Isaak Carter Augustus
commited on
Commit
•
20cb9f0
1
Parent(s):
3754562
Update README.md
Browse files
README.md
CHANGED
|
@@ -17,6 +17,395 @@ JOSIE_Beta-3-7B-slerp is a merge of the following models using [LazyMergekit](ht
|
|
| 17 |
* [Locutusque/Hercules-3.1-Mistral-7B](https://huggingface.co/Locutusque/Hercules-3.1-Mistral-7B)
|
| 18 |
* [cognitivecomputations/dolphin-2.8-experiment26-7b](https://huggingface.co/cognitivecomputations/dolphin-2.8-experiment26-7b)
|
| 19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
## 🧩 Configuration
|
| 21 |
|
| 22 |
```yaml
|
|
|
|
| 17 |
* [Locutusque/Hercules-3.1-Mistral-7B](https://huggingface.co/Locutusque/Hercules-3.1-Mistral-7B)
|
| 18 |
* [cognitivecomputations/dolphin-2.8-experiment26-7b](https://huggingface.co/cognitivecomputations/dolphin-2.8-experiment26-7b)
|
| 19 |
|
| 20 |
+
# IMPORTANT!!!
|
| 21 |
+
|
| 22 |
+
upon sseing the eval bechmarks on the LLM Leaderboard, this is the best performing model, but it's not uncensored, and it's answers are not really good when chatting with it.
|
| 23 |
+
I will further train it one datasets like dolphin and other.
|
| 24 |
+
|
| 25 |
+
```json
|
| 26 |
+
{
|
| 27 |
+
"all": {
|
| 28 |
+
"acc": 0.6432209013684985,
|
| 29 |
+
"acc_stderr": 0.03221665824377992,
|
| 30 |
+
"acc_norm": 0.6450099678239628,
|
| 31 |
+
"acc_norm_stderr": 0.032867717920871294,
|
| 32 |
+
"mc1": 0.3353733170134639,
|
| 33 |
+
"mc1_stderr": 0.01652753403966899,
|
| 34 |
+
"mc2": 0.48804542326643174,
|
| 35 |
+
"mc2_stderr": 0.015087630632446147
|
| 36 |
+
},
|
| 37 |
+
"harness|arc:challenge|25": {
|
| 38 |
+
"acc": 0.6083617747440273,
|
| 39 |
+
"acc_stderr": 0.014264122124938217,
|
| 40 |
+
"acc_norm": 0.6339590443686007,
|
| 41 |
+
"acc_norm_stderr": 0.014077223108470139
|
| 42 |
+
},
|
| 43 |
+
"harness|hellaswag|10": {
|
| 44 |
+
"acc": 0.6618203545110536,
|
| 45 |
+
"acc_stderr": 0.0047212316370927225,
|
| 46 |
+
"acc_norm": 0.8456482772356104,
|
| 47 |
+
"acc_norm_stderr": 0.0036054721167622867
|
| 48 |
+
},
|
| 49 |
+
"harness|hendrycksTest-abstract_algebra|5": {
|
| 50 |
+
"acc": 0.3,
|
| 51 |
+
"acc_stderr": 0.046056618647183814,
|
| 52 |
+
"acc_norm": 0.3,
|
| 53 |
+
"acc_norm_stderr": 0.046056618647183814
|
| 54 |
+
},
|
| 55 |
+
"harness|hendrycksTest-anatomy|5": {
|
| 56 |
+
"acc": 0.6074074074074074,
|
| 57 |
+
"acc_stderr": 0.04218506215368879,
|
| 58 |
+
"acc_norm": 0.6074074074074074,
|
| 59 |
+
"acc_norm_stderr": 0.04218506215368879
|
| 60 |
+
},
|
| 61 |
+
"harness|hendrycksTest-astronomy|5": {
|
| 62 |
+
"acc": 0.6710526315789473,
|
| 63 |
+
"acc_stderr": 0.03823428969926605,
|
| 64 |
+
"acc_norm": 0.6710526315789473,
|
| 65 |
+
"acc_norm_stderr": 0.03823428969926605
|
| 66 |
+
},
|
| 67 |
+
"harness|hendrycksTest-business_ethics|5": {
|
| 68 |
+
"acc": 0.61,
|
| 69 |
+
"acc_stderr": 0.04902071300001975,
|
| 70 |
+
"acc_norm": 0.61,
|
| 71 |
+
"acc_norm_stderr": 0.04902071300001975
|
| 72 |
+
},
|
| 73 |
+
"harness|hendrycksTest-clinical_knowledge|5": {
|
| 74 |
+
"acc": 0.7018867924528301,
|
| 75 |
+
"acc_stderr": 0.02815283794249387,
|
| 76 |
+
"acc_norm": 0.7018867924528301,
|
| 77 |
+
"acc_norm_stderr": 0.02815283794249387
|
| 78 |
+
},
|
| 79 |
+
"harness|hendrycksTest-college_biology|5": {
|
| 80 |
+
"acc": 0.7638888888888888,
|
| 81 |
+
"acc_stderr": 0.03551446610810826,
|
| 82 |
+
"acc_norm": 0.7638888888888888,
|
| 83 |
+
"acc_norm_stderr": 0.03551446610810826
|
| 84 |
+
},
|
| 85 |
+
"harness|hendrycksTest-college_chemistry|5": {
|
| 86 |
+
"acc": 0.44,
|
| 87 |
+
"acc_stderr": 0.04988876515698589,
|
| 88 |
+
"acc_norm": 0.44,
|
| 89 |
+
"acc_norm_stderr": 0.04988876515698589
|
| 90 |
+
},
|
| 91 |
+
"harness|hendrycksTest-college_computer_science|5": {
|
| 92 |
+
"acc": 0.52,
|
| 93 |
+
"acc_stderr": 0.050211673156867795,
|
| 94 |
+
"acc_norm": 0.52,
|
| 95 |
+
"acc_norm_stderr": 0.050211673156867795
|
| 96 |
+
},
|
| 97 |
+
"harness|hendrycksTest-college_mathematics|5": {
|
| 98 |
+
"acc": 0.37,
|
| 99 |
+
"acc_stderr": 0.04852365870939099,
|
| 100 |
+
"acc_norm": 0.37,
|
| 101 |
+
"acc_norm_stderr": 0.04852365870939099
|
| 102 |
+
},
|
| 103 |
+
"harness|hendrycksTest-college_medicine|5": {
|
| 104 |
+
"acc": 0.6473988439306358,
|
| 105 |
+
"acc_stderr": 0.03643037168958548,
|
| 106 |
+
"acc_norm": 0.6473988439306358,
|
| 107 |
+
"acc_norm_stderr": 0.03643037168958548
|
| 108 |
+
},
|
| 109 |
+
"harness|hendrycksTest-college_physics|5": {
|
| 110 |
+
"acc": 0.38235294117647056,
|
| 111 |
+
"acc_stderr": 0.04835503696107223,
|
| 112 |
+
"acc_norm": 0.38235294117647056,
|
| 113 |
+
"acc_norm_stderr": 0.04835503696107223
|
| 114 |
+
},
|
| 115 |
+
"harness|hendrycksTest-computer_security|5": {
|
| 116 |
+
"acc": 0.75,
|
| 117 |
+
"acc_stderr": 0.04351941398892446,
|
| 118 |
+
"acc_norm": 0.75,
|
| 119 |
+
"acc_norm_stderr": 0.04351941398892446
|
| 120 |
+
},
|
| 121 |
+
"harness|hendrycksTest-conceptual_physics|5": {
|
| 122 |
+
"acc": 0.574468085106383,
|
| 123 |
+
"acc_stderr": 0.03232146916224468,
|
| 124 |
+
"acc_norm": 0.574468085106383,
|
| 125 |
+
"acc_norm_stderr": 0.03232146916224468
|
| 126 |
+
},
|
| 127 |
+
"harness|hendrycksTest-econometrics|5": {
|
| 128 |
+
"acc": 0.4649122807017544,
|
| 129 |
+
"acc_stderr": 0.046920083813689104,
|
| 130 |
+
"acc_norm": 0.4649122807017544,
|
| 131 |
+
"acc_norm_stderr": 0.046920083813689104
|
| 132 |
+
},
|
| 133 |
+
"harness|hendrycksTest-electrical_engineering|5": {
|
| 134 |
+
"acc": 0.5310344827586206,
|
| 135 |
+
"acc_stderr": 0.04158632762097828,
|
| 136 |
+
"acc_norm": 0.5310344827586206,
|
| 137 |
+
"acc_norm_stderr": 0.04158632762097828
|
| 138 |
+
},
|
| 139 |
+
"harness|hendrycksTest-elementary_mathematics|5": {
|
| 140 |
+
"acc": 0.41534391534391535,
|
| 141 |
+
"acc_stderr": 0.025379524910778405,
|
| 142 |
+
"acc_norm": 0.41534391534391535,
|
| 143 |
+
"acc_norm_stderr": 0.025379524910778405
|
| 144 |
+
},
|
| 145 |
+
"harness|hendrycksTest-formal_logic|5": {
|
| 146 |
+
"acc": 0.46825396825396826,
|
| 147 |
+
"acc_stderr": 0.04463112720677171,
|
| 148 |
+
"acc_norm": 0.46825396825396826,
|
| 149 |
+
"acc_norm_stderr": 0.04463112720677171
|
| 150 |
+
},
|
| 151 |
+
"harness|hendrycksTest-global_facts|5": {
|
| 152 |
+
"acc": 0.35,
|
| 153 |
+
"acc_stderr": 0.0479372485441102,
|
| 154 |
+
"acc_norm": 0.35,
|
| 155 |
+
"acc_norm_stderr": 0.0479372485441102
|
| 156 |
+
},
|
| 157 |
+
"harness|hendrycksTest-high_school_biology|5": {
|
| 158 |
+
"acc": 0.7709677419354839,
|
| 159 |
+
"acc_stderr": 0.023904914311782648,
|
| 160 |
+
"acc_norm": 0.7709677419354839,
|
| 161 |
+
"acc_norm_stderr": 0.023904914311782648
|
| 162 |
+
},
|
| 163 |
+
"harness|hendrycksTest-high_school_chemistry|5": {
|
| 164 |
+
"acc": 0.5073891625615764,
|
| 165 |
+
"acc_stderr": 0.035176035403610105,
|
| 166 |
+
"acc_norm": 0.5073891625615764,
|
| 167 |
+
"acc_norm_stderr": 0.035176035403610105
|
| 168 |
+
},
|
| 169 |
+
"harness|hendrycksTest-high_school_computer_science|5": {
|
| 170 |
+
"acc": 0.69,
|
| 171 |
+
"acc_stderr": 0.04648231987117316,
|
| 172 |
+
"acc_norm": 0.69,
|
| 173 |
+
"acc_norm_stderr": 0.04648231987117316
|
| 174 |
+
},
|
| 175 |
+
"harness|hendrycksTest-high_school_european_history|5": {
|
| 176 |
+
"acc": 0.7757575757575758,
|
| 177 |
+
"acc_stderr": 0.032568666616811015,
|
| 178 |
+
"acc_norm": 0.7757575757575758,
|
| 179 |
+
"acc_norm_stderr": 0.032568666616811015
|
| 180 |
+
},
|
| 181 |
+
"harness|hendrycksTest-high_school_geography|5": {
|
| 182 |
+
"acc": 0.797979797979798,
|
| 183 |
+
"acc_stderr": 0.028606204289229872,
|
| 184 |
+
"acc_norm": 0.797979797979798,
|
| 185 |
+
"acc_norm_stderr": 0.028606204289229872
|
| 186 |
+
},
|
| 187 |
+
"harness|hendrycksTest-high_school_government_and_politics|5": {
|
| 188 |
+
"acc": 0.8963730569948186,
|
| 189 |
+
"acc_stderr": 0.02199531196364424,
|
| 190 |
+
"acc_norm": 0.8963730569948186,
|
| 191 |
+
"acc_norm_stderr": 0.02199531196364424
|
| 192 |
+
},
|
| 193 |
+
"harness|hendrycksTest-high_school_macroeconomics|5": {
|
| 194 |
+
"acc": 0.6641025641025641,
|
| 195 |
+
"acc_stderr": 0.023946724741563973,
|
| 196 |
+
"acc_norm": 0.6641025641025641,
|
| 197 |
+
"acc_norm_stderr": 0.023946724741563973
|
| 198 |
+
},
|
| 199 |
+
"harness|hendrycksTest-high_school_mathematics|5": {
|
| 200 |
+
"acc": 0.3888888888888889,
|
| 201 |
+
"acc_stderr": 0.029723278961476664,
|
| 202 |
+
"acc_norm": 0.3888888888888889,
|
| 203 |
+
"acc_norm_stderr": 0.029723278961476664
|
| 204 |
+
},
|
| 205 |
+
"harness|hendrycksTest-high_school_microeconomics|5": {
|
| 206 |
+
"acc": 0.680672268907563,
|
| 207 |
+
"acc_stderr": 0.030283995525884396,
|
| 208 |
+
"acc_norm": 0.680672268907563,
|
| 209 |
+
"acc_norm_stderr": 0.030283995525884396
|
| 210 |
+
},
|
| 211 |
+
"harness|hendrycksTest-high_school_physics|5": {
|
| 212 |
+
"acc": 0.3443708609271523,
|
| 213 |
+
"acc_stderr": 0.038796870240733264,
|
| 214 |
+
"acc_norm": 0.3443708609271523,
|
| 215 |
+
"acc_norm_stderr": 0.038796870240733264
|
| 216 |
+
},
|
| 217 |
+
"harness|hendrycksTest-high_school_psychology|5": {
|
| 218 |
+
"acc": 0.8422018348623853,
|
| 219 |
+
"acc_stderr": 0.01563002297009245,
|
| 220 |
+
"acc_norm": 0.8422018348623853,
|
| 221 |
+
"acc_norm_stderr": 0.01563002297009245
|
| 222 |
+
},
|
| 223 |
+
"harness|hendrycksTest-high_school_statistics|5": {
|
| 224 |
+
"acc": 0.5,
|
| 225 |
+
"acc_stderr": 0.034099716973523674,
|
| 226 |
+
"acc_norm": 0.5,
|
| 227 |
+
"acc_norm_stderr": 0.034099716973523674
|
| 228 |
+
},
|
| 229 |
+
"harness|hendrycksTest-high_school_us_history|5": {
|
| 230 |
+
"acc": 0.7990196078431373,
|
| 231 |
+
"acc_stderr": 0.028125972265654366,
|
| 232 |
+
"acc_norm": 0.7990196078431373,
|
| 233 |
+
"acc_norm_stderr": 0.028125972265654366
|
| 234 |
+
},
|
| 235 |
+
"harness|hendrycksTest-high_school_world_history|5": {
|
| 236 |
+
"acc": 0.7890295358649789,
|
| 237 |
+
"acc_stderr": 0.02655837250266192,
|
| 238 |
+
"acc_norm": 0.7890295358649789,
|
| 239 |
+
"acc_norm_stderr": 0.02655837250266192
|
| 240 |
+
},
|
| 241 |
+
"harness|hendrycksTest-human_aging|5": {
|
| 242 |
+
"acc": 0.695067264573991,
|
| 243 |
+
"acc_stderr": 0.030898610882477515,
|
| 244 |
+
"acc_norm": 0.695067264573991,
|
| 245 |
+
"acc_norm_stderr": 0.030898610882477515
|
| 246 |
+
},
|
| 247 |
+
"harness|hendrycksTest-human_sexuality|5": {
|
| 248 |
+
"acc": 0.7862595419847328,
|
| 249 |
+
"acc_stderr": 0.0359546161177469,
|
| 250 |
+
"acc_norm": 0.7862595419847328,
|
| 251 |
+
"acc_norm_stderr": 0.0359546161177469
|
| 252 |
+
},
|
| 253 |
+
"harness|hendrycksTest-international_law|5": {
|
| 254 |
+
"acc": 0.8016528925619835,
|
| 255 |
+
"acc_stderr": 0.036401182719909476,
|
| 256 |
+
"acc_norm": 0.8016528925619835,
|
| 257 |
+
"acc_norm_stderr": 0.036401182719909476
|
| 258 |
+
},
|
| 259 |
+
"harness|hendrycksTest-jurisprudence|5": {
|
| 260 |
+
"acc": 0.7870370370370371,
|
| 261 |
+
"acc_stderr": 0.0395783547198098,
|
| 262 |
+
"acc_norm": 0.7870370370370371,
|
| 263 |
+
"acc_norm_stderr": 0.0395783547198098
|
| 264 |
+
},
|
| 265 |
+
"harness|hendrycksTest-logical_fallacies|5": {
|
| 266 |
+
"acc": 0.754601226993865,
|
| 267 |
+
"acc_stderr": 0.03380939813943354,
|
| 268 |
+
"acc_norm": 0.754601226993865,
|
| 269 |
+
"acc_norm_stderr": 0.03380939813943354
|
| 270 |
+
},
|
| 271 |
+
"harness|hendrycksTest-machine_learning|5": {
|
| 272 |
+
"acc": 0.5535714285714286,
|
| 273 |
+
"acc_stderr": 0.04718471485219587,
|
| 274 |
+
"acc_norm": 0.5535714285714286,
|
| 275 |
+
"acc_norm_stderr": 0.04718471485219587
|
| 276 |
+
},
|
| 277 |
+
"harness|hendrycksTest-management|5": {
|
| 278 |
+
"acc": 0.7766990291262136,
|
| 279 |
+
"acc_stderr": 0.04123553189891431,
|
| 280 |
+
"acc_norm": 0.7766990291262136,
|
| 281 |
+
"acc_norm_stderr": 0.04123553189891431
|
| 282 |
+
},
|
| 283 |
+
"harness|hendrycksTest-marketing|5": {
|
| 284 |
+
"acc": 0.8760683760683761,
|
| 285 |
+
"acc_stderr": 0.021586494001281376,
|
| 286 |
+
"acc_norm": 0.8760683760683761,
|
| 287 |
+
"acc_norm_stderr": 0.021586494001281376
|
| 288 |
+
},
|
| 289 |
+
"harness|hendrycksTest-medical_genetics|5": {
|
| 290 |
+
"acc": 0.74,
|
| 291 |
+
"acc_stderr": 0.04408440022768079,
|
| 292 |
+
"acc_norm": 0.74,
|
| 293 |
+
"acc_norm_stderr": 0.04408440022768079
|
| 294 |
+
},
|
| 295 |
+
"harness|hendrycksTest-miscellaneous|5": {
|
| 296 |
+
"acc": 0.8186462324393359,
|
| 297 |
+
"acc_stderr": 0.01377869377846408,
|
| 298 |
+
"acc_norm": 0.8186462324393359,
|
| 299 |
+
"acc_norm_stderr": 0.01377869377846408
|
| 300 |
+
},
|
| 301 |
+
"harness|hendrycksTest-moral_disputes|5": {
|
| 302 |
+
"acc": 0.7225433526011561,
|
| 303 |
+
"acc_stderr": 0.024105712607754307,
|
| 304 |
+
"acc_norm": 0.7225433526011561,
|
| 305 |
+
"acc_norm_stderr": 0.024105712607754307
|
| 306 |
+
},
|
| 307 |
+
"harness|hendrycksTest-moral_scenarios|5": {
|
| 308 |
+
"acc": 0.288268156424581,
|
| 309 |
+
"acc_stderr": 0.015149132860209432,
|
| 310 |
+
"acc_norm": 0.288268156424581,
|
| 311 |
+
"acc_norm_stderr": 0.015149132860209432
|
| 312 |
+
},
|
| 313 |
+
"harness|hendrycksTest-nutrition|5": {
|
| 314 |
+
"acc": 0.7189542483660131,
|
| 315 |
+
"acc_stderr": 0.025738854797818733,
|
| 316 |
+
"acc_norm": 0.7189542483660131,
|
| 317 |
+
"acc_norm_stderr": 0.025738854797818733
|
| 318 |
+
},
|
| 319 |
+
"harness|hendrycksTest-philosophy|5": {
|
| 320 |
+
"acc": 0.7170418006430869,
|
| 321 |
+
"acc_stderr": 0.025583062489984813,
|
| 322 |
+
"acc_norm": 0.7170418006430869,
|
| 323 |
+
"acc_norm_stderr": 0.025583062489984813
|
| 324 |
+
},
|
| 325 |
+
"harness|hendrycksTest-prehistory|5": {
|
| 326 |
+
"acc": 0.7407407407407407,
|
| 327 |
+
"acc_stderr": 0.024383665531035457,
|
| 328 |
+
"acc_norm": 0.7407407407407407,
|
| 329 |
+
"acc_norm_stderr": 0.024383665531035457
|
| 330 |
+
},
|
| 331 |
+
"harness|hendrycksTest-professional_accounting|5": {
|
| 332 |
+
"acc": 0.5035460992907801,
|
| 333 |
+
"acc_stderr": 0.02982674915328092,
|
| 334 |
+
"acc_norm": 0.5035460992907801,
|
| 335 |
+
"acc_norm_stderr": 0.02982674915328092
|
| 336 |
+
},
|
| 337 |
+
"harness|hendrycksTest-professional_law|5": {
|
| 338 |
+
"acc": 0.4680573663624511,
|
| 339 |
+
"acc_stderr": 0.012744149704869647,
|
| 340 |
+
"acc_norm": 0.4680573663624511,
|
| 341 |
+
"acc_norm_stderr": 0.012744149704869647
|
| 342 |
+
},
|
| 343 |
+
"harness|hendrycksTest-professional_medicine|5": {
|
| 344 |
+
"acc": 0.6838235294117647,
|
| 345 |
+
"acc_stderr": 0.028245687391462927,
|
| 346 |
+
"acc_norm": 0.6838235294117647,
|
| 347 |
+
"acc_norm_stderr": 0.028245687391462927
|
| 348 |
+
},
|
| 349 |
+
"harness|hendrycksTest-professional_psychology|5": {
|
| 350 |
+
"acc": 0.6633986928104575,
|
| 351 |
+
"acc_stderr": 0.019117213911495158,
|
| 352 |
+
"acc_norm": 0.6633986928104575,
|
| 353 |
+
"acc_norm_stderr": 0.019117213911495158
|
| 354 |
+
},
|
| 355 |
+
"harness|hendrycksTest-public_relations|5": {
|
| 356 |
+
"acc": 0.6636363636363637,
|
| 357 |
+
"acc_stderr": 0.04525393596302506,
|
| 358 |
+
"acc_norm": 0.6636363636363637,
|
| 359 |
+
"acc_norm_stderr": 0.04525393596302506
|
| 360 |
+
},
|
| 361 |
+
"harness|hendrycksTest-security_studies|5": {
|
| 362 |
+
"acc": 0.7428571428571429,
|
| 363 |
+
"acc_stderr": 0.027979823538744546,
|
| 364 |
+
"acc_norm": 0.7428571428571429,
|
| 365 |
+
"acc_norm_stderr": 0.027979823538744546
|
| 366 |
+
},
|
| 367 |
+
"harness|hendrycksTest-sociology|5": {
|
| 368 |
+
"acc": 0.845771144278607,
|
| 369 |
+
"acc_stderr": 0.025538433368578337,
|
| 370 |
+
"acc_norm": 0.845771144278607,
|
| 371 |
+
"acc_norm_stderr": 0.025538433368578337
|
| 372 |
+
},
|
| 373 |
+
"harness|hendrycksTest-us_foreign_policy|5": {
|
| 374 |
+
"acc": 0.87,
|
| 375 |
+
"acc_stderr": 0.033799766898963086,
|
| 376 |
+
"acc_norm": 0.87,
|
| 377 |
+
"acc_norm_stderr": 0.033799766898963086
|
| 378 |
+
},
|
| 379 |
+
"harness|hendrycksTest-virology|5": {
|
| 380 |
+
"acc": 0.5301204819277109,
|
| 381 |
+
"acc_stderr": 0.03885425420866767,
|
| 382 |
+
"acc_norm": 0.5301204819277109,
|
| 383 |
+
"acc_norm_stderr": 0.03885425420866767
|
| 384 |
+
},
|
| 385 |
+
"harness|hendrycksTest-world_religions|5": {
|
| 386 |
+
"acc": 0.8128654970760234,
|
| 387 |
+
"acc_stderr": 0.02991312723236804,
|
| 388 |
+
"acc_norm": 0.8128654970760234,
|
| 389 |
+
"acc_norm_stderr": 0.02991312723236804
|
| 390 |
+
},
|
| 391 |
+
"harness|truthfulqa:mc|0": {
|
| 392 |
+
"mc1": 0.3353733170134639,
|
| 393 |
+
"mc1_stderr": 0.01652753403966899,
|
| 394 |
+
"mc2": 0.48804542326643174,
|
| 395 |
+
"mc2_stderr": 0.015087630632446147
|
| 396 |
+
},
|
| 397 |
+
"harness|winogrande|5": {
|
| 398 |
+
"acc": 0.8042620363062352,
|
| 399 |
+
"acc_stderr": 0.011151145042218319
|
| 400 |
+
},
|
| 401 |
+
"harness|gsm8k|5": {
|
| 402 |
+
"acc": 0.5860500379075056,
|
| 403 |
+
"acc_stderr": 0.013566991960151778
|
| 404 |
+
}
|
| 405 |
+
}
|
| 406 |
+
|
| 407 |
+
```
|
| 408 |
+
|
| 409 |
## 🧩 Configuration
|
| 410 |
|
| 411 |
```yaml
|