Commit
•
f9930b0
1
Parent(s):
3ed7a2e
Update README.md
Browse files
README.md
CHANGED
|
@@ -49,4 +49,395 @@ prompt.
|
|
| 49 |
>>> prompt = "Gracie and Joe are choosing numbers on the complex plane. Joe chooses the point $1+2i$. Gracie chooses $-1+i$. How far apart are Gracie and Joe's points?"
|
| 50 |
>>> res = generate_text(prompt)
|
| 51 |
>>> res
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
```
|
|
|
|
| 49 |
>>> prompt = "Gracie and Joe are choosing numbers on the complex plane. Joe chooses the point $1+2i$. Gracie chooses $-1+i$. How far apart are Gracie and Joe's points?"
|
| 50 |
>>> res = generate_text(prompt)
|
| 51 |
>>> res
|
| 52 |
+
```
|
| 53 |
+
|
| 54 |
+
# Benchmark / Evaluation
|
| 55 |
+
|
| 56 |
+
| Model | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8k |
|
| 57 |
+
| ------- | -------- | -------- | ------- | -------- | -------- | ------- | -------- |
|
| 58 |
+
| Sharathhebbar24/math_gpt2_sft | 28.503 | 22.87 | 30.41 | 25.06 | 37.62 | 51.54 | 0.68 |
|
| 59 |
+
|
| 60 |
+
```python
|
| 61 |
+
{
|
| 62 |
+
"all": {
|
| 63 |
+
"acc": 0.25082189621988066,
|
| 64 |
+
"acc_stderr": 0.030526589726831692,
|
| 65 |
+
"acc_norm": 0.25112870356236633,
|
| 66 |
+
"acc_norm_stderr": 0.03129390389566968,
|
| 67 |
+
"mc1": 0.24112607099143207,
|
| 68 |
+
"mc1_stderr": 0.014974827279752334,
|
| 69 |
+
"mc2": 0.3762297840067963,
|
| 70 |
+
"mc2_stderr": 0.01445991036363257
|
| 71 |
+
},
|
| 72 |
+
"harness|arc:challenge|25": {
|
| 73 |
+
"acc": 0.20563139931740615,
|
| 74 |
+
"acc_stderr": 0.01181074526074258,
|
| 75 |
+
"acc_norm": 0.22866894197952217,
|
| 76 |
+
"acc_norm_stderr": 0.012272853582540799
|
| 77 |
+
},
|
| 78 |
+
"harness|hellaswag|10": {
|
| 79 |
+
"acc": 0.2884883489344752,
|
| 80 |
+
"acc_stderr": 0.004521334761709224,
|
| 81 |
+
"acc_norm": 0.30412268472415854,
|
| 82 |
+
"acc_norm_stderr": 0.00459094683972719
|
| 83 |
+
},
|
| 84 |
+
"harness|hendrycksTest-abstract_algebra|5": {
|
| 85 |
+
"acc": 0.19,
|
| 86 |
+
"acc_stderr": 0.03942772444036625,
|
| 87 |
+
"acc_norm": 0.19,
|
| 88 |
+
"acc_norm_stderr": 0.03942772444036625
|
| 89 |
+
},
|
| 90 |
+
"harness|hendrycksTest-anatomy|5": {
|
| 91 |
+
"acc": 0.2074074074074074,
|
| 92 |
+
"acc_stderr": 0.03502553170678319,
|
| 93 |
+
"acc_norm": 0.2074074074074074,
|
| 94 |
+
"acc_norm_stderr": 0.03502553170678319
|
| 95 |
+
},
|
| 96 |
+
"harness|hendrycksTest-astronomy|5": {
|
| 97 |
+
"acc": 0.17763157894736842,
|
| 98 |
+
"acc_stderr": 0.031103182383123398,
|
| 99 |
+
"acc_norm": 0.17763157894736842,
|
| 100 |
+
"acc_norm_stderr": 0.031103182383123398
|
| 101 |
+
},
|
| 102 |
+
"harness|hendrycksTest-business_ethics|5": {
|
| 103 |
+
"acc": 0.19,
|
| 104 |
+
"acc_stderr": 0.03942772444036622,
|
| 105 |
+
"acc_norm": 0.19,
|
| 106 |
+
"acc_norm_stderr": 0.03942772444036622
|
| 107 |
+
},
|
| 108 |
+
"harness|hendrycksTest-clinical_knowledge|5": {
|
| 109 |
+
"acc": 0.2188679245283019,
|
| 110 |
+
"acc_stderr": 0.025447863825108618,
|
| 111 |
+
"acc_norm": 0.2188679245283019,
|
| 112 |
+
"acc_norm_stderr": 0.025447863825108618
|
| 113 |
+
},
|
| 114 |
+
"harness|hendrycksTest-college_biology|5": {
|
| 115 |
+
"acc": 0.25,
|
| 116 |
+
"acc_stderr": 0.03621034121889507,
|
| 117 |
+
"acc_norm": 0.25,
|
| 118 |
+
"acc_norm_stderr": 0.03621034121889507
|
| 119 |
+
},
|
| 120 |
+
"harness|hendrycksTest-college_chemistry|5": {
|
| 121 |
+
"acc": 0.2,
|
| 122 |
+
"acc_stderr": 0.04020151261036845,
|
| 123 |
+
"acc_norm": 0.2,
|
| 124 |
+
"acc_norm_stderr": 0.04020151261036845
|
| 125 |
+
},
|
| 126 |
+
"harness|hendrycksTest-college_computer_science|5": {
|
| 127 |
+
"acc": 0.32,
|
| 128 |
+
"acc_stderr": 0.046882617226215034,
|
| 129 |
+
"acc_norm": 0.32,
|
| 130 |
+
"acc_norm_stderr": 0.046882617226215034
|
| 131 |
+
},
|
| 132 |
+
"harness|hendrycksTest-college_mathematics|5": {
|
| 133 |
+
"acc": 0.24,
|
| 134 |
+
"acc_stderr": 0.042923469599092816,
|
| 135 |
+
"acc_norm": 0.24,
|
| 136 |
+
"acc_norm_stderr": 0.042923469599092816
|
| 137 |
+
},
|
| 138 |
+
"harness|hendrycksTest-college_medicine|5": {
|
| 139 |
+
"acc": 0.21965317919075145,
|
| 140 |
+
"acc_stderr": 0.031568093627031744,
|
| 141 |
+
"acc_norm": 0.21965317919075145,
|
| 142 |
+
"acc_norm_stderr": 0.031568093627031744
|
| 143 |
+
},
|
| 144 |
+
"harness|hendrycksTest-college_physics|5": {
|
| 145 |
+
"acc": 0.23529411764705882,
|
| 146 |
+
"acc_stderr": 0.04220773659171453,
|
| 147 |
+
"acc_norm": 0.23529411764705882,
|
| 148 |
+
"acc_norm_stderr": 0.04220773659171453
|
| 149 |
+
},
|
| 150 |
+
"harness|hendrycksTest-computer_security|5": {
|
| 151 |
+
"acc": 0.23,
|
| 152 |
+
"acc_stderr": 0.04229525846816505,
|
| 153 |
+
"acc_norm": 0.23,
|
| 154 |
+
"acc_norm_stderr": 0.04229525846816505
|
| 155 |
+
},
|
| 156 |
+
"harness|hendrycksTest-conceptual_physics|5": {
|
| 157 |
+
"acc": 0.2680851063829787,
|
| 158 |
+
"acc_stderr": 0.028957342788342347,
|
| 159 |
+
"acc_norm": 0.2680851063829787,
|
| 160 |
+
"acc_norm_stderr": 0.028957342788342347
|
| 161 |
+
},
|
| 162 |
+
"harness|hendrycksTest-econometrics|5": {
|
| 163 |
+
"acc": 0.24561403508771928,
|
| 164 |
+
"acc_stderr": 0.040493392977481404,
|
| 165 |
+
"acc_norm": 0.24561403508771928,
|
| 166 |
+
"acc_norm_stderr": 0.040493392977481404
|
| 167 |
+
},
|
| 168 |
+
"harness|hendrycksTest-electrical_engineering|5": {
|
| 169 |
+
"acc": 0.2482758620689655,
|
| 170 |
+
"acc_stderr": 0.036001056927277716,
|
| 171 |
+
"acc_norm": 0.2482758620689655,
|
| 172 |
+
"acc_norm_stderr": 0.036001056927277716
|
| 173 |
+
},
|
| 174 |
+
"harness|hendrycksTest-elementary_mathematics|5": {
|
| 175 |
+
"acc": 0.24074074074074073,
|
| 176 |
+
"acc_stderr": 0.0220190800122179,
|
| 177 |
+
"acc_norm": 0.24074074074074073,
|
| 178 |
+
"acc_norm_stderr": 0.0220190800122179
|
| 179 |
+
},
|
| 180 |
+
"harness|hendrycksTest-formal_logic|5": {
|
| 181 |
+
"acc": 0.23015873015873015,
|
| 182 |
+
"acc_stderr": 0.03764950879790605,
|
| 183 |
+
"acc_norm": 0.23015873015873015,
|
| 184 |
+
"acc_norm_stderr": 0.03764950879790605
|
| 185 |
+
},
|
| 186 |
+
"harness|hendrycksTest-global_facts|5": {
|
| 187 |
+
"acc": 0.18,
|
| 188 |
+
"acc_stderr": 0.038612291966536934,
|
| 189 |
+
"acc_norm": 0.18,
|
| 190 |
+
"acc_norm_stderr": 0.038612291966536934
|
| 191 |
+
},
|
| 192 |
+
"harness|hendrycksTest-high_school_biology|5": {
|
| 193 |
+
"acc": 0.25483870967741934,
|
| 194 |
+
"acc_stderr": 0.024790118459332208,
|
| 195 |
+
"acc_norm": 0.25483870967741934,
|
| 196 |
+
"acc_norm_stderr": 0.024790118459332208
|
| 197 |
+
},
|
| 198 |
+
"harness|hendrycksTest-high_school_chemistry|5": {
|
| 199 |
+
"acc": 0.19704433497536947,
|
| 200 |
+
"acc_stderr": 0.02798672466673622,
|
| 201 |
+
"acc_norm": 0.19704433497536947,
|
| 202 |
+
"acc_norm_stderr": 0.02798672466673622
|
| 203 |
+
},
|
| 204 |
+
"harness|hendrycksTest-high_school_computer_science|5": {
|
| 205 |
+
"acc": 0.22,
|
| 206 |
+
"acc_stderr": 0.041633319989322695,
|
| 207 |
+
"acc_norm": 0.22,
|
| 208 |
+
"acc_norm_stderr": 0.041633319989322695
|
| 209 |
+
},
|
| 210 |
+
"harness|hendrycksTest-high_school_european_history|5": {
|
| 211 |
+
"acc": 0.19393939393939394,
|
| 212 |
+
"acc_stderr": 0.0308741451365621,
|
| 213 |
+
"acc_norm": 0.19393939393939394,
|
| 214 |
+
"acc_norm_stderr": 0.0308741451365621
|
| 215 |
+
},
|
| 216 |
+
"harness|hendrycksTest-high_school_geography|5": {
|
| 217 |
+
"acc": 0.3484848484848485,
|
| 218 |
+
"acc_stderr": 0.033948539651564025,
|
| 219 |
+
"acc_norm": 0.3484848484848485,
|
| 220 |
+
"acc_norm_stderr": 0.033948539651564025
|
| 221 |
+
},
|
| 222 |
+
"harness|hendrycksTest-high_school_government_and_politics|5": {
|
| 223 |
+
"acc": 0.32124352331606215,
|
| 224 |
+
"acc_stderr": 0.033699508685490674,
|
| 225 |
+
"acc_norm": 0.32124352331606215,
|
| 226 |
+
"acc_norm_stderr": 0.033699508685490674
|
| 227 |
+
},
|
| 228 |
+
"harness|hendrycksTest-high_school_macroeconomics|5": {
|
| 229 |
+
"acc": 0.23333333333333334,
|
| 230 |
+
"acc_stderr": 0.021444547301560476,
|
| 231 |
+
"acc_norm": 0.23333333333333334,
|
| 232 |
+
"acc_norm_stderr": 0.021444547301560476
|
| 233 |
+
},
|
| 234 |
+
"harness|hendrycksTest-high_school_mathematics|5": {
|
| 235 |
+
"acc": 0.2851851851851852,
|
| 236 |
+
"acc_stderr": 0.027528599210340492,
|
| 237 |
+
"acc_norm": 0.2851851851851852,
|
| 238 |
+
"acc_norm_stderr": 0.027528599210340492
|
| 239 |
+
},
|
| 240 |
+
"harness|hendrycksTest-high_school_microeconomics|5": {
|
| 241 |
+
"acc": 0.29831932773109243,
|
| 242 |
+
"acc_stderr": 0.029719142876342856,
|
| 243 |
+
"acc_norm": 0.29831932773109243,
|
| 244 |
+
"acc_norm_stderr": 0.029719142876342856
|
| 245 |
+
},
|
| 246 |
+
"harness|hendrycksTest-high_school_physics|5": {
|
| 247 |
+
"acc": 0.2781456953642384,
|
| 248 |
+
"acc_stderr": 0.03658603262763744,
|
| 249 |
+
"acc_norm": 0.2781456953642384,
|
| 250 |
+
"acc_norm_stderr": 0.03658603262763744
|
| 251 |
+
},
|
| 252 |
+
"harness|hendrycksTest-high_school_psychology|5": {
|
| 253 |
+
"acc": 0.26788990825688075,
|
| 254 |
+
"acc_stderr": 0.018987462257978652,
|
| 255 |
+
"acc_norm": 0.26788990825688075,
|
| 256 |
+
"acc_norm_stderr": 0.018987462257978652
|
| 257 |
+
},
|
| 258 |
+
"harness|hendrycksTest-high_school_statistics|5": {
|
| 259 |
+
"acc": 0.4351851851851852,
|
| 260 |
+
"acc_stderr": 0.03381200005643525,
|
| 261 |
+
"acc_norm": 0.4351851851851852,
|
| 262 |
+
"acc_norm_stderr": 0.03381200005643525
|
| 263 |
+
},
|
| 264 |
+
"harness|hendrycksTest-high_school_us_history|5": {
|
| 265 |
+
"acc": 0.2647058823529412,
|
| 266 |
+
"acc_stderr": 0.0309645179269234,
|
| 267 |
+
"acc_norm": 0.2647058823529412,
|
| 268 |
+
"acc_norm_stderr": 0.0309645179269234
|
| 269 |
+
},
|
| 270 |
+
"harness|hendrycksTest-high_school_world_history|5": {
|
| 271 |
+
"acc": 0.28270042194092826,
|
| 272 |
+
"acc_stderr": 0.029312814153955927,
|
| 273 |
+
"acc_norm": 0.28270042194092826,
|
| 274 |
+
"acc_norm_stderr": 0.029312814153955927
|
| 275 |
+
},
|
| 276 |
+
"harness|hendrycksTest-human_aging|5": {
|
| 277 |
+
"acc": 0.31390134529147984,
|
| 278 |
+
"acc_stderr": 0.031146796482972465,
|
| 279 |
+
"acc_norm": 0.31390134529147984,
|
| 280 |
+
"acc_norm_stderr": 0.031146796482972465
|
| 281 |
+
},
|
| 282 |
+
"harness|hendrycksTest-human_sexuality|5": {
|
| 283 |
+
"acc": 0.2595419847328244,
|
| 284 |
+
"acc_stderr": 0.03844876139785271,
|
| 285 |
+
"acc_norm": 0.2595419847328244,
|
| 286 |
+
"acc_norm_stderr": 0.03844876139785271
|
| 287 |
+
},
|
| 288 |
+
"harness|hendrycksTest-international_law|5": {
|
| 289 |
+
"acc": 0.2231404958677686,
|
| 290 |
+
"acc_stderr": 0.03800754475228733,
|
| 291 |
+
"acc_norm": 0.2231404958677686,
|
| 292 |
+
"acc_norm_stderr": 0.03800754475228733
|
| 293 |
+
},
|
| 294 |
+
"harness|hendrycksTest-jurisprudence|5": {
|
| 295 |
+
"acc": 0.25925925925925924,
|
| 296 |
+
"acc_stderr": 0.042365112580946336,
|
| 297 |
+
"acc_norm": 0.25925925925925924,
|
| 298 |
+
"acc_norm_stderr": 0.042365112580946336
|
| 299 |
+
},
|
| 300 |
+
"harness|hendrycksTest-logical_fallacies|5": {
|
| 301 |
+
"acc": 0.25153374233128833,
|
| 302 |
+
"acc_stderr": 0.03408997886857529,
|
| 303 |
+
"acc_norm": 0.25153374233128833,
|
| 304 |
+
"acc_norm_stderr": 0.03408997886857529
|
| 305 |
+
},
|
| 306 |
+
"harness|hendrycksTest-machine_learning|5": {
|
| 307 |
+
"acc": 0.29464285714285715,
|
| 308 |
+
"acc_stderr": 0.043270409325787296,
|
| 309 |
+
"acc_norm": 0.29464285714285715,
|
| 310 |
+
"acc_norm_stderr": 0.043270409325787296
|
| 311 |
+
},
|
| 312 |
+
"harness|hendrycksTest-management|5": {
|
| 313 |
+
"acc": 0.17475728155339806,
|
| 314 |
+
"acc_stderr": 0.037601780060266224,
|
| 315 |
+
"acc_norm": 0.17475728155339806,
|
| 316 |
+
"acc_norm_stderr": 0.037601780060266224
|
| 317 |
+
},
|
| 318 |
+
"harness|hendrycksTest-marketing|5": {
|
| 319 |
+
"acc": 0.20085470085470086,
|
| 320 |
+
"acc_stderr": 0.026246772946890488,
|
| 321 |
+
"acc_norm": 0.20085470085470086,
|
| 322 |
+
"acc_norm_stderr": 0.026246772946890488
|
| 323 |
+
},
|
| 324 |
+
"harness|hendrycksTest-medical_genetics|5": {
|
| 325 |
+
"acc": 0.3,
|
| 326 |
+
"acc_stderr": 0.046056618647183814,
|
| 327 |
+
"acc_norm": 0.3,
|
| 328 |
+
"acc_norm_stderr": 0.046056618647183814
|
| 329 |
+
},
|
| 330 |
+
"harness|hendrycksTest-miscellaneous|5": {
|
| 331 |
+
"acc": 0.23499361430395913,
|
| 332 |
+
"acc_stderr": 0.01516202415227844,
|
| 333 |
+
"acc_norm": 0.23499361430395913,
|
| 334 |
+
"acc_norm_stderr": 0.01516202415227844
|
| 335 |
+
},
|
| 336 |
+
"harness|hendrycksTest-moral_disputes|5": {
|
| 337 |
+
"acc": 0.23699421965317918,
|
| 338 |
+
"acc_stderr": 0.02289408248992599,
|
| 339 |
+
"acc_norm": 0.23699421965317918,
|
| 340 |
+
"acc_norm_stderr": 0.02289408248992599
|
| 341 |
+
},
|
| 342 |
+
"harness|hendrycksTest-moral_scenarios|5": {
|
| 343 |
+
"acc": 0.23798882681564246,
|
| 344 |
+
"acc_stderr": 0.014242630070574915,
|
| 345 |
+
"acc_norm": 0.23798882681564246,
|
| 346 |
+
"acc_norm_stderr": 0.014242630070574915
|
| 347 |
+
},
|
| 348 |
+
"harness|hendrycksTest-nutrition|5": {
|
| 349 |
+
"acc": 0.23202614379084968,
|
| 350 |
+
"acc_stderr": 0.024170840879341005,
|
| 351 |
+
"acc_norm": 0.23202614379084968,
|
| 352 |
+
"acc_norm_stderr": 0.024170840879341005
|
| 353 |
+
},
|
| 354 |
+
"harness|hendrycksTest-philosophy|5": {
|
| 355 |
+
"acc": 0.1864951768488746,
|
| 356 |
+
"acc_stderr": 0.02212243977248077,
|
| 357 |
+
"acc_norm": 0.1864951768488746,
|
| 358 |
+
"acc_norm_stderr": 0.02212243977248077
|
| 359 |
+
},
|
| 360 |
+
"harness|hendrycksTest-prehistory|5": {
|
| 361 |
+
"acc": 0.24074074074074073,
|
| 362 |
+
"acc_stderr": 0.02378858355165854,
|
| 363 |
+
"acc_norm": 0.24074074074074073,
|
| 364 |
+
"acc_norm_stderr": 0.02378858355165854
|
| 365 |
+
},
|
| 366 |
+
"harness|hendrycksTest-professional_accounting|5": {
|
| 367 |
+
"acc": 0.2695035460992908,
|
| 368 |
+
"acc_stderr": 0.026469036818590627,
|
| 369 |
+
"acc_norm": 0.2695035460992908,
|
| 370 |
+
"acc_norm_stderr": 0.026469036818590627
|
| 371 |
+
},
|
| 372 |
+
"harness|hendrycksTest-professional_law|5": {
|
| 373 |
+
"acc": 0.2529335071707953,
|
| 374 |
+
"acc_stderr": 0.011102268713839989,
|
| 375 |
+
"acc_norm": 0.2529335071707953,
|
| 376 |
+
"acc_norm_stderr": 0.011102268713839989
|
| 377 |
+
},
|
| 378 |
+
"harness|hendrycksTest-professional_medicine|5": {
|
| 379 |
+
"acc": 0.4411764705882353,
|
| 380 |
+
"acc_stderr": 0.030161911930767102,
|
| 381 |
+
"acc_norm": 0.4411764705882353,
|
| 382 |
+
"acc_norm_stderr": 0.030161911930767102
|
| 383 |
+
},
|
| 384 |
+
"harness|hendrycksTest-professional_psychology|5": {
|
| 385 |
+
"acc": 0.25,
|
| 386 |
+
"acc_stderr": 0.01751781884501444,
|
| 387 |
+
"acc_norm": 0.25,
|
| 388 |
+
"acc_norm_stderr": 0.01751781884501444
|
| 389 |
+
},
|
| 390 |
+
"harness|hendrycksTest-public_relations|5": {
|
| 391 |
+
"acc": 0.21818181818181817,
|
| 392 |
+
"acc_stderr": 0.03955932861795833,
|
| 393 |
+
"acc_norm": 0.21818181818181817,
|
| 394 |
+
"acc_norm_stderr": 0.03955932861795833
|
| 395 |
+
},
|
| 396 |
+
"harness|hendrycksTest-security_studies|5": {
|
| 397 |
+
"acc": 0.20408163265306123,
|
| 398 |
+
"acc_stderr": 0.025801283475090506,
|
| 399 |
+
"acc_norm": 0.20408163265306123,
|
| 400 |
+
"acc_norm_stderr": 0.025801283475090506
|
| 401 |
+
},
|
| 402 |
+
"harness|hendrycksTest-sociology|5": {
|
| 403 |
+
"acc": 0.24378109452736318,
|
| 404 |
+
"acc_stderr": 0.03036049015401465,
|
| 405 |
+
"acc_norm": 0.24378109452736318,
|
| 406 |
+
"acc_norm_stderr": 0.03036049015401465
|
| 407 |
+
},
|
| 408 |
+
"harness|hendrycksTest-us_foreign_policy|5": {
|
| 409 |
+
"acc": 0.24,
|
| 410 |
+
"acc_stderr": 0.04292346959909281,
|
| 411 |
+
"acc_norm": 0.24,
|
| 412 |
+
"acc_norm_stderr": 0.04292346959909281
|
| 413 |
+
},
|
| 414 |
+
"harness|hendrycksTest-virology|5": {
|
| 415 |
+
"acc": 0.22289156626506024,
|
| 416 |
+
"acc_stderr": 0.03240004825594687,
|
| 417 |
+
"acc_norm": 0.22289156626506024,
|
| 418 |
+
"acc_norm_stderr": 0.03240004825594687
|
| 419 |
+
},
|
| 420 |
+
"harness|hendrycksTest-world_religions|5": {
|
| 421 |
+
"acc": 0.3216374269005848,
|
| 422 |
+
"acc_stderr": 0.03582529442573122,
|
| 423 |
+
"acc_norm": 0.3216374269005848,
|
| 424 |
+
"acc_norm_stderr": 0.03582529442573122
|
| 425 |
+
},
|
| 426 |
+
"harness|truthfulqa:mc|0": {
|
| 427 |
+
"mc1": 0.24112607099143207,
|
| 428 |
+
"mc1_stderr": 0.014974827279752334,
|
| 429 |
+
"mc2": 0.3762297840067963,
|
| 430 |
+
"mc2_stderr": 0.01445991036363257
|
| 431 |
+
},
|
| 432 |
+
"harness|winogrande|5": {
|
| 433 |
+
"acc": 0.5153906866614049,
|
| 434 |
+
"acc_stderr": 0.014045826789783668
|
| 435 |
+
},
|
| 436 |
+
"harness|gsm8k|5": {
|
| 437 |
+
"acc": 0.006823351023502654,
|
| 438 |
+
"acc_stderr": 0.0022675371022544823
|
| 439 |
+
}
|
| 440 |
+
}
|
| 441 |
+
|
| 442 |
+
|
| 443 |
```
|