1592 lines
37 KiB
JSON
1592 lines
37 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 10,
|
||
|
|
"global_step": 198,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.005050505050505051,
|
||
|
|
"grad_norm": 9.0,
|
||
|
|
"learning_rate": 0.0,
|
||
|
|
"loss": 2.2482,
|
||
|
|
"step": 1
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010101010101010102,
|
||
|
|
"grad_norm": 10.8125,
|
||
|
|
"learning_rate": 2.0000000000000003e-06,
|
||
|
|
"loss": 2.3539,
|
||
|
|
"step": 2
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.015151515151515152,
|
||
|
|
"grad_norm": 9.3125,
|
||
|
|
"learning_rate": 4.000000000000001e-06,
|
||
|
|
"loss": 2.2254,
|
||
|
|
"step": 3
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.020202020202020204,
|
||
|
|
"grad_norm": 9.5,
|
||
|
|
"learning_rate": 6e-06,
|
||
|
|
"loss": 2.2148,
|
||
|
|
"step": 4
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.025252525252525252,
|
||
|
|
"grad_norm": 7.25,
|
||
|
|
"learning_rate": 8.000000000000001e-06,
|
||
|
|
"loss": 2.1703,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.030303030303030304,
|
||
|
|
"grad_norm": 6.09375,
|
||
|
|
"learning_rate": 1e-05,
|
||
|
|
"loss": 2.0749,
|
||
|
|
"step": 6
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.03535353535353535,
|
||
|
|
"grad_norm": 5.25,
|
||
|
|
"learning_rate": 9.94818652849741e-06,
|
||
|
|
"loss": 1.9353,
|
||
|
|
"step": 7
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.04040404040404041,
|
||
|
|
"grad_norm": 4.875,
|
||
|
|
"learning_rate": 9.89637305699482e-06,
|
||
|
|
"loss": 2.023,
|
||
|
|
"step": 8
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.045454545454545456,
|
||
|
|
"grad_norm": 4.53125,
|
||
|
|
"learning_rate": 9.844559585492228e-06,
|
||
|
|
"loss": 2.0044,
|
||
|
|
"step": 9
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.050505050505050504,
|
||
|
|
"grad_norm": 4.34375,
|
||
|
|
"learning_rate": 9.792746113989638e-06,
|
||
|
|
"loss": 1.938,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.050505050505050504,
|
||
|
|
"eval_loss": 1.9176450967788696,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.319,
|
||
|
|
"eval_samples_per_second": 45.577,
|
||
|
|
"eval_steps_per_second": 22.789,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.05555555555555555,
|
||
|
|
"grad_norm": 3.96875,
|
||
|
|
"learning_rate": 9.740932642487048e-06,
|
||
|
|
"loss": 1.895,
|
||
|
|
"step": 11
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.06060606060606061,
|
||
|
|
"grad_norm": 3.984375,
|
||
|
|
"learning_rate": 9.689119170984456e-06,
|
||
|
|
"loss": 1.772,
|
||
|
|
"step": 12
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.06565656565656566,
|
||
|
|
"grad_norm": 3.875,
|
||
|
|
"learning_rate": 9.637305699481867e-06,
|
||
|
|
"loss": 1.8059,
|
||
|
|
"step": 13
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0707070707070707,
|
||
|
|
"grad_norm": 4.375,
|
||
|
|
"learning_rate": 9.585492227979275e-06,
|
||
|
|
"loss": 1.8952,
|
||
|
|
"step": 14
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.07575757575757576,
|
||
|
|
"grad_norm": 3.9375,
|
||
|
|
"learning_rate": 9.533678756476683e-06,
|
||
|
|
"loss": 1.844,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.08080808080808081,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 9.481865284974095e-06,
|
||
|
|
"loss": 1.8882,
|
||
|
|
"step": 16
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.08585858585858586,
|
||
|
|
"grad_norm": 3.578125,
|
||
|
|
"learning_rate": 9.430051813471504e-06,
|
||
|
|
"loss": 1.7425,
|
||
|
|
"step": 17
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.09090909090909091,
|
||
|
|
"grad_norm": 3.96875,
|
||
|
|
"learning_rate": 9.378238341968912e-06,
|
||
|
|
"loss": 1.7552,
|
||
|
|
"step": 18
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.09595959595959595,
|
||
|
|
"grad_norm": 4.21875,
|
||
|
|
"learning_rate": 9.326424870466322e-06,
|
||
|
|
"loss": 1.7315,
|
||
|
|
"step": 19
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10101010101010101,
|
||
|
|
"grad_norm": 3.734375,
|
||
|
|
"learning_rate": 9.27461139896373e-06,
|
||
|
|
"loss": 1.7405,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10101010101010101,
|
||
|
|
"eval_loss": 1.7153369188308716,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.375,
|
||
|
|
"eval_samples_per_second": 45.177,
|
||
|
|
"eval_steps_per_second": 22.588,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10606060606060606,
|
||
|
|
"grad_norm": 3.9375,
|
||
|
|
"learning_rate": 9.22279792746114e-06,
|
||
|
|
"loss": 1.6805,
|
||
|
|
"step": 21
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1111111111111111,
|
||
|
|
"grad_norm": 4.21875,
|
||
|
|
"learning_rate": 9.17098445595855e-06,
|
||
|
|
"loss": 1.7294,
|
||
|
|
"step": 22
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.11616161616161616,
|
||
|
|
"grad_norm": 3.875,
|
||
|
|
"learning_rate": 9.11917098445596e-06,
|
||
|
|
"loss": 1.8044,
|
||
|
|
"step": 23
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.12121212121212122,
|
||
|
|
"grad_norm": 4.59375,
|
||
|
|
"learning_rate": 9.06735751295337e-06,
|
||
|
|
"loss": 1.6795,
|
||
|
|
"step": 24
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.12626262626262627,
|
||
|
|
"grad_norm": 3.75,
|
||
|
|
"learning_rate": 9.015544041450778e-06,
|
||
|
|
"loss": 1.5566,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.13131313131313133,
|
||
|
|
"grad_norm": 3.90625,
|
||
|
|
"learning_rate": 8.963730569948186e-06,
|
||
|
|
"loss": 1.6053,
|
||
|
|
"step": 26
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.13636363636363635,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 8.911917098445596e-06,
|
||
|
|
"loss": 1.5785,
|
||
|
|
"step": 27
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1414141414141414,
|
||
|
|
"grad_norm": 3.765625,
|
||
|
|
"learning_rate": 8.860103626943006e-06,
|
||
|
|
"loss": 1.5936,
|
||
|
|
"step": 28
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.14646464646464646,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 8.808290155440415e-06,
|
||
|
|
"loss": 1.5388,
|
||
|
|
"step": 29
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.15151515151515152,
|
||
|
|
"grad_norm": 3.375,
|
||
|
|
"learning_rate": 8.756476683937825e-06,
|
||
|
|
"loss": 1.5542,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.15151515151515152,
|
||
|
|
"eval_loss": 1.617945909500122,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.365,
|
||
|
|
"eval_samples_per_second": 45.247,
|
||
|
|
"eval_steps_per_second": 22.624,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.15656565656565657,
|
||
|
|
"grad_norm": 3.3125,
|
||
|
|
"learning_rate": 8.704663212435233e-06,
|
||
|
|
"loss": 1.4746,
|
||
|
|
"step": 31
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16161616161616163,
|
||
|
|
"grad_norm": 3.703125,
|
||
|
|
"learning_rate": 8.652849740932643e-06,
|
||
|
|
"loss": 1.6288,
|
||
|
|
"step": 32
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16666666666666666,
|
||
|
|
"grad_norm": 3.8125,
|
||
|
|
"learning_rate": 8.601036269430052e-06,
|
||
|
|
"loss": 1.6978,
|
||
|
|
"step": 33
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1717171717171717,
|
||
|
|
"grad_norm": 4.3125,
|
||
|
|
"learning_rate": 8.549222797927462e-06,
|
||
|
|
"loss": 1.6176,
|
||
|
|
"step": 34
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.17676767676767677,
|
||
|
|
"grad_norm": 4.0,
|
||
|
|
"learning_rate": 8.497409326424872e-06,
|
||
|
|
"loss": 1.6729,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.18181818181818182,
|
||
|
|
"grad_norm": 3.84375,
|
||
|
|
"learning_rate": 8.44559585492228e-06,
|
||
|
|
"loss": 1.5704,
|
||
|
|
"step": 36
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.18686868686868688,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 8.393782383419689e-06,
|
||
|
|
"loss": 1.6134,
|
||
|
|
"step": 37
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1919191919191919,
|
||
|
|
"grad_norm": 3.703125,
|
||
|
|
"learning_rate": 8.341968911917099e-06,
|
||
|
|
"loss": 1.4351,
|
||
|
|
"step": 38
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.19696969696969696,
|
||
|
|
"grad_norm": 3.828125,
|
||
|
|
"learning_rate": 8.290155440414507e-06,
|
||
|
|
"loss": 1.5933,
|
||
|
|
"step": 39
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.20202020202020202,
|
||
|
|
"grad_norm": 3.59375,
|
||
|
|
"learning_rate": 8.238341968911918e-06,
|
||
|
|
"loss": 1.4878,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.20202020202020202,
|
||
|
|
"eval_loss": 1.5477628707885742,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.3147,
|
||
|
|
"eval_samples_per_second": 45.608,
|
||
|
|
"eval_steps_per_second": 22.804,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.20707070707070707,
|
||
|
|
"grad_norm": 3.4375,
|
||
|
|
"learning_rate": 8.186528497409328e-06,
|
||
|
|
"loss": 1.4662,
|
||
|
|
"step": 41
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21212121212121213,
|
||
|
|
"grad_norm": 3.84375,
|
||
|
|
"learning_rate": 8.134715025906736e-06,
|
||
|
|
"loss": 1.6173,
|
||
|
|
"step": 42
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21717171717171718,
|
||
|
|
"grad_norm": 4.0625,
|
||
|
|
"learning_rate": 8.082901554404146e-06,
|
||
|
|
"loss": 1.5095,
|
||
|
|
"step": 43
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2222222222222222,
|
||
|
|
"grad_norm": 3.609375,
|
||
|
|
"learning_rate": 8.031088082901555e-06,
|
||
|
|
"loss": 1.4314,
|
||
|
|
"step": 44
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.22727272727272727,
|
||
|
|
"grad_norm": 3.703125,
|
||
|
|
"learning_rate": 7.979274611398965e-06,
|
||
|
|
"loss": 1.4399,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.23232323232323232,
|
||
|
|
"grad_norm": 3.375,
|
||
|
|
"learning_rate": 7.927461139896375e-06,
|
||
|
|
"loss": 1.5381,
|
||
|
|
"step": 46
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.23737373737373738,
|
||
|
|
"grad_norm": 4.375,
|
||
|
|
"learning_rate": 7.875647668393783e-06,
|
||
|
|
"loss": 1.5451,
|
||
|
|
"step": 47
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.24242424242424243,
|
||
|
|
"grad_norm": 3.859375,
|
||
|
|
"learning_rate": 7.823834196891192e-06,
|
||
|
|
"loss": 1.4232,
|
||
|
|
"step": 48
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2474747474747475,
|
||
|
|
"grad_norm": 3.890625,
|
||
|
|
"learning_rate": 7.772020725388602e-06,
|
||
|
|
"loss": 1.552,
|
||
|
|
"step": 49
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.25252525252525254,
|
||
|
|
"grad_norm": 3.375,
|
||
|
|
"learning_rate": 7.72020725388601e-06,
|
||
|
|
"loss": 1.4916,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.25252525252525254,
|
||
|
|
"eval_loss": 1.4986205101013184,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.2877,
|
||
|
|
"eval_samples_per_second": 45.804,
|
||
|
|
"eval_steps_per_second": 22.902,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.25757575757575757,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 7.66839378238342e-06,
|
||
|
|
"loss": 1.5456,
|
||
|
|
"step": 51
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.26262626262626265,
|
||
|
|
"grad_norm": 3.5,
|
||
|
|
"learning_rate": 7.61658031088083e-06,
|
||
|
|
"loss": 1.5277,
|
||
|
|
"step": 52
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2676767676767677,
|
||
|
|
"grad_norm": 3.640625,
|
||
|
|
"learning_rate": 7.564766839378239e-06,
|
||
|
|
"loss": 1.3363,
|
||
|
|
"step": 53
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2727272727272727,
|
||
|
|
"grad_norm": 3.734375,
|
||
|
|
"learning_rate": 7.512953367875648e-06,
|
||
|
|
"loss": 1.5053,
|
||
|
|
"step": 54
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2777777777777778,
|
||
|
|
"grad_norm": 3.640625,
|
||
|
|
"learning_rate": 7.461139896373057e-06,
|
||
|
|
"loss": 1.4982,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2828282828282828,
|
||
|
|
"grad_norm": 4.0,
|
||
|
|
"learning_rate": 7.409326424870467e-06,
|
||
|
|
"loss": 1.5154,
|
||
|
|
"step": 56
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2878787878787879,
|
||
|
|
"grad_norm": 3.890625,
|
||
|
|
"learning_rate": 7.357512953367876e-06,
|
||
|
|
"loss": 1.4665,
|
||
|
|
"step": 57
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.29292929292929293,
|
||
|
|
"grad_norm": 3.65625,
|
||
|
|
"learning_rate": 7.305699481865286e-06,
|
||
|
|
"loss": 1.428,
|
||
|
|
"step": 58
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.29797979797979796,
|
||
|
|
"grad_norm": 3.5625,
|
||
|
|
"learning_rate": 7.253886010362695e-06,
|
||
|
|
"loss": 1.47,
|
||
|
|
"step": 59
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.30303030303030304,
|
||
|
|
"grad_norm": 4.5,
|
||
|
|
"learning_rate": 7.2020725388601045e-06,
|
||
|
|
"loss": 1.3728,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.30303030303030304,
|
||
|
|
"eval_loss": 1.458791732788086,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.4644,
|
||
|
|
"eval_samples_per_second": 44.552,
|
||
|
|
"eval_steps_per_second": 22.276,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.30808080808080807,
|
||
|
|
"grad_norm": 3.59375,
|
||
|
|
"learning_rate": 7.150259067357514e-06,
|
||
|
|
"loss": 1.4653,
|
||
|
|
"step": 61
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.31313131313131315,
|
||
|
|
"grad_norm": 3.796875,
|
||
|
|
"learning_rate": 7.098445595854922e-06,
|
||
|
|
"loss": 1.3996,
|
||
|
|
"step": 62
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3181818181818182,
|
||
|
|
"grad_norm": 4.28125,
|
||
|
|
"learning_rate": 7.0466321243523315e-06,
|
||
|
|
"loss": 1.5639,
|
||
|
|
"step": 63
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.32323232323232326,
|
||
|
|
"grad_norm": 4.625,
|
||
|
|
"learning_rate": 6.994818652849742e-06,
|
||
|
|
"loss": 1.4079,
|
||
|
|
"step": 64
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3282828282828283,
|
||
|
|
"grad_norm": 4.0625,
|
||
|
|
"learning_rate": 6.943005181347151e-06,
|
||
|
|
"loss": 1.4742,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3333333333333333,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 6.89119170984456e-06,
|
||
|
|
"loss": 1.4492,
|
||
|
|
"step": 66
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3383838383838384,
|
||
|
|
"grad_norm": 4.34375,
|
||
|
|
"learning_rate": 6.839378238341969e-06,
|
||
|
|
"loss": 1.3779,
|
||
|
|
"step": 67
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3434343434343434,
|
||
|
|
"grad_norm": 3.84375,
|
||
|
|
"learning_rate": 6.787564766839379e-06,
|
||
|
|
"loss": 1.4715,
|
||
|
|
"step": 68
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3484848484848485,
|
||
|
|
"grad_norm": 4.15625,
|
||
|
|
"learning_rate": 6.735751295336788e-06,
|
||
|
|
"loss": 1.3221,
|
||
|
|
"step": 69
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.35353535353535354,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 6.683937823834198e-06,
|
||
|
|
"loss": 1.4824,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.35353535353535354,
|
||
|
|
"eval_loss": 1.4250022172927856,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.3356,
|
||
|
|
"eval_samples_per_second": 45.458,
|
||
|
|
"eval_steps_per_second": 22.729,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.35858585858585856,
|
||
|
|
"grad_norm": 3.734375,
|
||
|
|
"learning_rate": 6.632124352331607e-06,
|
||
|
|
"loss": 1.4255,
|
||
|
|
"step": 71
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.36363636363636365,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 6.5803108808290166e-06,
|
||
|
|
"loss": 1.4265,
|
||
|
|
"step": 72
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3686868686868687,
|
||
|
|
"grad_norm": 3.421875,
|
||
|
|
"learning_rate": 6.528497409326425e-06,
|
||
|
|
"loss": 1.3832,
|
||
|
|
"step": 73
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.37373737373737376,
|
||
|
|
"grad_norm": 3.578125,
|
||
|
|
"learning_rate": 6.476683937823834e-06,
|
||
|
|
"loss": 1.4193,
|
||
|
|
"step": 74
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3787878787878788,
|
||
|
|
"grad_norm": 3.953125,
|
||
|
|
"learning_rate": 6.4248704663212435e-06,
|
||
|
|
"loss": 1.3977,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3838383838383838,
|
||
|
|
"grad_norm": 3.875,
|
||
|
|
"learning_rate": 6.373056994818654e-06,
|
||
|
|
"loss": 1.3253,
|
||
|
|
"step": 76
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3888888888888889,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 6.321243523316063e-06,
|
||
|
|
"loss": 1.3591,
|
||
|
|
"step": 77
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3939393939393939,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 6.269430051813472e-06,
|
||
|
|
"loss": 1.3304,
|
||
|
|
"step": 78
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.398989898989899,
|
||
|
|
"grad_norm": 3.484375,
|
||
|
|
"learning_rate": 6.217616580310881e-06,
|
||
|
|
"loss": 1.2627,
|
||
|
|
"step": 79
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.40404040404040403,
|
||
|
|
"grad_norm": 4.0625,
|
||
|
|
"learning_rate": 6.165803108808291e-06,
|
||
|
|
"loss": 1.4429,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.40404040404040403,
|
||
|
|
"eval_loss": 1.397228717803955,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.2387,
|
||
|
|
"eval_samples_per_second": 46.163,
|
||
|
|
"eval_steps_per_second": 23.082,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4090909090909091,
|
||
|
|
"grad_norm": 4.09375,
|
||
|
|
"learning_rate": 6.113989637305699e-06,
|
||
|
|
"loss": 1.4821,
|
||
|
|
"step": 81
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.41414141414141414,
|
||
|
|
"grad_norm": 3.78125,
|
||
|
|
"learning_rate": 6.06217616580311e-06,
|
||
|
|
"loss": 1.3956,
|
||
|
|
"step": 82
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.41919191919191917,
|
||
|
|
"grad_norm": 3.796875,
|
||
|
|
"learning_rate": 6.0103626943005185e-06,
|
||
|
|
"loss": 1.3448,
|
||
|
|
"step": 83
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.42424242424242425,
|
||
|
|
"grad_norm": 3.671875,
|
||
|
|
"learning_rate": 5.958549222797928e-06,
|
||
|
|
"loss": 1.3221,
|
||
|
|
"step": 84
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4292929292929293,
|
||
|
|
"grad_norm": 3.984375,
|
||
|
|
"learning_rate": 5.906735751295337e-06,
|
||
|
|
"loss": 1.4765,
|
||
|
|
"step": 85
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.43434343434343436,
|
||
|
|
"grad_norm": 3.8125,
|
||
|
|
"learning_rate": 5.854922279792746e-06,
|
||
|
|
"loss": 1.4065,
|
||
|
|
"step": 86
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4393939393939394,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 5.8031088082901555e-06,
|
||
|
|
"loss": 1.4029,
|
||
|
|
"step": 87
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4444444444444444,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 5.751295336787566e-06,
|
||
|
|
"loss": 1.2915,
|
||
|
|
"step": 88
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4494949494949495,
|
||
|
|
"grad_norm": 3.984375,
|
||
|
|
"learning_rate": 5.699481865284975e-06,
|
||
|
|
"loss": 1.3394,
|
||
|
|
"step": 89
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.45454545454545453,
|
||
|
|
"grad_norm": 4.0,
|
||
|
|
"learning_rate": 5.647668393782384e-06,
|
||
|
|
"loss": 1.336,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.45454545454545453,
|
||
|
|
"eval_loss": 1.3768280744552612,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.3603,
|
||
|
|
"eval_samples_per_second": 45.281,
|
||
|
|
"eval_steps_per_second": 22.64,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4595959595959596,
|
||
|
|
"grad_norm": 3.875,
|
||
|
|
"learning_rate": 5.5958549222797934e-06,
|
||
|
|
"loss": 1.3226,
|
||
|
|
"step": 91
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.46464646464646464,
|
||
|
|
"grad_norm": 3.9375,
|
||
|
|
"learning_rate": 5.544041450777202e-06,
|
||
|
|
"loss": 1.3859,
|
||
|
|
"step": 92
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4696969696969697,
|
||
|
|
"grad_norm": 4.5,
|
||
|
|
"learning_rate": 5.492227979274611e-06,
|
||
|
|
"loss": 1.4014,
|
||
|
|
"step": 93
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.47474747474747475,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 5.440414507772021e-06,
|
||
|
|
"loss": 1.3535,
|
||
|
|
"step": 94
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4797979797979798,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 5.3886010362694305e-06,
|
||
|
|
"loss": 1.3476,
|
||
|
|
"step": 95
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.48484848484848486,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 5.33678756476684e-06,
|
||
|
|
"loss": 1.3842,
|
||
|
|
"step": 96
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4898989898989899,
|
||
|
|
"grad_norm": 3.875,
|
||
|
|
"learning_rate": 5.284974093264249e-06,
|
||
|
|
"loss": 1.3107,
|
||
|
|
"step": 97
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.494949494949495,
|
||
|
|
"grad_norm": 3.96875,
|
||
|
|
"learning_rate": 5.233160621761658e-06,
|
||
|
|
"loss": 1.2887,
|
||
|
|
"step": 98
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 5.1813471502590676e-06,
|
||
|
|
"loss": 1.3819,
|
||
|
|
"step": 99
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5050505050505051,
|
||
|
|
"grad_norm": 3.859375,
|
||
|
|
"learning_rate": 5.129533678756478e-06,
|
||
|
|
"loss": 1.363,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5050505050505051,
|
||
|
|
"eval_loss": 1.359228253364563,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.2505,
|
||
|
|
"eval_samples_per_second": 46.076,
|
||
|
|
"eval_steps_per_second": 23.038,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.51010101010101,
|
||
|
|
"grad_norm": 3.953125,
|
||
|
|
"learning_rate": 5.077720207253887e-06,
|
||
|
|
"loss": 1.3335,
|
||
|
|
"step": 101
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5151515151515151,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 5.025906735751296e-06,
|
||
|
|
"loss": 1.2918,
|
||
|
|
"step": 102
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5202020202020202,
|
||
|
|
"grad_norm": 4.4375,
|
||
|
|
"learning_rate": 4.974093264248705e-06,
|
||
|
|
"loss": 1.3342,
|
||
|
|
"step": 103
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5252525252525253,
|
||
|
|
"grad_norm": 4.21875,
|
||
|
|
"learning_rate": 4.922279792746114e-06,
|
||
|
|
"loss": 1.3497,
|
||
|
|
"step": 104
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5303030303030303,
|
||
|
|
"grad_norm": 4.28125,
|
||
|
|
"learning_rate": 4.870466321243524e-06,
|
||
|
|
"loss": 1.3082,
|
||
|
|
"step": 105
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5353535353535354,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 4.818652849740933e-06,
|
||
|
|
"loss": 1.3279,
|
||
|
|
"step": 106
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5404040404040404,
|
||
|
|
"grad_norm": 4.0,
|
||
|
|
"learning_rate": 4.766839378238342e-06,
|
||
|
|
"loss": 1.342,
|
||
|
|
"step": 107
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5454545454545454,
|
||
|
|
"grad_norm": 3.734375,
|
||
|
|
"learning_rate": 4.715025906735752e-06,
|
||
|
|
"loss": 1.2986,
|
||
|
|
"step": 108
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5505050505050505,
|
||
|
|
"grad_norm": 3.96875,
|
||
|
|
"learning_rate": 4.663212435233161e-06,
|
||
|
|
"loss": 1.4217,
|
||
|
|
"step": 109
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5555555555555556,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 4.61139896373057e-06,
|
||
|
|
"loss": 1.2404,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5555555555555556,
|
||
|
|
"eval_loss": 1.3445496559143066,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.2605,
|
||
|
|
"eval_samples_per_second": 46.003,
|
||
|
|
"eval_steps_per_second": 23.001,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5606060606060606,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 4.55958549222798e-06,
|
||
|
|
"loss": 1.3445,
|
||
|
|
"step": 111
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5656565656565656,
|
||
|
|
"grad_norm": 3.828125,
|
||
|
|
"learning_rate": 4.507772020725389e-06,
|
||
|
|
"loss": 1.3496,
|
||
|
|
"step": 112
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5707070707070707,
|
||
|
|
"grad_norm": 4.0625,
|
||
|
|
"learning_rate": 4.455958549222798e-06,
|
||
|
|
"loss": 1.2852,
|
||
|
|
"step": 113
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5757575757575758,
|
||
|
|
"grad_norm": 4.25,
|
||
|
|
"learning_rate": 4.404145077720207e-06,
|
||
|
|
"loss": 1.3657,
|
||
|
|
"step": 114
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5808080808080808,
|
||
|
|
"grad_norm": 3.625,
|
||
|
|
"learning_rate": 4.352331606217617e-06,
|
||
|
|
"loss": 1.2133,
|
||
|
|
"step": 115
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5858585858585859,
|
||
|
|
"grad_norm": 3.796875,
|
||
|
|
"learning_rate": 4.300518134715026e-06,
|
||
|
|
"loss": 1.2591,
|
||
|
|
"step": 116
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5909090909090909,
|
||
|
|
"grad_norm": 4.375,
|
||
|
|
"learning_rate": 4.248704663212436e-06,
|
||
|
|
"loss": 1.4292,
|
||
|
|
"step": 117
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5959595959595959,
|
||
|
|
"grad_norm": 4.09375,
|
||
|
|
"learning_rate": 4.1968911917098444e-06,
|
||
|
|
"loss": 1.3218,
|
||
|
|
"step": 118
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.601010101010101,
|
||
|
|
"grad_norm": 3.65625,
|
||
|
|
"learning_rate": 4.145077720207254e-06,
|
||
|
|
"loss": 1.2819,
|
||
|
|
"step": 119
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6060606060606061,
|
||
|
|
"grad_norm": 4.3125,
|
||
|
|
"learning_rate": 4.093264248704664e-06,
|
||
|
|
"loss": 1.3704,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6060606060606061,
|
||
|
|
"eval_loss": 1.3300037384033203,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.4258,
|
||
|
|
"eval_samples_per_second": 44.82,
|
||
|
|
"eval_steps_per_second": 22.41,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6111111111111112,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 4.041450777202073e-06,
|
||
|
|
"loss": 1.3604,
|
||
|
|
"step": 121
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6161616161616161,
|
||
|
|
"grad_norm": 3.890625,
|
||
|
|
"learning_rate": 3.989637305699482e-06,
|
||
|
|
"loss": 1.3483,
|
||
|
|
"step": 122
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6212121212121212,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 3.937823834196892e-06,
|
||
|
|
"loss": 1.2874,
|
||
|
|
"step": 123
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6262626262626263,
|
||
|
|
"grad_norm": 4.25,
|
||
|
|
"learning_rate": 3.886010362694301e-06,
|
||
|
|
"loss": 1.2912,
|
||
|
|
"step": 124
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6313131313131313,
|
||
|
|
"grad_norm": 4.0625,
|
||
|
|
"learning_rate": 3.83419689119171e-06,
|
||
|
|
"loss": 1.3282,
|
||
|
|
"step": 125
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6363636363636364,
|
||
|
|
"grad_norm": 4.65625,
|
||
|
|
"learning_rate": 3.7823834196891194e-06,
|
||
|
|
"loss": 1.4102,
|
||
|
|
"step": 126
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6414141414141414,
|
||
|
|
"grad_norm": 3.859375,
|
||
|
|
"learning_rate": 3.7305699481865287e-06,
|
||
|
|
"loss": 1.3152,
|
||
|
|
"step": 127
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6464646464646465,
|
||
|
|
"grad_norm": 3.828125,
|
||
|
|
"learning_rate": 3.678756476683938e-06,
|
||
|
|
"loss": 1.283,
|
||
|
|
"step": 128
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6515151515151515,
|
||
|
|
"grad_norm": 4.09375,
|
||
|
|
"learning_rate": 3.6269430051813476e-06,
|
||
|
|
"loss": 1.2783,
|
||
|
|
"step": 129
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6565656565656566,
|
||
|
|
"grad_norm": 4.0,
|
||
|
|
"learning_rate": 3.575129533678757e-06,
|
||
|
|
"loss": 1.3097,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6565656565656566,
|
||
|
|
"eval_loss": 1.3192509412765503,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.4215,
|
||
|
|
"eval_samples_per_second": 44.849,
|
||
|
|
"eval_steps_per_second": 22.425,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6616161616161617,
|
||
|
|
"grad_norm": 4.8125,
|
||
|
|
"learning_rate": 3.5233160621761657e-06,
|
||
|
|
"loss": 1.337,
|
||
|
|
"step": 131
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6666666666666666,
|
||
|
|
"grad_norm": 4.15625,
|
||
|
|
"learning_rate": 3.4715025906735754e-06,
|
||
|
|
"loss": 1.337,
|
||
|
|
"step": 132
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6717171717171717,
|
||
|
|
"grad_norm": 4.28125,
|
||
|
|
"learning_rate": 3.4196891191709847e-06,
|
||
|
|
"loss": 1.2899,
|
||
|
|
"step": 133
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6767676767676768,
|
||
|
|
"grad_norm": 4.15625,
|
||
|
|
"learning_rate": 3.367875647668394e-06,
|
||
|
|
"loss": 1.3355,
|
||
|
|
"step": 134
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6818181818181818,
|
||
|
|
"grad_norm": 4.0,
|
||
|
|
"learning_rate": 3.3160621761658036e-06,
|
||
|
|
"loss": 1.261,
|
||
|
|
"step": 135
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6868686868686869,
|
||
|
|
"grad_norm": 3.984375,
|
||
|
|
"learning_rate": 3.2642487046632125e-06,
|
||
|
|
"loss": 1.2363,
|
||
|
|
"step": 136
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6919191919191919,
|
||
|
|
"grad_norm": 4.46875,
|
||
|
|
"learning_rate": 3.2124352331606218e-06,
|
||
|
|
"loss": 1.2569,
|
||
|
|
"step": 137
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.696969696969697,
|
||
|
|
"grad_norm": 3.96875,
|
||
|
|
"learning_rate": 3.1606217616580314e-06,
|
||
|
|
"loss": 1.2929,
|
||
|
|
"step": 138
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.702020202020202,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 3.1088082901554407e-06,
|
||
|
|
"loss": 1.3791,
|
||
|
|
"step": 139
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7070707070707071,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 3.0569948186528495e-06,
|
||
|
|
"loss": 1.2555,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7070707070707071,
|
||
|
|
"eval_loss": 1.3109263181686401,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.3328,
|
||
|
|
"eval_samples_per_second": 45.478,
|
||
|
|
"eval_steps_per_second": 22.739,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7121212121212122,
|
||
|
|
"grad_norm": 4.4375,
|
||
|
|
"learning_rate": 3.0051813471502592e-06,
|
||
|
|
"loss": 1.3653,
|
||
|
|
"step": 141
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7171717171717171,
|
||
|
|
"grad_norm": 3.953125,
|
||
|
|
"learning_rate": 2.9533678756476685e-06,
|
||
|
|
"loss": 1.3375,
|
||
|
|
"step": 142
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7222222222222222,
|
||
|
|
"grad_norm": 3.671875,
|
||
|
|
"learning_rate": 2.9015544041450778e-06,
|
||
|
|
"loss": 1.2255,
|
||
|
|
"step": 143
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7272727272727273,
|
||
|
|
"grad_norm": 3.9375,
|
||
|
|
"learning_rate": 2.8497409326424875e-06,
|
||
|
|
"loss": 1.2816,
|
||
|
|
"step": 144
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7323232323232324,
|
||
|
|
"grad_norm": 3.90625,
|
||
|
|
"learning_rate": 2.7979274611398967e-06,
|
||
|
|
"loss": 1.1868,
|
||
|
|
"step": 145
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7373737373737373,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 2.7461139896373056e-06,
|
||
|
|
"loss": 1.363,
|
||
|
|
"step": 146
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7424242424242424,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 2.6943005181347152e-06,
|
||
|
|
"loss": 1.2757,
|
||
|
|
"step": 147
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7474747474747475,
|
||
|
|
"grad_norm": 4.34375,
|
||
|
|
"learning_rate": 2.6424870466321245e-06,
|
||
|
|
"loss": 1.2534,
|
||
|
|
"step": 148
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7525252525252525,
|
||
|
|
"grad_norm": 3.921875,
|
||
|
|
"learning_rate": 2.5906735751295338e-06,
|
||
|
|
"loss": 1.3228,
|
||
|
|
"step": 149
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7575757575757576,
|
||
|
|
"grad_norm": 3.8125,
|
||
|
|
"learning_rate": 2.5388601036269435e-06,
|
||
|
|
"loss": 1.2467,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7575757575757576,
|
||
|
|
"eval_loss": 1.304114818572998,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.371,
|
||
|
|
"eval_samples_per_second": 45.205,
|
||
|
|
"eval_steps_per_second": 22.602,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7626262626262627,
|
||
|
|
"grad_norm": 4.375,
|
||
|
|
"learning_rate": 2.4870466321243523e-06,
|
||
|
|
"loss": 1.2795,
|
||
|
|
"step": 151
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7676767676767676,
|
||
|
|
"grad_norm": 4.59375,
|
||
|
|
"learning_rate": 2.435233160621762e-06,
|
||
|
|
"loss": 1.2662,
|
||
|
|
"step": 152
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7727272727272727,
|
||
|
|
"grad_norm": 4.5,
|
||
|
|
"learning_rate": 2.383419689119171e-06,
|
||
|
|
"loss": 1.3531,
|
||
|
|
"step": 153
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7777777777777778,
|
||
|
|
"grad_norm": 3.734375,
|
||
|
|
"learning_rate": 2.3316062176165805e-06,
|
||
|
|
"loss": 1.2902,
|
||
|
|
"step": 154
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7828282828282829,
|
||
|
|
"grad_norm": 3.984375,
|
||
|
|
"learning_rate": 2.27979274611399e-06,
|
||
|
|
"loss": 1.3014,
|
||
|
|
"step": 155
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7878787878787878,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 2.227979274611399e-06,
|
||
|
|
"loss": 1.3647,
|
||
|
|
"step": 156
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7929292929292929,
|
||
|
|
"grad_norm": 4.21875,
|
||
|
|
"learning_rate": 2.1761658031088083e-06,
|
||
|
|
"loss": 1.271,
|
||
|
|
"step": 157
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.797979797979798,
|
||
|
|
"grad_norm": 3.75,
|
||
|
|
"learning_rate": 2.124352331606218e-06,
|
||
|
|
"loss": 1.278,
|
||
|
|
"step": 158
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.803030303030303,
|
||
|
|
"grad_norm": 4.59375,
|
||
|
|
"learning_rate": 2.072538860103627e-06,
|
||
|
|
"loss": 1.3744,
|
||
|
|
"step": 159
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8080808080808081,
|
||
|
|
"grad_norm": 4.25,
|
||
|
|
"learning_rate": 2.0207253886010365e-06,
|
||
|
|
"loss": 1.2584,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8080808080808081,
|
||
|
|
"eval_loss": 1.2974194288253784,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.2253,
|
||
|
|
"eval_samples_per_second": 46.263,
|
||
|
|
"eval_steps_per_second": 23.131,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8131313131313131,
|
||
|
|
"grad_norm": 4.15625,
|
||
|
|
"learning_rate": 1.968911917098446e-06,
|
||
|
|
"loss": 1.3147,
|
||
|
|
"step": 161
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8181818181818182,
|
||
|
|
"grad_norm": 4.3125,
|
||
|
|
"learning_rate": 1.917098445595855e-06,
|
||
|
|
"loss": 1.3394,
|
||
|
|
"step": 162
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8232323232323232,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 1.8652849740932643e-06,
|
||
|
|
"loss": 1.2421,
|
||
|
|
"step": 163
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8282828282828283,
|
||
|
|
"grad_norm": 4.59375,
|
||
|
|
"learning_rate": 1.8134715025906738e-06,
|
||
|
|
"loss": 1.264,
|
||
|
|
"step": 164
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8333333333333334,
|
||
|
|
"grad_norm": 4.15625,
|
||
|
|
"learning_rate": 1.7616580310880829e-06,
|
||
|
|
"loss": 1.3201,
|
||
|
|
"step": 165
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8383838383838383,
|
||
|
|
"grad_norm": 5.125,
|
||
|
|
"learning_rate": 1.7098445595854923e-06,
|
||
|
|
"loss": 1.3013,
|
||
|
|
"step": 166
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8434343434343434,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 1.6580310880829018e-06,
|
||
|
|
"loss": 1.3202,
|
||
|
|
"step": 167
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8484848484848485,
|
||
|
|
"grad_norm": 4.3125,
|
||
|
|
"learning_rate": 1.6062176165803109e-06,
|
||
|
|
"loss": 1.3266,
|
||
|
|
"step": 168
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8535353535353535,
|
||
|
|
"grad_norm": 5.03125,
|
||
|
|
"learning_rate": 1.5544041450777204e-06,
|
||
|
|
"loss": 1.327,
|
||
|
|
"step": 169
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8585858585858586,
|
||
|
|
"grad_norm": 3.859375,
|
||
|
|
"learning_rate": 1.5025906735751296e-06,
|
||
|
|
"loss": 1.2611,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8585858585858586,
|
||
|
|
"eval_loss": 1.2933073043823242,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.2679,
|
||
|
|
"eval_samples_per_second": 45.948,
|
||
|
|
"eval_steps_per_second": 22.974,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8636363636363636,
|
||
|
|
"grad_norm": 4.875,
|
||
|
|
"learning_rate": 1.4507772020725389e-06,
|
||
|
|
"loss": 1.3112,
|
||
|
|
"step": 171
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8686868686868687,
|
||
|
|
"grad_norm": 4.28125,
|
||
|
|
"learning_rate": 1.3989637305699484e-06,
|
||
|
|
"loss": 1.3099,
|
||
|
|
"step": 172
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8737373737373737,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 1.3471502590673576e-06,
|
||
|
|
"loss": 1.3223,
|
||
|
|
"step": 173
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8787878787878788,
|
||
|
|
"grad_norm": 4.25,
|
||
|
|
"learning_rate": 1.2953367875647669e-06,
|
||
|
|
"loss": 1.2892,
|
||
|
|
"step": 174
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8838383838383839,
|
||
|
|
"grad_norm": 4.90625,
|
||
|
|
"learning_rate": 1.2435233160621762e-06,
|
||
|
|
"loss": 1.2813,
|
||
|
|
"step": 175
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8888888888888888,
|
||
|
|
"grad_norm": 4.25,
|
||
|
|
"learning_rate": 1.1917098445595854e-06,
|
||
|
|
"loss": 1.3404,
|
||
|
|
"step": 176
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8939393939393939,
|
||
|
|
"grad_norm": 4.1875,
|
||
|
|
"learning_rate": 1.139896373056995e-06,
|
||
|
|
"loss": 1.3433,
|
||
|
|
"step": 177
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.898989898989899,
|
||
|
|
"grad_norm": 4.03125,
|
||
|
|
"learning_rate": 1.0880829015544042e-06,
|
||
|
|
"loss": 1.3131,
|
||
|
|
"step": 178
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9040404040404041,
|
||
|
|
"grad_norm": 3.71875,
|
||
|
|
"learning_rate": 1.0362694300518134e-06,
|
||
|
|
"loss": 1.2493,
|
||
|
|
"step": 179
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9090909090909091,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 9.84455958549223e-07,
|
||
|
|
"loss": 1.2856,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9090909090909091,
|
||
|
|
"eval_loss": 1.2907087802886963,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.4924,
|
||
|
|
"eval_samples_per_second": 44.36,
|
||
|
|
"eval_steps_per_second": 22.18,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9141414141414141,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 9.326424870466322e-07,
|
||
|
|
"loss": 1.2629,
|
||
|
|
"step": 181
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9191919191919192,
|
||
|
|
"grad_norm": 4.53125,
|
||
|
|
"learning_rate": 8.808290155440414e-07,
|
||
|
|
"loss": 1.376,
|
||
|
|
"step": 182
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9242424242424242,
|
||
|
|
"grad_norm": 3.953125,
|
||
|
|
"learning_rate": 8.290155440414509e-07,
|
||
|
|
"loss": 1.2141,
|
||
|
|
"step": 183
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9292929292929293,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 7.772020725388602e-07,
|
||
|
|
"loss": 1.2419,
|
||
|
|
"step": 184
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9343434343434344,
|
||
|
|
"grad_norm": 4.09375,
|
||
|
|
"learning_rate": 7.253886010362694e-07,
|
||
|
|
"loss": 1.2521,
|
||
|
|
"step": 185
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9393939393939394,
|
||
|
|
"grad_norm": 4.21875,
|
||
|
|
"learning_rate": 6.735751295336788e-07,
|
||
|
|
"loss": 1.2741,
|
||
|
|
"step": 186
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9444444444444444,
|
||
|
|
"grad_norm": 4.375,
|
||
|
|
"learning_rate": 6.217616580310881e-07,
|
||
|
|
"loss": 1.2866,
|
||
|
|
"step": 187
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9494949494949495,
|
||
|
|
"grad_norm": 4.5,
|
||
|
|
"learning_rate": 5.699481865284974e-07,
|
||
|
|
"loss": 1.2895,
|
||
|
|
"step": 188
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9545454545454546,
|
||
|
|
"grad_norm": 4.15625,
|
||
|
|
"learning_rate": 5.181347150259067e-07,
|
||
|
|
"loss": 1.2494,
|
||
|
|
"step": 189
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9595959595959596,
|
||
|
|
"grad_norm": 4.0625,
|
||
|
|
"learning_rate": 4.663212435233161e-07,
|
||
|
|
"loss": 1.3102,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9595959595959596,
|
||
|
|
"eval_loss": 1.290001392364502,
|
||
|
|
"eval_model_preparation_time": 0.018,
|
||
|
|
"eval_runtime": 6.3975,
|
||
|
|
"eval_samples_per_second": 45.018,
|
||
|
|
"eval_steps_per_second": 22.509,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9646464646464646,
|
||
|
|
"grad_norm": 4.15625,
|
||
|
|
"learning_rate": 4.1450777202072546e-07,
|
||
|
|
"loss": 1.2886,
|
||
|
|
"step": 191
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9696969696969697,
|
||
|
|
"grad_norm": 3.9375,
|
||
|
|
"learning_rate": 3.626943005181347e-07,
|
||
|
|
"loss": 1.2093,
|
||
|
|
"step": 192
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9747474747474747,
|
||
|
|
"grad_norm": 4.46875,
|
||
|
|
"learning_rate": 3.1088082901554404e-07,
|
||
|
|
"loss": 1.2681,
|
||
|
|
"step": 193
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9797979797979798,
|
||
|
|
"grad_norm": 3.84375,
|
||
|
|
"learning_rate": 2.5906735751295336e-07,
|
||
|
|
"loss": 1.2424,
|
||
|
|
"step": 194
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9848484848484849,
|
||
|
|
"grad_norm": 4.5625,
|
||
|
|
"learning_rate": 2.0725388601036273e-07,
|
||
|
|
"loss": 1.2644,
|
||
|
|
"step": 195
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.98989898989899,
|
||
|
|
"grad_norm": 3.84375,
|
||
|
|
"learning_rate": 1.5544041450777202e-07,
|
||
|
|
"loss": 1.2725,
|
||
|
|
"step": 196
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9949494949494949,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 1.0362694300518136e-07,
|
||
|
|
"loss": 1.2174,
|
||
|
|
"step": 197
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 4.28125,
|
||
|
|
"learning_rate": 5.181347150259068e-08,
|
||
|
|
"loss": 1.2857,
|
||
|
|
"step": 198
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 1,
|
||
|
|
"max_steps": 198,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"save_steps": 5000,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 3.443359132709683e+16,
|
||
|
|
"train_batch_size": 2,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|