Update README.md
Browse files
README.md
CHANGED
|
@@ -16,4 +16,386 @@ As this was an test run, we just tested it and heres the Data, the model hasnt I
|
|
| 16 |
| gpt-3.5-turbo | 8.075000 | 7.943750 | 7.943750 |
|
| 17 |
| claude-v1 | 8.150000 | 7.900000 | 8.025000 |
|
| 18 |
| **LexGPT-V3** | **8.14375** | 7.719355 | 7.926667 |
|
| 19 |
-
| vicuna-13b-v1.3 | 6.812500 | 5.962500 | 6.387500 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
| gpt-3.5-turbo | 8.075000 | 7.943750 | 7.943750 |
|
| 17 |
| claude-v1 | 8.150000 | 7.900000 | 8.025000 |
|
| 18 |
| **LexGPT-V3** | **8.14375** | 7.719355 | 7.926667 |
|
| 19 |
+
| vicuna-13b-v1.3 | 6.812500 | 5.962500 | 6.387500 |
|
| 20 |
+
|
| 21 |
+
Open-LLM Leaderboard Results:
|
| 22 |
+
{
|
| 23 |
+
"all": {
|
| 24 |
+
"acc": 0.647154984215818,
|
| 25 |
+
"acc_stderr": 0.03221441224437104,
|
| 26 |
+
"acc_norm": 0.6487599114885558,
|
| 27 |
+
"acc_norm_stderr": 0.032860268812293904,
|
| 28 |
+
"mc1": 0.4283965728274174,
|
| 29 |
+
"mc1_stderr": 0.017323088597314757,
|
| 30 |
+
"mc2": 0.5998074537794252,
|
| 31 |
+
"mc2_stderr": 0.015494960379071198
|
| 32 |
+
},
|
| 33 |
+
"harness|arc:challenge|25": {
|
| 34 |
+
"acc": 0.64419795221843,
|
| 35 |
+
"acc_stderr": 0.01399057113791876,
|
| 36 |
+
"acc_norm": 0.6646757679180887,
|
| 37 |
+
"acc_norm_stderr": 0.013796182947785562
|
| 38 |
+
},
|
| 39 |
+
"harness|hellaswag|10": {
|
| 40 |
+
"acc": 0.6782513443537144,
|
| 41 |
+
"acc_stderr": 0.004661924314756093,
|
| 42 |
+
"acc_norm": 0.8590918143796057,
|
| 43 |
+
"acc_norm_stderr": 0.003472157511639361
|
| 44 |
+
},
|
| 45 |
+
"harness|hendrycksTest-abstract_algebra|5": {
|
| 46 |
+
"acc": 0.29,
|
| 47 |
+
"acc_stderr": 0.045604802157206845,
|
| 48 |
+
"acc_norm": 0.29,
|
| 49 |
+
"acc_norm_stderr": 0.045604802157206845
|
| 50 |
+
},
|
| 51 |
+
"harness|hendrycksTest-anatomy|5": {
|
| 52 |
+
"acc": 0.5925925925925926,
|
| 53 |
+
"acc_stderr": 0.04244633238353227,
|
| 54 |
+
"acc_norm": 0.5925925925925926,
|
| 55 |
+
"acc_norm_stderr": 0.04244633238353227
|
| 56 |
+
},
|
| 57 |
+
"harness|hendrycksTest-astronomy|5": {
|
| 58 |
+
"acc": 0.6973684210526315,
|
| 59 |
+
"acc_stderr": 0.03738520676119667,
|
| 60 |
+
"acc_norm": 0.6973684210526315,
|
| 61 |
+
"acc_norm_stderr": 0.03738520676119667
|
| 62 |
+
},
|
| 63 |
+
"harness|hendrycksTest-business_ethics|5": {
|
| 64 |
+
"acc": 0.59,
|
| 65 |
+
"acc_stderr": 0.04943110704237102,
|
| 66 |
+
"acc_norm": 0.59,
|
| 67 |
+
"acc_norm_stderr": 0.04943110704237102
|
| 68 |
+
},
|
| 69 |
+
"harness|hendrycksTest-clinical_knowledge|5": {
|
| 70 |
+
"acc": 0.7132075471698113,
|
| 71 |
+
"acc_stderr": 0.027834912527544057,
|
| 72 |
+
"acc_norm": 0.7132075471698113,
|
| 73 |
+
"acc_norm_stderr": 0.027834912527544057
|
| 74 |
+
},
|
| 75 |
+
"harness|hendrycksTest-college_biology|5": {
|
| 76 |
+
"acc": 0.7847222222222222,
|
| 77 |
+
"acc_stderr": 0.03437079344106135,
|
| 78 |
+
"acc_norm": 0.7847222222222222,
|
| 79 |
+
"acc_norm_stderr": 0.03437079344106135
|
| 80 |
+
},
|
| 81 |
+
"harness|hendrycksTest-college_chemistry|5": {
|
| 82 |
+
"acc": 0.49,
|
| 83 |
+
"acc_stderr": 0.05024183937956912,
|
| 84 |
+
"acc_norm": 0.49,
|
| 85 |
+
"acc_norm_stderr": 0.05024183937956912
|
| 86 |
+
},
|
| 87 |
+
"harness|hendrycksTest-college_computer_science|5": {
|
| 88 |
+
"acc": 0.54,
|
| 89 |
+
"acc_stderr": 0.05009082659620332,
|
| 90 |
+
"acc_norm": 0.54,
|
| 91 |
+
"acc_norm_stderr": 0.05009082659620332
|
| 92 |
+
},
|
| 93 |
+
"harness|hendrycksTest-college_mathematics|5": {
|
| 94 |
+
"acc": 0.34,
|
| 95 |
+
"acc_stderr": 0.04760952285695235,
|
| 96 |
+
"acc_norm": 0.34,
|
| 97 |
+
"acc_norm_stderr": 0.04760952285695235
|
| 98 |
+
},
|
| 99 |
+
"harness|hendrycksTest-college_medicine|5": {
|
| 100 |
+
"acc": 0.653179190751445,
|
| 101 |
+
"acc_stderr": 0.036291466701596636,
|
| 102 |
+
"acc_norm": 0.653179190751445,
|
| 103 |
+
"acc_norm_stderr": 0.036291466701596636
|
| 104 |
+
},
|
| 105 |
+
"harness|hendrycksTest-college_physics|5": {
|
| 106 |
+
"acc": 0.4019607843137255,
|
| 107 |
+
"acc_stderr": 0.04878608714466996,
|
| 108 |
+
"acc_norm": 0.4019607843137255,
|
| 109 |
+
"acc_norm_stderr": 0.04878608714466996
|
| 110 |
+
},
|
| 111 |
+
"harness|hendrycksTest-computer_security|5": {
|
| 112 |
+
"acc": 0.76,
|
| 113 |
+
"acc_stderr": 0.042923469599092816,
|
| 114 |
+
"acc_norm": 0.76,
|
| 115 |
+
"acc_norm_stderr": 0.042923469599092816
|
| 116 |
+
},
|
| 117 |
+
"harness|hendrycksTest-conceptual_physics|5": {
|
| 118 |
+
"acc": 0.5914893617021276,
|
| 119 |
+
"acc_stderr": 0.032134180267015755,
|
| 120 |
+
"acc_norm": 0.5914893617021276,
|
| 121 |
+
"acc_norm_stderr": 0.032134180267015755
|
| 122 |
+
},
|
| 123 |
+
"harness|hendrycksTest-econometrics|5": {
|
| 124 |
+
"acc": 0.45614035087719296,
|
| 125 |
+
"acc_stderr": 0.04685473041907789,
|
| 126 |
+
"acc_norm": 0.45614035087719296,
|
| 127 |
+
"acc_norm_stderr": 0.04685473041907789
|
| 128 |
+
},
|
| 129 |
+
"harness|hendrycksTest-electrical_engineering|5": {
|
| 130 |
+
"acc": 0.5379310344827586,
|
| 131 |
+
"acc_stderr": 0.041546596717075474,
|
| 132 |
+
"acc_norm": 0.5379310344827586,
|
| 133 |
+
"acc_norm_stderr": 0.041546596717075474
|
| 134 |
+
},
|
| 135 |
+
"harness|hendrycksTest-elementary_mathematics|5": {
|
| 136 |
+
"acc": 0.4126984126984127,
|
| 137 |
+
"acc_stderr": 0.02535574126305527,
|
| 138 |
+
"acc_norm": 0.4126984126984127,
|
| 139 |
+
"acc_norm_stderr": 0.02535574126305527
|
| 140 |
+
},
|
| 141 |
+
"harness|hendrycksTest-formal_logic|5": {
|
| 142 |
+
"acc": 0.4603174603174603,
|
| 143 |
+
"acc_stderr": 0.04458029125470973,
|
| 144 |
+
"acc_norm": 0.4603174603174603,
|
| 145 |
+
"acc_norm_stderr": 0.04458029125470973
|
| 146 |
+
},
|
| 147 |
+
"harness|hendrycksTest-global_facts|5": {
|
| 148 |
+
"acc": 0.39,
|
| 149 |
+
"acc_stderr": 0.04902071300001975,
|
| 150 |
+
"acc_norm": 0.39,
|
| 151 |
+
"acc_norm_stderr": 0.04902071300001975
|
| 152 |
+
},
|
| 153 |
+
"harness|hendrycksTest-high_school_biology|5": {
|
| 154 |
+
"acc": 0.7967741935483871,
|
| 155 |
+
"acc_stderr": 0.02289168798455496,
|
| 156 |
+
"acc_norm": 0.7967741935483871,
|
| 157 |
+
"acc_norm_stderr": 0.02289168798455496
|
| 158 |
+
},
|
| 159 |
+
"harness|hendrycksTest-high_school_chemistry|5": {
|
| 160 |
+
"acc": 0.5073891625615764,
|
| 161 |
+
"acc_stderr": 0.035176035403610105,
|
| 162 |
+
"acc_norm": 0.5073891625615764,
|
| 163 |
+
"acc_norm_stderr": 0.035176035403610105
|
| 164 |
+
},
|
| 165 |
+
"harness|hendrycksTest-high_school_computer_science|5": {
|
| 166 |
+
"acc": 0.71,
|
| 167 |
+
"acc_stderr": 0.045604802157206845,
|
| 168 |
+
"acc_norm": 0.71,
|
| 169 |
+
"acc_norm_stderr": 0.045604802157206845
|
| 170 |
+
},
|
| 171 |
+
"harness|hendrycksTest-high_school_european_history|5": {
|
| 172 |
+
"acc": 0.7878787878787878,
|
| 173 |
+
"acc_stderr": 0.031922715695483,
|
| 174 |
+
"acc_norm": 0.7878787878787878,
|
| 175 |
+
"acc_norm_stderr": 0.031922715695483
|
| 176 |
+
},
|
| 177 |
+
"harness|hendrycksTest-high_school_geography|5": {
|
| 178 |
+
"acc": 0.7878787878787878,
|
| 179 |
+
"acc_stderr": 0.029126522834586815,
|
| 180 |
+
"acc_norm": 0.7878787878787878,
|
| 181 |
+
"acc_norm_stderr": 0.029126522834586815
|
| 182 |
+
},
|
| 183 |
+
"harness|hendrycksTest-high_school_government_and_politics|5": {
|
| 184 |
+
"acc": 0.8963730569948186,
|
| 185 |
+
"acc_stderr": 0.021995311963644237,
|
| 186 |
+
"acc_norm": 0.8963730569948186,
|
| 187 |
+
"acc_norm_stderr": 0.021995311963644237
|
| 188 |
+
},
|
| 189 |
+
"harness|hendrycksTest-high_school_macroeconomics|5": {
|
| 190 |
+
"acc": 0.6846153846153846,
|
| 191 |
+
"acc_stderr": 0.02355964698318994,
|
| 192 |
+
"acc_norm": 0.6846153846153846,
|
| 193 |
+
"acc_norm_stderr": 0.02355964698318994
|
| 194 |
+
},
|
| 195 |
+
"harness|hendrycksTest-high_school_mathematics|5": {
|
| 196 |
+
"acc": 0.35185185185185186,
|
| 197 |
+
"acc_stderr": 0.02911661760608301,
|
| 198 |
+
"acc_norm": 0.35185185185185186,
|
| 199 |
+
"acc_norm_stderr": 0.02911661760608301
|
| 200 |
+
},
|
| 201 |
+
"harness|hendrycksTest-high_school_microeconomics|5": {
|
| 202 |
+
"acc": 0.7100840336134454,
|
| 203 |
+
"acc_stderr": 0.029472485833136094,
|
| 204 |
+
"acc_norm": 0.7100840336134454,
|
| 205 |
+
"acc_norm_stderr": 0.029472485833136094
|
| 206 |
+
},
|
| 207 |
+
"harness|hendrycksTest-high_school_physics|5": {
|
| 208 |
+
"acc": 0.3708609271523179,
|
| 209 |
+
"acc_stderr": 0.03943966699183629,
|
| 210 |
+
"acc_norm": 0.3708609271523179,
|
| 211 |
+
"acc_norm_stderr": 0.03943966699183629
|
| 212 |
+
},
|
| 213 |
+
"harness|hendrycksTest-high_school_psychology|5": {
|
| 214 |
+
"acc": 0.8293577981651377,
|
| 215 |
+
"acc_stderr": 0.016129271025099857,
|
| 216 |
+
"acc_norm": 0.8293577981651377,
|
| 217 |
+
"acc_norm_stderr": 0.016129271025099857
|
| 218 |
+
},
|
| 219 |
+
"harness|hendrycksTest-high_school_statistics|5": {
|
| 220 |
+
"acc": 0.5462962962962963,
|
| 221 |
+
"acc_stderr": 0.033953227263757976,
|
| 222 |
+
"acc_norm": 0.5462962962962963,
|
| 223 |
+
"acc_norm_stderr": 0.033953227263757976
|
| 224 |
+
},
|
| 225 |
+
"harness|hendrycksTest-high_school_us_history|5": {
|
| 226 |
+
"acc": 0.8333333333333334,
|
| 227 |
+
"acc_stderr": 0.026156867523931045,
|
| 228 |
+
"acc_norm": 0.8333333333333334,
|
| 229 |
+
"acc_norm_stderr": 0.026156867523931045
|
| 230 |
+
},
|
| 231 |
+
"harness|hendrycksTest-high_school_world_history|5": {
|
| 232 |
+
"acc": 0.8227848101265823,
|
| 233 |
+
"acc_stderr": 0.024856364184503224,
|
| 234 |
+
"acc_norm": 0.8227848101265823,
|
| 235 |
+
"acc_norm_stderr": 0.024856364184503224
|
| 236 |
+
},
|
| 237 |
+
"harness|hendrycksTest-human_aging|5": {
|
| 238 |
+
"acc": 0.7130044843049327,
|
| 239 |
+
"acc_stderr": 0.03036037971029195,
|
| 240 |
+
"acc_norm": 0.7130044843049327,
|
| 241 |
+
"acc_norm_stderr": 0.03036037971029195
|
| 242 |
+
},
|
| 243 |
+
"harness|hendrycksTest-human_sexuality|5": {
|
| 244 |
+
"acc": 0.7709923664122137,
|
| 245 |
+
"acc_stderr": 0.036853466317118506,
|
| 246 |
+
"acc_norm": 0.7709923664122137,
|
| 247 |
+
"acc_norm_stderr": 0.036853466317118506
|
| 248 |
+
},
|
| 249 |
+
"harness|hendrycksTest-international_law|5": {
|
| 250 |
+
"acc": 0.7520661157024794,
|
| 251 |
+
"acc_stderr": 0.03941897526516302,
|
| 252 |
+
"acc_norm": 0.7520661157024794,
|
| 253 |
+
"acc_norm_stderr": 0.03941897526516302
|
| 254 |
+
},
|
| 255 |
+
"harness|hendrycksTest-jurisprudence|5": {
|
| 256 |
+
"acc": 0.7592592592592593,
|
| 257 |
+
"acc_stderr": 0.04133119440243839,
|
| 258 |
+
"acc_norm": 0.7592592592592593,
|
| 259 |
+
"acc_norm_stderr": 0.04133119440243839
|
| 260 |
+
},
|
| 261 |
+
"harness|hendrycksTest-logical_fallacies|5": {
|
| 262 |
+
"acc": 0.754601226993865,
|
| 263 |
+
"acc_stderr": 0.03380939813943354,
|
| 264 |
+
"acc_norm": 0.754601226993865,
|
| 265 |
+
"acc_norm_stderr": 0.03380939813943354
|
| 266 |
+
},
|
| 267 |
+
"harness|hendrycksTest-machine_learning|5": {
|
| 268 |
+
"acc": 0.5089285714285714,
|
| 269 |
+
"acc_stderr": 0.04745033255489123,
|
| 270 |
+
"acc_norm": 0.5089285714285714,
|
| 271 |
+
"acc_norm_stderr": 0.04745033255489123
|
| 272 |
+
},
|
| 273 |
+
"harness|hendrycksTest-management|5": {
|
| 274 |
+
"acc": 0.7378640776699029,
|
| 275 |
+
"acc_stderr": 0.043546310772605956,
|
| 276 |
+
"acc_norm": 0.7378640776699029,
|
| 277 |
+
"acc_norm_stderr": 0.043546310772605956
|
| 278 |
+
},
|
| 279 |
+
"harness|hendrycksTest-marketing|5": {
|
| 280 |
+
"acc": 0.8547008547008547,
|
| 281 |
+
"acc_stderr": 0.023086635086841407,
|
| 282 |
+
"acc_norm": 0.8547008547008547,
|
| 283 |
+
"acc_norm_stderr": 0.023086635086841407
|
| 284 |
+
},
|
| 285 |
+
"harness|hendrycksTest-medical_genetics|5": {
|
| 286 |
+
"acc": 0.72,
|
| 287 |
+
"acc_stderr": 0.045126085985421276,
|
| 288 |
+
"acc_norm": 0.72,
|
| 289 |
+
"acc_norm_stderr": 0.045126085985421276
|
| 290 |
+
},
|
| 291 |
+
"harness|hendrycksTest-miscellaneous|5": {
|
| 292 |
+
"acc": 0.8186462324393359,
|
| 293 |
+
"acc_stderr": 0.01377869377846408,
|
| 294 |
+
"acc_norm": 0.8186462324393359,
|
| 295 |
+
"acc_norm_stderr": 0.01377869377846408
|
| 296 |
+
},
|
| 297 |
+
"harness|hendrycksTest-moral_disputes|5": {
|
| 298 |
+
"acc": 0.7341040462427746,
|
| 299 |
+
"acc_stderr": 0.023786203255508283,
|
| 300 |
+
"acc_norm": 0.7341040462427746,
|
| 301 |
+
"acc_norm_stderr": 0.023786203255508283
|
| 302 |
+
},
|
| 303 |
+
"harness|hendrycksTest-moral_scenarios|5": {
|
| 304 |
+
"acc": 0.329608938547486,
|
| 305 |
+
"acc_stderr": 0.01572153107518388,
|
| 306 |
+
"acc_norm": 0.329608938547486,
|
| 307 |
+
"acc_norm_stderr": 0.01572153107518388
|
| 308 |
+
},
|
| 309 |
+
"harness|hendrycksTest-nutrition|5": {
|
| 310 |
+
"acc": 0.7450980392156863,
|
| 311 |
+
"acc_stderr": 0.02495418432487991,
|
| 312 |
+
"acc_norm": 0.7450980392156863,
|
| 313 |
+
"acc_norm_stderr": 0.02495418432487991
|
| 314 |
+
},
|
| 315 |
+
"harness|hendrycksTest-philosophy|5": {
|
| 316 |
+
"acc": 0.6913183279742765,
|
| 317 |
+
"acc_stderr": 0.026236965881153266,
|
| 318 |
+
"acc_norm": 0.6913183279742765,
|
| 319 |
+
"acc_norm_stderr": 0.026236965881153266
|
| 320 |
+
},
|
| 321 |
+
"harness|hendrycksTest-prehistory|5": {
|
| 322 |
+
"acc": 0.7314814814814815,
|
| 323 |
+
"acc_stderr": 0.02465968518596728,
|
| 324 |
+
"acc_norm": 0.7314814814814815,
|
| 325 |
+
"acc_norm_stderr": 0.02465968518596728
|
| 326 |
+
},
|
| 327 |
+
"harness|hendrycksTest-professional_accounting|5": {
|
| 328 |
+
"acc": 0.49645390070921985,
|
| 329 |
+
"acc_stderr": 0.02982674915328092,
|
| 330 |
+
"acc_norm": 0.49645390070921985,
|
| 331 |
+
"acc_norm_stderr": 0.02982674915328092
|
| 332 |
+
},
|
| 333 |
+
"harness|hendrycksTest-professional_law|5": {
|
| 334 |
+
"acc": 0.4817470664928292,
|
| 335 |
+
"acc_stderr": 0.012761723960595472,
|
| 336 |
+
"acc_norm": 0.4817470664928292,
|
| 337 |
+
"acc_norm_stderr": 0.012761723960595472
|
| 338 |
+
},
|
| 339 |
+
"harness|hendrycksTest-professional_medicine|5": {
|
| 340 |
+
"acc": 0.6875,
|
| 341 |
+
"acc_stderr": 0.02815637344037142,
|
| 342 |
+
"acc_norm": 0.6875,
|
| 343 |
+
"acc_norm_stderr": 0.02815637344037142
|
| 344 |
+
},
|
| 345 |
+
"harness|hendrycksTest-professional_psychology|5": {
|
| 346 |
+
"acc": 0.6454248366013072,
|
| 347 |
+
"acc_stderr": 0.0193533605475537,
|
| 348 |
+
"acc_norm": 0.6454248366013072,
|
| 349 |
+
"acc_norm_stderr": 0.0193533605475537
|
| 350 |
+
},
|
| 351 |
+
"harness|hendrycksTest-public_relations|5": {
|
| 352 |
+
"acc": 0.6727272727272727,
|
| 353 |
+
"acc_stderr": 0.0449429086625209,
|
| 354 |
+
"acc_norm": 0.6727272727272727,
|
| 355 |
+
"acc_norm_stderr": 0.0449429086625209
|
| 356 |
+
},
|
| 357 |
+
"harness|hendrycksTest-security_studies|5": {
|
| 358 |
+
"acc": 0.7306122448979592,
|
| 359 |
+
"acc_stderr": 0.02840125202902294,
|
| 360 |
+
"acc_norm": 0.7306122448979592,
|
| 361 |
+
"acc_norm_stderr": 0.02840125202902294
|
| 362 |
+
},
|
| 363 |
+
"harness|hendrycksTest-sociology|5": {
|
| 364 |
+
"acc": 0.8159203980099502,
|
| 365 |
+
"acc_stderr": 0.027403859410786845,
|
| 366 |
+
"acc_norm": 0.8159203980099502,
|
| 367 |
+
"acc_norm_stderr": 0.027403859410786845
|
| 368 |
+
},
|
| 369 |
+
"harness|hendrycksTest-us_foreign_policy|5": {
|
| 370 |
+
"acc": 0.88,
|
| 371 |
+
"acc_stderr": 0.03265986323710906,
|
| 372 |
+
"acc_norm": 0.88,
|
| 373 |
+
"acc_norm_stderr": 0.03265986323710906
|
| 374 |
+
},
|
| 375 |
+
"harness|hendrycksTest-virology|5": {
|
| 376 |
+
"acc": 0.5301204819277109,
|
| 377 |
+
"acc_stderr": 0.03885425420866767,
|
| 378 |
+
"acc_norm": 0.5301204819277109,
|
| 379 |
+
"acc_norm_stderr": 0.03885425420866767
|
| 380 |
+
},
|
| 381 |
+
"harness|hendrycksTest-world_religions|5": {
|
| 382 |
+
"acc": 0.8538011695906432,
|
| 383 |
+
"acc_stderr": 0.02709729011807081,
|
| 384 |
+
"acc_norm": 0.8538011695906432,
|
| 385 |
+
"acc_norm_stderr": 0.02709729011807081
|
| 386 |
+
},
|
| 387 |
+
"harness|truthfulqa:mc|0": {
|
| 388 |
+
"mc1": 0.4283965728274174,
|
| 389 |
+
"mc1_stderr": 0.017323088597314757,
|
| 390 |
+
"mc2": 0.5998074537794252,
|
| 391 |
+
"mc2_stderr": 0.015494960379071198
|
| 392 |
+
},
|
| 393 |
+
"harness|winogrande|5": {
|
| 394 |
+
"acc": 0.7853196527229677,
|
| 395 |
+
"acc_stderr": 0.011539912734345403
|
| 396 |
+
},
|
| 397 |
+
"harness|gsm8k|5": {
|
| 398 |
+
"acc": 0.6156178923426838,
|
| 399 |
+
"acc_stderr": 0.013399219253698186
|
| 400 |
+
}
|
| 401 |
+
}
|