Commit
•
b147afc
1
Parent(s):
f364909
Update README.md
Browse files
README.md
CHANGED
|
@@ -73,4 +73,389 @@ pipeline = transformers.pipeline(
|
|
| 73 |
|
| 74 |
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 75 |
print(outputs[0]["generated_text"])
|
| 76 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
|
| 74 |
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 75 |
print(outputs[0]["generated_text"])
|
| 76 |
+
```
|
| 77 |
+
|
| 78 |
+
# Evaluation results:
|
| 79 |
+
|
| 80 |
+
```json
|
| 81 |
+
{
|
| 82 |
+
"all": {
|
| 83 |
+
"acc": 0.635008846776534,
|
| 84 |
+
"acc_stderr": 0.03244450973873997,
|
| 85 |
+
"acc_norm": 0.6365238167399629,
|
| 86 |
+
"acc_norm_stderr": 0.033101612504829854,
|
| 87 |
+
"mc1": 0.397796817625459,
|
| 88 |
+
"mc1_stderr": 0.017133934248559635,
|
| 89 |
+
"mc2": 0.5816259277988214,
|
| 90 |
+
"mc2_stderr": 0.01521267822060948
|
| 91 |
+
},
|
| 92 |
+
"harness|arc:challenge|25": {
|
| 93 |
+
"acc": 0.6220136518771331,
|
| 94 |
+
"acc_stderr": 0.0141696645203031,
|
| 95 |
+
"acc_norm": 0.6459044368600683,
|
| 96 |
+
"acc_norm_stderr": 0.013975454122756557
|
| 97 |
+
},
|
| 98 |
+
"harness|hellaswag|10": {
|
| 99 |
+
"acc": 0.6512646883091018,
|
| 100 |
+
"acc_stderr": 0.004755960559929163,
|
| 101 |
+
"acc_norm": 0.8397729535949015,
|
| 102 |
+
"acc_norm_stderr": 0.003660668242740655
|
| 103 |
+
},
|
| 104 |
+
"harness|hendrycksTest-abstract_algebra|5": {
|
| 105 |
+
"acc": 0.4,
|
| 106 |
+
"acc_stderr": 0.04923659639173309,
|
| 107 |
+
"acc_norm": 0.4,
|
| 108 |
+
"acc_norm_stderr": 0.04923659639173309
|
| 109 |
+
},
|
| 110 |
+
"harness|hendrycksTest-anatomy|5": {
|
| 111 |
+
"acc": 0.5703703703703704,
|
| 112 |
+
"acc_stderr": 0.042763494943765995,
|
| 113 |
+
"acc_norm": 0.5703703703703704,
|
| 114 |
+
"acc_norm_stderr": 0.042763494943765995
|
| 115 |
+
},
|
| 116 |
+
"harness|hendrycksTest-astronomy|5": {
|
| 117 |
+
"acc": 0.6842105263157895,
|
| 118 |
+
"acc_stderr": 0.0378272898086547,
|
| 119 |
+
"acc_norm": 0.6842105263157895,
|
| 120 |
+
"acc_norm_stderr": 0.0378272898086547
|
| 121 |
+
},
|
| 122 |
+
"harness|hendrycksTest-business_ethics|5": {
|
| 123 |
+
"acc": 0.58,
|
| 124 |
+
"acc_stderr": 0.049604496374885836,
|
| 125 |
+
"acc_norm": 0.58,
|
| 126 |
+
"acc_norm_stderr": 0.049604496374885836
|
| 127 |
+
},
|
| 128 |
+
"harness|hendrycksTest-clinical_knowledge|5": {
|
| 129 |
+
"acc": 0.6792452830188679,
|
| 130 |
+
"acc_stderr": 0.028727502957880267,
|
| 131 |
+
"acc_norm": 0.6792452830188679,
|
| 132 |
+
"acc_norm_stderr": 0.028727502957880267
|
| 133 |
+
},
|
| 134 |
+
"harness|hendrycksTest-college_biology|5": {
|
| 135 |
+
"acc": 0.7361111111111112,
|
| 136 |
+
"acc_stderr": 0.03685651095897532,
|
| 137 |
+
"acc_norm": 0.7361111111111112,
|
| 138 |
+
"acc_norm_stderr": 0.03685651095897532
|
| 139 |
+
},
|
| 140 |
+
"harness|hendrycksTest-college_chemistry|5": {
|
| 141 |
+
"acc": 0.54,
|
| 142 |
+
"acc_stderr": 0.05009082659620332,
|
| 143 |
+
"acc_norm": 0.54,
|
| 144 |
+
"acc_norm_stderr": 0.05009082659620332
|
| 145 |
+
},
|
| 146 |
+
"harness|hendrycksTest-college_computer_science|5": {
|
| 147 |
+
"acc": 0.51,
|
| 148 |
+
"acc_stderr": 0.05024183937956912,
|
| 149 |
+
"acc_norm": 0.51,
|
| 150 |
+
"acc_norm_stderr": 0.05024183937956912
|
| 151 |
+
},
|
| 152 |
+
"harness|hendrycksTest-college_mathematics|5": {
|
| 153 |
+
"acc": 0.29,
|
| 154 |
+
"acc_stderr": 0.04560480215720684,
|
| 155 |
+
"acc_norm": 0.29,
|
| 156 |
+
"acc_norm_stderr": 0.04560480215720684
|
| 157 |
+
},
|
| 158 |
+
"harness|hendrycksTest-college_medicine|5": {
|
| 159 |
+
"acc": 0.6416184971098265,
|
| 160 |
+
"acc_stderr": 0.036563436533531585,
|
| 161 |
+
"acc_norm": 0.6416184971098265,
|
| 162 |
+
"acc_norm_stderr": 0.036563436533531585
|
| 163 |
+
},
|
| 164 |
+
"harness|hendrycksTest-college_physics|5": {
|
| 165 |
+
"acc": 0.3235294117647059,
|
| 166 |
+
"acc_stderr": 0.04655010411319619,
|
| 167 |
+
"acc_norm": 0.3235294117647059,
|
| 168 |
+
"acc_norm_stderr": 0.04655010411319619
|
| 169 |
+
},
|
| 170 |
+
"harness|hendrycksTest-computer_security|5": {
|
| 171 |
+
"acc": 0.76,
|
| 172 |
+
"acc_stderr": 0.04292346959909283,
|
| 173 |
+
"acc_norm": 0.76,
|
| 174 |
+
"acc_norm_stderr": 0.04292346959909283
|
| 175 |
+
},
|
| 176 |
+
"harness|hendrycksTest-conceptual_physics|5": {
|
| 177 |
+
"acc": 0.5829787234042553,
|
| 178 |
+
"acc_stderr": 0.03223276266711712,
|
| 179 |
+
"acc_norm": 0.5829787234042553,
|
| 180 |
+
"acc_norm_stderr": 0.03223276266711712
|
| 181 |
+
},
|
| 182 |
+
"harness|hendrycksTest-econometrics|5": {
|
| 183 |
+
"acc": 0.4649122807017544,
|
| 184 |
+
"acc_stderr": 0.046920083813689104,
|
| 185 |
+
"acc_norm": 0.4649122807017544,
|
| 186 |
+
"acc_norm_stderr": 0.046920083813689104
|
| 187 |
+
},
|
| 188 |
+
"harness|hendrycksTest-electrical_engineering|5": {
|
| 189 |
+
"acc": 0.5517241379310345,
|
| 190 |
+
"acc_stderr": 0.04144311810878152,
|
| 191 |
+
"acc_norm": 0.5517241379310345,
|
| 192 |
+
"acc_norm_stderr": 0.04144311810878152
|
| 193 |
+
},
|
| 194 |
+
"harness|hendrycksTest-elementary_mathematics|5": {
|
| 195 |
+
"acc": 0.42063492063492064,
|
| 196 |
+
"acc_stderr": 0.025424835086924006,
|
| 197 |
+
"acc_norm": 0.42063492063492064,
|
| 198 |
+
"acc_norm_stderr": 0.025424835086924006
|
| 199 |
+
},
|
| 200 |
+
"harness|hendrycksTest-formal_logic|5": {
|
| 201 |
+
"acc": 0.4444444444444444,
|
| 202 |
+
"acc_stderr": 0.044444444444444495,
|
| 203 |
+
"acc_norm": 0.4444444444444444,
|
| 204 |
+
"acc_norm_stderr": 0.044444444444444495
|
| 205 |
+
},
|
| 206 |
+
"harness|hendrycksTest-global_facts|5": {
|
| 207 |
+
"acc": 0.44,
|
| 208 |
+
"acc_stderr": 0.04988876515698589,
|
| 209 |
+
"acc_norm": 0.44,
|
| 210 |
+
"acc_norm_stderr": 0.04988876515698589
|
| 211 |
+
},
|
| 212 |
+
"harness|hendrycksTest-high_school_biology|5": {
|
| 213 |
+
"acc": 0.7548387096774194,
|
| 214 |
+
"acc_stderr": 0.024472243840895525,
|
| 215 |
+
"acc_norm": 0.7548387096774194,
|
| 216 |
+
"acc_norm_stderr": 0.024472243840895525
|
| 217 |
+
},
|
| 218 |
+
"harness|hendrycksTest-high_school_chemistry|5": {
|
| 219 |
+
"acc": 0.5024630541871922,
|
| 220 |
+
"acc_stderr": 0.035179450386910616,
|
| 221 |
+
"acc_norm": 0.5024630541871922,
|
| 222 |
+
"acc_norm_stderr": 0.035179450386910616
|
| 223 |
+
},
|
| 224 |
+
"harness|hendrycksTest-high_school_computer_science|5": {
|
| 225 |
+
"acc": 0.66,
|
| 226 |
+
"acc_stderr": 0.04760952285695237,
|
| 227 |
+
"acc_norm": 0.66,
|
| 228 |
+
"acc_norm_stderr": 0.04760952285695237
|
| 229 |
+
},
|
| 230 |
+
"harness|hendrycksTest-high_school_european_history|5": {
|
| 231 |
+
"acc": 0.7818181818181819,
|
| 232 |
+
"acc_stderr": 0.03225078108306289,
|
| 233 |
+
"acc_norm": 0.7818181818181819,
|
| 234 |
+
"acc_norm_stderr": 0.03225078108306289
|
| 235 |
+
},
|
| 236 |
+
"harness|hendrycksTest-high_school_geography|5": {
|
| 237 |
+
"acc": 0.797979797979798,
|
| 238 |
+
"acc_stderr": 0.02860620428922988,
|
| 239 |
+
"acc_norm": 0.797979797979798,
|
| 240 |
+
"acc_norm_stderr": 0.02860620428922988
|
| 241 |
+
},
|
| 242 |
+
"harness|hendrycksTest-high_school_government_and_politics|5": {
|
| 243 |
+
"acc": 0.8756476683937824,
|
| 244 |
+
"acc_stderr": 0.023814477086593552,
|
| 245 |
+
"acc_norm": 0.8756476683937824,
|
| 246 |
+
"acc_norm_stderr": 0.023814477086593552
|
| 247 |
+
},
|
| 248 |
+
"harness|hendrycksTest-high_school_macroeconomics|5": {
|
| 249 |
+
"acc": 0.658974358974359,
|
| 250 |
+
"acc_stderr": 0.02403548967633509,
|
| 251 |
+
"acc_norm": 0.658974358974359,
|
| 252 |
+
"acc_norm_stderr": 0.02403548967633509
|
| 253 |
+
},
|
| 254 |
+
"harness|hendrycksTest-high_school_mathematics|5": {
|
| 255 |
+
"acc": 0.32592592592592595,
|
| 256 |
+
"acc_stderr": 0.02857834836547308,
|
| 257 |
+
"acc_norm": 0.32592592592592595,
|
| 258 |
+
"acc_norm_stderr": 0.02857834836547308
|
| 259 |
+
},
|
| 260 |
+
"harness|hendrycksTest-high_school_microeconomics|5": {
|
| 261 |
+
"acc": 0.6638655462184874,
|
| 262 |
+
"acc_stderr": 0.030684737115135363,
|
| 263 |
+
"acc_norm": 0.6638655462184874,
|
| 264 |
+
"acc_norm_stderr": 0.030684737115135363
|
| 265 |
+
},
|
| 266 |
+
"harness|hendrycksTest-high_school_physics|5": {
|
| 267 |
+
"acc": 0.304635761589404,
|
| 268 |
+
"acc_stderr": 0.03757949922943344,
|
| 269 |
+
"acc_norm": 0.304635761589404,
|
| 270 |
+
"acc_norm_stderr": 0.03757949922943344
|
| 271 |
+
},
|
| 272 |
+
"harness|hendrycksTest-high_school_psychology|5": {
|
| 273 |
+
"acc": 0.8238532110091743,
|
| 274 |
+
"acc_stderr": 0.016332882393431353,
|
| 275 |
+
"acc_norm": 0.8238532110091743,
|
| 276 |
+
"acc_norm_stderr": 0.016332882393431353
|
| 277 |
+
},
|
| 278 |
+
"harness|hendrycksTest-high_school_statistics|5": {
|
| 279 |
+
"acc": 0.5092592592592593,
|
| 280 |
+
"acc_stderr": 0.03409386946992699,
|
| 281 |
+
"acc_norm": 0.5092592592592593,
|
| 282 |
+
"acc_norm_stderr": 0.03409386946992699
|
| 283 |
+
},
|
| 284 |
+
"harness|hendrycksTest-high_school_us_history|5": {
|
| 285 |
+
"acc": 0.7990196078431373,
|
| 286 |
+
"acc_stderr": 0.02812597226565437,
|
| 287 |
+
"acc_norm": 0.7990196078431373,
|
| 288 |
+
"acc_norm_stderr": 0.02812597226565437
|
| 289 |
+
},
|
| 290 |
+
"harness|hendrycksTest-high_school_world_history|5": {
|
| 291 |
+
"acc": 0.759493670886076,
|
| 292 |
+
"acc_stderr": 0.027820781981149685,
|
| 293 |
+
"acc_norm": 0.759493670886076,
|
| 294 |
+
"acc_norm_stderr": 0.027820781981149685
|
| 295 |
+
},
|
| 296 |
+
"harness|hendrycksTest-human_aging|5": {
|
| 297 |
+
"acc": 0.6681614349775785,
|
| 298 |
+
"acc_stderr": 0.03160295143776679,
|
| 299 |
+
"acc_norm": 0.6681614349775785,
|
| 300 |
+
"acc_norm_stderr": 0.03160295143776679
|
| 301 |
+
},
|
| 302 |
+
"harness|hendrycksTest-human_sexuality|5": {
|
| 303 |
+
"acc": 0.7404580152671756,
|
| 304 |
+
"acc_stderr": 0.03844876139785271,
|
| 305 |
+
"acc_norm": 0.7404580152671756,
|
| 306 |
+
"acc_norm_stderr": 0.03844876139785271
|
| 307 |
+
},
|
| 308 |
+
"harness|hendrycksTest-international_law|5": {
|
| 309 |
+
"acc": 0.8016528925619835,
|
| 310 |
+
"acc_stderr": 0.036401182719909456,
|
| 311 |
+
"acc_norm": 0.8016528925619835,
|
| 312 |
+
"acc_norm_stderr": 0.036401182719909456
|
| 313 |
+
},
|
| 314 |
+
"harness|hendrycksTest-jurisprudence|5": {
|
| 315 |
+
"acc": 0.7777777777777778,
|
| 316 |
+
"acc_stderr": 0.040191074725573483,
|
| 317 |
+
"acc_norm": 0.7777777777777778,
|
| 318 |
+
"acc_norm_stderr": 0.040191074725573483
|
| 319 |
+
},
|
| 320 |
+
"harness|hendrycksTest-logical_fallacies|5": {
|
| 321 |
+
"acc": 0.754601226993865,
|
| 322 |
+
"acc_stderr": 0.03380939813943354,
|
| 323 |
+
"acc_norm": 0.754601226993865,
|
| 324 |
+
"acc_norm_stderr": 0.03380939813943354
|
| 325 |
+
},
|
| 326 |
+
"harness|hendrycksTest-machine_learning|5": {
|
| 327 |
+
"acc": 0.4732142857142857,
|
| 328 |
+
"acc_stderr": 0.047389751192741546,
|
| 329 |
+
"acc_norm": 0.4732142857142857,
|
| 330 |
+
"acc_norm_stderr": 0.047389751192741546
|
| 331 |
+
},
|
| 332 |
+
"harness|hendrycksTest-management|5": {
|
| 333 |
+
"acc": 0.7766990291262136,
|
| 334 |
+
"acc_stderr": 0.04123553189891431,
|
| 335 |
+
"acc_norm": 0.7766990291262136,
|
| 336 |
+
"acc_norm_stderr": 0.04123553189891431
|
| 337 |
+
},
|
| 338 |
+
"harness|hendrycksTest-marketing|5": {
|
| 339 |
+
"acc": 0.8632478632478633,
|
| 340 |
+
"acc_stderr": 0.022509033937077802,
|
| 341 |
+
"acc_norm": 0.8632478632478633,
|
| 342 |
+
"acc_norm_stderr": 0.022509033937077802
|
| 343 |
+
},
|
| 344 |
+
"harness|hendrycksTest-medical_genetics|5": {
|
| 345 |
+
"acc": 0.69,
|
| 346 |
+
"acc_stderr": 0.04648231987117316,
|
| 347 |
+
"acc_norm": 0.69,
|
| 348 |
+
"acc_norm_stderr": 0.04648231987117316
|
| 349 |
+
},
|
| 350 |
+
"harness|hendrycksTest-miscellaneous|5": {
|
| 351 |
+
"acc": 0.8173690932311622,
|
| 352 |
+
"acc_stderr": 0.013816335389973141,
|
| 353 |
+
"acc_norm": 0.8173690932311622,
|
| 354 |
+
"acc_norm_stderr": 0.013816335389973141
|
| 355 |
+
},
|
| 356 |
+
"harness|hendrycksTest-moral_disputes|5": {
|
| 357 |
+
"acc": 0.7254335260115607,
|
| 358 |
+
"acc_stderr": 0.02402774515526502,
|
| 359 |
+
"acc_norm": 0.7254335260115607,
|
| 360 |
+
"acc_norm_stderr": 0.02402774515526502
|
| 361 |
+
},
|
| 362 |
+
"harness|hendrycksTest-moral_scenarios|5": {
|
| 363 |
+
"acc": 0.27039106145251396,
|
| 364 |
+
"acc_stderr": 0.014854993938010071,
|
| 365 |
+
"acc_norm": 0.27039106145251396,
|
| 366 |
+
"acc_norm_stderr": 0.014854993938010071
|
| 367 |
+
},
|
| 368 |
+
"harness|hendrycksTest-nutrition|5": {
|
| 369 |
+
"acc": 0.7516339869281046,
|
| 370 |
+
"acc_stderr": 0.02473998135511359,
|
| 371 |
+
"acc_norm": 0.7516339869281046,
|
| 372 |
+
"acc_norm_stderr": 0.02473998135511359
|
| 373 |
+
},
|
| 374 |
+
"harness|hendrycksTest-philosophy|5": {
|
| 375 |
+
"acc": 0.7331189710610932,
|
| 376 |
+
"acc_stderr": 0.025122637608816653,
|
| 377 |
+
"acc_norm": 0.7331189710610932,
|
| 378 |
+
"acc_norm_stderr": 0.025122637608816653
|
| 379 |
+
},
|
| 380 |
+
"harness|hendrycksTest-prehistory|5": {
|
| 381 |
+
"acc": 0.7222222222222222,
|
| 382 |
+
"acc_stderr": 0.024922001168886324,
|
| 383 |
+
"acc_norm": 0.7222222222222222,
|
| 384 |
+
"acc_norm_stderr": 0.024922001168886324
|
| 385 |
+
},
|
| 386 |
+
"harness|hendrycksTest-professional_accounting|5": {
|
| 387 |
+
"acc": 0.46099290780141844,
|
| 388 |
+
"acc_stderr": 0.02973659252642444,
|
| 389 |
+
"acc_norm": 0.46099290780141844,
|
| 390 |
+
"acc_norm_stderr": 0.02973659252642444
|
| 391 |
+
},
|
| 392 |
+
"harness|hendrycksTest-professional_law|5": {
|
| 393 |
+
"acc": 0.4680573663624511,
|
| 394 |
+
"acc_stderr": 0.012744149704869647,
|
| 395 |
+
"acc_norm": 0.4680573663624511,
|
| 396 |
+
"acc_norm_stderr": 0.012744149704869647
|
| 397 |
+
},
|
| 398 |
+
"harness|hendrycksTest-professional_medicine|5": {
|
| 399 |
+
"acc": 0.6801470588235294,
|
| 400 |
+
"acc_stderr": 0.02833295951403121,
|
| 401 |
+
"acc_norm": 0.6801470588235294,
|
| 402 |
+
"acc_norm_stderr": 0.02833295951403121
|
| 403 |
+
},
|
| 404 |
+
"harness|hendrycksTest-professional_psychology|5": {
|
| 405 |
+
"acc": 0.6470588235294118,
|
| 406 |
+
"acc_stderr": 0.01933314202079716,
|
| 407 |
+
"acc_norm": 0.6470588235294118,
|
| 408 |
+
"acc_norm_stderr": 0.01933314202079716
|
| 409 |
+
},
|
| 410 |
+
"harness|hendrycksTest-public_relations|5": {
|
| 411 |
+
"acc": 0.6727272727272727,
|
| 412 |
+
"acc_stderr": 0.0449429086625209,
|
| 413 |
+
"acc_norm": 0.6727272727272727,
|
| 414 |
+
"acc_norm_stderr": 0.0449429086625209
|
| 415 |
+
},
|
| 416 |
+
"harness|hendrycksTest-security_studies|5": {
|
| 417 |
+
"acc": 0.6816326530612244,
|
| 418 |
+
"acc_stderr": 0.029822533793982062,
|
| 419 |
+
"acc_norm": 0.6816326530612244,
|
| 420 |
+
"acc_norm_stderr": 0.029822533793982062
|
| 421 |
+
},
|
| 422 |
+
"harness|hendrycksTest-sociology|5": {
|
| 423 |
+
"acc": 0.8507462686567164,
|
| 424 |
+
"acc_stderr": 0.025196929874827072,
|
| 425 |
+
"acc_norm": 0.8507462686567164,
|
| 426 |
+
"acc_norm_stderr": 0.025196929874827072
|
| 427 |
+
},
|
| 428 |
+
"harness|hendrycksTest-us_foreign_policy|5": {
|
| 429 |
+
"acc": 0.85,
|
| 430 |
+
"acc_stderr": 0.035887028128263734,
|
| 431 |
+
"acc_norm": 0.85,
|
| 432 |
+
"acc_norm_stderr": 0.035887028128263734
|
| 433 |
+
},
|
| 434 |
+
"harness|hendrycksTest-virology|5": {
|
| 435 |
+
"acc": 0.5180722891566265,
|
| 436 |
+
"acc_stderr": 0.03889951252827216,
|
| 437 |
+
"acc_norm": 0.5180722891566265,
|
| 438 |
+
"acc_norm_stderr": 0.03889951252827216
|
| 439 |
+
},
|
| 440 |
+
"harness|hendrycksTest-world_religions|5": {
|
| 441 |
+
"acc": 0.8362573099415205,
|
| 442 |
+
"acc_stderr": 0.028380919596145866,
|
| 443 |
+
"acc_norm": 0.8362573099415205,
|
| 444 |
+
"acc_norm_stderr": 0.028380919596145866
|
| 445 |
+
},
|
| 446 |
+
"harness|truthfulqa:mc|0": {
|
| 447 |
+
"mc1": 0.397796817625459,
|
| 448 |
+
"mc1_stderr": 0.017133934248559635,
|
| 449 |
+
"mc2": 0.5816259277988214,
|
| 450 |
+
"mc2_stderr": 0.01521267822060948
|
| 451 |
+
},
|
| 452 |
+
"harness|winogrande|5": {
|
| 453 |
+
"acc": 0.7963693764798737,
|
| 454 |
+
"acc_stderr": 0.011317798781626913
|
| 455 |
+
},
|
| 456 |
+
"harness|gsm8k|5": {
|
| 457 |
+
"acc": 0.5966641394996209,
|
| 458 |
+
"acc_stderr": 0.013512654781814702
|
| 459 |
+
}
|
| 460 |
+
}
|
| 461 |
+
```
|