llava-v1.5-7b-selective-23k-lora / trainer_state.json
yuezih
upload
dfa8781
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"global_step": 182,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 3.3333333333333335e-05,
"loss": 1.0555,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 6.666666666666667e-05,
"loss": 1.0519,
"step": 2
},
{
"epoch": 0.02,
"learning_rate": 0.0001,
"loss": 1.0504,
"step": 3
},
{
"epoch": 0.02,
"learning_rate": 0.00013333333333333334,
"loss": 1.0691,
"step": 4
},
{
"epoch": 0.03,
"learning_rate": 0.0001666666666666667,
"loss": 1.0723,
"step": 5
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0722,
"step": 6
},
{
"epoch": 0.04,
"learning_rate": 0.00019998406937250034,
"loss": 1.0411,
"step": 7
},
{
"epoch": 0.04,
"learning_rate": 0.0001999362825656992,
"loss": 1.0547,
"step": 8
},
{
"epoch": 0.05,
"learning_rate": 0.0001998566548050729,
"loss": 1.0558,
"step": 9
},
{
"epoch": 0.05,
"learning_rate": 0.00019974521146102537,
"loss": 1.0537,
"step": 10
},
{
"epoch": 0.06,
"learning_rate": 0.0001996019880408046,
"loss": 1.054,
"step": 11
},
{
"epoch": 0.07,
"learning_rate": 0.00019942703017718975,
"loss": 1.0643,
"step": 12
},
{
"epoch": 0.07,
"learning_rate": 0.00019922039361395185,
"loss": 1.0553,
"step": 13
},
{
"epoch": 0.08,
"learning_rate": 0.0001989821441880933,
"loss": 1.0878,
"step": 14
},
{
"epoch": 0.08,
"learning_rate": 0.00019871235780887113,
"loss": 1.0588,
"step": 15
},
{
"epoch": 0.09,
"learning_rate": 0.0001984111204336116,
"loss": 1.0319,
"step": 16
},
{
"epoch": 0.09,
"learning_rate": 0.00019807852804032305,
"loss": 1.0609,
"step": 17
},
{
"epoch": 0.1,
"learning_rate": 0.00019771468659711595,
"loss": 1.0511,
"step": 18
},
{
"epoch": 0.1,
"learning_rate": 0.00019731971202844036,
"loss": 1.0569,
"step": 19
},
{
"epoch": 0.11,
"learning_rate": 0.00019689373017815073,
"loss": 1.0452,
"step": 20
},
{
"epoch": 0.12,
"learning_rate": 0.00019643687676941068,
"loss": 1.0627,
"step": 21
},
{
"epoch": 0.12,
"learning_rate": 0.00019594929736144976,
"loss": 1.0351,
"step": 22
},
{
"epoch": 0.13,
"learning_rate": 0.0001954311473031864,
"loss": 1.0515,
"step": 23
},
{
"epoch": 0.13,
"learning_rate": 0.00019488259168373197,
"loss": 1.0502,
"step": 24
},
{
"epoch": 0.14,
"learning_rate": 0.00019430380527979123,
"loss": 1.058,
"step": 25
},
{
"epoch": 0.14,
"learning_rate": 0.0001936949724999762,
"loss": 1.0678,
"step": 26
},
{
"epoch": 0.15,
"learning_rate": 0.00019305628732605137,
"loss": 1.0382,
"step": 27
},
{
"epoch": 0.15,
"learning_rate": 0.0001923879532511287,
"loss": 1.0493,
"step": 28
},
{
"epoch": 0.16,
"learning_rate": 0.00019169018321483198,
"loss": 1.0555,
"step": 29
},
{
"epoch": 0.16,
"learning_rate": 0.00019096319953545185,
"loss": 1.0413,
"step": 30
},
{
"epoch": 0.17,
"learning_rate": 0.00019020723383911215,
"loss": 1.0521,
"step": 31
},
{
"epoch": 0.18,
"learning_rate": 0.00018942252698597113,
"loss": 1.0523,
"step": 32
},
{
"epoch": 0.18,
"learning_rate": 0.00018860932899348028,
"loss": 1.0782,
"step": 33
},
{
"epoch": 0.19,
"learning_rate": 0.00018776789895672558,
"loss": 1.0492,
"step": 34
},
{
"epoch": 0.19,
"learning_rate": 0.00018689850496587674,
"loss": 1.0628,
"step": 35
},
{
"epoch": 0.2,
"learning_rate": 0.00018600142402077006,
"loss": 1.0818,
"step": 36
},
{
"epoch": 0.2,
"learning_rate": 0.0001850769419426531,
"loss": 1.0533,
"step": 37
},
{
"epoch": 0.21,
"learning_rate": 0.00018412535328311814,
"loss": 1.061,
"step": 38
},
{
"epoch": 0.21,
"learning_rate": 0.00018314696123025454,
"loss": 1.0414,
"step": 39
},
{
"epoch": 0.22,
"learning_rate": 0.00018214207751204918,
"loss": 1.0515,
"step": 40
},
{
"epoch": 0.23,
"learning_rate": 0.0001811110222970659,
"loss": 1.0519,
"step": 41
},
{
"epoch": 0.23,
"learning_rate": 0.00018005412409243606,
"loss": 1.0528,
"step": 42
},
{
"epoch": 0.24,
"learning_rate": 0.0001789717196391916,
"loss": 1.046,
"step": 43
},
{
"epoch": 0.24,
"learning_rate": 0.00017786415380497553,
"loss": 1.0489,
"step": 44
},
{
"epoch": 0.25,
"learning_rate": 0.00017673177947416258,
"loss": 1.0856,
"step": 45
},
{
"epoch": 0.25,
"learning_rate": 0.00017557495743542585,
"loss": 1.0508,
"step": 46
},
{
"epoch": 0.26,
"learning_rate": 0.00017439405626678496,
"loss": 1.0709,
"step": 47
},
{
"epoch": 0.26,
"learning_rate": 0.00017318945221817255,
"loss": 1.0621,
"step": 48
},
{
"epoch": 0.27,
"learning_rate": 0.00017196152909155628,
"loss": 1.074,
"step": 49
},
{
"epoch": 0.27,
"learning_rate": 0.00017071067811865476,
"loss": 1.0546,
"step": 50
},
{
"epoch": 0.28,
"learning_rate": 0.00016943729783628608,
"loss": 1.0616,
"step": 51
},
{
"epoch": 0.29,
"learning_rate": 0.00016814179395938913,
"loss": 1.0568,
"step": 52
},
{
"epoch": 0.29,
"learning_rate": 0.00016682457925175763,
"loss": 1.0421,
"step": 53
},
{
"epoch": 0.3,
"learning_rate": 0.00016548607339452853,
"loss": 1.0618,
"step": 54
},
{
"epoch": 0.3,
"learning_rate": 0.0001641267028524661,
"loss": 1.0637,
"step": 55
},
{
"epoch": 0.31,
"learning_rate": 0.0001627469007380852,
"loss": 1.023,
"step": 56
},
{
"epoch": 0.31,
"learning_rate": 0.00016134710667365596,
"loss": 1.0577,
"step": 57
},
{
"epoch": 0.32,
"learning_rate": 0.0001599277666511347,
"loss": 1.0542,
"step": 58
},
{
"epoch": 0.32,
"learning_rate": 0.0001584893328900653,
"loss": 1.0607,
"step": 59
},
{
"epoch": 0.33,
"learning_rate": 0.0001570322636934964,
"loss": 1.0664,
"step": 60
},
{
"epoch": 0.34,
"learning_rate": 0.00015555702330196023,
"loss": 1.0557,
"step": 61
},
{
"epoch": 0.34,
"learning_rate": 0.00015406408174555976,
"loss": 1.0704,
"step": 62
},
{
"epoch": 0.35,
"learning_rate": 0.00015255391469421128,
"loss": 1.0444,
"step": 63
},
{
"epoch": 0.35,
"learning_rate": 0.00015102700330609,
"loss": 1.0549,
"step": 64
},
{
"epoch": 0.36,
"learning_rate": 0.00014948383407432678,
"loss": 1.0403,
"step": 65
},
{
"epoch": 0.36,
"learning_rate": 0.0001479248986720057,
"loss": 1.0352,
"step": 66
},
{
"epoch": 0.37,
"learning_rate": 0.00014635069379551055,
"loss": 1.0534,
"step": 67
},
{
"epoch": 0.37,
"learning_rate": 0.00014476172100627127,
"loss": 1.0444,
"step": 68
},
{
"epoch": 0.38,
"learning_rate": 0.00014315848657096004,
"loss": 1.0637,
"step": 69
},
{
"epoch": 0.38,
"learning_rate": 0.00014154150130018866,
"loss": 1.0581,
"step": 70
},
{
"epoch": 0.39,
"learning_rate": 0.00013991128038575741,
"loss": 1.0488,
"step": 71
},
{
"epoch": 0.4,
"learning_rate": 0.000138268343236509,
"loss": 1.076,
"step": 72
},
{
"epoch": 0.4,
"learning_rate": 0.00013661321331283796,
"loss": 1.0295,
"step": 73
},
{
"epoch": 0.41,
"learning_rate": 0.00013494641795990986,
"loss": 1.0569,
"step": 74
},
{
"epoch": 0.41,
"learning_rate": 0.00013326848823964243,
"loss": 1.0782,
"step": 75
},
{
"epoch": 0.42,
"learning_rate": 0.0001315799587615025,
"loss": 1.07,
"step": 76
},
{
"epoch": 0.42,
"learning_rate": 0.00012988136751217291,
"loss": 1.0647,
"step": 77
},
{
"epoch": 0.43,
"learning_rate": 0.00012817325568414297,
"loss": 1.0527,
"step": 78
},
{
"epoch": 0.43,
"learning_rate": 0.0001264561675032779,
"loss": 1.052,
"step": 79
},
{
"epoch": 0.44,
"learning_rate": 0.00012473065005542155,
"loss": 1.0569,
"step": 80
},
{
"epoch": 0.45,
"learning_rate": 0.00012299725311208808,
"loss": 1.0412,
"step": 81
},
{
"epoch": 0.45,
"learning_rate": 0.00012125652895529766,
"loss": 1.0668,
"step": 82
},
{
"epoch": 0.46,
"learning_rate": 0.00011950903220161285,
"loss": 1.0387,
"step": 83
},
{
"epoch": 0.46,
"learning_rate": 0.00011775531962543036,
"loss": 1.0593,
"step": 84
},
{
"epoch": 0.47,
"learning_rate": 0.00011599594998158602,
"loss": 1.0429,
"step": 85
},
{
"epoch": 0.47,
"learning_rate": 0.00011423148382732853,
"loss": 1.0486,
"step": 86
},
{
"epoch": 0.48,
"learning_rate": 0.0001124624833437186,
"loss": 1.0505,
"step": 87
},
{
"epoch": 0.48,
"learning_rate": 0.00011068951215651132,
"loss": 1.0576,
"step": 88
},
{
"epoch": 0.49,
"learning_rate": 0.0001089131351565776,
"loss": 1.0575,
"step": 89
},
{
"epoch": 0.49,
"learning_rate": 0.00010713391831992323,
"loss": 1.054,
"step": 90
},
{
"epoch": 0.5,
"learning_rate": 0.00010535242852736151,
"loss": 1.0454,
"step": 91
},
{
"epoch": 0.51,
"learning_rate": 0.00010356923338389806,
"loss": 1.043,
"step": 92
},
{
"epoch": 0.51,
"learning_rate": 0.0001017849010378846,
"loss": 1.0382,
"step": 93
},
{
"epoch": 0.52,
"learning_rate": 0.0001,
"loss": 1.0645,
"step": 94
},
{
"epoch": 0.52,
"learning_rate": 9.821509896211539e-05,
"loss": 1.0499,
"step": 95
},
{
"epoch": 0.53,
"learning_rate": 9.643076661610196e-05,
"loss": 1.0205,
"step": 96
},
{
"epoch": 0.53,
"learning_rate": 9.464757147263849e-05,
"loss": 1.0618,
"step": 97
},
{
"epoch": 0.54,
"learning_rate": 9.286608168007678e-05,
"loss": 1.0752,
"step": 98
},
{
"epoch": 0.54,
"learning_rate": 9.108686484342241e-05,
"loss": 1.0555,
"step": 99
},
{
"epoch": 0.55,
"learning_rate": 8.931048784348875e-05,
"loss": 1.0637,
"step": 100
},
{
"epoch": 0.55,
"learning_rate": 8.753751665628141e-05,
"loss": 1.0502,
"step": 101
},
{
"epoch": 0.56,
"learning_rate": 8.57685161726715e-05,
"loss": 1.0569,
"step": 102
},
{
"epoch": 0.57,
"learning_rate": 8.400405001841399e-05,
"loss": 1.0415,
"step": 103
},
{
"epoch": 0.57,
"learning_rate": 8.224468037456969e-05,
"loss": 1.0497,
"step": 104
},
{
"epoch": 0.58,
"learning_rate": 8.049096779838719e-05,
"loss": 1.0516,
"step": 105
},
{
"epoch": 0.58,
"learning_rate": 7.874347104470234e-05,
"loss": 1.0502,
"step": 106
},
{
"epoch": 0.59,
"learning_rate": 7.700274688791196e-05,
"loss": 1.0394,
"step": 107
},
{
"epoch": 0.59,
"learning_rate": 7.526934994457844e-05,
"loss": 1.0585,
"step": 108
},
{
"epoch": 0.6,
"learning_rate": 7.354383249672212e-05,
"loss": 1.0485,
"step": 109
},
{
"epoch": 0.6,
"learning_rate": 7.182674431585704e-05,
"loss": 1.0539,
"step": 110
},
{
"epoch": 0.61,
"learning_rate": 7.011863248782711e-05,
"loss": 1.0591,
"step": 111
},
{
"epoch": 0.62,
"learning_rate": 6.842004123849752e-05,
"loss": 1.0528,
"step": 112
},
{
"epoch": 0.62,
"learning_rate": 6.673151176035762e-05,
"loss": 1.0342,
"step": 113
},
{
"epoch": 0.63,
"learning_rate": 6.505358204009017e-05,
"loss": 1.0435,
"step": 114
},
{
"epoch": 0.63,
"learning_rate": 6.338678668716209e-05,
"loss": 1.0469,
"step": 115
},
{
"epoch": 0.64,
"learning_rate": 6.173165676349103e-05,
"loss": 1.0297,
"step": 116
},
{
"epoch": 0.64,
"learning_rate": 6.008871961424258e-05,
"loss": 1.0337,
"step": 117
},
{
"epoch": 0.65,
"learning_rate": 5.845849869981137e-05,
"loss": 1.0793,
"step": 118
},
{
"epoch": 0.65,
"learning_rate": 5.684151342903992e-05,
"loss": 1.0456,
"step": 119
},
{
"epoch": 0.66,
"learning_rate": 5.5238278993728756e-05,
"loss": 1.0519,
"step": 120
},
{
"epoch": 0.66,
"learning_rate": 5.364930620448946e-05,
"loss": 1.0143,
"step": 121
},
{
"epoch": 0.67,
"learning_rate": 5.207510132799436e-05,
"loss": 1.0377,
"step": 122
},
{
"epoch": 0.68,
"learning_rate": 5.051616592567323e-05,
"loss": 1.044,
"step": 123
},
{
"epoch": 0.68,
"learning_rate": 4.8972996693910054e-05,
"loss": 1.0583,
"step": 124
},
{
"epoch": 0.69,
"learning_rate": 4.7446085305788725e-05,
"loss": 1.0382,
"step": 125
},
{
"epoch": 0.69,
"learning_rate": 4.593591825444028e-05,
"loss": 1.0654,
"step": 126
},
{
"epoch": 0.7,
"learning_rate": 4.444297669803981e-05,
"loss": 1.0555,
"step": 127
},
{
"epoch": 0.7,
"learning_rate": 4.296773630650358e-05,
"loss": 1.0428,
"step": 128
},
{
"epoch": 0.71,
"learning_rate": 4.15106671099347e-05,
"loss": 1.0236,
"step": 129
},
{
"epoch": 0.71,
"learning_rate": 4.007223334886531e-05,
"loss": 1.0258,
"step": 130
},
{
"epoch": 0.72,
"learning_rate": 3.865289332634407e-05,
"loss": 1.0655,
"step": 131
},
{
"epoch": 0.73,
"learning_rate": 3.725309926191479e-05,
"loss": 1.0677,
"step": 132
},
{
"epoch": 0.73,
"learning_rate": 3.5873297147533915e-05,
"loss": 1.0256,
"step": 133
},
{
"epoch": 0.74,
"learning_rate": 3.45139266054715e-05,
"loss": 1.0398,
"step": 134
},
{
"epoch": 0.74,
"learning_rate": 3.3175420748242406e-05,
"loss": 1.0504,
"step": 135
},
{
"epoch": 0.75,
"learning_rate": 3.185820604061088e-05,
"loss": 1.0413,
"step": 136
},
{
"epoch": 0.75,
"learning_rate": 3.056270216371395e-05,
"loss": 1.0678,
"step": 137
},
{
"epoch": 0.76,
"learning_rate": 2.9289321881345254e-05,
"loss": 1.0241,
"step": 138
},
{
"epoch": 0.76,
"learning_rate": 2.8038470908443714e-05,
"loss": 1.0376,
"step": 139
},
{
"epoch": 0.77,
"learning_rate": 2.681054778182748e-05,
"loss": 1.0484,
"step": 140
},
{
"epoch": 0.77,
"learning_rate": 2.5605943733215042e-05,
"loss": 1.0663,
"step": 141
},
{
"epoch": 0.78,
"learning_rate": 2.4425042564574184e-05,
"loss": 1.0423,
"step": 142
},
{
"epoch": 0.79,
"learning_rate": 2.3268220525837437e-05,
"loss": 1.064,
"step": 143
},
{
"epoch": 0.79,
"learning_rate": 2.2135846195024513e-05,
"loss": 1.0541,
"step": 144
},
{
"epoch": 0.8,
"learning_rate": 2.1028280360808407e-05,
"loss": 1.0343,
"step": 145
},
{
"epoch": 0.8,
"learning_rate": 1.994587590756397e-05,
"loss": 1.0331,
"step": 146
},
{
"epoch": 0.81,
"learning_rate": 1.8888977702934085e-05,
"loss": 1.0457,
"step": 147
},
{
"epoch": 0.81,
"learning_rate": 1.7857922487950874e-05,
"loss": 1.07,
"step": 148
},
{
"epoch": 0.82,
"learning_rate": 1.6853038769745467e-05,
"loss": 1.0616,
"step": 149
},
{
"epoch": 0.82,
"learning_rate": 1.587464671688187e-05,
"loss": 1.0651,
"step": 150
},
{
"epoch": 0.83,
"learning_rate": 1.4923058057346929e-05,
"loss": 1.0497,
"step": 151
},
{
"epoch": 0.84,
"learning_rate": 1.3998575979229944e-05,
"loss": 1.0409,
"step": 152
},
{
"epoch": 0.84,
"learning_rate": 1.3101495034123313e-05,
"loss": 1.04,
"step": 153
},
{
"epoch": 0.85,
"learning_rate": 1.2232101043274436e-05,
"loss": 1.0541,
"step": 154
},
{
"epoch": 0.85,
"learning_rate": 1.139067100651976e-05,
"loss": 1.0453,
"step": 155
},
{
"epoch": 0.86,
"learning_rate": 1.057747301402887e-05,
"loss": 1.0602,
"step": 156
},
{
"epoch": 0.86,
"learning_rate": 9.792766160887868e-06,
"loss": 1.05,
"step": 157
},
{
"epoch": 0.87,
"learning_rate": 9.036800464548157e-06,
"loss": 1.0546,
"step": 158
},
{
"epoch": 0.87,
"learning_rate": 8.309816785168034e-06,
"loss": 1.0632,
"step": 159
},
{
"epoch": 0.88,
"learning_rate": 7.612046748871327e-06,
"loss": 1.0394,
"step": 160
},
{
"epoch": 0.88,
"learning_rate": 6.943712673948644e-06,
"loss": 1.0306,
"step": 161
},
{
"epoch": 0.89,
"learning_rate": 6.3050275000238414e-06,
"loss": 1.0287,
"step": 162
},
{
"epoch": 0.9,
"learning_rate": 5.696194720208792e-06,
"loss": 1.0399,
"step": 163
},
{
"epoch": 0.9,
"learning_rate": 5.1174083162680465e-06,
"loss": 1.0401,
"step": 164
},
{
"epoch": 0.91,
"learning_rate": 4.568852696813619e-06,
"loss": 1.0361,
"step": 165
},
{
"epoch": 0.91,
"learning_rate": 4.050702638550275e-06,
"loss": 1.0339,
"step": 166
},
{
"epoch": 0.92,
"learning_rate": 3.5631232305893046e-06,
"loss": 1.0353,
"step": 167
},
{
"epoch": 0.92,
"learning_rate": 3.1062698218492724e-06,
"loss": 1.0418,
"step": 168
},
{
"epoch": 0.93,
"learning_rate": 2.6802879715596585e-06,
"loss": 1.0463,
"step": 169
},
{
"epoch": 0.93,
"learning_rate": 2.2853134028840594e-06,
"loss": 1.0664,
"step": 170
},
{
"epoch": 0.94,
"learning_rate": 1.921471959676957e-06,
"loss": 1.055,
"step": 171
},
{
"epoch": 0.95,
"learning_rate": 1.5888795663883904e-06,
"loss": 1.033,
"step": 172
},
{
"epoch": 0.95,
"learning_rate": 1.2876421911288905e-06,
"loss": 1.0441,
"step": 173
},
{
"epoch": 0.96,
"learning_rate": 1.0178558119067315e-06,
"loss": 1.0464,
"step": 174
},
{
"epoch": 0.96,
"learning_rate": 7.796063860481595e-07,
"loss": 1.0477,
"step": 175
},
{
"epoch": 0.97,
"learning_rate": 5.729698228102653e-07,
"loss": 1.0405,
"step": 176
},
{
"epoch": 0.97,
"learning_rate": 3.9801195919541014e-07,
"loss": 1.0329,
"step": 177
},
{
"epoch": 0.98,
"learning_rate": 2.547885389746485e-07,
"loss": 1.0379,
"step": 178
},
{
"epoch": 0.98,
"learning_rate": 1.4334519492711362e-07,
"loss": 1.0373,
"step": 179
},
{
"epoch": 0.99,
"learning_rate": 6.37174343008251e-08,
"loss": 1.0709,
"step": 180
},
{
"epoch": 0.99,
"learning_rate": 1.593062749967178e-08,
"loss": 1.0425,
"step": 181
},
{
"epoch": 1.0,
"learning_rate": 0.0,
"loss": 0.9978,
"step": 182
},
{
"epoch": 1.0,
"step": 182,
"total_flos": 94360717230080.0,
"train_loss": 1.0510600356610267,
"train_runtime": 3384.5855,
"train_samples_per_second": 6.866,
"train_steps_per_second": 0.054
}
],
"max_steps": 182,
"num_train_epochs": 1,
"total_flos": 94360717230080.0,
"trial_name": null,
"trial_params": null
}