tpoisonooo
init(project): update
bf866da
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 137,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"grad_norm": 1.890625,
"learning_rate": 2e-05,
"loss": 0.9848,
"step": 1
},
{
"epoch": 0.01,
"grad_norm": 1.953125,
"learning_rate": 4e-05,
"loss": 0.9509,
"step": 2
},
{
"epoch": 0.02,
"grad_norm": 2.53125,
"learning_rate": 6e-05,
"loss": 1.0676,
"step": 3
},
{
"epoch": 0.03,
"grad_norm": 1.953125,
"learning_rate": 8e-05,
"loss": 0.823,
"step": 4
},
{
"epoch": 0.04,
"grad_norm": 2.515625,
"learning_rate": 0.0001,
"loss": 0.6513,
"step": 5
},
{
"epoch": 0.04,
"grad_norm": 1.2421875,
"learning_rate": 0.00012,
"loss": 0.2973,
"step": 6
},
{
"epoch": 0.05,
"grad_norm": 0.51171875,
"learning_rate": 0.00014,
"loss": 0.1631,
"step": 7
},
{
"epoch": 0.06,
"grad_norm": 0.396484375,
"learning_rate": 0.00016,
"loss": 0.1212,
"step": 8
},
{
"epoch": 0.07,
"grad_norm": 2.390625,
"learning_rate": 0.00018,
"loss": 0.2939,
"step": 9
},
{
"epoch": 0.07,
"grad_norm": 0.734375,
"learning_rate": 0.0002,
"loss": 0.2165,
"step": 10
},
{
"epoch": 0.08,
"grad_norm": 0.8671875,
"learning_rate": 0.0001999694057253083,
"loss": 0.2527,
"step": 11
},
{
"epoch": 0.09,
"grad_norm": 0.1923828125,
"learning_rate": 0.00019987764162142613,
"loss": 0.1284,
"step": 12
},
{
"epoch": 0.09,
"grad_norm": 0.255859375,
"learning_rate": 0.00019972476383747748,
"loss": 0.1244,
"step": 13
},
{
"epoch": 0.1,
"grad_norm": 0.09521484375,
"learning_rate": 0.0001995108659171607,
"loss": 0.1175,
"step": 14
},
{
"epoch": 0.11,
"grad_norm": 0.27734375,
"learning_rate": 0.00019923607874151032,
"loss": 0.1832,
"step": 15
},
{
"epoch": 0.12,
"grad_norm": 0.09228515625,
"learning_rate": 0.00019890057044881306,
"loss": 0.1227,
"step": 16
},
{
"epoch": 0.12,
"grad_norm": 0.115234375,
"learning_rate": 0.00019850454633172631,
"loss": 0.1211,
"step": 17
},
{
"epoch": 0.13,
"grad_norm": 0.064453125,
"learning_rate": 0.00019804824871166255,
"loss": 0.1122,
"step": 18
},
{
"epoch": 0.14,
"grad_norm": 0.10400390625,
"learning_rate": 0.00019753195679051628,
"loss": 0.1077,
"step": 19
},
{
"epoch": 0.15,
"grad_norm": 0.051025390625,
"learning_rate": 0.00019695598647982468,
"loss": 0.104,
"step": 20
},
{
"epoch": 0.15,
"grad_norm": 0.09326171875,
"learning_rate": 0.00019632069020746572,
"loss": 0.1143,
"step": 21
},
{
"epoch": 0.16,
"grad_norm": 0.125,
"learning_rate": 0.00019562645670201276,
"loss": 0.1172,
"step": 22
},
{
"epoch": 0.17,
"grad_norm": 0.130859375,
"learning_rate": 0.00019487371075487713,
"loss": 0.1083,
"step": 23
},
{
"epoch": 0.18,
"grad_norm": 0.08349609375,
"learning_rate": 0.0001940629129603844,
"loss": 0.1051,
"step": 24
},
{
"epoch": 0.18,
"grad_norm": 0.146484375,
"learning_rate": 0.00019319455943394347,
"loss": 0.1076,
"step": 25
},
{
"epoch": 0.19,
"grad_norm": 0.09423828125,
"learning_rate": 0.00019226918150848068,
"loss": 0.0868,
"step": 26
},
{
"epoch": 0.2,
"grad_norm": 0.275390625,
"learning_rate": 0.00019128734540932495,
"loss": 0.1223,
"step": 27
},
{
"epoch": 0.2,
"grad_norm": 0.318359375,
"learning_rate": 0.00019024965190774263,
"loss": 0.1346,
"step": 28
},
{
"epoch": 0.21,
"grad_norm": 0.46484375,
"learning_rate": 0.00018915673595333444,
"loss": 0.1361,
"step": 29
},
{
"epoch": 0.22,
"grad_norm": 0.10498046875,
"learning_rate": 0.00018800926628551886,
"loss": 0.1178,
"step": 30
},
{
"epoch": 0.23,
"grad_norm": 0.1044921875,
"learning_rate": 0.00018680794502434018,
"loss": 0.1085,
"step": 31
},
{
"epoch": 0.23,
"grad_norm": 0.0634765625,
"learning_rate": 0.00018555350724085162,
"loss": 0.0985,
"step": 32
},
{
"epoch": 0.24,
"grad_norm": 0.2119140625,
"learning_rate": 0.00018424672050733576,
"loss": 0.1486,
"step": 33
},
{
"epoch": 0.25,
"grad_norm": 0.158203125,
"learning_rate": 0.00018288838442763838,
"loss": 0.1285,
"step": 34
},
{
"epoch": 0.26,
"grad_norm": 0.099609375,
"learning_rate": 0.00018147933014790244,
"loss": 0.1147,
"step": 35
},
{
"epoch": 0.26,
"grad_norm": 0.08056640625,
"learning_rate": 0.00018002041984800174,
"loss": 0.0984,
"step": 36
},
{
"epoch": 0.27,
"grad_norm": 0.07763671875,
"learning_rate": 0.0001785125462139855,
"loss": 0.1147,
"step": 37
},
{
"epoch": 0.28,
"grad_norm": 0.062255859375,
"learning_rate": 0.000176956631891857,
"loss": 0.1016,
"step": 38
},
{
"epoch": 0.28,
"grad_norm": 0.053466796875,
"learning_rate": 0.00017535362892301954,
"loss": 0.1013,
"step": 39
},
{
"epoch": 0.29,
"grad_norm": 0.0673828125,
"learning_rate": 0.0001737045181617364,
"loss": 0.1001,
"step": 40
},
{
"epoch": 0.3,
"grad_norm": 0.0908203125,
"learning_rate": 0.00017201030867496005,
"loss": 0.12,
"step": 41
},
{
"epoch": 0.31,
"grad_norm": 0.12060546875,
"learning_rate": 0.000170272037124899,
"loss": 0.1234,
"step": 42
},
{
"epoch": 0.31,
"grad_norm": 0.0703125,
"learning_rate": 0.00016849076713469914,
"loss": 0.1168,
"step": 43
},
{
"epoch": 0.32,
"grad_norm": 0.07177734375,
"learning_rate": 0.00016666758863762793,
"loss": 0.094,
"step": 44
},
{
"epoch": 0.33,
"grad_norm": 0.038818359375,
"learning_rate": 0.00016480361721016054,
"loss": 0.1026,
"step": 45
},
{
"epoch": 0.34,
"grad_norm": 0.080078125,
"learning_rate": 0.00016289999338937427,
"loss": 0.0922,
"step": 46
},
{
"epoch": 0.34,
"grad_norm": 0.072265625,
"learning_rate": 0.00016095788197507081,
"loss": 0.1234,
"step": 47
},
{
"epoch": 0.35,
"grad_norm": 0.07080078125,
"learning_rate": 0.00015897847131705195,
"loss": 0.1014,
"step": 48
},
{
"epoch": 0.36,
"grad_norm": 0.10205078125,
"learning_rate": 0.0001569629725879857,
"loss": 0.0768,
"step": 49
},
{
"epoch": 0.36,
"grad_norm": 0.166015625,
"learning_rate": 0.00015491261904230727,
"loss": 0.0872,
"step": 50
},
{
"epoch": 0.37,
"grad_norm": 0.1796875,
"learning_rate": 0.00015282866526160837,
"loss": 0.1247,
"step": 51
},
{
"epoch": 0.38,
"grad_norm": 0.07958984375,
"learning_rate": 0.00015071238638697732,
"loss": 0.071,
"step": 52
},
{
"epoch": 0.39,
"grad_norm": 0.1943359375,
"learning_rate": 0.00014856507733875836,
"loss": 0.1064,
"step": 53
},
{
"epoch": 0.39,
"grad_norm": 0.23046875,
"learning_rate": 0.00014638805202420895,
"loss": 0.0965,
"step": 54
},
{
"epoch": 0.4,
"grad_norm": 0.17578125,
"learning_rate": 0.0001441826425335387,
"loss": 0.0919,
"step": 55
},
{
"epoch": 0.41,
"grad_norm": 0.1611328125,
"learning_rate": 0.0001419501983248229,
"loss": 0.0636,
"step": 56
},
{
"epoch": 0.42,
"grad_norm": 0.400390625,
"learning_rate": 0.00013969208539828872,
"loss": 0.0864,
"step": 57
},
{
"epoch": 0.42,
"grad_norm": 0.130859375,
"learning_rate": 0.00013740968546047935,
"loss": 0.0747,
"step": 58
},
{
"epoch": 0.43,
"grad_norm": 0.373046875,
"learning_rate": 0.00013510439507880776,
"loss": 0.1053,
"step": 59
},
{
"epoch": 0.44,
"grad_norm": 0.283203125,
"learning_rate": 0.00013277762482701767,
"loss": 0.0828,
"step": 60
},
{
"epoch": 0.45,
"grad_norm": 0.318359375,
"learning_rate": 0.0001304307984220736,
"loss": 0.0796,
"step": 61
},
{
"epoch": 0.45,
"grad_norm": 0.328125,
"learning_rate": 0.0001280653518530093,
"loss": 0.0879,
"step": 62
},
{
"epoch": 0.46,
"grad_norm": 0.1474609375,
"learning_rate": 0.0001256827325022668,
"loss": 0.1077,
"step": 63
},
{
"epoch": 0.47,
"grad_norm": 0.2734375,
"learning_rate": 0.00012328439826006415,
"loss": 0.0839,
"step": 64
},
{
"epoch": 0.47,
"grad_norm": 0.134765625,
"learning_rate": 0.00012087181663233354,
"loss": 0.0732,
"step": 65
},
{
"epoch": 0.48,
"grad_norm": 0.322265625,
"learning_rate": 0.0001184464638427756,
"loss": 0.0832,
"step": 66
},
{
"epoch": 0.49,
"grad_norm": 0.6484375,
"learning_rate": 0.00011600982392957978,
"loss": 0.0903,
"step": 67
},
{
"epoch": 0.5,
"grad_norm": 0.36328125,
"learning_rate": 0.00011356338783736255,
"loss": 0.0936,
"step": 68
},
{
"epoch": 0.5,
"grad_norm": 0.0732421875,
"learning_rate": 0.00011110865250488047,
"loss": 0.0622,
"step": 69
},
{
"epoch": 0.51,
"grad_norm": 0.28125,
"learning_rate": 0.00010864711994907458,
"loss": 0.0873,
"step": 70
},
{
"epoch": 0.52,
"grad_norm": 0.1494140625,
"learning_rate": 0.00010618029634600843,
"loss": 0.1168,
"step": 71
},
{
"epoch": 0.53,
"grad_norm": 0.185546875,
"learning_rate": 0.00010370969110926052,
"loss": 0.0881,
"step": 72
},
{
"epoch": 0.53,
"grad_norm": 0.27734375,
"learning_rate": 0.00010123681596633629,
"loss": 0.1136,
"step": 73
},
{
"epoch": 0.54,
"grad_norm": 0.1884765625,
"learning_rate": 9.876318403366372e-05,
"loss": 0.0861,
"step": 74
},
{
"epoch": 0.55,
"grad_norm": 0.080078125,
"learning_rate": 9.629030889073949e-05,
"loss": 0.0819,
"step": 75
},
{
"epoch": 0.55,
"grad_norm": 0.1318359375,
"learning_rate": 9.38197036539916e-05,
"loss": 0.0784,
"step": 76
},
{
"epoch": 0.56,
"grad_norm": 0.240234375,
"learning_rate": 9.135288005092546e-05,
"loss": 0.0631,
"step": 77
},
{
"epoch": 0.57,
"grad_norm": 0.1826171875,
"learning_rate": 8.889134749511955e-05,
"loss": 0.1073,
"step": 78
},
{
"epoch": 0.58,
"grad_norm": 0.1953125,
"learning_rate": 8.643661216263743e-05,
"loss": 0.0909,
"step": 79
},
{
"epoch": 0.58,
"grad_norm": 0.298828125,
"learning_rate": 8.399017607042025e-05,
"loss": 0.1,
"step": 80
},
{
"epoch": 0.59,
"grad_norm": 0.0703125,
"learning_rate": 8.155353615722442e-05,
"loss": 0.0581,
"step": 81
},
{
"epoch": 0.6,
"grad_norm": 0.06689453125,
"learning_rate": 7.91281833676665e-05,
"loss": 0.0695,
"step": 82
},
{
"epoch": 0.61,
"grad_norm": 0.1572265625,
"learning_rate": 7.671560173993587e-05,
"loss": 0.1054,
"step": 83
},
{
"epoch": 0.61,
"grad_norm": 0.052734375,
"learning_rate": 7.431726749773322e-05,
"loss": 0.0946,
"step": 84
},
{
"epoch": 0.62,
"grad_norm": 0.08154296875,
"learning_rate": 7.193464814699073e-05,
"loss": 0.0561,
"step": 85
},
{
"epoch": 0.63,
"grad_norm": 0.181640625,
"learning_rate": 6.956920157792639e-05,
"loss": 0.0902,
"step": 86
},
{
"epoch": 0.64,
"grad_norm": 0.0908203125,
"learning_rate": 6.722237517298232e-05,
"loss": 0.0694,
"step": 87
},
{
"epoch": 0.64,
"grad_norm": 0.11767578125,
"learning_rate": 6.489560492119225e-05,
"loss": 0.1068,
"step": 88
},
{
"epoch": 0.65,
"grad_norm": 0.125,
"learning_rate": 6.259031453952069e-05,
"loss": 0.0671,
"step": 89
},
{
"epoch": 0.66,
"grad_norm": 0.0732421875,
"learning_rate": 6.0307914601711305e-05,
"loss": 0.0891,
"step": 90
},
{
"epoch": 0.66,
"grad_norm": 0.08056640625,
"learning_rate": 5.8049801675177115e-05,
"loss": 0.0782,
"step": 91
},
{
"epoch": 0.67,
"grad_norm": 0.2060546875,
"learning_rate": 5.5817357466461336e-05,
"loss": 0.1028,
"step": 92
},
{
"epoch": 0.68,
"grad_norm": 0.125,
"learning_rate": 5.361194797579108e-05,
"loss": 0.0581,
"step": 93
},
{
"epoch": 0.69,
"grad_norm": 0.08935546875,
"learning_rate": 5.1434922661241635e-05,
"loss": 0.0619,
"step": 94
},
{
"epoch": 0.69,
"grad_norm": 0.07373046875,
"learning_rate": 4.928761361302269e-05,
"loss": 0.0529,
"step": 95
},
{
"epoch": 0.7,
"grad_norm": 0.0703125,
"learning_rate": 4.717133473839163e-05,
"loss": 0.0397,
"step": 96
},
{
"epoch": 0.71,
"grad_norm": 0.11767578125,
"learning_rate": 4.5087380957692784e-05,
"loss": 0.0459,
"step": 97
},
{
"epoch": 0.72,
"grad_norm": 0.203125,
"learning_rate": 4.303702741201431e-05,
"loss": 0.0752,
"step": 98
},
{
"epoch": 0.72,
"grad_norm": 0.15625,
"learning_rate": 4.1021528682948066e-05,
"loss": 0.0988,
"step": 99
},
{
"epoch": 0.73,
"grad_norm": 0.09814453125,
"learning_rate": 3.904211802492922e-05,
"loss": 0.0739,
"step": 100
},
{
"epoch": 0.74,
"grad_norm": 0.1298828125,
"learning_rate": 3.7100006610625784e-05,
"loss": 0.0634,
"step": 101
},
{
"epoch": 0.74,
"grad_norm": 0.1181640625,
"learning_rate": 3.519638278983948e-05,
"loss": 0.057,
"step": 102
},
{
"epoch": 0.75,
"grad_norm": 0.1123046875,
"learning_rate": 3.333241136237206e-05,
"loss": 0.0556,
"step": 103
},
{
"epoch": 0.76,
"grad_norm": 0.32421875,
"learning_rate": 3.150923286530089e-05,
"loss": 0.1233,
"step": 104
},
{
"epoch": 0.77,
"grad_norm": 0.193359375,
"learning_rate": 2.9727962875101e-05,
"loss": 0.1074,
"step": 105
},
{
"epoch": 0.77,
"grad_norm": 0.1484375,
"learning_rate": 2.798969132503997e-05,
"loss": 0.0993,
"step": 106
},
{
"epoch": 0.78,
"grad_norm": 0.0771484375,
"learning_rate": 2.6295481838263626e-05,
"loss": 0.0537,
"step": 107
},
{
"epoch": 0.79,
"grad_norm": 0.0869140625,
"learning_rate": 2.4646371076980457e-05,
"loss": 0.0704,
"step": 108
},
{
"epoch": 0.8,
"grad_norm": 0.10205078125,
"learning_rate": 2.3043368108143047e-05,
"loss": 0.0825,
"step": 109
},
{
"epoch": 0.8,
"grad_norm": 0.09912109375,
"learning_rate": 2.1487453786014512e-05,
"loss": 0.0422,
"step": 110
},
{
"epoch": 0.81,
"grad_norm": 0.1123046875,
"learning_rate": 1.997958015199829e-05,
"loss": 0.0726,
"step": 111
},
{
"epoch": 0.82,
"grad_norm": 0.1982421875,
"learning_rate": 1.8520669852097573e-05,
"loss": 0.0518,
"step": 112
},
{
"epoch": 0.82,
"grad_norm": 0.091796875,
"learning_rate": 1.7111615572361628e-05,
"loss": 0.0884,
"step": 113
},
{
"epoch": 0.83,
"grad_norm": 0.1181640625,
"learning_rate": 1.5753279492664262e-05,
"loss": 0.0892,
"step": 114
},
{
"epoch": 0.84,
"grad_norm": 0.07080078125,
"learning_rate": 1.4446492759148411e-05,
"loss": 0.0478,
"step": 115
},
{
"epoch": 0.85,
"grad_norm": 0.087890625,
"learning_rate": 1.319205497565983e-05,
"loss": 0.0786,
"step": 116
},
{
"epoch": 0.85,
"grad_norm": 0.134765625,
"learning_rate": 1.1990733714481184e-05,
"loss": 0.0429,
"step": 117
},
{
"epoch": 0.86,
"grad_norm": 0.083984375,
"learning_rate": 1.0843264046665557e-05,
"loss": 0.0455,
"step": 118
},
{
"epoch": 0.87,
"grad_norm": 0.1982421875,
"learning_rate": 9.750348092257367e-06,
"loss": 0.0896,
"step": 119
},
{
"epoch": 0.88,
"grad_norm": 0.09716796875,
"learning_rate": 8.712654590675085e-06,
"loss": 0.0727,
"step": 120
},
{
"epoch": 0.88,
"grad_norm": 0.0751953125,
"learning_rate": 7.730818491519343e-06,
"loss": 0.0284,
"step": 121
},
{
"epoch": 0.89,
"grad_norm": 0.11181640625,
"learning_rate": 6.805440566056553e-06,
"loss": 0.0484,
"step": 122
},
{
"epoch": 0.9,
"grad_norm": 0.130859375,
"learning_rate": 5.937087039615619e-06,
"loss": 0.0573,
"step": 123
},
{
"epoch": 0.91,
"grad_norm": 0.1142578125,
"learning_rate": 5.126289245122906e-06,
"loss": 0.071,
"step": 124
},
{
"epoch": 0.91,
"grad_norm": 0.09375,
"learning_rate": 4.37354329798726e-06,
"loss": 0.0695,
"step": 125
},
{
"epoch": 0.92,
"grad_norm": 0.09375,
"learning_rate": 3.679309792534291e-06,
"loss": 0.0453,
"step": 126
},
{
"epoch": 0.93,
"grad_norm": 0.1298828125,
"learning_rate": 3.0440135201753374e-06,
"loss": 0.0906,
"step": 127
},
{
"epoch": 0.93,
"grad_norm": 0.09912109375,
"learning_rate": 2.468043209483739e-06,
"loss": 0.0617,
"step": 128
},
{
"epoch": 0.94,
"grad_norm": 0.142578125,
"learning_rate": 1.951751288337467e-06,
"loss": 0.0667,
"step": 129
},
{
"epoch": 0.95,
"grad_norm": 0.13671875,
"learning_rate": 1.4954536682736719e-06,
"loss": 0.0686,
"step": 130
},
{
"epoch": 0.96,
"grad_norm": 0.103515625,
"learning_rate": 1.0994295511869257e-06,
"loss": 0.0415,
"step": 131
},
{
"epoch": 0.96,
"grad_norm": 0.12255859375,
"learning_rate": 7.639212584897082e-07,
"loss": 0.0573,
"step": 132
},
{
"epoch": 0.97,
"grad_norm": 0.1689453125,
"learning_rate": 4.891340828393487e-07,
"loss": 0.0876,
"step": 133
},
{
"epoch": 0.98,
"grad_norm": 0.0712890625,
"learning_rate": 2.752361625225297e-07,
"loss": 0.0599,
"step": 134
},
{
"epoch": 0.99,
"grad_norm": 0.072265625,
"learning_rate": 1.2235837857387246e-07,
"loss": 0.0593,
"step": 135
},
{
"epoch": 0.99,
"grad_norm": 0.123046875,
"learning_rate": 3.059427469168652e-08,
"loss": 0.101,
"step": 136
},
{
"epoch": 1.0,
"grad_norm": 0.146484375,
"learning_rate": 0.0,
"loss": 0.0724,
"step": 137
},
{
"epoch": 1.0,
"eval_loss": 0.05070766434073448,
"eval_runtime": 6.5801,
"eval_samples_per_second": 17.477,
"eval_steps_per_second": 1.216,
"step": 137
}
],
"logging_steps": 1,
"max_steps": 137,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"total_flos": 9.602610483743949e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}