215,
"loss": 0.2834,
"step": 203000
},
{
"epoch": 4.53,
"eval_loss": 0.26848986744880676,
"eval_runtime": 2.068,
"eval_samples_per_second": 1110.727,
"eval_steps_per_second": 17.408,
"step": 203000
},
{
"epoch": 4.54,
"learning_rate": 0.00021015667285984336,
"loss": 0.2834,
"step": 203500
},
{
"epoch": 4.55,
"learning_rate": 0.00020971291231610707,
"loss": 0.283,
"step": 204000
},
{
"epoch": 4.55,
"eval_loss": 0.26503807306289673,
"eval_runtime": 1.9718,
"eval_samples_per_second": 1164.914,
"eval_steps_per_second": 18.257,
"step": 204000
},
{
"epoch": 4.56,
"learning_rate": 0.0002092685534407274,
"loss": 0.2829,
"step": 204500
},
{
"epoch": 4.57,
"learning_rate": 0.00020882360109314197,
"loss": 0.2828,
"step": 205000
},
{
"epoch": 4.57,
"eval_loss": 0.2652031183242798,
"eval_runtime": 2.0127,
"eval_samples_per_second": 1141.23,
"eval_steps_per_second": 17.886,
"step": 205000
},
{
"epoch": 4.58,
"learning_rate": 0.0002083780601392783,
"loss": 0.2832,
"step": 205500
},
{
"epoch": 4.59,
"learning_rate": 0.0002079319354515008,
"loss": 0.2827,
"step": 206000
},
{
"epoch": 4.59,
"eval_loss": 0.26393648982048035,
"eval_runtime": 2.0365,
"eval_samples_per_second": 1127.908,
"eval_steps_per_second": 17.677,
"step": 206000
},
{
"epoch": 4.61,
"learning_rate": 0.00020748523190855772,
"loss": 0.2827,
"step": 206500
},
{
"epoch": 4.62,
"learning_rate": 0.0002070379543955273,
"loss": 0.2826,
"step": 207000
},
{
"epoch": 4.62,
"eval_loss": 0.26564860343933105,
"eval_runtime": 1.9676,
"eval_samples_per_second": 1167.423,
"eval_steps_per_second": 18.297,
"step": 207000
},
{
"epoch": 4.63,
"learning_rate": 0.00020659010780376487,
"loss": 0.2825,
"step": 207500
},
{
"epoch": 4.64,
"learning_rate": 0.00020614169703084896,
"loss": 0.2824,
"step": 208000
},
{
"epoch": 4.64,
"eval_loss": 0.26574334502220154,
"eval_runtime": 2.0204,
"eval_samples_per_second": 1136.893,
"eval_steps_per_second": 17.818,
"step": 208000
},
{
"epoch": 4.65,
"learning_rate": 0.0002056927269805279,
"loss": 0.2822,
"step": 208500
},
{
"epoch": 4.66,
"learning_rate": 0.00020524320256266635,
"loss": 0.2823,
"step": 209000
},
{
"epoch": 4.66,
"eval_loss": 0.26812705397605896,
"eval_runtime": 2.0173,
"eval_samples_per_second": 1138.645,
"eval_steps_per_second": 17.846,
"step": 209000
},
{
"epoch": 4.67,
"learning_rate": 0.0002047931286931912,
"loss": 0.2821,
"step": 209500
},
{
"epoch": 4.68,
"learning_rate": 0.00020434251029403824,
"loss": 0.2819,
"step": 210000
},
{
"epoch": 4.68,
"eval_loss": 0.2622651159763336,
"eval_runtime": 1.9772,
"eval_samples_per_second": 1161.716,
"eval_steps_per_second": 18.207,
"step": 210000
},
{
"epoch": 4.7,
"learning_rate": 0.00020389135229309803,
"loss": 0.2819,
"step": 210500
},
{
"epoch": 4.71,
"learning_rate": 0.00020343965962416229,
"loss": 0.2818,
"step": 211000
},
{
"epoch": 4.71,
"eval_loss": 0.26698386669158936,
"eval_runtime": 1.9801,
"eval_samples_per_second": 1160.056,
"eval_steps_per_second": 18.181,
"step": 211000
},
{
"epoch": 4.72,
"learning_rate": 0.00020298743722686958,
"loss": 0.2818,
"step": 211500
},
{
"epoch": 4.73,
"learning_rate": 0.0002025346900466516,
"loss": 0.2818,
"step": 212000
},
{
"epoch": 4.73,
"eval_loss": 0.2666832506656647,
"eval_runtime": 2.0181,
"eval_samples_per_second": 1138.224,
"eval_steps_per_second": 17.839,
"step": 212000
},
{
"epoch": 4.74,
"learning_rate": 0.0002020814230346791,
"loss": 0.2815,
"step": 212500
},
{
"epoch": 4.75,
"learning_rate": 0.00020162764114780733,
"loss": 0.2816,
"step": 213000
},
{
"epoch": 4.75,
"eval_loss": 0.2648046016693115,
"eval_runtime": 1.9978,
"eval_samples_per_second": 1149.778,
"eval_steps_per_second": 18.02,
"step": 213000
},
{
"epoch": 4.76,
"learning_rate": 0.0002011733493485224,
"loss": 0.2817,
"step": 213500
},
{
"epoch": 4.77,
"learning_rate": 0.00020071855260488664,
"loss": 0.2815,
"step": 214000
},
{
"epoch": 4.77,
"eval_loss": 0.2653491497039795,
"eval_runtime": 1.9416,
"eval_samples_per_second": 1183.015,
"eval_steps_per_second": 18.541,
"step": 214000
},
{
"epoch": 4.78,
"learning_rate": 0.0002002632558904843,
"loss": 0.2813,
"step": 214500
},
{
"epoch": 4.8,
"learning_rate": 0.00019980746418436736,
"loss": 0.2814,
"step": 215000
},
{
"epoch": 4.8,
"eval_loss": 0.2634630799293518,
"eval_runtime": 1.9884,
"eval_samples_per_second": 1155.227,
"eval_steps_per_second": 18.105,
"step": 215000
},
{
"epoch": 4.81,
"learning_rate": 0.00019935118247100088,
"loss": 0.6253,
"step": 215500
},
{
"epoch": 4.82,
"learning_rate": 0.00019889441574020864,
"loss": 0.6773,
"step": 216000
},
{
"epoch": 4.82,
"eval_loss": 0.6766608357429504,
"eval_runtime": 2.0373,
"eval_samples_per_second": 1127.446,
"eval_steps_per_second": 17.67,
"step": 216000
},
{
"epoch": 4.83,
"learning_rate": 0.0001984371689871183,
"loss": 0.6773,
"step": 216500
},
{
"epoch": 4.84,
"learning_rate": 0.00019797944721210725,
"loss": 0.6251,
"step": 217000
},
{
"epoch": 4.84,
"eval_loss": 0.29495057463645935,
"eval_runtime": 1.9016,
"eval_samples_per_second": 1207.906,
"eval_steps_per_second": 18.931,
"step": 217000
},
{
"epoch": 4.85,
"learning_rate": 0.00019752125542074736,
"loss": 0.2897,
"step": 217500
},
{
"epoch": 4.86,
"learning_rate": 0.00019706259862375074,
"loss": 0.285,
"step": 218000
},
{
"epoch": 4.86,
"eval_loss": 0.2679499685764313,
"eval_runtime": 1.9722,
"eval_samples_per_second": 1164.712,
"eval_steps_per_second": 18.254,
"step": 218000
},
{
"epoch": 4.87,
"learning_rate": 0.00019660348183691453,
"loss": 0.2835,
"step": 218500
},
{
"epoch": 4.88,
"learning_rate": 0.0001961439100810664,
"loss": 0.2825,
"step": 219000
},
{
"epoch": 4.88,
"eval_loss": 0.26572567224502563,
"eval_runtime": 2.0053,
"eval_samples_per_second": 1145.441,
"eval_steps_per_second": 17.952,
"step": 219000
},
{
"epoch": 4.9,
"learning_rate": 0.00019568388838200952,
"loss": 0.2819,
"step": 219500
},
{
"epoch": 4.91,
"learning_rate": 0.00019522342177046744,
"loss": 0.2818,
"step": 220000
},
{
"epoch": 4.91,
"eval_loss": 0.26373621821403503,
"eval_runtime": 1.9567,
"eval_samples_per_second": 1173.894,
"eval_steps_per_second": 18.398,
"step": 220000
},
{
"epoch": 4.92,
"learning_rate": 0.00019476251528202922,
"loss": 0.2815,
"step": 220500
},
{
"epoch": 4.93,
"learning_rate": 0.0001943011739570944,
"loss": 0.2812,
"step": 221000
},
{
"epoch": 4.93,
"eval_loss": 0.26436519622802734,
"eval_runtime": 2.0008,
"eval_samples_per_second": 1148.012,
"eval_steps_per_second": 17.992,
"step": 221000
},
{
"epoch": 4.94,
"learning_rate": 0.00019383940284081774,
"loss": 0.2808,
"step": 221500
},
{
"epoch": 4.95,
"learning_rate": 0.00019337720698305431,
"loss": 0.2808,
"step": 222000
},
{
"epoch": 4.95,
"eval_loss": 0.26223430037498474,
"eval_runtime": 1.9438,
"eval_samples_per_second": 1181.718,
"eval_steps_per_second": 18.521,
"step": 222000
},
{
"epoch": 4.96,
"learning_rate": 0.0001929145914383038,
"loss": 0.2805,
"step": 222500
},
{
"epoch": 4.97,
"learning_rate": 0.00019245156126565586,
"loss": 0.2804,
"step": 223000
},
{
"epoch": 4.97,
"eval_loss": 0.26395902037620544,
"eval_runtime": 1.9723,
"eval_samples_per_second": 1164.61,
"eval_steps_per_second": 18.252,
"step": 223000
},
{
"epoch": 4.99,
"learning_rate": 0.00019198812152873416,
"loss": 0.2805,
"step": 223500
},
{
"epoch": 5.0,
"learning_rate": 0.00019152427729564144,
"loss": 0.2803,
"step": 224000
},
{
"epoch": 5.0,
"eval_loss": 0.2627279460430145,
"eval_runtime": 1.9464,
"eval_samples_per_second": 1180.147,
"eval_steps_per_second": 18.496,
"step": 224000
},
{
"epoch": 5.01,
"learning_rate": 0.00019106003363890395,
"loss": 0.2801,
"step": 224500
},
{
"epoch": 5.02,
"learning_rate": 0.00019059539563541584,
"loss": 0.28,
"step": 225000
},
{
"epoch": 5.02,
"eval_loss": 0.2637809216976166,
"eval_runtime": 1.9918,
"eval_samples_per_second": 1153.2,
"eval_steps_per_second": 18.074,
"step": 225000
},
{
"epoch": 5.03,
"learning_rate": 0.000190130368366384,
"loss": 0.2797,
"step": 225500
},
{
"epoch": 5.04,
"learning_rate": 0.00018966495691727207,
"loss": 0.2797,
"step": 226000
},
{
"epoch": 5.04,
"eval_loss": 0.26251962780952454,
"eval_runtime": 2.0094,
"eval_samples_per_second": 1143.149,
"eval_steps_per_second": 17.916,
"step": 226000
},
{
"epoch": 5.05,
"learning_rate": 0.0001891991663777451,
"loss": 0.2797,
"step": 226500
},
{
"epoch": 5.06,
"learning_rate": 0.00018873300184161387,
"loss": 0.2795,
"step": 227000
},
{
"epoch": 5.06,
"eval_loss": 0.26239344477653503,
"eval_runtime": 2.0026,
"eval_samples_per_second": 1146.993,
"eval_steps_per_second": 17.976,
"step": 227000
},
{
"epoch": 5.07,
"learning_rate": 0.00018826646840677894,
"loss": 0.2798,
"step": 227500
},
{
"epoch": 5.09,
"learning_rate": 0.00018779957117517532,
"loss": 0.2825,
"step": 228000
},
{
"epoch": 5.09,
"eval_loss": 0.2686799168586731,
"eval_runtime": 2.0437,
"eval_samples_per_second": 1123.939,
"eval_steps_per_second": 17.615,
"step": 228000
},
{
"epoch": 5.1,
"learning_rate": 0.00018733231525271625,
"loss": 0.2845,
"step": 228500
},
{
"epoch": 5.11,
"learning_rate": 0.00018686470574923766,
"loss": 0.283,
"step": 229000
},
{
"epoch": 5.11,
"eval_loss": 0.2658364176750183,
"eval_runtime": 1.9961,
"eval_samples_per_second": 1150.721,
"eval_steps_per_second": 18.035,
"step": 229000
},
{
"epoch": 5.12,
"learning_rate": 0.00018639674777844224,
"loss": 0.2821,
"step": 229500
},
{
"epoch": 5.13,
"learning_rate": 0.00018592844645784327,
"loss": 0.2791,
"step": 230000
},
{
"epoch": 5.13,
"eval_loss": 0.2603645920753479,
"eval_runtime": 2.047,
"eval_samples_per_second": 1122.145,
"eval_steps_per_second": 17.587,
"step": 230000
},
{
"epoch": 5.14,
"learning_rate": 0.00018545980690870903,
"loss": 0.2799,
"step": 230500
},
{
"epoch": 5.15,
"learning_rate": 0.00018499083425600648,
"loss": 0.2791,
"step": 231000
},
{
"epoch": 5.15,
"eval_loss": 0.26100829243659973,
"eval_runtime": 1.9738,
"eval_samples_per_second": 1163.773,
"eval_steps_per_second": 18.239,
"step": 231000
},
{
"epoch": 5.16,
"learning_rate": 0.00018452153362834552,
"loss": 0.2791,
"step": 231500
},
{
"epoch": 5.17,
"learning_rate": 0.00018405191015792254,
"loss": 0.2829,
"step": 232000
},
{
"epoch": 5.17,
"eval_loss": 0.26664701104164124,
"eval_runtime": 2.018,
"eval_samples_per_second": 1138.267,
"eval_steps_per_second": 17.84,
"step": 232000
},
{
"epoch": 5.19,
"learning_rate": 0.0001835819689804646,
"loss": 0.2818,
"step": 232500
},
{
"epoch": 5.2,
"learning_rate": 0.0001831117152351732,
"loss": 0.2789,
"step": 233000
},
{
"epoch": 5.2,
"eval_loss": 0.2605225443840027,
"eval_runtime": 1.994,
"eval_samples_per_second": 1151.94,
"eval_steps_per_second": 18.054,
"step": 233000
},
{
"epoch": 5.21,
"learning_rate": 0.00018264115406466778,
"loss": 0.2784,
"step": 233500
},
{
"epoch": 5.22,
"learning_rate": 0.00018217029061493007,
"loss": 0.2816,
"step": 234000
},
{
"epoch": 5.22,
"eval_loss": 0.26246902346611023,
"eval_runtime": 2.0381,
"eval_samples_per_second": 1127.052,
"eval_steps_per_second": 17.664,
"step": 234000
},
{
"epoch": 5.23,
"learning_rate": 0.00018169913003524717,
"loss": 0.2818,
"step": 234500
},
{
"epoch": 5.24,
"learning_rate": 0.00018122767747815594,
"loss": 0.2791,
"step": 235000
},
{
"epoch": 5.24,
"eval_loss": 0.26137277483940125,
"eval_runtime": 2.0611,
"eval_samples_per_second": 1114.437,
"eval_steps_per_second": 17.466,
"step": 235000
},
{
"epoch": 5.25,
"learning_rate": 0.00018075593809938574,
"loss": 0.2783,
"step": 235500
},
{
"epoch": 5.26,
"learning_rate": 0.00018028391705780295,
"loss": 0.2782,
"step": 236000
},
{
"epoch": 5.26,
"eval_loss": 0.26220422983169556,
"eval_runtime": 2.0409,
"eval_samples_per_second": 1125.474,
"eval_steps_per_second": 17.639,
"step": 236000
},
{
"epoch": 5.28,
"learning_rate": 0.0001798116195153541,
"loss": 0.2782,
"step": 236500
},
{
"epoch": 5.29,
"learning_rate": 0.0001793390506370094,
"loss": 0.2816,
"step": 237000
},
{
"epoch": 5.29,
"eval_loss": 0.2643316388130188,
"eval_runtime": 2.0022,
"eval_samples_per_second": 1147.236,
"eval_steps_per_second": 17.98,
"step": 237000
},
{
"epoch": 5.3,
"learning_rate": 0.00017886621559070638,
"loss": 0.281,
"step": 237500
},
{
"epoch": 5.31,
"learning_rate": 0.00017839311954729337,
"loss": 0.2799,
"step": 238000
},
{
"epoch": 5.31,
"eval_loss": 0.2624029219150543,
"eval_runtime": 1.9308,
"eval_samples_per_second": 1189.661,
"eval_steps_per_second": 18.645,
"step": 238000
},
{
"epoch": 5.32,
"learning_rate": 0.00017791976768047292,
"loss": 0.28,
"step": 238500
},
{
"epoch": 5.33,
"learning_rate": 0.00017744616516674518,
"loss": 0.2792,
"step": 239000
},
{
"epoch": 5.33,
"eval_loss": 0.2625776529312134,
"eval_runtime": 1.9068,
"eval_samples_per_second": 1204.635,
"eval_steps_per_second": 18.88,
"step": 239000
},
{
"epoch": 5.34,
"learning_rate": 0.00017697231718535132,
"loss": 0.2807,
"step": 239500
},
{
"epoch": 5.35,
"learning_rate": 0.00017649822891821707,
"loss": 0.2794,
"step": 240000
},
{
"epoch": 5.35,
"eval_loss": 0.26089775562286377,
"eval_runtime": 2.0641,
"eval_samples_per_second": 1112.856,
"eval_steps_per_second": 17.441,
"step": 240000
},
{
"epoch": 5.36,
"learning_rate": 0.00017602390554989563,
"loss": 0.2782,
"step": 240500
},
{
"epoch": 5.38,
"learning_rate": 0.0001755493522675115,
"loss": 0.2778,
"step": 241000
},
{
"epoch": 5.38,
"eval_loss": 0.26353833079338074,
"eval_runtime": 2.0451,
"eval_samples_per_second": 1123.162,
"eval_steps_per_second": 17.603,
"step": 241000
},
{
"epoch": 5.39,
"learning_rate": 0.00017507457426070317,
"loss": 0.2778,
"step": 241500
},
{
"epoch": 5.4,
"learning_rate": 0.00017459957672156704,
"loss": 0.2843,
"step": 242000
},
{
"epoch": 5.4,
"eval_loss": 0.2617889940738678,
"eval_runtime": 2.1044,
"eval_samples_per_second": 1091.504,
"eval_steps_per_second": 17.107,
"step": 242000
},
{
"epoch": 5.41,
"learning_rate": 0.00017412436484459998,
"loss": 0.2799,
"step": 242500
},
{
"epoch": 5.42,
"learning_rate": 0.00017364894382664297,
"loss": 0.2778,
"step": 243000
},
{
"epoch": 5.42,
"eval_loss": 0.261848509311676,
"eval_runtime": 2.0011,
"eval_samples_per_second": 1147.894,
"eval_steps_per_second": 17.991,
"step": 243000
},
{
"epoch": 5.43,
"learning_rate": 0.00017317331886682408,
"loss": 0.2781,
"step": 243500
},
{
"epoch": 5.44,
"learning_rate": 0.0001726974951665017,
"loss": 0.2777,
"step": 244000
},
{
"epoch": 5.44,
"eval_loss": 0.2593342363834381,
"eval_runtime": 2.0597,
"eval_samples_per_second": 1115.201,
"eval_steps_per_second": 17.478,
"step": 244000
},
{
"epoch": 5.45,
"learning_rate": 0.0001722214779292076,
"loss": 0.2771,
"step": 244500
},
{
"epoch": 5.46,
"learning_rate": 0.00017174527236058998,
"loss": 0.2769,
"step": 245000
},
{
"epoch": 5.46,
"eval_loss": 0.2584320306777954,
"eval_runtime": 2.0154,
"eval_samples_per_second": 1139.721,
"eval_steps_per_second": 17.862,
"step": 245000
},
{
"epoch": 5.48,
"learning_rate": 0.00017126888366835662,
"loss": 0.278,
"step": 245500
},
{
"epoch": 5.49,
"learning_rate": 0.000170792317062218,
"loss": 0.2793,
"step": 246000
},
{
"epoch": 5.49,
"eval_loss": 0.2609806954860687,
"eval_runtime": 2.0172,
"eval_samples_per_second": 1138.717,
"eval_steps_per_second": 17.847,
"step": 246000
},
{
"epoch": 5.5,
"learning_rate": 0.00017031557775383011,
"loss": 0.277,
"step": 246500
},
{
"epoch": 5.51,
"learning_rate": 0.0001698386709567377,
"loss": 0.2775,
"step": 247000
},
{
"epoch": 5.51,
"eval_loss": 0.25816255807876587,
"eval_runtime": 2.0194,
"eval_samples_per_second": 1137.488,
"eval_steps_per_second": 17.827,
"step": 247000
},
{
"epoch": 5.52,
"learning_rate": 0.0001693616018863171,
"loss": 0.2768,
"step": 247500
},
{
"epoch": 5.53,
"learning_rate": 0.00016888437575971913,
"loss": 0.278,
"step": 248000
},
{
"epoch": 5.53,
"eval_loss": 0.2611255645751953,
"eval_runtime": 2.0505,
"eval_samples_per_second": 1120.189,
"eval_steps_per_second": 17.556,
"step": 248000
},
{
"epoch": 5.54,
"learning_rate": 0.00016840699779581238,
"loss": 0.2768,
"step": 248500
},
{
"epoch": 5.55,
"learning_rate": 0.00016792947321512573,
"loss": 0.2765,
"step": 249000
},
{
"epoch": 5.55,
"eval_loss": 0.25990262627601624,
"eval_runtime": 2.0037,
"eval_samples_per_second": 1146.355,
"eval_steps_per_second": 17.966,
"step": 249000
},
{
"epoch": 5.57,
"learning_rate": 0.00016745180723979144,
"loss": 0.2796,
"step": 249500
},
{
"epoch": 5.58,
"learning_rate": 0.00016697400509348818,
"loss": 0.2767,
"step": 250000
},
{
"epoch": 5.58,
"eval_loss": 0.25914427638053894,
"eval_runtime": 2.0331,
"eval_samples_per_second": 1129.79,
"eval_steps_per_second": 17.707,
"step": 250000
},
{
"epoch": 5.59,
"learning_rate": 0.00016649607200138356,
"loss": 0.2772,
"step": 250500
},
{
"epoch": 5.6,
"learning_rate": 0.00016601801319007743,
"loss": 0.2766,
"step": 251000
},
{
"epoch": 5.6,
"eval_loss": 0.25976353883743286,
"eval_runtime": 2.0645,
"eval_samples_per_second": 1112.603,
"eval_steps_per_second": 17.437,
"step": 251000
},
{
"epoch": 5.61,
"learning_rate": 0.00016553983388754428,
"loss": 0.276,
"step": 251500
},
{
"epoch": 5.62,
"learning_rate": 0.00016506153932307636,
"loss": 0.2776,
"step": 252000
},
{
"epoch": 5.62,
"eval_loss": 0.2604348659515381,
"eval_runtime": 1.9875,
"eval_samples_per_second": 1155.746,
"eval_steps_per_second": 18.114,
"step": 252000
},
{
"epoch": 5.63,
"learning_rate": 0.00016458313472722638,
"loss": 0.2761,
"step": 252500
},
{
"epoch": 5.64,
"learning_rate": 0.00016410462533175045,
"loss": 0.2759,
"step": 253000
},
{
"epoch": 5.64,
"eval_loss": 0.25958681106567383,
"eval_runtime": 2.0685,
"eval_samples_per_second": 1110.487,
"eval_steps_per_second": 17.404,
"step": 253000
},
{
"epoch": 5.65,
"learning_rate": 0.00016362601636955049,
"loss": 0.2757,
"step": 253500
},
{
"epoch": 5.67,
"learning_rate": 0.00016314731307461754,
"loss": 0.2762,
"step": 254000
},
{
"epoch": 5.67,
"eval_loss": 0.2618536949157715,
"eval_runtime": 2.0252,
"eval_samples_per_second": 1134.221,
"eval_steps_per_second": 17.776,
"step": 254000
},
{
"epoch": 5.68,
"learning_rate": 0.0001626685206819742,
"loss": 0.2775,
"step": 254500
},
{
"epoch": 5.69,
"learning_rate": 0.0001621896444276172,
"loss": 0.2762,
"step": 255000
},
{
"epoch": 5.69,
"eval_loss": 0.26094868779182434,
"eval_runtime": 2.0113,
"eval_samples_per_second": 1142.074,
"eval_steps_per_second": 17.899,
"step": 255000
},
{
"epoch": 5.7,
"learning_rate": 0.00016171068954846067,
"loss": 0.276,
"step": 255500
},
{
"epoch": 5.71,
"learning_rate": 0.00016123166128227835,
"loss": 0.2754,
"step": 256000
},
{
"epoch": 5.71,
"eval_loss": 0.26099157333374023,
"eval_runtime": 1.9698,
"eval_samples_per_second": 1166.087,
"eval_steps_per_second": 18.276,
"step": 256000
},
{
"epoch": 5.72,
"learning_rate": 0.0001607525648676467,
"loss": 0.2756,
"step": 256500
},
{
"epoch": 5.73,
"learning_rate": 0.0001602734055438873,
"loss": 0.2757,
"step": 257000
},
{
"epoch": 5.73,
"eval_loss": 0.2581404149532318,
"eval_runtime": 1.9737,
"eval_samples_per_second": 1163.806,
"eval_steps_per_second": 18.24,
"step": 257000
},
{
"epoch": 5.74,
"learning_rate": 0.00015979418855100963,
"loss": 0.2757,
"step": 257500
},
{
"epoch": 5.75,
"learning_rate": 0.00015931491912965417,
"loss": 0.2751,
"step": 258000
},
{
"epoch": 5.75,
"eval_loss": 0.2564995288848877,
"eval_runtime": 1.9044,
"eval_samples_per_second": 1206.164,
"eval_steps_per_second": 18.904,
"step": 258000
},
{
"epoch": 5.77,
"learning_rate": 0.0001588356025210344,
"loss": 0.2752,
"step": 258500
},
{
"epoch": 5.78,
"learning_rate": 0.00015835624396688,
"loss": 0.275,
"step": 259000
},
{
"epoch": 5.78,
"eval_loss": 0.2554072439670563,
"eval_runtime": 1.984,
"eval_samples_per_second": 1157.747,
"eval_steps_per_second": 18.145,
"step": 259000
},
{
"epoch": 5.79,
"learning_rate": 0.00015787684870937924,
"loss": 0.2747,
"step": 259500
},
{
"epoch": 5.8,
"learning_rate": 0.00015739742199112196,
"loss": 0.2748,
"step": 260000
},
{
"epoch": 5.8,
"eval_loss": 0.2579441964626312,
"eval_runtime": 1.8886,
"eval_samples_per_second": 1216.255,
"eval_steps_per_second": 19.062,
"step": 260000
},
{
"epoch": 5.81,
"learning_rate": 0.00015691796905504187,
"loss": 0.2748,
"step": 260500
},
{
"epoch": 5.82,
"learning_rate": 0.00015643849514435944,
"loss": 0.2747,
"step": 261000
},
{
"epoch": 5.82,
"eval_loss": 0.2571495771408081,
"eval_runtime": 1.9462,
"eval_samples_per_second": 1180.236,
"eval_steps_per_second": 18.497,
"step": 261000
},
{
"epoch": 5.83,
"learning_rate": 0.00015595900550252463,
"loss": 0.2746,
"step": 261500
},
{
"epoch": 5.84,
"learning_rate": 0.00015547950537315926,
"loss": 0.2747,
"step": 262000
},
{
"epoch": 5.84,
"eval_loss": 0.25983506441116333,
"eval_runtime": 1.9787,
"eval_samples_per_second": 1160.871,
"eval_steps_per_second": 18.194,
"step": 262000
},
{
"epoch": 5.86,
"learning_rate": 0.00015499999999999997,
"loss": 0.2744,
"step": 262500
},
{
"epoch": 5.87,
"learning_rate": 0.00015452049462684068,
"loss": 0.2744,
"step": 263000
},
{
"epoch": 5.87,
"eval_loss": 0.2598707377910614,
"eval_runtime": 2.0257,
"eval_samples_per_second": 1133.91,
"eval_steps_per_second": 17.771,
"step": 263000
},
{
"epoch": 5.88,
"learning_rate": 0.00015404099449747535,
"loss": 0.2742,
"step": 263500
},
{
"epoch": 5.89,
"learning_rate": 0.0001535615048556405,
"loss": 0.2746,
"step": 264000
},
{
"epoch": 5.89,
"eval_loss": 0.2571488916873932,
"eval_runtime": 1.9862,
"eval_samples_per_second": 1156.504,
"eval_steps_per_second": 18.125,
"step": 264000
},
{
"epoch": 5.9,
"learning_rate": 0.0001530820309449581,
"loss": 0.274,
"step": 264500
},
{
"epoch": 5.91,
"learning_rate": 0.00015260257800887798,
"loss": 0.2743,
"step": 265000
},
{
"epoch": 5.91,
"eval_loss": 0.2586234211921692,
"eval_runtime": 1.9966,
"eval_samples_per_second": 1150.434,
"eval_steps_per_second": 18.03,
"step": 265000
},
{
"epoch": 5.92,
"learning_rate": 0.0001521231512906207,
"loss": 0.274,
"step": 265500
},
{
"epoch": 5.93,
"learning_rate": 0.00015164375603311998,
"loss": 0.274,
"step": 266000
},
{
"epoch": 5.93,
"eval_loss": 0.2586924433708191,
"eval_runtime": 1.9388,
"eval_samples_per_second": 1184.767,
"eval_steps_per_second": 18.568,
"step": 266000
},
{
"epoch": 5.94,
"learning_rate": 0.00015116439747896553,
"loss": 0.2739,
"step": 266500
},
{
"epoch": 5.96,
"learning_rate": 0.00015068508087034578,
"loss": 0.2741,
"step": 267000
},
{
"epoch": 5.96,
"eval_loss": 0.25905266404151917,
"eval_runtime": 1.9258,
"eval_samples_per_second": 1192.738,
"eval_steps_per_second": 18.693,
"step": 267000
},
{
"epoch": 5.97,
"learning_rate": 0.00015020581144899027,
"loss": 0.2737,
"step": 267500
},
{
"epoch": 5.98,
"learning_rate": 0.0001497265944561127,
"loss": 0.2735,
"step": 268000
},
{
"epoch": 5.98,
"eval_loss": 0.2568492293357849,
"eval_runtime": 2.0451,
"eval_samples_per_second": 1123.172,
"eval_steps_per_second": 17.603,
"step": 268000
},
{
"epoch": 5.99,
"learning_rate": 0.00014924743513235327,
"loss": 0.2735,
"step": 268500
},
{
"epoch": 6.0,
"learning_rate": 0.0001487683387177216,
"loss": 0.2736,
"step": 269000
},
{
"epoch": 6.0,
"eval_loss": 0.25593629479408264,
"eval_runtime": 2.0056,
"eval_samples_per_second": 1145.295,
"eval_steps_per_second": 17.95,
"step": 269000
},
{
"epoch": 6.01,
"learning_rate": 0.00014828931045153928,
"loss": 0.2736,
"step": 269500
},
{
"epoch": 6.02,
"learning_rate": 0.00014781035557238272,
"loss": 0.2735,
"step": 270000
},
{
"epoch": 6.02,
"eval_loss": 0.2567874789237976,
"eval_runtime": 1.9639,
"eval_samples_per_second": 1169.619,
"eval_steps_per_second": 18.331,
"step": 270000
},
{
"epoch": 6.03,
"learning_rate": 0.00014733147931802578,
"loss": 0.2733,
"step": 270500
},
{
"epoch": 6.04,
"learning_rate": 0.00014685268692538238,
"loss": 0.273,
"step": 271000
},
{
"epoch": 6.04,
"eval_loss": 0.2563098967075348,
"eval_runtime": 1.9603,
"eval_samples_per_second": 1171.784,
"eval_steps_per_second": 18.365,
"step": 271000
},
{
"epoch": 6.06,
"learning_rate": 0.00014637398363044946,
"loss": 0.2732,
"step": 271500
},
{
"epoch": 6.07,
"learning_rate": 0.00014589537466824955,
"loss": 0.2732,
"step": 272000
},
{
"epoch": 6.07,
"eval_loss": 0.2550007402896881,
"eval_runtime": 1.9875,
"eval_samples_per_second": 1155.744,
"eval_steps_per_second": 18.114,
"step": 272000
},
{
"epoch": 6.08,
"learning_rate": 0.00014541686527277356,
"loss": 0.2729,
"step": 272500
},
{
"epoch": 6.09,
"learning_rate": 0.00014493846067692358,
"loss": 0.2759,
"step": 273000
},
{
"epoch": 6.09,
"eval_loss": 0.2564890682697296,
"eval_runtime": 1.9877,
"eval_samples_per_second": 1155.615,
"eval_steps_per_second": 18.112,
"step": 273000
},
{
"epoch": 6.1,
"learning_rate": 0.00014446016611245567,
"loss": 0.279,
"step": 273500
},
{
"epoch": 6.11,
"learning_rate": 0.00014398198680992252,
"loss": 0.2778,
"step": 274000
},
{
"epoch": 6.11,
"eval_loss": 0.25714361667633057,
"eval_runtime": 2.0307,
"eval_samples_per_second": 1131.141,
"eval_steps_per_second": 17.728,
"step": 274000
},
{
"epoch": 6.12,
"learning_rate": 0.00014350392799861636,
"loss": 0.2758,
"step": 274500
},
{
"epoch": 6.13,
"learning_rate": 0.0001430259949065118,
"loss": 0.275,
"step": 275000
},
{
"epoch": 6.13,
"eval_loss": 0.258448988199234,
"eval_runtime": 1.9657,
"eval_samples_per_second": 1168.538,
"eval_steps_per_second": 18.314,
"step": 275000
},
{
"epoch": 6.15,
"learning_rate": 0.0001425481927602085,
"loss": 0.2749,
"step": 275500
},
{
"epoch": 6.16,
"learning_rate": 0.0001420705267848743,
"loss": 0.2742,
"step": 276000
},
{
"epoch": 6.16,
"eval_loss": 0.25592052936553955,
"eval_runtime": 1.9618,
"eval_samples_per_second": 1170.879,
"eval_steps_per_second": 18.351,
"step": 276000
},
{
"epoch": 6.17,
"learning_rate": 0.00014159300220418757,
"loss": 0.2742,
"step": 276500
},
{
"epoch": 6.18,
"learning_rate": 0.0001411156242402808,
"loss": 0.2741,
"step": 277000
},
{
"epoch": 6.18,
"eval_loss": 0.2545378506183624,
"eval_runtime": 1.9657,
"eval_samples_per_second": 1168.513,
"eval_steps_per_second": 18.314,
"step": 277000
},
{
"epoch": 6.19,
"learning_rate": 0.0001406383981136829,
"loss": 0.2726,
"step": 277500
},
{
"epoch": 6.2,
"learning_rate": 0.00014016132904326226,
"loss": 0.2723,
"step": 278000
},
{
"epoch": 6.2,
"eval_loss": 0.25342997908592224,
"eval_runtime": 1.92,
"eval_samples_per_second": 1196.358,
"eval_steps_per_second": 18.75,
"step": 278000
},
{
"epoch": 6.21,
"learning_rate": 0.00013968442224616989,
"loss": 0.2721,
"step": 278500
},
{
"epoch": 6.22,
"learning_rate": 0.00013920768293778195,
"loss": 0.2722,
"step": 279000
},
{
"epoch": 6.22,
"eval_loss": 0.25338229537010193,
"eval_runtime": 1.9729,
"eval_samples_per_second": 1164.267,
"eval_steps_per_second": 18.247,
"step": 279000
},
{
"epoch": 6.23,
"learning_rate": 0.00013873111633164336,
"loss": 0.2722,
"step": 279500
},
{
"epoch": 6.25,
"learning_rate": 0.00013825472763941,
"loss": 0.2722,
"step": 280000
},
{
"epoch": 6.25,
"eval_loss": 0.25419288873672485,
"eval_runtime": 1.9659,
"eval_samples_per_second": 1168.404,
"eval_steps_per_second": 18.312,
"step": 280000
},
{
"epoch": 6.26,
"learning_rate": 0.00013777852207079235,
"loss": 0.2716,
"step": 280500
},
{
"epoch": 6.27,
"learning_rate": 0.00013730250483349825,
"loss": 0.2718,
"step": 281000
},
{
"epoch": 6.27,
"eval_loss": 0.25364091992378235,
"eval_runtime": 2.0184,
"eval_samples_per_second": 1138.006,
"eval_steps_per_second": 17.836,
"step": 281000
},
{
"epoch": 6.28,
"learning_rate": 0.00013682668113317584,
"loss": 0.2718,
"step": 281500
},
{
"epoch": 6.29,
"learning_rate": 0.00013635105617335703,
"loss": 0.2717,
"step": 282000
},
{
"epoch": 6.29,
"eval_loss": 0.2546054422855377,
"eval_runtime": 1.9353,
"eval_samples_per_second": 1186.875,
"eval_steps_per_second": 18.601,
"step": 282000
},
{
"epoch": 6.3,
"learning_rate": 0.00013587563515539996,
"loss": 0.2718,
"step": 282500
},
{
"epoch": 6.31,
"learning_rate": 0.00013540042327843296,
"loss": 0.2716,
"step": 283000
},
{
"epoch": 6.31,
"eval_loss": 0.25594207644462585,
"eval_runtime": 1.9903,
"eval_samples_per_second": 1154.11,
"eval_steps_per_second": 18.088,
"step": 283000
},
{
"epoch": 6.32,
"learning_rate": 0.00013492542573929678,
"loss": 0.2718,
"step": 283500
},
{
"epoch": 6.33,
"learning_rate": 0.00013445064773248846,
"loss": 0.2718,
"step": 284000
},
{
"epoch": 6.33,
"eval_loss": 0.25390535593032837,
"eval_runtime": 1.9996,
"eval_samples_per_second": 1148.745,
"eval_steps_per_second": 18.004,
"step": 284000
},
{
"epoch": 6.35,
"learning_rate": 0.00013397609445010432,
"loss": 0.2711,
"step": 284500
},
{
"epoch": 6.36,
"learning_rate": 0.00013350177108178288,
"loss": 0.2715,
"step": 285000
},
{
"epoch": 6.36,
"eval_loss": 0.2538110613822937,
"eval_runtime": 2.0098,
"eval_samples_per_second": 1142.879,
"eval_steps_per_second": 17.912,
"step": 285000
},
{
"epoch": 6.37,
"learning_rate": 0.00013302768281464863,
"loss": 0.2713,
"step": 285500
},
{
"epoch": 6.38,
"learning_rate": 0.0001325538348332548,
"loss": 0.2713,
"step": 286000
},
{
"epoch": 6.38,
"eval_loss": 0.2533688545227051,
"eval_runtime": 1.9973,
"eval_samples_per_second": 1150.057,
"eval_steps_per_second": 18.024,
"step": 286000
},
{
"epoch": 6.39,
"learning_rate": 0.00013208023231952706,
"loss": 0.2711,
"step": 286500
},
{
"epoch": 6.4,
"learning_rate": 0.0001316068804527066,
"loss": 0.271,
"step": 287000
},
{
"epoch": 6.4,
"eval_loss": 0.25663408637046814,
"eval_runtime": 2.0033,
"eval_samples_per_second": 1146.595,
"eval_steps_per_second": 17.97,
"step": 287000
},
{
"epoch": 6.41,
"learning_rate": 0.00013113378440929353,
"loss": 0.2709,
"step": 287500
},
{
"epoch": 6.42,
"learning_rate": 0.00013066094936299056,
"loss": 0.2707,
"step": 288000
},
{
"epoch": 6.42,
"eval_loss": 0.2554568946361542,
"eval_runtime": 1.9451,
"eval_samples_per_second": 1180.915,
"eval_steps_per_second": 18.508,
"step": 288000
},
{
"epoch": 6.44,
"learning_rate": 0.00013018838048464582,
"loss": 0.271,
"step": 288500
},
{
"epoch": 6.45,
"learning_rate": 0.00012971608294219702,
"loss": 0.2709,
"step": 289000
},
{
"epoch": 6.45,
"eval_loss": 0.2539980113506317,
"eval_runtime": 2.0049,
"eval_samples_per_second": 1145.67,
"eval_steps_per_second": 17.956,
"step": 289000
},
{
"epoch": 6.46,
"learning_rate": 0.00012924406190061423,
"loss": 0.2707,
"step": 289500
},
{
"epoch": 6.47,
"learning_rate": 0.0001287723225218441,
"loss": 0.2706,
"step": 290000
},
{
"epoch": 6.47,
"eval_loss": 0.2525833547115326,
"eval_runtime": 1.9728,
"eval_samples_per_second": 1164.356,
"eval_steps_per_second": 18.249,
"step": 290000
},
{
"epoch": 6.48,
"learning_rate": 0.00012830086996475274,
"loss": 0.2707,
"step": 290500
},
{
"epoch": 6.49,
"learning_rate": 0.00012782970938506988,
"loss": 0.2707,
"step": 291000
},
{
"epoch": 6.49,
"eval_loss": 0.2514384388923645,
"eval_runtime": 2.0121,
"eval_samples_per_second": 1141.574,
"eval_steps_per_second": 17.891,
"step": 291000
},
{
"epoch": 6.5,
"learning_rate": 0.00012735884593533222,
"loss": 0.2706,
"step": 291500
},
{
"epoch": 6.51,
"learning_rate": 0.0001268882847648268,
"loss": 0.2702,
"step": 292000
},
{
"epoch": 6.51,
"eval_loss": 0.2527428865432739,
"eval_runtime": 2.1423,
"eval_samples_per_second": 1072.197,
"eval_steps_per_second": 16.804,
"step": 292000
},
{
"epoch": 6.52,
"learning_rate": 0.00012641803101953535,
"loss": 0.271,
"step": 292500
},
{
"epoch": 6.54,
"learning_rate": 0.00012594808984207743,
"loss": 0.2702,
"step": 293000
},
{
"epoch": 6.54,
"eval_loss": 0.2542712688446045,
"eval_runtime": 2.0253,
"eval_samples_per_second": 1134.154,
"eval_steps_per_second": 17.775,
"step": 293000
},
{
"epoch": 6.55,
"learning_rate": 0.00012547846637165445,
"loss": 0.2698,
"step": 293500
},
{
"epoch": 6.56,
"learning_rate": 0.00012500916574399346,
"loss": 0.27,
"step": 294000
},
{
"epoch": 6.56,
"eval_loss": 0.25084540247917175,
"eval_runtime": 1.9928,
"eval_samples_per_second": 1152.639,
"eval_steps_per_second": 18.065,
"step": 294000
},
{
"epoch": 6.57,
"learning_rate": 0.00012454019309129095,
"loss": 0.2698,
"step": 294500
},
{
"epoch": 6.58,
"learning_rate": 0.0001240715535421567,
"loss": 0.27,
"step": 295000
},
{
"epoch": 6.58,
"eval_loss": 0.252863347530365,
"eval_runtime": 1.9718,
"eval_samples_per_second": 1164.945,
"eval_steps_per_second": 18.258,
"step": 295000
},
{
"epoch": 6.59,
"learning_rate": 0.00012360325222155773,
"loss": 0.2701,
"step": 295500
},
{
"epoch": 6.6,
"learning_rate": 0.00012313529425076228,
"loss": 0.2697,
"step": 296000
},
{
"epoch": 6.6,
"eval_loss": 0.25461703538894653,
"eval_runtime": 1.9521,
"eval_samples_per_second": 1176.688,
"eval_steps_per_second": 18.442,
"step": 296000
},
{
"epoch": 6.61,
"learning_rate": 0.00012266768474728372,
"loss": 0.2698,
"step": 296500
},
{
"epoch": 6.62,
"learning_rate": 0.00012220042882482468,
"loss": 0.2703,
"step": 297000
},
{
"epoch": 6.62,
"eval_loss": 0.2524978220462799,
"eval_runtime": 2.026,
"eval_samples_per_second": 1133.755,
"eval_steps_per_second": 17.769,
"step": 297000
},
{
"epoch": 6.64,
"learning_rate": 0.00012173353159322102,
"loss": 0.2695,
"step": 297500
},
{
"epoch": 6.65,
"learning_rate": 0.00012126699815838609,
"loss": 0.2697,
"step": 298000
},
{
"epoch": 6.65,
"eval_loss": 0.2543928623199463,
"eval_runtime": 1.9625,
"eval_samples_per_second": 1170.438,
"eval_steps_per_second": 18.344,
"step": 298000
},
{
"epoch": 6.66,
"learning_rate": 0.00012080083362225484,
"loss": 0.2696,
"step": 298500
},
{
"epoch": 6.67,
"learning_rate": 0.00012033504308272786,
"loss": 0.2698,
"step": 299000
},
{
"epoch": 6.67,
"eval_loss": 0.24954737722873688,
"eval_runtime": 1.9913,
"eval_samples_per_second": 1153.505,
"eval_steps_per_second": 18.078,
"step": 299000
},
{
"epoch": 6.68,
"learning_rate": 0.00011986963163361598,
"loss": 0.2715,
"step": 299500
},
{
"epoch": 6.69,
"learning_rate": 0.0001194046043645841,
"loss": 0.274,
"step": 300000
},
{
"epoch": 6.69,
"eval_loss": 0.2508510649204254,
"eval_runtime": 1.967,
"eval_samples_per_second": 1167.773,
"eval_steps_per_second": 18.302,
"step": 300000
},
{
"epoch": 6.7,
"learning_rate": 0.00011893996636109606,
"loss": 0.2728,
"step": 300500
},
{
"epoch": 6.71,
"learning_rate": 0.00011847572270435852,
"loss": 0.2726,
"step": 301000
},
{
"epoch": 6.71,
"eval_loss": 0.2552023231983185,
"eval_runtime": 5.3426,
"eval_samples_per_second": 429.941,
"eval_steps_per_second": 6.738,
"step": 301000
},
{
"epoch": 6.73,
"learning_rate": 0.00011801187847126579,
"loss": 0.2696,
"step": 301500
},
{
"epoch": 6.74,
"learning_rate": 0.00011754843873434411,
"loss": 0.2691,
"step": 302000
},
{
"epoch": 6.74,
"eval_loss": 0.25297147035598755,
"eval_runtime": 1.9927,
"eval_samples_per_second": 1152.684,
"eval_steps_per_second": 18.066,
"step": 302000
},
{
"epoch": 6.75,
"learning_rate": 0.00011708540856169612,
"loss": 0.2702,
"step": 302500
},
{
"epoch": 6.76,
"learning_rate": 0.00011662279301694567,
"loss": 0.2698,
"step": 303000
},
{
"epoch": 6.76,
"eval_loss": 0.253802090883255,
"eval_runtime": 1.9502,
"eval_samples_per_second": 1177.813,
"eval_steps_per_second": 18.459,
"step": 303000
},
{
"epoch": 6.77,
"learning_rate": 0.0001161605971591822,
"loss": 0.2697,
"step": 303500
},
{
"epoch": 6.78,
"learning_rate": 0.00011569882604290559,
"loss": 0.2706,
"step": 304000
},
{
"epoch": 6.78,
"eval_loss": 0.252233624458313,
"eval_runtime": 2.0068,
"eval_samples_per_second": 1144.633,
"eval_steps_per_second": 17.939,
"step": 304000
},
{
"epoch": 6.79,
"learning_rate": 0.00011523748471797075,
"loss": 0.2704,
"step": 304500
},
{
"epoch": 6.8,
"learning_rate": 0.00011477657822953255,
"loss": 0.2705,
"step": 305000
},
{
"epoch": 6.8,
"eval_loss": 0.25286948680877686,
"eval_runtime": 2.0237,
"eval_samples_per_second": 1135.028,
"eval_steps_per_second": 17.789,
"step": 305000
},
{
"epoch": 6.81,
"learning_rate": 0.00011431611161799043,
"loss": 0.2732,
"step": 305500
},
{
"epoch": 6.83,
"learning_rate": 0.0001138560899189335,
"loss": 0.2707,
"step": 306000
},
{
"epoch": 6.83,
"eval_loss": 0.2536955773830414,
"eval_runtime": 2.005,
"eval_samples_per_second": 1145.645,
"eval_steps_per_second": 17.955,
"step": 306000
},
{
"epoch": 6.84,
"learning_rate": 0.00011339651816308543,
"loss": 0.271,
"step": 306500
},
{
"epoch": 6.85,
"learning_rate": 0.00011293740137624925,
"loss": 0.2713,
"step": 307000
},
{
"epoch": 6.85,
"eval_loss": 0.25167515873908997,
"eval_runtime": 1.9694,
"eval_samples_per_second": 1166.352,
"eval_steps_per_second": 18.28,
"step": 307000
},
{
"epoch": 6.86,
"learning_rate": 0.00011247874457925261,
"loss": 0.2716,
"step": 307500
},
{
"epoch": 6.87,
"learning_rate": 0.0001120205527878927,
"loss": 0.2696,
"step": 308000
},
{
"epoch": 6.87,
"eval_loss": 0.25594934821128845,
"eval_runtime": 1.9543,
"eval_samples_per_second": 1175.377,
"eval_steps_per_second": 18.421,
"step": 308000
},
{
"epoch": 6.88,
"learning_rate": 0.00011156283101288165,
"loss": 0.2701,
"step": 308500
},
{
"epoch": 6.89,
"learning_rate": 0.00011110558425979132,
"loss": 0.2702,
"step": 309000
},
{
"epoch": 6.89,
"eval_loss": 0.2543644607067108,
"eval_runtime": 1.9412,
"eval_samples_per_second": 1183.27,
"eval_steps_per_second": 18.545,
"step": 309000
},
{
"epoch": 6.9,
"learning_rate": 0.00011064881752899906,
"loss": 0.2698,
"step": 309500
},
{
"epoch": 6.91,
"learning_rate": 0.00011019253581563262,
"loss": 0.2695,
"step": 310000
},
{
"epoch": 6.91,
"eval_loss": 0.2494657188653946,
"eval_runtime": 2.0069,
"eval_samples_per_second": 1144.556,
"eval_steps_per_second": 17.938,
"step": 310000
},
{
"epoch": 6.93,
"learning_rate": 0.00010973674410951567,
"loss": 0.2688,
"step": 310500
},
{
"epoch": 6.94,
"learning_rate": 0.00010928144739511337,
"loss": 0.2688,
"step": 311000
},
{
"epoch": 6.94,
"eval_loss": 0.2559347450733185,
"eval_runtime": 1.9953,
"eval_samples_per_second": 1151.191,
"eval_steps_per_second": 18.042,
"step": 311000
},
{
"epoch": 6.95,
"learning_rate": 0.00010882665065147757,
"loss": 0.269,
"step": 311500
},
{
"epoch": 6.96,
"learning_rate": 0.00010837235885219267,
"loss": 0.2681,
"step": 312000
},
{
"epoch": 6.96,
"eval_loss": 0.2514026165008545,
"eval_runtime": 2.0268,
"eval_samples_per_second": 1133.324,
"eval_steps_per_second": 17.762,
"step": 312000
},
{
"epoch": 6.97,
"learning_rate": 0.00010791857696532089,
"loss": 0.2679,
"step": 312500
},
{
"epoch": 6.98,
"learning_rate": 0.00010746530995334832,
"loss": 0.2678,
"step": 313000
},
{
"epoch": 6.98,
"eval_loss": 0.25468307733535767,
"eval_runtime": 1.9525,
"eval_samples_per_second": 1176.424,
"eval_steps_per_second": 18.438,
"step": 313000
},
{
"epoch": 6.99,
"learning_rate": 0.0001070125627731304,
"loss": 0.268,
"step": 313500
},
{
"epoch": 7.0,
"learning_rate": 0.0001065603403758377,
"loss": 0.2682,
"step": 314000
},
{
"epoch": 7.0,
"eval_loss": 0.2510579824447632,
"eval_runtime": 2.0207,
"eval_samples_per_second": 1136.748,
"eval_steps_per_second": 17.816,
"step": 314000
},
{
"epoch": 7.02,
"learning_rate": 0.00010610864770690196,
"loss": 0.2676,
"step": 314500
},
{
"epoch": 7.03,
"learning_rate": 0.00010565748970596172,
"loss": 0.2675,
"step": 315000
},
{
"epoch": 7.03,
"eval_loss": 0.25261634588241577,
"eval_runtime": 2.04,
"eval_samples_per_second": 1125.977,
"eval_steps_per_second": 17.647,
"step": 315000
},
{
"epoch": 7.04,
"learning_rate": 0.00010520687130680884,
"loss": 0.2676,
"step": 315500
},
{
"epoch": 7.05,
"learning_rate": 0.00010475679743733364,
"loss": 0.2672,
"step": 316000
},
{
"epoch": 7.05,
"eval_loss": 0.25101763010025024,
"eval_runtime": 1.9764,
"eval_samples_per_second": 1162.216,
"eval_steps_per_second": 18.215,
"step": 316000
},
{
"epoch": 7.06,
"learning_rate": 0.00010430727301947202,
"loss": 0.2673,
"step": 316500
},
{
"epoch": 7.07,
"learning_rate": 0.00010385830296915104,
"loss": 0.2674,
"step": 317000
},
{
"epoch": 7.07,
"eval_loss": 0.2475864142179489,
"eval_runtime": 2.013,
"eval_samples_per_second": 1141.106,
"eval_steps_per_second": 17.884,
"step": 317000
},
{
"epoch": 7.08,
"learning_rate": 0.00010340989219623508,
"loss": 0.2674,
"step": 317500
},
{
"epoch": 7.09,
"learning_rate": 0.0001029620456044727,
"loss": 0.2672,
"step": 318000
},
{
"epoch": 7.09,
"eval_loss": 0.2495851069688797,
"eval_runtime": 2.0072,
"eval_samples_per_second": 1144.404,
"eval_steps_per_second": 17.936,
"step": 318000
},
{
"epoch": 7.1,
"learning_rate": 0.00010251476809144226,
"loss": 0.267,
"step": 318500
},
{
"epoch": 7.12,
"learning_rate": 0.00010206806454849917,
"loss": 0.2668,
"step": 319000
},
{
"epoch": 7.12,
"eval_loss": 0.24910998344421387,
"eval_runtime": 2.0019,
"eval_samples_per_second": 1147.411,
"eval_steps_per_second": 17.983,
"step": 319000
},
{
"epoch": 7.13,
"learning_rate": 0.00010162193986072167,
"loss": 0.267,
"step": 319500
},
{
"epoch": 7.14,
"learning_rate": 0.00010117639890685795,
"loss": 0.2671,
"step": 320000
},
{
"epoch": 7.14,
"eval_loss": 0.2508242130279541,
"eval_runtime": 1.9741,
"eval_samples_per_second": 1163.57,
"eval_steps_per_second": 18.236,
"step": 320000
},
{
"epoch": 7.15,
"learning_rate": 0.00010073144655927253,
"loss": 0.2668,
"step": 320500
},
{
"epoch": 7.16,
"learning_rate": 0.0001002870876838929,
"loss": 0.2666,
"step": 321000
},
{
"epoch": 7.16,
"eval_loss": 0.24993054568767548,
"eval_runtime": 2.0461,
"eval_samples_per_second": 1122.605,
"eval_steps_per_second": 17.594,
"step": 321000
},
{
"epoch": 7.17,
"learning_rate": 9.984332714015662e-05,
"loss": 0.2668,
"step": 321500
},
{
"epoch": 7.18,
"learning_rate": 9.94001697809578e-05,
"loss": 0.2666,
"step": 322000
},
{
"epoch": 7.18,
"eval_loss": 0.2504477798938751,
"eval_runtime": 1.9968,
"eval_samples_per_second": 1150.317,
"eval_steps_per_second": 18.028,
"step": 322000
},
{
"epoch": 7.19,
"learning_rate": 9.895762045259445e-05,
"loss": 0.2667,
"step": 322500
},
{
"epoch": 7.2,
"learning_rate": 9.851568399471498e-05,
"loss": 0.2667,
"step": 323000
},
{
"epoch": 7.2,
"eval_loss": 0.2511838674545288,
"eval_runtime": 1.9771,
"eval_samples_per_second": 1161.828,
"eval_steps_per_second": 18.209,
"step": 323000
},
{
"epoch": 7.22,
"learning_rate": 9.807436524026574e-05,
"loss": 0.2665,
"step": 323500
},
{
"epoch": 7.23,
"learning_rate": 9.763366901543801e-05,
"loss": 0.2665,
"step": 324000
},
{
"epoch": 7.23,
"eval_loss": 0.2511294484138489,
"eval_runtime": 2.0259,
"eval_samples_per_second": 1133.828,
"eval_steps_per_second": 17.77,
"step": 324000
},
{
"epoch": 7.24,
"learning_rate": 9.719360013961495e-05,
"loss": 0.2658,
"step": 324500
},
{
"epoch": 7.25,
"learning_rate": 9.675416342531944e-05,
"loss": 0.2661,
"step": 325000
},
{
"epoch": 7.25,
"eval_loss": 0.25169792771339417,
"eval_runtime": 1.9684,
"eval_samples_per_second": 1166.965,
"eval_steps_per_second": 18.289,
"step": 325000
},
{
"epoch": 7.26,
"learning_rate": 9.631536367816086e-05,
"loss": 0.2661,
"step": 325500
},
{
"epoch": 7.27,
"learning_rate": 9.587720569678299e-05,
"loss": 0.2661,
"step": 326000
},
{
"epoch": 7.27,
"eval_loss": 0.2492215633392334,
"eval_runtime": 1.9633,
"eval_samples_per_second": 1169.965,
"eval_steps_per_second": 18.336,
"step": 326000
},
{
"epoch": 7.28,
"learning_rate": 9.543969427281131e-05,
"loss": 0.2662,
"step": 326500
},
{
"epoch": 7.29,
"learning_rate": 9.500283419080062e-05,
"loss": 0.2665,
"step": 327000
},
{
"epoch": 7.29,
"eval_loss": 0.252391517162323,
"eval_runtime": 2.0026,
"eval_samples_per_second": 1147.012,
"eval_steps_per_second": 17.977,
"step": 327000
},
{
"epoch": 7.31,
"learning_rate": 9.45666302281829e-05,
"loss": 0.2657,
"step": 327500
},
{
"epoch": 7.32,
"learning_rate": 9.413108715521467e-05,
"loss": 0.266,
"step": 328000
},
{
"epoch": 7.32,
"eval_loss": 0.24788686633110046,
"eval_runtime": 1.9452,
"eval_samples_per_second": 1180.833,
"eval_steps_per_second": 18.507,
"step": 328000
},
{
"epoch": 7.33,
"learning_rate": 9.369620973492525e-05,
"loss": 0.2658,
"step": 328500
},
{
"epoch": 7.34,
"learning_rate": 9.326200272306445e-05,
"loss": 0.2654,
"step": 329000
},
{
"epoch": 7.34,
"eval_loss": 0.2492542862892151,
"eval_runtime": 2.0015,
"eval_samples_per_second": 1147.652,
"eval_steps_per_second": 17.987,
"step": 329000
},
{
"epoch": 7.35,
"learning_rate": 9.282847086805059e-05,
"loss": 0.2657,
"step": 329500
},
{
"epoch": 7.36,
"learning_rate": 9.239561891091853e-05,
"loss": 0.2656,
"step": 330000
},
{
"epoch": 7.36,
"eval_loss": 0.2490701973438263,
"eval_runtime": 2.0564,
"eval_samples_per_second": 1117.004,
"eval_steps_per_second": 17.506,
"step": 330000
},
{
"epoch": 7.37,
"learning_rate": 9.196345158526793e-05,
"loss": 0.2655,
"step": 330500
},
{
"epoch": 7.38,
"learning_rate": 9.153197361721149e-05,
"loss": 0.2653,
"step": 331000
},
{
"epoch": 7.38,
"eval_loss": 0.24651852250099182,
"eval_runtime": 1.9849,
"eval_samples_per_second": 1157.249,
"eval_steps_per_second": 18.137,
"step": 331000
},
{
"epoch": 7.39,
"learning_rate": 9.110118972532302e-05,
"loss": 0.2652,
"step": 331500
},
{
"epoch": 7.41,
"learning_rate": 9.067110462058634e-05,
"loss": 0.2656,
"step": 332000
},
{
"epoch": 7.41,
"eval_loss": 0.24983063340187073,
"eval_runtime": 1.9337,
"eval_samples_per_second": 1187.894,
"eval_steps_per_second": 18.617,
"step": 332000
},
{
"epoch": 7.42,
"learning_rate": 9.024172300634305e-05,
"loss": 0.2655,
"step": 332500
},
{
"epoch": 7.43,
"learning_rate": 8.981304957824182e-05,
"loss": 0.2655,
"step": 333000
},
{
"epoch": 7.43,
"eval_loss": 0.2502962350845337,
"eval_runtime": 2.0089,
"eval_samples_per_second": 1143.431,
"eval_steps_per_second": 17.921,
"step": 333000
},
{
"epoch": 7.44,
"learning_rate": 8.938508902418643e-05,
"loss": 0.2651,
"step": 333500
},
{
"epoch": 7.45,
"learning_rate": 8.89578460242851e-05,
"loss": 0.265,
"step": 334000
},
{
"epoch": 7.45,
"eval_loss": 0.2484452873468399,
"eval_runtime": 1.9509,
"eval_samples_per_second": 1177.421,
"eval_steps_per_second": 18.453,
"step": 334000
},
{
"epoch": 7.46,
"learning_rate": 8.85313252507988e-05,
"loss": 0.265,
"step": 334500
},
{
"epoch": 7.47,
"learning_rate": 8.810553136809027e-05,
"loss": 0.2651,
"step": 335000
},
{
"epoch": 7.47,
"eval_loss": 0.24745772778987885,
"eval_runtime": 2.0372,
"eval_samples_per_second": 1127.544,
"eval_steps_per_second": 17.672,
"step": 335000
},
{
"epoch": 7.48,
"learning_rate": 8.76804690325733e-05,
"loss": 0.2649,
"step": 335500
},
{
"epoch": 7.49,
"learning_rate": 8.725614289266137e-05,
"loss": 0.2647,
"step": 336000
},
{
"epoch": 7.49,
"eval_loss": 0.24857033789157867,
"eval_runtime": 2.0353,
"eval_samples_per_second": 1128.559,
"eval_steps_per_second": 17.687,
"step": 336000
},
{
"epoch": 7.51,
"learning_rate": 8.683255758871734e-05,
"loss": 0.2649,
"step": 336500
},
{
"epoch": 7.52,
"learning_rate": 8.640971775300207e-05,
"loss": 0.2654,
"step": 337000
},
{
"epoch": 7.52,
"eval_loss": 0.24552865326404572,
"eval_runtime": 2.0136,
"eval_samples_per_second": 1140.751,
"eval_steps_per_second": 17.879,
"step": 337000
},
{
"epoch": 7.53,
"learning_rate": 8.598762800962431e-05,
"loss": 0.2767,
"step": 337500
},
{
"epoch": 7.54,
"learning_rate": 8.55662929744899e-05,
"loss": 0.2655,
"step": 338000
},
{
"epoch": 7.54,
"eval_loss": 0.24788717925548553,
"eval_runtime": 2.0497,
"eval_samples_per_second": 1120.627,
"eval_steps_per_second": 17.563,
"step": 338000
},
{
"epoch": 7.55,
"learning_rate": 8.514571725525124e-05,
"loss": 0.2666,
"step": 338500
},
{
"epoch": 7.56,
"learning_rate": 8.47259054512571e-05,
"loss": 0.2678,
"step": 339000
},
{
"epoch": 7.56,
"eval_loss": 0.24549083411693573,
"eval_runtime": 2.0079,
"eval_samples_per_second": 1143.957,
"eval_steps_per_second": 17.929,
"step": 339000
},
{
"epoch": 7.57,
"learning_rate": 8.430686215350198e-05,
"loss": 0.265,
"step": 339500
},
{
"epoch": 7.58,
"learning_rate": 8.388859194457636e-05,
"loss": 0.265,
"step": 340000
},
{
"epoch": 7.58,
"eval_loss": 0.24893943965435028,
"eval_runtime": 2.0535,
"eval_samples_per_second": 1118.576,
"eval_steps_per_second": 17.531,
"step": 340000
},
{
"epoch": 7.6,
"learning_rate": 8.347109939861605e-05,
"loss": 0.2675,
"step": 340500
},
{
"epoch": 7.61,
"learning_rate": 8.305438908125285e-05,
"loss": 0.2649,
"step": 341000
},
{
"epoch": 7.61,
"eval_loss": 0.2475811243057251,
"eval_runtime": 2.0091,
"eval_samples_per_second": 1143.275,
"eval_steps_per_second": 17.918,
"step": 341000
},
{
"epoch": 7.62,
"learning_rate": 8.263846554956402e-05,
"loss": 0.2659,
"step": 341500
},
{
"epoch": 7.63,
"learning_rate": 8.222333335202254e-05,
"loss": 0.2649,
"step": 342000
},
{
"epoch": 7.63,
"eval_loss": 0.2466391623020172,
"eval_runtime": 2.0704,
"eval_samples_per_second": 1109.463,
"eval_steps_per_second": 17.388,
"step": 342000
},
{
"epoch": 7.64,
"learning_rate": 8.18089970284477e-05,
"loss": 0.2646,
"step": 342500
},
{
"epoch": 7.65,
"learning_rate": 8.1395461109955e-05,
"loss": 0.2645,
"step": 343000
},
{
"epoch": 7.65,
"eval_loss": 0.24718651175498962,
"eval_runtime": 2.0893,
"eval_samples_per_second": 1099.419,
"eval_steps_per_second": 17.231,
"step": 343000
},
{
"epoch": 7.66,
"learning_rate": 8.098273011890726e-05,
"loss": 0.2654,
"step": 343500
},
{
"epoch": 7.67,
"learning_rate": 8.057080856886426e-05,
"loss": 0.2676,
"step": 344000
},
{
"epoch": 7.67,
"eval_loss": 0.2473945915699005,
"eval_runtime": 1.9984,
"eval_samples_per_second": 1149.403,
"eval_steps_per_second": 18.014,
"step": 344000
},
{
"epoch": 7.68,
"learning_rate": 8.015970096453414e-05,
"loss": 0.2655,
"step": 344500
},
{
"epoch": 7.7,
"learning_rate": 7.974941180172382e-05,
"loss": 0.265,
"step": 345000
},
{
"epoch": 7.7,
"eval_loss": 0.24670176208019257,
"eval_runtime": 2.0516,
"eval_samples_per_second": 1119.641,
"eval_steps_per_second": 17.548,
"step": 345000
},
{
"epoch": 7.71,
"learning_rate": 7.933994556728976e-05,
"loss": 0.2648,
"step": 345500
},
{
"epoch": 7.72,
"learning_rate": 7.893130673908927e-05,
"loss": 0.2666,
"step": 346000
},
{
"epoch": 7.72,
"eval_loss": 0.24692493677139282,
"eval_runtime": 2.0097,
"eval_samples_per_second": 1142.931,
"eval_steps_per_second": 17.913,
"step": 346000
},
{
"epoch": 7.73,
"learning_rate": 7.852349978593091e-05,
"loss": 0.2645,
"step": 346500
},
{
"epoch": 7.74,
"learning_rate": 7.811652916752633e-05,
"loss": 0.2645,
"step": 347000
},
{
"epoch": 7.74,
"eval_loss": 0.24660193920135498,
"eval_runtime": 2.0554,
"eval_samples_per_second": 1117.529,
"eval_steps_per_second": 17.515,
"step": 347000
},
{
"epoch": 7.75,
"learning_rate": 7.771039933444092e-05,
"loss": 0.2644,
"step": 347500
},
{
"epoch": 7.76,
"learning_rate": 7.730511472804544e-05,
"loss": 0.2636,
"step": 348000
},
{
"epoch": 7.76,
"eval_loss": 0.248375803232193,
"eval_runtime": 1.9897,
"eval_samples_per_second": 1154.427,
"eval_steps_per_second": 18.093,
"step": 348000
},
{
"epoch": 7.77,
"learning_rate": 7.690067978046758e-05,
"loss": 0.2637,
"step": 348500
},
{
"epoch": 7.78,
"learning_rate": 7.649709891454298e-05,
"loss": 0.2632,
"step": 349000
},
{
"epoch": 7.78,
"eval_loss": 0.24848568439483643,
"eval_runtime": 2.0056,
"eval_samples_per_second": 1145.314,
"eval_steps_per_second": 17.95,
"step": 349000
},
{
"epoch": 7.8,
"learning_rate": 7.609437654376742e-05,
"loss": 0.2636,
"step": 349500
},
{
"epoch": 7.81,
"learning_rate": 7.569251707224812e-05,
"loss": 0.2635,
"step": 350000
},
{
"epoch": 7.81,
"eval_loss": 0.24438388645648956,
"eval_runtime": 2.0031,
"eval_samples_per_second": 1146.703,
"eval_steps_per_second": 17.972,
"step": 350000
},
{
"epoch": 7.82,
"learning_rate": 7.529152489465592e-05,
"loss": 0.2638,
"step": 350500
},
{
"epoch": 7.83,
"learning_rate": 7.489140439617708e-05,
"loss": 0.2632,
"step": 351000
},
{
"epoch": 7.83,
"eval_loss": 0.2466663420200348,
"eval_runtime": 1.9675,
"eval_samples_per_second": 1167.492,
"eval_steps_per_second": 18.298,
"step": 351000
},
{
"epoch": 7.84,
"learning_rate": 7.449215995246522e-05,
"loss": 0.263,
"step": 351500
},
{
"epoch": 7.85,
"learning_rate": 7.409379592959367e-05,
"loss": 0.2631,
"step": 352000
},
{
"epoch": 7.85,
"eval_loss": 0.24644367396831512,
"eval_runtime": 1.9998,
"eval_samples_per_second": 1148.612,
"eval_steps_per_second": 18.002,
"step": 352000
},
{
"epoch": 7.86,
"learning_rate": 7.369631668400746e-05,
"loss": 0.2632,
"step": 352500
},
{
"epoch": 7.87,
"learning_rate": 7.3299726562476e-05,
"loss": 0.2629,
"step": 353000
},
{
"epoch": 7.87,
"eval_loss": 0.24639040231704712,
"eval_runtime": 1.996,
"eval_samples_per_second": 1150.775,
"eval_steps_per_second": 18.036,
"step": 353000
},
{
"epoch": 7.88,
"learning_rate": 7.290402990204531e-05,
"loss": 0.2628,
"step": 353500
},
{
"epoch": 7.9,
"learning_rate": 7.250923102999073e-05,
"loss": 0.2629,
"step": 354000
},
{
"epoch": 7.9,
"eval_loss": 0.24618536233901978,
"eval_runtime": 1.9784,
"eval_samples_per_second": 1161.045,
"eval_steps_per_second": 18.197,
"step": 354000
},
{
"epoch": 7.91,
"learning_rate": 7.211533426376934e-05,
"loss": 0.2629,
"step": 354500
},
{
"epoch": 7.92,
"learning_rate": 7.172234391097317e-05,
"loss": 0.2625,
"step": 355000
},
{
"epoch": 7.92,
"eval_loss": 0.24593985080718994,
"eval_runtime": 2.059,
"eval_samples_per_second": 1115.596,
"eval_steps_per_second": 17.484,
"step": 355000
},
{
"epoch": 7.93,
"learning_rate": 7.133026426928173e-05,
"loss": 0.2626,
"step": 355500
},
{
"epoch": 7.94,
"learning_rate": 7.093909962641514e-05,
"loss": 0.2626,
"step": 356000
},
{
"epoch": 7.94,
"eval_loss": 0.24694356322288513,
"eval_runtime": 1.9849,
"eval_samples_per_second": 1157.222,
"eval_steps_per_second": 18.137,
"step": 356000
},
{
"epoch": 7.95,
"learning_rate": 7.054885426008737e-05,
"loss": 0.2624,
"step": 356500
},
{
"epoch": 7.96,
"learning_rate": 7.015953243795907e-05,
"loss": 0.2625,
"step": 357000
},
{
"epoch": 7.96,
"eval_loss": 0.24489082396030426,
"eval_runtime": 2.0439,
"eval_samples_per_second": 1123.842,
"eval_steps_per_second": 17.614,
"step": 357000
},
{
"epoch": 7.97,
"learning_rate": 6.97711384175914e-05,
"loss": 0.2623,
"step": 357500
},
{
"epoch": 7.99,
"learning_rate": 6.938367644639911e-05,
"loss": 0.4432,
"step": 358000
},
{
"epoch": 7.99,
"eval_loss": 0.6783205270767212,
"eval_runtime": 2.0172,
"eval_samples_per_second": 1138.683,
"eval_steps_per_second": 17.846,
"step": 358000
},
{
"epoch": 8.0,
"learning_rate": 6.899715076160425e-05,
"loss": 0.6785,
"step": 358500
},
{
"epoch": 8.01,
"learning_rate": 6.861156559018986e-05,
"loss": 0.6774,
"step": 359000
},
{
"epoch": 8.01,
"eval_loss": 0.6766601204872131,
"eval_runtime": 2.0317,
"eval_samples_per_second": 1130.554,
"eval_steps_per_second": 17.719,
"step": 359000
},
{
"epoch": 8.02,
"learning_rate": 6.822692514885346e-05,
"loss": 0.6773,
"step": 359500
},
{
"epoch": 8.03,
"learning_rate": 6.784323364396135e-05,
"loss": 0.6773,
"step": 360000
},
{
"epoch": 8.03,
"eval_loss": 0.6772929430007935,
"eval_runtime": 2.0141,
"eval_samples_per_second": 1140.463,
"eval_steps_per_second": 17.874,
"step": 360000
},
{
"epoch": 8.04,
"learning_rate": 6.746049527150238e-05,
"loss": 0.6774,
"step": 360500
},
{
"epoch": 8.05,
"learning_rate": 6.707871421704209e-05,
"loss": 0.4655,
"step": 361000
},
{
"epoch": 8.05,
"eval_loss": 0.2545197010040283,
"eval_runtime": 2.0031,
"eval_samples_per_second": 1146.74,
"eval_steps_per_second": 17.972,
"step": 361000
},
{
"epoch": 8.06,
"learning_rate": 6.669789465567683e-05,
"loss": 0.2697,
"step": 361500
},
{
"epoch": 8.07,
"learning_rate": 6.631804075198838e-05,
"loss": 0.2677,
"step": 362000
},
{
"epoch": 8.07,
"eval_loss": 0.2466827929019928,
"eval_runtime": 1.9817,
"eval_samples_per_second": 1159.082,
"eval_steps_per_second": 18.166,
"step": 362000
},
{
"epoch": 8.09,
"learning_rate": 6.593915665999816e-05,
"loss": 0.2646,
"step": 362500
},
{
"epoch": 8.1,
"learning_rate": 6.55612465231219e-05,
"loss": 0.2638,
"step": 363000
},
{
"epoch": 8.1,
"eval_loss": 0.24644041061401367,
"eval_runtime": 1.9701,
"eval_samples_per_second": 1165.949,
"eval_steps_per_second": 18.273,
"step": 363000
},
{
"epoch": 8.11,
"learning_rate": 6.518431447412434e-05,
"loss": 0.2633,
"step": 363500
},
{
"epoch": 8.12,
"learning_rate": 6.480836463507392e-05,
"loss": 0.263,
"step": 364000
},
{
"epoch": 8.12,
"eval_loss": 0.24640262126922607,
"eval_runtime": 1.9937,
"eval_samples_per_second": 1152.148,
"eval_steps_per_second": 18.057,
"step": 364000
},
{
"epoch": 8.13,
"learning_rate": 6.443340111729786e-05,
"loss": 0.2626,
"step": 364500
},
{
"epoch": 8.14,
"learning_rate": 6.405942802133713e-05,
"loss": 0.2624,
"step": 365000
},
{
"epoch": 8.14,
"eval_loss": 0.2466081976890564,
"eval_runtime": 1.9386,
"eval_samples_per_second": 1184.893,
"eval_steps_per_second": 18.57,
"step": 365000
},
{
"epoch": 8.15,
"learning_rate": 6.36864494369016e-05,
"loss": 0.2622,
"step": 365500
},
{
"epoch": 8.16,
"learning_rate": 6.331446944282534e-05,
"loss": 0.2615,
"step": 366000
},
{
"epoch": 8.16,
"eval_loss": 0.24654769897460938,
"eval_runtime": 1.9437,
"eval_samples_per_second": 1181.759,
"eval_steps_per_second": 18.521,
"step": 366000
},
{
"epoch": 8.17,
"learning_rate": 6.294349210702188e-05,
"loss": 0.2617,
"step": 366500
},
{
"epoch": 8.19,
"learning_rate": 6.257352148643998e-05,
"loss": 0.262,
"step": 367000
},
{
"epoch": 8.19,
"eval_loss": 0.2468617558479309,
"eval_runtime": 1.8987,
"eval_samples_per_second": 1209.803,
"eval_steps_per_second": 18.961,
"step": 367000
},
{
"epoch": 8.2,
"learning_rate": 6.220456162701908e-05,
"loss": 0.2617,
"step": 367500
},
{
"epoch": 8.21,
"learning_rate": 6.183661656364515e-05,
"loss": 0.2615,
"step": 368000
},
{
"epoch": 8.21,
"eval_loss": 0.24478143453598022,
"eval_runtime": 1.9857,
"eval_samples_per_second": 1156.78,
"eval_steps_per_second": 18.13,
"step": 368000
},
{
"epoch": 8.22,
"learning_rate": 6.146969032010631e-05,
"loss": 0.2616,
"step": 368500
},
{
"epoch": 8.23,
"learning_rate": 6.110378690904928e-05,
"loss": 0.2617,
"step": 369000
},
{
"epoch": 8.23,
"eval_loss": 0.2461312711238861,
"eval_runtime": 2.0304,
"eval_samples_per_second": 1131.313,
"eval_steps_per_second": 17.731,
"step": 369000
},
{
"epoch": 8.24,
"learning_rate": 6.073891033193507e-05,
"loss": 0.2613,
"step": 369500
},
{
"epoch": 8.25,
"learning_rate": 6.037506457899553e-05,
"loss": 0.2611,
"step": 370000
},
{
"epoch": 8.25,
"eval_loss": 0.24196158349514008,
"eval_runtime": 1.9951,
"eval_samples_per_second": 1151.317,
"eval_steps_per_second": 18.044,
"step": 370000
},
{
"epoch": 8.26,
"learning_rate": 6.0012253629189544e-05,
"loss": 0.261,
"step": 370500
},
{
"epoch": 8.28,
"learning_rate": 5.965048145015944e-05,
"loss": 0.2611,
"step": 371000
},
{
"epoch": 8.28,
"eval_loss": 0.24591800570487976,
"eval_runtime": 1.9624,
"eval_samples_per_second": 1170.514,
"eval_steps_per_second": 18.345,
"step": 371000
},
{
"epoch": 8.29,
"learning_rate": 5.928975199818785e-05,
"loss": 0.2611,
"step": 371500
},
{
"epoch": 8.3,
"learning_rate": 5.893006921815428e-05,
"loss": 0.2608,
"step": 372000
},
{
"epoch": 8.3,
"eval_loss": 0.2432386726140976,
"eval_runtime": 2.0093,
"eval_samples_per_second": 1143.209,
"eval_steps_per_second": 17.917,
"step": 372000
},
{
"epoch": 8.31,
"learning_rate": 5.857143704349198e-05,
"loss": 0.2608,
"step": 372500
},
{
"epoch": 8.32,
"learning_rate": 5.8213859396144986e-05,
"loss": 0.2605,
"step": 373000
},
{
"epoch": 8.32,
"eval_loss": 0.2445555180311203,
"eval_runtime": 1.9525,
"eval_samples_per_second": 1176.451,
"eval_steps_per_second": 18.438,
"step": 373000
},
{
"epoch": 8.33,
"learning_rate": 5.785734018652507e-05,
"loss": 0.2609,
"step": 373500
},
{
"epoch": 8.34,
"learning_rate": 5.750188331346927e-05,
"loss": 0.2609,
"step": 374000
},
{
"epoch": 8.34,
"eval_loss": 0.24143685400485992,
"eval_runtime": 1.9945,
"eval_samples_per_second": 1151.665,
"eval_steps_per_second": 18.05,
"step": 374000
},
{
"epoch": 8.35,
"learning_rate": 5.714749266419695e-05,
"loss": 0.2605,
"step": 374500
},
{
"epoch": 8.36,
"learning_rate": 5.6794172114267566e-05,
"loss": 0.2614,
"step": 375000
},
{
"epoch": 8.36,
"eval_loss": 0.2436528205871582,
"eval_runtime": 1.9628,
"eval_samples_per_second": 1170.242,
"eval_steps_per_second": 18.341,
"step": 375000
},
{
"epoch": 8.38,
"learning_rate": 5.6441925527537914e-05,
"loss": 0.2614,
"step": 375500
},
{
"epoch": 8.39,
"learning_rate": 5.60907567561203e-05,
"loss": 0.2624,
"step": 376000
},
{
"epoch": 8.39,
"eval_loss": 0.24615313112735748,
"eval_runtime": 1.9915,
"eval_samples_per_second": 1153.415,
"eval_steps_per_second": 18.077,
"step": 376000
},
{
"epoch": 8.4,
"learning_rate": 5.574066964034012e-05,
"loss": 0.2614,
"step": 376500
},
{
"epoch": 8.41,
"learning_rate": 5.539166800869402e-05,
"loss": 0.2611,
"step": 377000
},
{
"epoch": 8.41,
"eval_loss": 0.24158482253551483,
"eval_runtime": 1.9896,
"eval_samples_per_second": 1154.526,
"eval_steps_per_second": 18.094,
"step": 377000
},
{
"epoch": 8.42,
"learning_rate": 5.5043755677807955e-05,
"loss": 0.261,
"step": 377500
},
{
"epoch": 8.43,
"learning_rate": 5.4696936452395344e-05,
"loss": 0.2604,
"step": 378000
},
{
"epoch": 8.43,
"eval_loss": 0.2432401180267334,
"eval_runtime": 1.9707,
"eval_samples_per_second": 1165.6,
"eval_steps_per_second": 18.268,
"step": 378000
},
{
"epoch": 8.44,
"learning_rate": 5.435121412521576e-05,
"loss": 0.2604,
"step": 378500
},
{
"epoch": 8.45,
"learning_rate": 5.400659247703307e-05,
"loss": 0.2605,
"step": 379000
},
{
"epoch": 8.45,
"eval_loss": 0.24426017701625824,
"eval_runtime": 2.017,
"eval_samples_per_second": 1138.839,
"eval_steps_per_second": 17.849,
"step": 379000
},
{
"epoch": 8.46,
"learning_rate": 5.36630752765745e-05,
"loss": 0.2605,
"step": 379500
},
{
"epoch": 8.48,
"learning_rate": 5.3320666280489146e-05,
"loss": 0.26,
"step": 380000
},
{
"epoch": 8.48,
"eval_loss": 0.24248941242694855,
"eval_runtime": 2.0075,
"eval_samples_per_second": 1144.188,
"eval_steps_per_second": 17.932,
"step": 380000
},
{
"epoch": 8.49,
"learning_rate": 5.2979369233306834e-05,
"loss": 0.2597,
"step": 380500
},
{
"epoch": 8.5,
"learning_rate": 5.26391878673975e-05,
"loss": 0.2598,
"step": 381000
},
{
"epoch": 8.5,
"eval_loss": 0.24129272997379303,
"eval_runtime": 1.9892,
"eval_samples_per_second": 1154.716,
"eval_steps_per_second": 18.097,
"step": 381000
},
{
"epoch": 8.51,
"learning_rate": 5.230012590292987e-05,
"loss": 0.26,
"step": 381500
},
{
"epoch": 8.52,
"learning_rate": 5.1962187047831517e-05,
"loss": 0.2597,
"step": 382000
},
{
"epoch": 8.52,
"eval_loss": 0.2438046783208847,
"eval_runtime": 2.0783,
"eval_samples_per_second": 1105.244,
"eval_steps_per_second": 17.322,
"step": 382000
},
{
"epoch": 8.53,
"learning_rate": 5.162537499774743e-05,
"loss": 0.2603,
"step": 382500
},
{
"epoch": 8.54,
"learning_rate": 5.128969343600032e-05,
"loss": 0.2602,
"step": 383000
},
{
"epoch": 8.54,
"eval_loss": 0.24265888333320618,
"eval_runtime": 1.9571,
"eval_samples_per_second": 1173.666,
"eval_steps_per_second": 18.394,
"step": 383000
},
{
"epoch": 8.55,
"learning_rate": 5.09551460335499e-05,
"loss": 0.2599,
"step": 383500
},
{
"epoch": 8.57,
"learning_rate": 5.062173644895296e-05,
"loss": 0.2599,
"step": 384000
},
{
"epoch": 8.57,
"eval_loss": 0.24097691476345062,
"eval_runtime": 1.9672,
"eval_samples_per_second": 1167.657,
"eval_steps_per_second": 18.3,
"step": 384000
},
{
"epoch": 8.58,
"learning_rate": 5.0289468328323434e-05,
"loss": 0.2591,
"step": 384500
},
{
"epoch": 8.59,
"learning_rate": 4.995834530529208e-05,
"loss": 0.2596,
"step": 385000
},
{
"epoch": 8.59,
"eval_loss": 0.24366655945777893,
"eval_runtime": 2.0603,
"eval_samples_per_second": 1114.876,
"eval_steps_per_second": 17.473,
"step": 385000
},
{
"epoch": 8.6,
"learning_rate": 4.9628371000967394e-05,
"loss": 0.2628,
"step": 385500
},
{
"epoch": 8.61,
"learning_rate": 4.929954902389534e-05,
"loss": 0.2605,
"step": 386000
},
{
"epoch": 8.61,
"eval_loss": 0.2400692254304886,
"eval_runtime": 2.0808,
"eval_samples_per_second": 1103.885,
"eval_steps_per_second": 17.301,
"step": 386000
},
{
"epoch": 8.62,
"learning_rate": 4.897188297002046e-05,
"loss": 0.2606,
"step": 386500
},
{
"epoch": 8.63,
"learning_rate": 4.8645376422646226e-05,
"loss": 0.2595,
"step": 387000
},
{
"epoch": 8.63,
"eval_loss": 0.24330011010169983,
"eval_runtime": 2.0153,
"eval_samples_per_second": 1139.773,
"eval_steps_per_second": 17.863,
"step": 387000
},
{
"epoch": 8.64,
"learning_rate": 4.832003295239591e-05,
"loss": 0.2601,
"step": 387500
},
{
"epoch": 8.65,
"learning_rate": 4.7995856117173624e-05,
"loss": 0.2597,
"step": 388000
},
{
"epoch": 8.65,
"eval_loss": 0.24292755126953125,
"eval_runtime": 1.9613,
"eval_samples_per_second": 1171.185,
"eval_steps_per_second": 18.356,
"step": 388000
},
{
"epoch": 8.67,
"learning_rate": 4.767284946212521e-05,
"loss": 0.2604,
"step": 388500
},
{
"epoch": 8.68,
"learning_rate": 4.735101651959977e-05,
"loss": 0.2598,
"step": 389000
},
{
"epoch": 8.68,
"eval_loss": 0.24331249296665192,
"eval_runtime": 1.9683,
"eval_samples_per_second": 1166.976,
"eval_steps_per_second": 18.29,
"step": 389000
},
{
"epoch": 8.69,
"learning_rate": 4.7030360809110754e-05,
"loss": 0.2593,
"step": 389500
},
{
"epoch": 8.7,
"learning_rate": 4.6710885837297726e-05,
"loss": 0.2594,
"step": 390000
},
{
"epoch": 8.7,
"eval_loss": 0.24239999055862427,
"eval_runtime": 1.9699,
"eval_samples_per_second": 1166.029,
"eval_steps_per_second": 18.275,
"step": 390000
},
{
"epoch": 8.71,
"learning_rate": 4.639259509788768e-05,
"loss": 0.26,
"step": 390500
},
{
"epoch": 8.72,
"learning_rate": 4.60754920716572e-05,
"loss": 0.259,
"step": 391000
},
{
"epoch": 8.72,
"eval_loss": 0.2440621256828308,
"eval_runtime": 2.0127,
"eval_samples_per_second": 1141.226,
"eval_steps_per_second": 17.886,
"step": 391000
},
{
"epoch": 8.73,
"learning_rate": 4.5759580226394167e-05,
"loss": 0.2593,
"step": 391500
},
{
"epoch": 8.74,
"learning_rate": 4.544486301685993e-05,
"loss": 0.2594,
"step": 392000
},
{
"epoch": 8.74,
"eval_loss": 0.23992255330085754,
"eval_runtime": 1.971,
"eval_samples_per_second": 1165.424,
"eval_steps_per_second": 18.265,
"step": 392000
},
{
"epoch": 8.75,
"learning_rate": 4.5131343884751484e-05,
"loss": 0.2591,
"step": 392500
},
{
"epoch": 8.77,
"learning_rate": 4.4819026258663774e-05,
"loss": 0.2591,
"step": 393000
},
{
"epoch": 8.77,
"eval_loss": 0.2458277940750122,
"eval_runtime": 1.9966,
"eval_samples_per_second": 1150.457,
"eval_steps_per_second": 18.031,
"step": 393000
},
{
"epoch": 8.78,
"learning_rate": 4.450791355405234e-05,
"loss": 0.2592,
"step": 393500
},
{
"epoch": 8.79,
"learning_rate": 4.419800917319588e-05,
"loss": 0.2591,
"step": 394000
},
{
"epoch": 8.79,
"eval_loss": 0.2452658861875534,
"eval_runtime": 1.9635,
"eval_samples_per_second": 1169.821,
"eval_steps_per_second": 18.334,
"step": 394000
},
{
"epoch": 8.8,
"learning_rate": 4.3889316505159056e-05,
"loss": 0.2601,
"step": 394500
},
{
"epoch": 8.81,
"learning_rate": 4.3581838925755465e-05,
"loss": 0.2591,
"step": 395000
},
{
"epoch": 8.81,
"eval_loss": 0.24037905037403107,
"eval_runtime": 2.0778,
"eval_samples_per_second": 1105.5,
"eval_steps_per_second": 17.326,
"step": 395000
},
{
"epoch": 8.82,
"learning_rate": 4.327557979751057e-05,
"loss": 0.2589,
"step": 395500
},
{
"epoch": 8.83,
"learning_rate": 4.297054246962517e-05,
"loss": 0.259,
"step": 396000
},
{
"epoch": 8.83,
"eval_loss": 0.2426002323627472,
"eval_runtime": 1.9871,
"eval_samples_per_second": 1155.977,
"eval_steps_per_second": 18.117,
"step": 396000
},
{
"epoch": 8.84,
"learning_rate": 4.266673027793864e-05,
"loss": 0.2583,
"step": 396500
},
{
"epoch": 8.86,
"learning_rate": 4.236414654489242e-05,
"loss": 0.2583,
"step": 397000
},
{
"epoch": 8.86,
"eval_loss": 0.24136394262313843,
"eval_runtime": 2.043,
"eval_samples_per_second": 1124.309,
"eval_steps_per_second": 17.621,
"step": 397000
},
{
"epoch": 8.87,
"learning_rate": 4.206279457949371e-05,
"loss": 0.2581,
"step": 397500
},
{
"epoch": 8.88,
"learning_rate": 4.1762677677279335e-05,
"loss": 0.2584,
"step": 398000
},
{
"epoch": 8.88,
"eval_loss": 0.24172720313072205,
"eval_runtime": 1.9971,
"eval_samples_per_second": 1150.177,
"eval_steps_per_second": 18.026,
"step": 398000
},
{
"epoch": 8.89,
"learning_rate": 4.146379912027964e-05,
"loss": 0.2582,
"step": 398500
},
{
"epoch": 8.9,
"learning_rate": 4.1166162176982664e-05,
"loss": 0.2584,
"step": 399000
},
{
"epoch": 8.9,
"eval_loss": 0.2435595542192459,
"eval_runtime": 2.0606,
"eval_samples_per_second": 1114.743,
"eval_steps_per_second": 17.471,
"step": 399000
},
{
"epoch": 8.91,
"learning_rate": 4.086977010229838e-05,
"loss": 0.2583,
"step": 399500
},
{
"epoch": 8.92,
"learning_rate": 4.057462613752294e-05,
"loss": 0.2615,
"step": 400000
},
{
"epoch": 8.92,
"eval_loss": 0.24273519217967987,
"eval_runtime": 1.9831,
"eval_samples_per_second": 1158.313,
"eval_steps_per_second": 18.154,
"step": 400000
},
{
"epoch": 8.93,
"learning_rate": 4.0280733510303475e-05,
"loss": 0.2627,
"step": 400500
},
{
"epoch": 8.94,
"learning_rate": 3.9988095434602716e-05,
"loss": 0.2591,
"step": 401000
},
{
"epoch": 8.94,
"eval_loss": 0.24216598272323608,
"eval_runtime": 1.9765,
"eval_samples_per_second": 1162.139,
"eval_steps_per_second": 18.214,
"step": 401000
},
{
"epoch": 8.96,
"learning_rate": 3.9696715110663726e-05,
"loss": 0.2588,
"step": 401500
},
{
"epoch": 8.97,
"learning_rate": 3.9406595724975116e-05,
"loss": 0.2585,
"step": 402000
},
{
"epoch": 8.97,
"eval_loss": 0.245298832654953,
"eval_runtime": 2.023,
"eval_samples_per_second": 1135.425,
"eval_steps_per_second": 17.795,
"step": 402000
},
{
"epoch": 8.98,
"learning_rate": 3.9117740450235914e-05,
"loss": 0.2586,
"step": 402500
},
{
"epoch": 8.99,
"learning_rate": 3.8830152445321163e-05,
"loss": 0.258,
"step": 403000
},
{
"epoch": 8.99,
"eval_loss": 0.24233770370483398,
"eval_runtime": 2.0488,
"eval_samples_per_second": 1121.145,
"eval_steps_per_second": 17.571,
"step": 403000
},
{
"epoch": 9.0,
"learning_rate": 3.854383485524724e-05,
"loss": 0.2583,
"step": 403500
},
{
"epoch": 9.01,
"learning_rate": 3.8258790811137425e-05,
"loss": 0.2577,
"step": 404000
},
{
"epoch": 9.01,
"eval_loss": 0.24036923050880432,
"eval_runtime": 2.0024,
"eval_samples_per_second": 1147.117,
"eval_steps_per_second": 17.978,
"step": 404000
},
{
"epoch": 9.02,
"learning_rate": 3.7975023430187676e-05,
"loss": 0.2581,
"step": 404500
},
{
"epoch": 9.03,
"learning_rate": 3.7692535815632624e-05,
"loss": 0.2578,
"step": 405000
},
{
"epoch": 9.03,
"eval_loss": 0.24242642521858215,
"eval_runtime": 2.0428,
"eval_samples_per_second": 1124.43,
"eval_steps_per_second": 17.623,
"step": 405000
},
{
"epoch": 9.04,
"learning_rate": 3.741133105671159e-05,
"loss": 0.2575,
"step": 405500
},
{
"epoch": 9.06,
"learning_rate": 3.713141222863474e-05,
"loss": 0.2576,
"step": 406000
},
{
"epoch": 9.06,
"eval_loss": 0.2440737634897232,
"eval_runtime": 2.0157,
"eval_samples_per_second": 1139.547,
"eval_steps_per_second": 17.86,
"step": 406000
},
{
"epoch": 9.07,
"learning_rate": 3.6852782392549584e-05,
"loss": 0.2575,
"step": 406500
},
{
"epoch": 9.08,
"learning_rate": 3.657544459550729e-05,
"loss": 0.2574,
"step": 407000
},
{
"epoch": 9.08,
"eval_loss": 0.2409050464630127,
"eval_runtime": 2.0313,
"eval_samples_per_second": 1130.817,
"eval_steps_per_second": 17.723,
"step": 407000
},
{
"epoch": 9.09,
"learning_rate": 3.6299401870429606e-05,
"loss": 0.2574,
"step": 407500
},
{
"epoch": 9.1,
"learning_rate": 3.6024657236075546e-05,
"loss": 0.2573,
"step": 408000
},
{
"epoch": 9.1,
"eval_loss": 0.24264054000377655,
"eval_runtime": 2.0373,
"eval_samples_per_second": 1127.471,
"eval_steps_per_second": 17.67,
"step": 408000
},
{
"epoch": 9.11,
"learning_rate": 3.575121369700841e-05,
"loss": 0.2576,
"step": 408500
},
{
"epoch": 9.12,
"learning_rate": 3.5479074243562995e-05,
"loss": 0.2576,
"step": 409000
},
{
"epoch": 9.12,
"eval_loss": 0.24062800407409668,
"eval_runtime": 2.0012,
"eval_samples_per_second": 1147.804,
"eval_steps_per_second": 17.989,
"step": 409000
},
{
"epoch": 9.13,
"learning_rate": 3.5208241851812644e-05,
"loss": 0.2572,
"step": 409500
},
{
"epoch": 9.15,
"learning_rate": 3.493871948353709e-05,
"loss": 0.2574,
"step": 410000
},
{
"epoch": 9.15,
"eval_loss": 0.23946772515773773,
"eval_runtime": 2.1141,
"eval_samples_per_second": 1086.515,
"eval_steps_per_second": 17.029,
"step": 410000
},
{
"epoch": 9.16,
"learning_rate": 3.4670510086189736e-05,
"loss": 0.2575,
"step": 410500
},
{
"epoch": 9.17,
"learning_rate": 3.440361659286563e-05,
"loss": 0.2571,
"step": 411000
},
{
"epoch": 9.17,
"eval_loss": 0.23894140124320984,
"eval_runtime": 2.0061,
"eval_samples_per_second": 1145.029,
"eval_steps_per_second": 17.946,
"step": 411000
},
{
"epoch": 9.18,
"learning_rate": 3.413804192226918e-05,
"loss": 0.2573,
"step": 411500
},
{
"epoch": 9.19,
"learning_rate": 3.387378897868246e-05,
"loss": 0.257,
"step": 412000
},
{
"epoch": 9.19,
"eval_loss": 0.23886139690876007,
"eval_runtime": 2.0352,
"eval_samples_per_second": 1128.628,
"eval_steps_per_second": 17.689,
"step": 412000
},
{
"epoch": 9.2,
"learning_rate": 3.361086065193336e-05,
"loss": 0.257,
"step": 412500
},
{
"epoch": 9.21,
"learning_rate": 3.334925981736389e-05,
"loss": 0.257,
"step": 413000
},
{
"epoch": 9.21,
"eval_loss": 0.23928523063659668,
"eval_runtime": 2.0712,
"eval_samples_per_second": 1109.008,
"eval_steps_per_second": 17.381,
"step": 413000
},
{
"epoch": 9.22,
"learning_rate": 3.3088989335798925e-05,
"loss": 0.2572,
"step": 413500
},
{
"epoch": 9.23,
"learning_rate": 3.283005205351467e-05,
"loss": 0.2568,
"step": 414000
},
{
"epoch": 9.23,
"eval_loss": 0.24012628197669983,
"eval_runtime": 1.9771,
"eval_samples_per_second": 1161.82,
"eval_steps_per_second": 18.209,
"step": 414000
},
{
"epoch": 9.25,
"learning_rate": 3.2572450802207845e-05,
"loss": 0.2567,
"step": 414500
},
{
"epoch": 9.26,
"learning_rate": 3.2316188398964344e-05,
"loss": 0.2568,
"step": 415000
},
{
"epoch": 9.26,
"eval_loss": 0.24090546369552612,
"eval_runtime": 1.9932,
"eval_samples_per_second": 1152.439,
"eval_steps_per_second": 18.062,
"step": 415000
},
{
"epoch": 9.27,
"learning_rate": 3.206126764622888e-05,
"loss": 0.2571,
"step": 415500
},
{
"epoch": 9.28,
"learning_rate": 3.180769133177392e-05,
"loss": 0.2567,
"step": 416000
},
{
"epoch": 9.28,
"eval_loss": 0.24101045727729797,
"eval_runtime": 2.0158,
"eval_samples_per_second": 1139.515,
"eval_steps_per_second": 17.859,
"step": 416000
},
{
"epoch": 9.29,
"learning_rate": 3.155546222866939e-05,
"loss": 0.2567,
"step": 416500
},
{
"epoch": 9.3,
"learning_rate": 3.130458309525239e-05,
"loss": 0.2564,
"step": 417000
},
{
"epoch": 9.3,
"eval_loss": 0.2400181144475937,
"eval_runtime": 2.0304,
"eval_samples_per_second": 1131.292,
"eval_steps_per_second": 17.73,
"step": 417000
},
{
"epoch": 9.31,
"learning_rate": 3.1055056675096826e-05,
"loss": 0.2567,
"step": 417500
},
{
"epoch": 9.32,
"learning_rate": 3.0806885696983816e-05,
"loss": 0.2563,
"step": 418000
},
{
"epoch": 9.32,
"eval_loss": 0.24112293124198914,
"eval_runtime": 2.0404,
"eval_samples_per_second": 1125.741,
"eval_steps_per_second": 17.643,
"step": 418000
},
{
"epoch": 9.33,
"learning_rate": 3.056007287487128e-05,
"loss": 0.2564,
"step": 418500
},
{
"epoch": 9.35,
"learning_rate": 3.0314620907864744e-05,
"loss": 0.2561,
"step": 419000
},
{
"epoch": 9.35,
"eval_loss": 0.2410934865474701,
"eval_runtime": 2.0663,
"eval_samples_per_second": 1111.669,
"eval_steps_per_second": 17.423,
"step": 419000
},
{
"epoch": 9.36,
"learning_rate": 3.0070532480187637e-05,
"loss": 0.2564,
"step": 419500
},
{
"epoch": 9.37,
"learning_rate": 2.9827810261151784e-05,
"loss": 0.2562,
"step": 420000
},
{
"epoch": 9.37,
"eval_loss": 0.23920150101184845,
"eval_runtime": 2.0818,
"eval_samples_per_second": 1103.391,
"eval_steps_per_second": 17.293,
"step": 420000
},
{
"epoch": 9.38,
"learning_rate": 2.9586456905128618e-05,
"loss": 0.2562,
"step": 420500
},
{
"epoch": 9.39,
"learning_rate": 2.9346475051519687e-05,
"loss": 0.2583,
"step": 421000
},
{
"epoch": 9.39,
"eval_loss": 0.2419823408126831,
"eval_runtime": 2.1088,
"eval_samples_per_second": 1089.249,
"eval_steps_per_second": 17.071,
"step": 421000
},
{
"epoch": 9.4,
"learning_rate": 2.910786732472815e-05,
"loss": 0.257,
"step": 421500
},
{
"epoch": 9.41,
"learning_rate": 2.887063633412981e-05,
"loss": 0.2565,
"step": 422000
},
{
"epoch": 9.41,
"eval_loss": 0.240878626704216,
"eval_runtime": 2.008,
"eval_samples_per_second": 1143.9,
"eval_steps_per_second": 17.928,
"step": 422000
},
{
"epoch": 9.42,
"learning_rate": 2.863478467404478e-05,
"loss": 0.2563,
"step": 422500
},
{
"epoch": 9.44,
"learning_rate": 2.8400314923709112e-05,
"loss": 0.2562,
"step": 423000
},
{
"epoch": 9.44,
"eval_loss": 0.2374911606311798,
"eval_runtime": 2.1002,
"eval_samples_per_second": 1093.715,
"eval_steps_per_second": 17.141,
"step": 423000
},
{
"epoch": 9.45,
"learning_rate": 2.816722964724636e-05,
"loss": 0.256,
"step": 423500
},
{
"epoch": 9.46,
"learning_rate": 2.793553139363981e-05,
"loss": 0.2556,
"step": 424000
},
{
"epoch": 9.46,
"eval_loss": 0.2397317737340927,
"eval_runtime": 2.1055,
"eval_samples_per_second": 1090.975,
"eval_steps_per_second": 17.098,
"step": 424000
},
{
"epoch": 9.47,
"learning_rate": 2.7705222696704366e-05,
"loss": 0.256,
"step": 424500
},
{
"epoch": 9.48,
"learning_rate": 2.7476306075059096e-05,
"loss": 0.2562,
"step": 425000
},
{
"epoch": 9.48,
"eval_loss": 0.23977774381637573,
"eval_runtime": 2.06,
"eval_samples_per_second": 1115.059,
"eval_steps_per_second": 17.476,
"step": 425000
},
{
"epoch": 9.49,
"learning_rate": 2.7248784032099478e-05,
"loss": 0.2574,
"step": 425500
},
{
"epoch": 9.5,
"learning_rate": 2.7022659055970144e-05,
"loss": 0.2584,
"step": 426000
},
{
"epoch": 9.5,
"eval_loss": 0.2388191670179367,
"eval_runtime": 2.1089,
"eval_samples_per_second": 1089.195,
"eval_steps_per_second": 17.071,
"step": 426000
},
{
"epoch": 9.51,
"learning_rate": 2.6797933619537604e-05,
"loss": 0.2572,
"step": 426500
},
{
"epoch": 9.52,
"learning_rate": 2.6574610180363166e-05,
"loss": 0.2566,
"step": 427000
},
{
"epoch": 9.52,
"eval_loss": 0.24121782183647156,
"eval_runtime": 2.0738,
"eval_samples_per_second": 1107.617,
"eval_steps_per_second": 17.359,
"step": 427000
},
{
"epoch": 9.54,
"learning_rate": 2.6352691180676286e-05,
"loss": 0.2568,
"step": 427500
},
{
"epoch": 9.55,
"learning_rate": 2.6132179047347505e-05,
"loss": 0.256,
"step": 428000
},
{
"epoch": 9.55,
"eval_loss": 0.24006181955337524,
"eval_runtime": 1.9904,
"eval_samples_per_second": 1154.041,
"eval_steps_per_second": 18.087,
"step": 428000
},
{
"epoch": 9.56,
"learning_rate": 2.5913076191862238e-05,
"loss": 0.2564,
"step": 428500
},
{
"epoch": 9.57,
"learning_rate": 2.5695385010294165e-05,
"loss": 0.2564,
"step": 429000
},
{
"epoch": 9.57,
"eval_loss": 0.24071797728538513,
"eval_runtime": 1.9595,
"eval_samples_per_second": 1172.235,
"eval_steps_per_second": 18.372,
"step": 429000
},
{
"epoch": 9.58,
"learning_rate": 2.5479107883279144e-05,
"loss": 0.2564,
"step": 429500
},
{
"epoch": 9.59,
"learning_rate": 2.5264247175989292e-05,
"loss": 0.2564,
"step": 430000
},
{
"epoch": 9.59,
"eval_loss": 0.23974178731441498,
"eval_runtime": 2.0589,
"eval_samples_per_second": 1115.617,
"eval_steps_per_second": 17.485,
"step": 430000
},
{
"epoch": 9.6,
"learning_rate": 2.5050805238106804e-05,
"loss": 0.2561,
"step": 430500
},
{
"epoch": 9.61,
"learning_rate": 2.4838784403798542e-05,
"loss": 0.256,
"step": 431000
},
{
"epoch": 9.61,
"eval_loss": 0.23836444318294525,
"eval_runtime": 2.1211,
"eval_samples_per_second": 1082.938,
"eval_steps_per_second": 16.972,
"step": 431000
},
{
"epoch": 9.62,
"learning_rate": 2.4628186991690346e-05,
"loss": 0.256,
"step": 431500
},
{
"epoch": 9.64,
"learning_rate": 2.4419015304841797e-05,
"loss": 0.2562,
"step": 432000
},
{
"epoch": 9.64,
"eval_loss": 0.2407396286725998,
"eval_runtime": 2.084,
"eval_samples_per_second": 1102.19,
"eval_steps_per_second": 17.274,
"step": 432000
},
{
"epoch": 9.65,
"learning_rate": 2.4211271630720957e-05,
"loss": 0.2559,
"step": 432500
},
{
"epoch": 9.66,
"learning_rate": 2.4004958241179347e-05,
"loss": 0.2558,
"step": 433000
},
{
"epoch": 9.66,
"eval_loss": 0.23972494900226593,
"eval_runtime": 2.0364,
"eval_samples_per_second": 1127.958,
"eval_steps_per_second": 17.678,
"step": 433000
},
{
"epoch": 9.67,
"learning_rate": 2.3800077392427193e-05,
"loss": 0.2559,
"step": 433500
},
{
"epoch": 9.68,
"learning_rate": 2.3596631325008536e-05,
"loss": 0.256,
"step": 434000
},
{
"epoch": 9.68,
"eval_loss": 0.24173137545585632,
"eval_runtime": 2.0453,
"eval_samples_per_second": 1123.05,
"eval_steps_per_second": 17.601,
"step": 434000
},
{
"epoch": 9.69,
"learning_rate": 2.3394622263777042e-05,
"loss": 0.2566,
"step": 434500
},
{
"epoch": 9.7,
"learning_rate": 2.3194052417871433e-05,
"loss": 0.2558,
"step": 435000
},
{
"epoch": 9.7,
"eval_loss": 0.23999714851379395,
"eval_runtime": 2.0454,
"eval_samples_per_second": 1123.023,
"eval_steps_per_second": 17.601,
"step": 435000
},
{
"epoch": 9.71,
"learning_rate": 2.2994923980691425e-05,
"loss": 0.2556,
"step": 435500
},
{
"epoch": 9.73,
"learning_rate": 2.279723912987365e-05,
"loss": 0.2552,
"step": 436000
},
{
"epoch": 9.73,
"eval_loss": 0.23865634202957153,
"eval_runtime": 2.0599,
"eval_samples_per_second": 1115.091,
"eval_steps_per_second": 17.476,
"step": 436000
},
{
"epoch": 9.74,
"learning_rate": 2.2601000027268006e-05,
"loss": 0.2555,
"step": 436500
},
{
"epoch": 9.75,
"learning_rate": 2.2406208818913857e-05,
"loss": 0.2556,
"step": 437000
},
{
"epoch": 9.75,
"eval_loss": 0.2411411553621292,
"eval_runtime": 2.0137,
"eval_samples_per_second": 1140.659,
"eval_steps_per_second": 17.877,
"step": 437000
},
{
"epoch": 9.76,
"learning_rate": 2.221286763501666e-05,
"loss": 0.2571,
"step": 437500
},
{
"epoch": 9.77,
"learning_rate": 2.2020978589924673e-05,
"loss": 0.258,
"step": 438000
},
{
"epoch": 9.77,
"eval_loss": 0.24090658128261566,
"eval_runtime": 2.0077,
"eval_samples_per_second": 1144.123,
"eval_steps_per_second": 17.931,
"step": 438000
},
{
"epoch": 9.78,
"learning_rate": 2.1830543782105647e-05,
"loss": 0.2566,
"step": 438500
},
{
"epoch": 9.79,
"learning_rate": 2.1641565294124206e-05,
"loss": 0.2565,
"step": 439000
},
{
"epoch": 9.79,
"eval_loss": 0.23793531954288483,
"eval_runtime": 1.9748,
"eval_samples_per_second": 1163.127,
"eval_steps_per_second": 18.229,
"step": 439000
},
{
"epoch": 9.8,
"learning_rate": 2.1454045192618794e-05,
"loss": 0.2564,
"step": 439500
},
{
"epoch": 9.81,
"learning_rate": 2.1267985528279212e-05,
"loss": 0.2569,
"step": 440000
},
{
"epoch": 9.81,
"eval_loss": 0.23971830308437347,
"eval_runtime": 1.9911,
"eval_samples_per_second": 1153.611,
"eval_steps_per_second": 18.08,
"step": 440000
},
{
"epoch": 9.83,
"learning_rate": 2.1083388335824145e-05,
"loss": 0.2568,
"step": 440500
},
{
"epoch": 9.84,
"learning_rate": 2.0900255633978873e-05,
"loss": 0.257,
"step": 441000
},
{
"epoch": 9.84,
"eval_loss": 0.23978090286254883,
"eval_runtime": 2.0085,
"eval_samples_per_second": 1143.638,
"eval_steps_per_second": 17.924,
"step": 441000
},
{
"epoch": 9.85,
"learning_rate": 2.0718589425453314e-05,
"loss": 0.2559,
"step": 441500
},
{
"epoch": 9.86,
"learning_rate": 2.0538391696920015e-05,
"loss": 0.2559,
"step": 442000
},
{
"epoch": 9.86,
"eval_loss": 0.23992407321929932,
"eval_runtime": 1.9856,
"eval_samples_per_second": 1156.855,
"eval_steps_per_second": 18.131,
"step": 442000
},
{
"epoch": 9.87,
"learning_rate": 2.035966441899249e-05,
"loss": 0.2557,
"step": 442500
},
{
"epoch": 9.88,
"learning_rate": 2.0182409546203555e-05,
"loss": 0.2556,
"step": 443000
},
{
"epoch": 9.88,
"eval_loss": 0.23771096765995026,
"eval_runtime": 1.9588,
"eval_samples_per_second": 1172.656,
"eval_steps_per_second": 18.379,
"step": 443000
},
{
"epoch": 9.89,
"learning_rate": 2.000662901698415e-05,
"loss": 0.2562,
"step": 443500
},
{
"epoch": 9.9,
"learning_rate": 1.983232475364195e-05,
"loss": 0.2565,
"step": 444000
},
{
"epoch": 9.9,
"eval_loss": 0.24137504398822784,
"eval_runtime": 1.9524,
"eval_samples_per_second": 1176.51,
"eval_steps_per_second": 18.439,
"step": 444000
},
{
"epoch": 9.91,
"learning_rate": 1.9659498662340474e-05,
"loss": 0.2563,
"step": 444500
},
{
"epoch": 9.93,
"learning_rate": 1.948815263307819e-05,
"loss": 0.2562,
"step": 445000
},
{
"epoch": 9.93,
"eval_loss": 0.23801474273204803,
"eval_runtime": 2.0266,
"eval_samples_per_second": 1133.424,
"eval_steps_per_second": 17.764,
"step": 445000
},
{
"epoch": 9.94,
"learning_rate": 1.9318288539667765e-05,
"loss": 0.2562,
"step": 445500
},
{
"epoch": 9.95,
"learning_rate": 1.914990823971574e-05,
"loss": 0.2558,
"step": 446000
},
{
"epoch": 9.95,
"eval_loss": 0.23844870924949646,
"eval_runtime": 2.0906,
"eval_samples_per_second": 1098.726,
"eval_steps_per_second": 17.22,
"step": 446000
},
{
"epoch": 9.96,
"learning_rate": 1.8983013574602096e-05,
"loss": 0.2559,
"step": 446500
},
{
"epoch": 9.97,
"learning_rate": 1.8817606369460156e-05,
"loss": 0.2555,
"step": 447000
},
{
"epoch": 9.97,
"eval_loss": 0.23887751996517181,
"eval_runtime": 2.059,
"eval_samples_per_second": 1115.606,
"eval_steps_per_second": 17.484,
"step": 447000
},
{
"epoch": 9.98,
"learning_rate": 1.865368843315663e-05,
"loss": 0.2561,
"step": 447500
},
{
"epoch": 9.99,
"learning_rate": 1.8491261558271762e-05,
"loss": 0.2558,
"step": 448000
},
{
"epoch": 9.99,
"eval_loss": 0.2371719628572464,
"eval_runtime": 1.9897,
"eval_samples_per_second": 1154.434,
"eval_steps_per_second": 18.093,
"step": 448000
},
{
"epoch": 10.0,
"learning_rate": 1.833032752107986e-05,
"loss": 0.256,
"step": 448500
},
{
"epoch": 10.02,
"learning_rate": 1.817088808152978e-05,
"loss": 0.2563,
"step": 449000
},
{
"epoch": 10.02,
"eval_loss": 0.23607970774173737,
"eval_runtime": 2.0104,
"eval_samples_per_second": 1142.561,
"eval_steps_per_second": 17.907,
"step": 449000
},
{
"epoch": 10.03,
"learning_rate": 1.801294498322569e-05,
"loss": 0.2559,
"step": 449500
},
{
"epoch": 10.04,
"learning_rate": 1.7856499953407978e-05,
"loss": 0.2555,
"step": 450000
},
{
"epoch": 10.04,
"eval_loss": 0.238793283700943,
"eval_runtime": 2.0158,
"eval_samples_per_second": 1139.483,
"eval_steps_per_second": 17.859,
"step": 450000
},
{
"epoch": 10.05,
"learning_rate": 1.770155470293445e-05,
"loss": 0.2555,
"step": 450500
},
{
"epoch": 10.06,
"learning_rate": 1.7548110926261522e-05,
"loss": 0.2557,
"step": 451000
},
{
"epoch": 10.06,
"eval_loss": 0.23789702355861664,
"eval_runtime": 2.0142,
"eval_samples_per_second": 1140.419,
"eval_steps_per_second": 17.873,
"step": 451000
},
{
"epoch": 10.07,
"learning_rate": 1.7396170301425777e-05,
"loss": 0.2558,
"step": 451500
},
{
"epoch": 10.08,
"learning_rate": 1.7245734490025544e-05,
"loss": 0.2556,
"step": 452000
},
{
"epoch": 10.08,
"eval_loss": 0.2400251030921936,
"eval_runtime": 1.9794,
"eval_samples_per_second": 1160.467,
"eval_steps_per_second": 18.188,
"step": 452000
},
{
"epoch": 10.09,
"learning_rate": 1.7096805137202738e-05,
"loss": 0.2559,
"step": 452500
},
{
"epoch": 10.1,
"learning_rate": 1.6949383871624917e-05,
"loss": 0.2556,
"step": 453000
},
{
"epoch": 10.1,
"eval_loss": 0.24092479050159454,
"eval_runtime": 1.9634,
"eval_samples_per_second": 1169.917,
"eval_steps_per_second": 18.336,
"step": 453000
},
{
"epoch": 10.12,
"learning_rate": 1.6803472305467368e-05,
"loss": 0.2557,
"step": 453500
},
{
"epoch": 10.13,
"learning_rate": 1.665907203439568e-05,
"loss": 0.2557,
"step": 454000
},
{
"epoch": 10.13,
"eval_loss": 0.23987002670764923,
"eval_runtime": 1.9519,
"eval_samples_per_second": 1176.785,
"eval_steps_per_second": 18.443,
"step": 454000
},
{
"epoch": 10.14,
"learning_rate": 1.6516184637548058e-05,
"loss": 0.257,
"step": 454500
},
{
"epoch": 10.15,
"learning_rate": 1.6374811677518142e-05,
"loss": 0.2555,
"step": 455000
},
{
"epoch": 10.15,
"eval_loss": 0.2387952208518982,
"eval_runtime": 2.0231,
"eval_samples_per_second": 1135.362,
"eval_steps_per_second": 17.794,
"step": 455000
},
{
"epoch": 10.16,
"learning_rate": 1.6234954700338025e-05,
"loss": 0.2555,
"step": 455500
},
{
"epoch": 10.17,
"learning_rate": 1.6096615235461148e-05,
"loss": 0.2549,
"step": 456000
},
{
"epoch": 10.17,
"eval_loss": 0.23763985931873322,
"eval_runtime": 2.0435,
"eval_samples_per_second": 1124.037,
"eval_steps_per_second": 17.617,
"step": 456000
},
{
"epoch": 10.18,
"learning_rate": 1.59597947957458e-05,
"loss": 0.2557,
"step": 456500
},
{
"epoch": 10.19,
"learning_rate": 1.5824494877438344e-05,
"loss": 0.2551,
"step": 457000
},
{
"epoch": 10.19,
"eval_loss": 0.23833227157592773,
"eval_runtime": 2.01,
"eval_samples_per_second": 1142.798,
"eval_steps_per_second": 17.911,
"step": 457000
},
{
"epoch": 10.2,
"learning_rate": 1.569071696015702e-05,
"loss": 0.2549,
"step": 457500
},
{
"epoch": 10.22,
"learning_rate": 1.555846250687569e-05,
"loss": 0.2551,
"step": 458000
},
{
"epoch": 10.22,
"eval_loss": 0.23883755505084991,
"eval_runtime": 1.9432,
"eval_samples_per_second": 1182.062,
"eval_steps_per_second": 18.526,
"step": 458000
},
{
"epoch": 10.23,
"learning_rate": 1.542773296390789e-05,
"loss": 0.2555,
"step": 458500
},
{
"epoch": 10.24,
"learning_rate": 1.5298529760890945e-05,
"loss": 0.2559,
"step": 459000
},
{
"epoch": 10.24,
"eval_loss": 0.23782269656658173,
"eval_runtime": 1.9879,
"eval_samples_per_second": 1155.497,
"eval_steps_per_second": 18.11,
"step": 459000
},
{
"epoch": 10.25,
"learning_rate": 1.5170854310770376e-05,
"loss": 0.2557,
"step": 459500
},
{
"epoch": 10.26,
"learning_rate": 1.5044708009784457e-05,
"loss": 0.2555,
"step": 460000
},
{
"epoch": 10.26,
"eval_loss": 0.23906496167182922,
"eval_runtime": 1.9573,
"eval_samples_per_second": 1173.539,
"eval_steps_per_second": 18.392,
"step": 460000
},
{
"epoch": 10.27,
"learning_rate": 1.4920092237448903e-05,
"loss": 0.2554,
"step": 460500
},
{
"epoch": 10.28,
"learning_rate": 1.4797008356541874e-05,
"loss": 0.2548,
"step": 461000
},
{
"epoch": 10.28,
"eval_loss": 0.2381051927804947,
"eval_runtime": 1.9705,
"eval_samples_per_second": 1165.71,
"eval_steps_per_second": 18.27,
"step": 461000
},
{
"epoch": 10.29,
"learning_rate": 1.4675457713088947e-05,
"loss": 0.2547,
"step": 461500
},
{
"epoch": 10.31,
"learning_rate": 1.4555441636348494e-05,
"loss": 0.2549,
"step": 462000
},
{
"epoch": 10.31,
"eval_loss": 0.23718567192554474,
"eval_runtime": 2.0107,
"eval_samples_per_second": 1142.393,
"eval_steps_per_second": 17.904,
"step": 462000
},
{
"epoch": 10.32,
"learning_rate": 1.4436961438797095e-05,
"loss": 0.2547,
"step": 462500
},
{
"epoch": 10.33,
"learning_rate": 1.4320018416115206e-05,
"loss": 0.2548,
"step": 463000
},
{
"epoch": 10.33,
"eval_loss": 0.23715750873088837,
"eval_runtime": 2.0434,
"eval_samples_per_second": 1124.132,
"eval_steps_per_second": 17.618,
"step": 463000
},
{
"epoch": 10.34,
"learning_rate": 1.4204613847173003e-05,
"loss": 0.2547,
"step": 463500
},
{
"epoch": 10.35,
"learning_rate": 1.4090748994016354e-05,
"loss": 0.2547,
"step": 464000
},
{
"epoch": 10.35,
"eval_loss": 0.23930229246616364,
"eval_runtime": 2.02,
"eval_samples_per_second": 1137.119,
"eval_steps_per_second": 17.822,
"step": 464000
},
{
"epoch": 10.36,
"learning_rate": 1.3978425101853049e-05,
"loss": 0.2545,
"step": 464500
},
{
"epoch": 10.37,
"learning_rate": 1.3867643399039165e-05,
"loss": 0.2546,
"step": 465000
},
{
"epoch": 10.37,
"eval_loss": 0.24023665487766266,
"eval_runtime": 2.0161,
"eval_samples_per_second": 1139.319,
"eval_steps_per_second": 17.856,
"step": 465000
},
{
"epoch": 10.38,
"learning_rate": 1.3758405097065648e-05,
"loss": 0.2547,
"step": 465500
},
{
"epoch": 10.39,
"learning_rate": 1.3650711390545131e-05,
"loss": 0.2549,
"step": 466000
},
{
"epoch": 10.39,
"eval_loss": 0.2383406162261963,
"eval_runtime": 2.0372,
"eval_samples_per_second": 1127.524,
"eval_steps_per_second": 17.671,
"step": 466000
},
{
"epoch": 10.41,
"learning_rate": 1.3544563457198657e-05,
"loss": 0.2546,
"step": 466500
},
{
"epoch": 10.42,
"learning_rate": 1.343996245784307e-05,
"loss": 0.2545,
"step": 467000
},
{
"epoch": 10.42,
"eval_loss": 0.23841743171215057,
"eval_runtime": 1.9888,
"eval_samples_per_second": 1154.948,
"eval_steps_per_second": 18.101,
"step": 467000
},
{
"epoch": 10.43,
"learning_rate": 1.3336909536378107e-05,
"loss": 0.2549,
"step": 467500
},
{
"epoch": 10.44,
"learning_rate": 1.3235405819774022e-05,
"loss": 0.2544,
"step": 468000
},
{
"epoch": 10.44,
"eval_loss": 0.2374790459871292,
"eval_runtime": 2.0476,
"eval_samples_per_second": 1121.805,
"eval_steps_per_second": 17.582,
"step": 468000
},
{
"epoch": 10.45,
"learning_rate": 1.3135452418059208e-05,
"loss": 0.2543,
"step": 468500
},
{
"epoch": 10.46,
"learning_rate": 1.3037050424308027e-05,
"loss": 0.2544,
"step": 469000
},
{
"epoch": 10.46,
"eval_loss": 0.2366662174463272,
"eval_runtime": 2.0397,
"eval_samples_per_second": 1126.142,
"eval_steps_per_second": 17.65,
"step": 469000
},
{
"epoch": 10.47,
"learning_rate": 1.2940200914628945e-05,
"loss": 0.2547,
"step": 469500
},
{
"epoch": 10.48,
"learning_rate": 1.2844904948152644e-05,
"loss": 0.255,
"step": 470000
},
{
"epoch": 10.48,
"eval_loss": 0.23585031926631927,
"eval_runtime": 1.91,
"eval_samples_per_second": 1202.649,
"eval_steps_per_second": 18.849,
"step": 470000
},
{
"epoch": 10.49,
"learning_rate": 1.2751163567020592e-05,
"loss": 0.2544,
"step": 470500
},
{
"epoch": 10.51,
"learning_rate": 1.2658977796373478e-05,
"loss": 0.2546,
"step": 471000
},
{
"epoch": 10.51,
"eval_loss": 0.23835012316703796,
"eval_runtime": 2.0021,
"eval_samples_per_second": 1147.276,
"eval_steps_per_second": 17.981,
"step": 471000
},
{
"epoch": 10.52,
"learning_rate": 1.2568348644340153e-05,
"loss": 0.2545,
"step": 471500
},
{
"epoch": 10.53,
"learning_rate": 1.2479277102026465e-05,
"loss": 0.2544,
"step": 472000
},
{
"epoch": 10.53,
"eval_loss": 0.23700165748596191,
"eval_runtime": 2.0158,
"eval_samples_per_second": 1139.473,
"eval_steps_per_second": 17.859,
"step": 472000
},
{
"epoch": 10.54,
"learning_rate": 1.2391764143504556e-05,
"loss": 0.2542,
"step": 472500
},
{
"epoch": 10.55,
"learning_rate": 1.2305810725802118e-05,
"loss": 0.254,
"step": 473000
},
{
"epoch": 10.55,
"eval_loss": 0.23847134411334991,
"eval_runtime": 1.9758,
"eval_samples_per_second": 1162.557,
"eval_steps_per_second": 18.22,
"step": 473000
},
{
"epoch": 10.56,
"learning_rate": 1.222141778889195e-05,
"loss": 0.2538,
"step": 473500
},
{
"epoch": 10.57,
"learning_rate": 1.2138586255681707e-05,
"loss": 0.2539,
"step": 474000
},
{
"epoch": 10.57,
"eval_loss": 0.23491570353507996,
"eval_runtime": 1.99,
"eval_samples_per_second": 1154.271,
"eval_steps_per_second": 18.09,
"step": 474000
},
{
"epoch": 10.58,
"learning_rate": 1.2057317032003731e-05,
"loss": 0.2542,
"step": 474500
},
{
"epoch": 10.6,
"learning_rate": 1.1977611006605263e-05,
"loss": 0.2549,
"step": 475000
},
{
"epoch": 10.6,
"eval_loss": 0.2358667552471161,
"eval_runtime": 2.0083,
"eval_samples_per_second": 1143.743,
"eval_steps_per_second": 17.925,
"step": 475000
},
{
"epoch": 10.61,
"learning_rate": 1.1899469051138602e-05,
"loss": 0.2543,
"step": 475500
},
{
"epoch": 10.62,
"learning_rate": 1.1822892020151667e-05,
"loss": 0.2549,
"step": 476000
},
{
"epoch": 10.62,
"eval_loss": 0.2398059368133545,
"eval_runtime": 2.007,
"eval_samples_per_second": 1144.474,
"eval_steps_per_second": 17.937,
"step": 476000
},
{
"epoch": 10.63,
"learning_rate": 1.1747880751078614e-05,
"loss": 0.2554,
"step": 476500
},
{
"epoch": 10.64,
"learning_rate": 1.1674436064230637e-05,
"loss": 0.2548,
"step": 477000
},
{
"epoch": 10.64,
"eval_loss": 0.23931536078453064,
"eval_runtime": 1.9895,
"eval_samples_per_second": 1154.534,
"eval_steps_per_second": 18.095,
"step": 477000
},
{
"epoch": 10.65,
"learning_rate": 1.1602558762787069e-05,
"loss": 0.2547,
"step": 477500
},
{
"epoch": 10.66,
"learning_rate": 1.1532249632786582e-05,
"loss": 0.2542,
"step": 478000
},
{
"epoch": 10.66,
"eval_loss": 0.237422376871109,
"eval_runtime": 2.0815,
"eval_samples_per_second": 1103.524,
"eval_steps_per_second": 17.295,
"step": 478000
},
{
"epoch": 10.67,
"learning_rate": 1.1463509443118552e-05,
"loss": 0.2541,
"step": 478500
},
{
"epoch": 10.68,
"learning_rate": 1.1396338945514663e-05,
"loss": 0.2543,
"step": 479000
},
{
"epoch": 10.68,
"eval_loss": 0.2378796935081482,
"eval_runtime": 2.0025,
"eval_samples_per_second": 1147.061,
"eval_steps_per_second": 17.977,
"step": 479000
},
{
"epoch": 10.7,
"learning_rate": 1.133073887454072e-05,
"loss": 0.2538,
"step": 479500
},
{
"epoch": 10.71,
"learning_rate": 1.1266709947588599e-05,
"loss": 0.2539,
"step": 480000
},
{
"epoch": 10.71,
"eval_loss": 0.2366316020488739,
"eval_runtime": 1.9626,
"eval_samples_per_second": 1170.393,
"eval_steps_per_second": 18.343,
"step": 480000
},
{
"epoch": 10.72,
"learning_rate": 1.1204252864868377e-05,
"loss": 0.2538,
"step": 480500
},
{
"epoch": 10.73,
"learning_rate": 1.1143368309400725e-05,
"loss": 0.2539,
"step": 481000
},
{
"epoch": 10.73,
"eval_loss": 0.23603801429271698,
"eval_runtime": 1.9718,
"eval_samples_per_second": 1164.912,
"eval_steps_per_second": 18.257,
"step": 481000
},
{
"epoch": 10.74,
"learning_rate": 1.1084056947009348e-05,
"loss": 0.2538,
"step": 481500
},
{
"epoch": 10.75,
"learning_rate": 1.1026319426313837e-05,
"loss": 0.2538,
"step": 482000
},
{
"epoch": 10.75,
"eval_loss": 0.23858527839183807,
"eval_runtime": 1.961,
"eval_samples_per_second": 1171.312,
"eval_steps_per_second": 18.358,
"step": 482000
},
{
"epoch": 10.76,
"learning_rate": 1.097015637872247e-05,
"loss": 0.2538,
"step": 482500
},
{
"epoch": 10.77,
"learning_rate": 1.0915568418425301e-05,
"loss": 0.2537,
"step": 483000
},
{
"epoch": 10.77,
"eval_loss": 0.23714858293533325,
"eval_runtime": 2.009,
"eval_samples_per_second": 1143.375,
"eval_steps_per_second": 17.92,
"step": 483000
},
{
"epoch": 10.78,
"learning_rate": 1.0862556142387571e-05,
"loss": 0.2539,
"step": 483500
},
{
"epoch": 10.8,
"learning_rate": 1.081112013034298e-05,
"loss": 0.2537,
"step": 484000
},
{
"epoch": 10.8,
"eval_loss": 0.23877692222595215,
"eval_runtime": 1.9856,
"eval_samples_per_second": 1156.824,
"eval_steps_per_second": 18.13,
"step": 484000
},
{
"epoch": 10.81,
"learning_rate": 1.0761260944787561e-05,
"loss": 0.2551,
"step": 484500
},
{
"epoch": 10.82,
"learning_rate": 1.0712979130973347e-05,
"loss": 0.2542,
"step": 485000
},
{
"epoch": 10.82,
"eval_loss": 0.23765695095062256,
"eval_runtime": 1.9888,
"eval_samples_per_second": 1154.965,
"eval_steps_per_second": 18.101,
"step": 485000
},
{
"epoch": 10.83,
"learning_rate": 1.0666275216902535e-05,
"loss": 0.2539,
"step": 485500
},
{
"epoch": 10.84,
"learning_rate": 1.0621149713321656e-05,
"loss": 0.2539,
"step": 486000
},
{
"epoch": 10.84,
"eval_loss": 0.23621481657028198,
"eval_runtime": 1.9428,
"eval_samples_per_second": 1182.329,
"eval_steps_per_second": 18.53,
"step": 486000
},
{
"epoch": 10.85,
"learning_rate": 1.0577603113715964e-05,
"loss": 0.2539,
"step": 486500
},
{
"epoch": 10.86,
"learning_rate": 1.0535635894304106e-05,
"loss": 0.2535,
"step": 487000
},
{
"epoch": 10.86,
"eval_loss": 0.23733575642108917,
"eval_runtime": 1.9603,
"eval_samples_per_second": 1171.773,
"eval_steps_per_second": 18.365,
"step": 487000
},
{
"epoch": 10.87,
"learning_rate": 1.0495248514032875e-05,
"loss": 0.2539,
"step": 487500
},
{
"epoch": 10.89,
"learning_rate": 1.045644141457218e-05,
"loss": 0.2533,
"step": 488000
},
{
"epoch": 10.89,
"eval_loss": 0.23612964153289795,
"eval_runtime": 1.9923,
"eval_samples_per_second": 1152.93,
"eval_steps_per_second": 18.069,
"step": 488000
},
{
"epoch": 10.9,
"learning_rate": 1.0419215020310254e-05,
"loss": 0.2534,
"step": 488500
},
{
"epoch": 10.91,
"learning_rate": 1.0383569738348988e-05,
"loss": 0.2533,
"step": 489000
},
{
"epoch": 10.91,
"eval_loss": 0.2368190884590149,
"eval_runtime": 1.9507,
"eval_samples_per_second": 1177.524,
"eval_steps_per_second": 18.455,
"step": 489000
},
{
"epoch": 10.92,
"learning_rate": 1.0349505958499436e-05,
"loss": 0.2534,
"step": 489500
},
{
"epoch": 10.93,
"learning_rate": 1.0317024053277693e-05,
"loss": 0.2535,
"step": 490000
},
{
"epoch": 10.93,
"eval_loss": 0.23948417603969574,
"eval_runtime": 2.0351,
"eval_samples_per_second": 1128.67,
"eval_steps_per_second": 17.689,
"step": 490000
},
{
"epoch": 10.94,
"learning_rate": 1.0286124377900624e-05,
"loss": 0.2541,
"step": 490500
},
{
"epoch": 10.95,
"learning_rate": 1.0256807270282153e-05,
"loss": 0.2537,
"step": 491000
},
{
"epoch": 10.95,
"eval_loss": 0.23797546327114105,
"eval_runtime": 1.9751,
"eval_samples_per_second": 1162.982,
"eval_steps_per_second": 18.227,
"step": 491000
},
{
"epoch": 10.96,
"learning_rate": 1.0229073051029455e-05,
"loss": 0.2539,
"step": 491500
},
{
"epoch": 10.97,
"learning_rate": 1.020292202343952e-05,
"loss": 0.254,
"step": 492000
},
{
"epoch": 10.97,
"eval_loss": 0.23647533357143402,
"eval_runtime": 2.0099,
"eval_samples_per_second": 1142.868,
"eval_steps_per_second": 17.912,
"step": 492000
},
{
"epoch": 10.99,
"learning_rate": 1.0178354473495813e-05,
"loss": 0.2539,
"step": 492500
},
{
"epoch": 11.0,
"learning_rate": 1.0155370669865077e-05,
"loss": 0.254,
"step": 493000
},
{
"epoch": 11.0,
"eval_loss": 0.2374097853899002,
"eval_runtime": 1.9968,
"eval_samples_per_second": 1150.347,
"eval_steps_per_second": 18.029,
"step": 493000
},
{
"epoch": 11.01,
"learning_rate": 1.0133970863894557e-05,
"loss": 0.2537,
"step": 493500
},
{
"epoch": 11.02,
"learning_rate": 1.0114155289609061e-05,
"loss": 0.2535,
"step": 494000
},
{
"epoch": 11.02,
"eval_loss": 0.23701806366443634,
"eval_runtime": 2.0405,
"eval_samples_per_second": 1125.682,
"eval_steps_per_second": 17.642,
"step": 494000
},
{
"epoch": 11.03,
"learning_rate": 1.0095924163708572e-05,
"loss": 0.2542,
"step": 494500
},
{
"epoch": 11.04,
"learning_rate": 1.0079277685565724e-05,
"loss": 0.2538,
"step": 495000
},
{
"epoch": 11.04,
"eval_loss": 0.23448336124420166,
"eval_runtime": 1.9969,
"eval_samples_per_second": 1150.279,
"eval_steps_per_second": 18.028,
"step": 495000
},
{
"epoch": 11.05,
"learning_rate": 1.0064216037223772e-05,
"loss": 0.2536,
"step": 495500
},
{
"epoch": 11.06,
"learning_rate": 1.0050739383394454e-05,
"loss": 0.2539,
"step": 496000
},
{
"epoch": 11.06,
"eval_loss": 0.23559238016605377,
"eval_runtime": 2.0311,
"eval_samples_per_second": 1130.902,
"eval_steps_per_second": 17.724,
"step": 496000
},
{
"epoch": 11.07,
"learning_rate": 1.003884787145633e-05,
"loss": 0.2532,
"step": 496500
},
{
"epoch": 11.09,
"learning_rate": 1.002854163145305e-05,
"loss": 0.2533,
"step": 497000
},
{
"epoch": 11.09,
"eval_loss": 0.23612073063850403,
"eval_runtime": 2.0421,
"eval_samples_per_second": 1124.803,
"eval_steps_per_second": 17.629,
"step": 497000
},
{
"epoch": 11.1,
"learning_rate": 1.0019820776091995e-05,
"loss": 0.2531,
"step": 497500
},
{
"epoch": 11.11,
"learning_rate": 1.0012685400743077e-05,
"loss": 0.2533,
"step": 498000
},
{
"epoch": 11.11,
"eval_loss": 0.23812700808048248,
"eval_runtime": 2.036,
"eval_samples_per_second": 1128.176,
"eval_steps_per_second": 17.681,
"step": 498000
},
{
"epoch": 11.12,
"learning_rate": 1.0007135583437572e-05,
"loss": 0.2531,
"step": 498500
},
{
"epoch": 11.13,
"learning_rate": 1.0003171384867436e-05,
"loss": 0.2534,
"step": 499000
},
{
"epoch": 11.13,
"eval_loss": 0.23714645206928253,
"eval_runtime": 2.0313,
"eval_samples_per_second": 1130.821,
"eval_steps_per_second": 17.723,
"step": 499000
},
{
"epoch": 11.14,
"learning_rate": 1.0000792848384467e-05,
"loss": 0.2535,
"step": 499500
},
{
"epoch": 11.15,
"learning_rate": 1e-05,
"loss": 0.2535,
"step": 500000
},
{
"epoch": 11.15,
"eval_loss": 0.23556514084339142,
"eval_runtime": 1.9635,
"eval_samples_per_second": 1169.857,
"eval_steps_per_second": 18.335,
"step": 500000
}
],
"max_steps": 500000,
"num_train_epochs": 12,
"total_flos": 1.5974043941849432e+22,
"trial_name": null,
"trial_params": null
}