mistralic-expert-17 / checkpoint-600 /trainer_state.json
pharaouk's picture
Training in progress, step 600, checkpoint
0c9367e
raw
history blame
75.1 kB
{
"best_metric": 0.3515583574771881,
"best_model_checkpoint": "experts/mistralic-expert-17/checkpoint-600",
"epoch": 0.1243587750660656,
"eval_steps": 200,
"global_step": 600,
"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
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4524,
"step": 201
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4312,
"step": 202
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2717,
"step": 203
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3977,
"step": 204
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3864,
"step": 205
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2953,
"step": 206
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3302,
"step": 207
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4244,
"step": 208
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4016,
"step": 209
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4256,
"step": 210
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4162,
"step": 211
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.282,
"step": 212
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3207,
"step": 213
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2938,
"step": 214
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2823,
"step": 215
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2986,
"step": 216
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2869,
"step": 217
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.317,
"step": 218
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2975,
"step": 219
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2523,
"step": 220
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3691,
"step": 221
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.341,
"step": 222
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2603,
"step": 223
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4282,
"step": 224
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4877,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2787,
"step": 226
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3901,
"step": 227
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3039,
"step": 228
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4518,
"step": 229
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.325,
"step": 230
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2634,
"step": 231
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3732,
"step": 232
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3162,
"step": 233
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3761,
"step": 234
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2739,
"step": 235
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.344,
"step": 236
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2995,
"step": 237
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2882,
"step": 238
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2372,
"step": 239
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3331,
"step": 240
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6815,
"step": 241
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2672,
"step": 242
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4416,
"step": 243
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2813,
"step": 244
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3574,
"step": 245
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3237,
"step": 246
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2798,
"step": 247
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5538,
"step": 248
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3436,
"step": 249
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3082,
"step": 250
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5797,
"step": 251
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2435,
"step": 252
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4687,
"step": 253
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2921,
"step": 254
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3151,
"step": 255
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2787,
"step": 256
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3544,
"step": 257
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2891,
"step": 258
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4261,
"step": 259
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2829,
"step": 260
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2342,
"step": 261
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3482,
"step": 262
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3608,
"step": 263
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4406,
"step": 264
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3543,
"step": 265
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4826,
"step": 266
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3967,
"step": 267
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3463,
"step": 268
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.2837,
"step": 269
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3827,
"step": 270
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3583,
"step": 271
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3035,
"step": 272
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3266,
"step": 273
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5502,
"step": 274
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4045,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.2922,
"step": 276
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4068,
"step": 277
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.32,
"step": 278
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3039,
"step": 279
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5073,
"step": 280
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.357,
"step": 281
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.2757,
"step": 282
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3473,
"step": 283
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4409,
"step": 284
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4095,
"step": 285
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3177,
"step": 286
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5387,
"step": 287
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3647,
"step": 288
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4568,
"step": 289
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4054,
"step": 290
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3302,
"step": 291
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3426,
"step": 292
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4874,
"step": 293
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5772,
"step": 294
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.2785,
"step": 295
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3493,
"step": 296
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3909,
"step": 297
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4654,
"step": 298
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4774,
"step": 299
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.2985,
"step": 300
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4445,
"step": 301
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3078,
"step": 302
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.2923,
"step": 303
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.373,
"step": 304
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3412,
"step": 305
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.2629,
"step": 306
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.258,
"step": 307
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4849,
"step": 308
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4029,
"step": 309
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3758,
"step": 310
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4228,
"step": 311
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3841,
"step": 312
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3038,
"step": 313
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4246,
"step": 314
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3501,
"step": 315
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3886,
"step": 316
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3705,
"step": 317
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3549,
"step": 318
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3336,
"step": 319
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3522,
"step": 320
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.2667,
"step": 321
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4979,
"step": 322
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3397,
"step": 323
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3273,
"step": 324
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3509,
"step": 325
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3015,
"step": 326
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.2906,
"step": 327
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.2315,
"step": 328
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3104,
"step": 329
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4869,
"step": 330
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4246,
"step": 331
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3485,
"step": 332
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3974,
"step": 333
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4052,
"step": 334
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4124,
"step": 335
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4349,
"step": 336
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3406,
"step": 337
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3532,
"step": 338
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4889,
"step": 339
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5194,
"step": 340
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.514,
"step": 341
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.327,
"step": 342
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3088,
"step": 343
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3761,
"step": 344
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.2712,
"step": 345
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3488,
"step": 346
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4925,
"step": 347
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3292,
"step": 348
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.282,
"step": 349
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3427,
"step": 350
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3182,
"step": 351
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4053,
"step": 352
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3911,
"step": 353
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.2906,
"step": 354
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4012,
"step": 355
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3397,
"step": 356
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3832,
"step": 357
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3739,
"step": 358
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3106,
"step": 359
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3123,
"step": 360
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3197,
"step": 361
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3987,
"step": 362
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2746,
"step": 363
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3073,
"step": 364
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3593,
"step": 365
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3287,
"step": 366
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.354,
"step": 367
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3877,
"step": 368
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4417,
"step": 369
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2696,
"step": 370
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3225,
"step": 371
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3691,
"step": 372
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3359,
"step": 373
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2475,
"step": 374
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3173,
"step": 375
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3585,
"step": 376
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2903,
"step": 377
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3593,
"step": 378
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3115,
"step": 379
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5,
"step": 380
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4372,
"step": 381
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.319,
"step": 382
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2573,
"step": 383
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5916,
"step": 384
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4126,
"step": 385
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3318,
"step": 386
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2875,
"step": 387
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2891,
"step": 388
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3617,
"step": 389
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3148,
"step": 390
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4457,
"step": 391
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.407,
"step": 392
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4488,
"step": 393
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4066,
"step": 394
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3808,
"step": 395
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2792,
"step": 396
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4491,
"step": 397
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3646,
"step": 398
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3868,
"step": 399
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2754,
"step": 400
},
{
"epoch": 0.08,
"eval_loss": 0.358918160200119,
"eval_runtime": 159.4642,
"eval_samples_per_second": 6.271,
"eval_steps_per_second": 3.136,
"step": 400
},
{
"epoch": 0.08,
"mmlu_eval_accuracy": 0.5999132037466034,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.42857142857142855,
"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": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.6363636363636364,
"mmlu_eval_accuracy_conceptual_physics": 0.5,
"mmlu_eval_accuracy_econometrics": 0.6666666666666666,
"mmlu_eval_accuracy_electrical_engineering": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.4878048780487805,
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.59375,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
"mmlu_eval_accuracy_high_school_mathematics": 0.3448275862068966,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6923076923076923,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.85,
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
"mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.88,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7209302325581395,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.33,
"mmlu_eval_accuracy_nutrition": 0.7272727272727273,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.42857142857142855,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.3941176470588235,
"mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
"mmlu_eval_accuracy_professional_psychology": 0.6376811594202898,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"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": 1.4201551996774524,
"step": 400
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3131,
"step": 401
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3419,
"step": 402
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3984,
"step": 403
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3459,
"step": 404
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4061,
"step": 405
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.2641,
"step": 406
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.332,
"step": 407
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3048,
"step": 408
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3402,
"step": 409
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3617,
"step": 410
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4099,
"step": 411
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6116,
"step": 412
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4472,
"step": 413
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2897,
"step": 414
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3139,
"step": 415
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4072,
"step": 416
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3577,
"step": 417
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4119,
"step": 418
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3019,
"step": 419
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5519,
"step": 420
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2923,
"step": 421
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4213,
"step": 422
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3394,
"step": 423
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3897,
"step": 424
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2891,
"step": 425
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.285,
"step": 426
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3758,
"step": 427
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2731,
"step": 428
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.374,
"step": 429
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3646,
"step": 430
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4202,
"step": 431
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2353,
"step": 432
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.383,
"step": 433
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3019,
"step": 434
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5666,
"step": 435
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3783,
"step": 436
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2525,
"step": 437
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3956,
"step": 438
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.395,
"step": 439
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2849,
"step": 440
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3503,
"step": 441
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3204,
"step": 442
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3204,
"step": 443
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3083,
"step": 444
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4108,
"step": 445
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2786,
"step": 446
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5366,
"step": 447
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2785,
"step": 448
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3132,
"step": 449
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4062,
"step": 450
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3812,
"step": 451
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2555,
"step": 452
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4278,
"step": 453
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2947,
"step": 454
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3247,
"step": 455
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5202,
"step": 456
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3669,
"step": 457
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3395,
"step": 458
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.358,
"step": 459
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2811,
"step": 460
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4165,
"step": 461
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2591,
"step": 462
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3491,
"step": 463
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.33,
"step": 464
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4262,
"step": 465
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3111,
"step": 466
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4079,
"step": 467
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4417,
"step": 468
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.261,
"step": 469
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2752,
"step": 470
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2719,
"step": 471
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3703,
"step": 472
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2798,
"step": 473
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3553,
"step": 474
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3205,
"step": 475
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3406,
"step": 476
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3759,
"step": 477
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3276,
"step": 478
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3335,
"step": 479
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5614,
"step": 480
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2922,
"step": 481
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3785,
"step": 482
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3085,
"step": 483
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4565,
"step": 484
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4444,
"step": 485
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2936,
"step": 486
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3858,
"step": 487
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3834,
"step": 488
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3273,
"step": 489
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3334,
"step": 490
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3236,
"step": 491
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2944,
"step": 492
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5452,
"step": 493
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3507,
"step": 494
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3539,
"step": 495
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2818,
"step": 496
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4009,
"step": 497
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3196,
"step": 498
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4456,
"step": 499
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2854,
"step": 500
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.359,
"step": 501
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3794,
"step": 502
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.581,
"step": 503
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3004,
"step": 504
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3796,
"step": 505
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3091,
"step": 506
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4228,
"step": 507
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3734,
"step": 508
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2583,
"step": 509
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3844,
"step": 510
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3349,
"step": 511
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.347,
"step": 512
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2649,
"step": 513
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3511,
"step": 514
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.1902,
"step": 515
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2628,
"step": 516
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.258,
"step": 517
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4467,
"step": 518
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3565,
"step": 519
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3713,
"step": 520
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2867,
"step": 521
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2861,
"step": 522
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2694,
"step": 523
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4202,
"step": 524
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4187,
"step": 525
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4243,
"step": 526
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4473,
"step": 527
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4199,
"step": 528
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3368,
"step": 529
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3606,
"step": 530
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6544,
"step": 531
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3448,
"step": 532
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3672,
"step": 533
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3245,
"step": 534
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3069,
"step": 535
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.47,
"step": 536
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3799,
"step": 537
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3455,
"step": 538
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4227,
"step": 539
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3065,
"step": 540
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3319,
"step": 541
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5573,
"step": 542
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3221,
"step": 543
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2999,
"step": 544
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4778,
"step": 545
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4276,
"step": 546
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3707,
"step": 547
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3303,
"step": 548
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5229,
"step": 549
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3393,
"step": 550
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3118,
"step": 551
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.2864,
"step": 552
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3364,
"step": 553
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3086,
"step": 554
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3338,
"step": 555
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2945,
"step": 556
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3808,
"step": 557
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2725,
"step": 558
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2995,
"step": 559
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3137,
"step": 560
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3209,
"step": 561
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2393,
"step": 562
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3013,
"step": 563
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2315,
"step": 564
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3284,
"step": 565
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3422,
"step": 566
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4019,
"step": 567
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5752,
"step": 568
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.317,
"step": 569
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3557,
"step": 570
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3452,
"step": 571
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3375,
"step": 572
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2105,
"step": 573
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4988,
"step": 574
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2949,
"step": 575
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3991,
"step": 576
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3326,
"step": 577
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3207,
"step": 578
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3678,
"step": 579
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2938,
"step": 580
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3795,
"step": 581
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4644,
"step": 582
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4052,
"step": 583
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2996,
"step": 584
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.455,
"step": 585
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4427,
"step": 586
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3916,
"step": 587
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3169,
"step": 588
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.576,
"step": 589
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4151,
"step": 590
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2811,
"step": 591
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3179,
"step": 592
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2883,
"step": 593
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5965,
"step": 594
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3987,
"step": 595
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4434,
"step": 596
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5114,
"step": 597
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.2867,
"step": 598
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.255,
"step": 599
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.33,
"step": 600
},
{
"epoch": 0.12,
"eval_loss": 0.3515583574771881,
"eval_runtime": 159.3504,
"eval_samples_per_second": 6.275,
"eval_steps_per_second": 3.138,
"step": 600
},
{
"epoch": 0.12,
"mmlu_eval_accuracy": 0.6095696896582704,
"mmlu_eval_accuracy_abstract_algebra": 0.45454545454545453,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
"mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
"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": 0.7272727272727273,
"mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
"mmlu_eval_accuracy_econometrics": 0.6666666666666666,
"mmlu_eval_accuracy_electrical_engineering": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.4634146341463415,
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.625,
"mmlu_eval_accuracy_high_school_chemistry": 0.45454545454545453,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
"mmlu_eval_accuracy_high_school_mathematics": 0.3448275862068966,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.85,
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
"mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7558139534883721,
"mmlu_eval_accuracy_moral_disputes": 0.6052631578947368,
"mmlu_eval_accuracy_moral_scenarios": 0.36,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.6764705882352942,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.4838709677419355,
"mmlu_eval_accuracy_professional_law": 0.4,
"mmlu_eval_accuracy_professional_medicine": 0.6774193548387096,
"mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6666666666666666,
"mmlu_eval_accuracy_sociology": 0.9090909090909091,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
"mmlu_loss": 1.383797744780234,
"step": 600
}
],
"logging_steps": 1,
"max_steps": 14472,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 1.4948975984207462e+17,
"trial_name": null,
"trial_params": null
}