mistralic-expert-17 / checkpoint-200 /trainer_state.json
pharaouk's picture
Training in progress, step 200, checkpoint
dcce312
raw
history blame
25.3 kB
{
"best_metric": 0.36924177408218384,
"best_model_checkpoint": "experts/mistralic-expert-17/checkpoint-200",
"epoch": 0.041452925022021865,
"eval_steps": 200,
"global_step": 200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3764,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3983,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4823,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4139,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.341,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3615,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3709,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3184,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3923,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5588,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4881,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4644,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3299,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.404,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3933,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3784,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.406,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4184,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3953,
"step": 19
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4085,
"step": 20
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4547,
"step": 21
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4166,
"step": 22
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4566,
"step": 23
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3928,
"step": 24
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3639,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.386,
"step": 26
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3157,
"step": 27
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3111,
"step": 28
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5034,
"step": 29
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4071,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6236,
"step": 31
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3294,
"step": 32
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4722,
"step": 33
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3625,
"step": 34
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5145,
"step": 35
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.416,
"step": 36
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3519,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3509,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.416,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4899,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3193,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5121,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3911,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4805,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3315,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3784,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4312,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4413,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3399,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3861,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4977,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.515,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3766,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3144,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3097,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3277,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3147,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4264,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3957,
"step": 59
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3379,
"step": 60
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3163,
"step": 61
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3506,
"step": 62
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3888,
"step": 63
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.347,
"step": 64
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.522,
"step": 65
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3002,
"step": 66
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3523,
"step": 67
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6818,
"step": 68
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4075,
"step": 69
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4022,
"step": 70
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4122,
"step": 71
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4788,
"step": 72
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3144,
"step": 73
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4427,
"step": 74
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.453,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3033,
"step": 76
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.475,
"step": 77
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4291,
"step": 78
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4268,
"step": 79
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3052,
"step": 80
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.2165,
"step": 81
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5203,
"step": 82
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3173,
"step": 83
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3227,
"step": 84
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.387,
"step": 85
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3191,
"step": 86
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.2438,
"step": 87
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3296,
"step": 88
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.2889,
"step": 89
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3352,
"step": 90
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3428,
"step": 91
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3576,
"step": 92
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3154,
"step": 93
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4834,
"step": 94
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.2879,
"step": 95
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3734,
"step": 96
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4016,
"step": 97
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4421,
"step": 98
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3338,
"step": 99
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4342,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3754,
"step": 101
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5515,
"step": 102
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3301,
"step": 103
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3824,
"step": 104
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3394,
"step": 105
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.2945,
"step": 106
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3857,
"step": 107
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3952,
"step": 108
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3497,
"step": 109
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4295,
"step": 110
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3641,
"step": 111
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.2941,
"step": 112
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4429,
"step": 113
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.2982,
"step": 114
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.407,
"step": 115
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5319,
"step": 116
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3071,
"step": 117
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.353,
"step": 118
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5562,
"step": 119
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3285,
"step": 120
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3044,
"step": 121
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3912,
"step": 122
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3579,
"step": 123
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4041,
"step": 124
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4563,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3469,
"step": 126
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3381,
"step": 127
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3378,
"step": 128
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4174,
"step": 129
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3629,
"step": 130
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3426,
"step": 131
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5591,
"step": 132
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.321,
"step": 133
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3866,
"step": 134
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4362,
"step": 135
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3987,
"step": 136
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4104,
"step": 137
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3393,
"step": 138
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3664,
"step": 139
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2806,
"step": 140
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5358,
"step": 141
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5086,
"step": 142
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3179,
"step": 143
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2922,
"step": 144
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.413,
"step": 145
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3515,
"step": 146
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3137,
"step": 147
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3654,
"step": 148
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5826,
"step": 149
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4276,
"step": 150
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2985,
"step": 151
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3914,
"step": 152
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2938,
"step": 153
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2313,
"step": 154
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4092,
"step": 155
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4689,
"step": 156
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3192,
"step": 157
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3818,
"step": 158
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3851,
"step": 159
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2992,
"step": 160
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3687,
"step": 161
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3448,
"step": 162
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3684,
"step": 163
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2847,
"step": 164
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3367,
"step": 165
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.2416,
"step": 166
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4329,
"step": 167
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4203,
"step": 168
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.355,
"step": 169
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3465,
"step": 170
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2676,
"step": 171
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3554,
"step": 172
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2724,
"step": 173
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4885,
"step": 174
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.556,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.298,
"step": 176
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.286,
"step": 177
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3485,
"step": 178
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3193,
"step": 179
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3288,
"step": 180
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2939,
"step": 181
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3689,
"step": 182
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.324,
"step": 183
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3524,
"step": 184
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3494,
"step": 185
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3723,
"step": 186
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3995,
"step": 187
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4912,
"step": 188
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3704,
"step": 189
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3676,
"step": 190
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4459,
"step": 191
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.336,
"step": 192
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.627,
"step": 193
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5047,
"step": 194
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3175,
"step": 195
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4015,
"step": 196
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3854,
"step": 197
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3298,
"step": 198
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3249,
"step": 199
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3183,
"step": 200
},
{
"epoch": 0.04,
"eval_loss": 0.36924177408218384,
"eval_runtime": 159.4711,
"eval_samples_per_second": 6.271,
"eval_steps_per_second": 3.135,
"step": 200
},
{
"epoch": 0.04,
"mmlu_eval_accuracy": 0.5940031246983312,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.35714285714285715,
"mmlu_eval_accuracy_astronomy": 0.75,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
"mmlu_eval_accuracy_college_biology": 0.5625,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
"mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
"mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.5,
"mmlu_eval_accuracy_econometrics": 0.6666666666666666,
"mmlu_eval_accuracy_electrical_engineering": 0.625,
"mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.625,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5348837209302325,
"mmlu_eval_accuracy_high_school_mathematics": 0.3103448275862069,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6923076923076923,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
"mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
"mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.88,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.37,
"mmlu_eval_accuracy_nutrition": 0.7878787878787878,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.5161290322580645,
"mmlu_eval_accuracy_professional_law": 0.4,
"mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
"mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.7037037037037037,
"mmlu_eval_accuracy_sociology": 0.9545454545454546,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.8947368421052632,
"mmlu_loss": 1.4302824790888604,
"step": 200
}
],
"logging_steps": 1,
"max_steps": 14472,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 5.002448732690842e+16,
"trial_name": null,
"trial_params": null
}