Update README.md
Browse files
README.md
CHANGED
|
@@ -65,4 +65,395 @@ generated_ids = [
|
|
| 65 |
|
| 66 |
response = tokenizer.batch_decode(generated_ids)[0]
|
| 67 |
print(response)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 68 |
```
|
|
|
|
| 65 |
|
| 66 |
response = tokenizer.batch_decode(generated_ids)[0]
|
| 67 |
print(response)
|
| 68 |
+
```
|
| 69 |
+
|
| 70 |
+
## Benchmarks
|
| 71 |
+
|
| 72 |
+
Nous Benchmark:
|
| 73 |
+
```
|
| 74 |
+
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|
| 75 |
+
|---------------------------------------------------|------:|------:|---------:|-------:|------:|
|
| 76 |
+
|[Master-Yi-9B](https://huggingface.co/qnguyen3/Master-Yi-9B)| 43.55| 71.48| 48.54| 41.43| 51.25|
|
| 77 |
+
|
| 78 |
+
### AGIEval
|
| 79 |
+
| Task |Version| Metric |Value| |Stderr|
|
| 80 |
+
|------------------------------|------:|--------|----:|---|-----:|
|
| 81 |
+
|agieval_aqua_rat | 0|acc |35.83|± | 3.01|
|
| 82 |
+
| | |acc_norm|31.89|± | 2.93|
|
| 83 |
+
|agieval_logiqa_en | 0|acc |38.25|± | 1.91|
|
| 84 |
+
| | |acc_norm|37.79|± | 1.90|
|
| 85 |
+
|agieval_lsat_ar | 0|acc |23.04|± | 2.78|
|
| 86 |
+
| | |acc_norm|20.43|± | 2.66|
|
| 87 |
+
|agieval_lsat_lr | 0|acc |48.04|± | 2.21|
|
| 88 |
+
| | |acc_norm|42.75|± | 2.19|
|
| 89 |
+
|agieval_lsat_rc | 0|acc |61.34|± | 2.97|
|
| 90 |
+
| | |acc_norm|52.79|± | 3.05|
|
| 91 |
+
|agieval_sat_en | 0|acc |79.13|± | 2.84|
|
| 92 |
+
| | |acc_norm|72.33|± | 3.12|
|
| 93 |
+
|agieval_sat_en_without_passage| 0|acc |44.17|± | 3.47|
|
| 94 |
+
| | |acc_norm|42.72|± | 3.45|
|
| 95 |
+
|agieval_sat_math | 0|acc |52.27|± | 3.38|
|
| 96 |
+
| | |acc_norm|47.73|± | 3.38|
|
| 97 |
+
|
| 98 |
+
Average: 43.55%
|
| 99 |
+
|
| 100 |
+
### GPT4All
|
| 101 |
+
| Task |Version| Metric |Value| |Stderr|
|
| 102 |
+
|-------------|------:|--------|----:|---|-----:|
|
| 103 |
+
|arc_challenge| 0|acc |54.95|± | 1.45|
|
| 104 |
+
| | |acc_norm|58.70|± | 1.44|
|
| 105 |
+
|arc_easy | 0|acc |82.28|± | 0.78|
|
| 106 |
+
| | |acc_norm|81.10|± | 0.80|
|
| 107 |
+
|boolq | 1|acc |86.15|± | 0.60|
|
| 108 |
+
|hellaswag | 0|acc |59.16|± | 0.49|
|
| 109 |
+
| | |acc_norm|77.53|± | 0.42|
|
| 110 |
+
|openbookqa | 0|acc |37.40|± | 2.17|
|
| 111 |
+
| | |acc_norm|44.00|± | 2.22|
|
| 112 |
+
|piqa | 0|acc |79.00|± | 0.95|
|
| 113 |
+
| | |acc_norm|80.25|± | 0.93|
|
| 114 |
+
|winogrande | 0|acc |72.61|± | 1.25|
|
| 115 |
+
|
| 116 |
+
Average: 71.48%
|
| 117 |
+
|
| 118 |
+
### TruthfulQA
|
| 119 |
+
| Task |Version|Metric|Value| |Stderr|
|
| 120 |
+
|-------------|------:|------|----:|---|-----:|
|
| 121 |
+
|truthfulqa_mc| 1|mc1 |33.05|± | 1.65|
|
| 122 |
+
| | |mc2 |48.54|± | 1.54|
|
| 123 |
+
|
| 124 |
+
Average: 48.54%
|
| 125 |
+
|
| 126 |
+
### Bigbench
|
| 127 |
+
| Task |Version| Metric |Value| |Stderr|
|
| 128 |
+
|------------------------------------------------|------:|---------------------|----:|---|-----:|
|
| 129 |
+
|bigbench_causal_judgement | 0|multiple_choice_grade|54.74|± | 3.62|
|
| 130 |
+
|bigbench_date_understanding | 0|multiple_choice_grade|68.02|± | 2.43|
|
| 131 |
+
|bigbench_disambiguation_qa | 0|multiple_choice_grade|40.31|± | 3.06|
|
| 132 |
+
|bigbench_geometric_shapes | 0|multiple_choice_grade|30.36|± | 2.43|
|
| 133 |
+
| | |exact_str_match | 2.23|± | 0.78|
|
| 134 |
+
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|26.00|± | 1.96|
|
| 135 |
+
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|20.71|± | 1.53|
|
| 136 |
+
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|44.00|± | 2.87|
|
| 137 |
+
|bigbench_movie_recommendation | 0|multiple_choice_grade|35.00|± | 2.14|
|
| 138 |
+
|bigbench_navigate | 0|multiple_choice_grade|58.40|± | 1.56|
|
| 139 |
+
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|61.80|± | 1.09|
|
| 140 |
+
|bigbench_ruin_names | 0|multiple_choice_grade|42.41|± | 2.34|
|
| 141 |
+
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|31.56|± | 1.47|
|
| 142 |
+
|bigbench_snarks | 0|multiple_choice_grade|55.25|± | 3.71|
|
| 143 |
+
|bigbench_sports_understanding | 0|multiple_choice_grade|69.37|± | 1.47|
|
| 144 |
+
|bigbench_temporal_sequences | 0|multiple_choice_grade|27.70|± | 1.42|
|
| 145 |
+
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|21.36|± | 1.16|
|
| 146 |
+
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|14.69|± | 0.85|
|
| 147 |
+
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|44.00|± | 2.87|
|
| 148 |
+
|
| 149 |
+
Average: 41.43%
|
| 150 |
+
|
| 151 |
+
Average score: 51.25%
|
| 152 |
+
```
|
| 153 |
+
|
| 154 |
+
OpenLLM Benchmark:
|
| 155 |
+
```
|
| 156 |
+
| Model |ARC |HellaSwag|MMLU |TruthfulQA|Winogrande|GSM8K|Average|
|
| 157 |
+
|---------------------------------------------------|---:|--------:|----:|---------:|---------:|----:|------:|
|
| 158 |
+
|[Master-Yi-9B](https://huggingface.co/qnguyen3/Master-Yi-9B)|61.6| 79.89|69.95| 48.59| 77.35|67.48| 67.48|
|
| 159 |
+
|
| 160 |
+
### ARC
|
| 161 |
+
| Task |Version| Metric | Value | |Stderr|
|
| 162 |
+
|-------------|------:|--------------------|-------------|---|------|
|
| 163 |
+
|arc_challenge| 1|acc,none | 0.59| | |
|
| 164 |
+
| | |acc_stderr,none | 0.01| | |
|
| 165 |
+
| | |acc_norm,none | 0.62| | |
|
| 166 |
+
| | |acc_norm_stderr,none| 0.01| | |
|
| 167 |
+
| | |alias |arc_challenge| | |
|
| 168 |
+
|
| 169 |
+
Average: 61.6%
|
| 170 |
+
|
| 171 |
+
### HellaSwag
|
| 172 |
+
| Task |Version| Metric | Value | |Stderr|
|
| 173 |
+
|---------|------:|--------------------|---------|---|------|
|
| 174 |
+
|hellaswag| 1|acc,none | 0.61| | |
|
| 175 |
+
| | |acc_stderr,none | 0| | |
|
| 176 |
+
| | |acc_norm,none | 0.80| | |
|
| 177 |
+
| | |acc_norm_stderr,none| 0| | |
|
| 178 |
+
| | |alias |hellaswag| | |
|
| 179 |
+
|
| 180 |
+
Average: 79.89%
|
| 181 |
+
|
| 182 |
+
### MMLU
|
| 183 |
+
| Task |Version| Metric | Value | |Stderr|
|
| 184 |
+
|----------------------------------------|-------|---------------|---------------------------------------|---|------|
|
| 185 |
+
|mmlu |N/A |acc,none | 0.7| | |
|
| 186 |
+
| | |acc_stderr,none| 0| | |
|
| 187 |
+
| | |alias |mmlu | | |
|
| 188 |
+
|mmlu_abstract_algebra | 0|alias | - abstract_algebra | | |
|
| 189 |
+
| | |acc,none |0.46 | | |
|
| 190 |
+
| | |acc_stderr,none|0.05 | | |
|
| 191 |
+
|mmlu_anatomy | 0|alias | - anatomy | | |
|
| 192 |
+
| | |acc,none |0.64 | | |
|
| 193 |
+
| | |acc_stderr,none|0.04 | | |
|
| 194 |
+
|mmlu_astronomy | 0|alias | - astronomy | | |
|
| 195 |
+
| | |acc,none |0.77 | | |
|
| 196 |
+
| | |acc_stderr,none|0.03 | | |
|
| 197 |
+
|mmlu_business_ethics | 0|alias | - business_ethics | | |
|
| 198 |
+
| | |acc,none |0.76 | | |
|
| 199 |
+
| | |acc_stderr,none|0.04 | | |
|
| 200 |
+
|mmlu_clinical_knowledge | 0|alias | - clinical_knowledge | | |
|
| 201 |
+
| | |acc,none |0.71 | | |
|
| 202 |
+
| | |acc_stderr,none|0.03 | | |
|
| 203 |
+
|mmlu_college_biology | 0|alias | - college_biology | | |
|
| 204 |
+
| | |acc,none |0.82 | | |
|
| 205 |
+
| | |acc_stderr,none|0.03 | | |
|
| 206 |
+
|mmlu_college_chemistry | 0|alias | - college_chemistry | | |
|
| 207 |
+
| | |acc,none |0.52 | | |
|
| 208 |
+
| | |acc_stderr,none|0.05 | | |
|
| 209 |
+
|mmlu_college_computer_science | 0|alias | - college_computer_science | | |
|
| 210 |
+
| | |acc,none |0.56 | | |
|
| 211 |
+
| | |acc_stderr,none|0.05 | | |
|
| 212 |
+
|mmlu_college_mathematics | 0|alias | - college_mathematics | | |
|
| 213 |
+
| | |acc,none |0.44 | | |
|
| 214 |
+
| | |acc_stderr,none|0.05 | | |
|
| 215 |
+
|mmlu_college_medicine | 0|alias | - college_medicine | | |
|
| 216 |
+
| | |acc,none |0.72 | | |
|
| 217 |
+
| | |acc_stderr,none|0.03 | | |
|
| 218 |
+
|mmlu_college_physics | 0|alias | - college_physics | | |
|
| 219 |
+
| | |acc,none |0.45 | | |
|
| 220 |
+
| | |acc_stderr,none|0.05 | | |
|
| 221 |
+
|mmlu_computer_security | 0|alias | - computer_security | | |
|
| 222 |
+
| | |acc,none |0.81 | | |
|
| 223 |
+
| | |acc_stderr,none|0.04 | | |
|
| 224 |
+
|mmlu_conceptual_physics | 0|alias | - conceptual_physics | | |
|
| 225 |
+
| | |acc,none |0.74 | | |
|
| 226 |
+
| | |acc_stderr,none|0.03 | | |
|
| 227 |
+
|mmlu_econometrics | 0|alias | - econometrics | | |
|
| 228 |
+
| | |acc,none |0.65 | | |
|
| 229 |
+
| | |acc_stderr,none|0.04 | | |
|
| 230 |
+
|mmlu_electrical_engineering | 0|alias | - electrical_engineering | | |
|
| 231 |
+
| | |acc,none |0.72 | | |
|
| 232 |
+
| | |acc_stderr,none|0.04 | | |
|
| 233 |
+
|mmlu_elementary_mathematics | 0|alias | - elementary_mathematics | | |
|
| 234 |
+
| | |acc,none |0.62 | | |
|
| 235 |
+
| | |acc_stderr,none|0.02 | | |
|
| 236 |
+
|mmlu_formal_logic | 0|alias | - formal_logic | | |
|
| 237 |
+
| | |acc,none |0.57 | | |
|
| 238 |
+
| | |acc_stderr,none|0.04 | | |
|
| 239 |
+
|mmlu_global_facts | 0|alias | - global_facts | | |
|
| 240 |
+
| | |acc,none |0.46 | | |
|
| 241 |
+
| | |acc_stderr,none|0.05 | | |
|
| 242 |
+
|mmlu_high_school_biology | 0|alias | - high_school_biology | | |
|
| 243 |
+
| | |acc,none |0.86 | | |
|
| 244 |
+
| | |acc_stderr,none|0.02 | | |
|
| 245 |
+
|mmlu_high_school_chemistry | 0|alias | - high_school_chemistry | | |
|
| 246 |
+
| | |acc,none |0.67 | | |
|
| 247 |
+
| | |acc_stderr,none|0.03 | | |
|
| 248 |
+
|mmlu_high_school_computer_science | 0|alias | - high_school_computer_science | | |
|
| 249 |
+
| | |acc,none |0.84 | | |
|
| 250 |
+
| | |acc_stderr,none|0.04 | | |
|
| 251 |
+
|mmlu_high_school_european_history | 0|alias | - high_school_european_history | | |
|
| 252 |
+
| | |acc,none |0.82 | | |
|
| 253 |
+
| | |acc_stderr,none|0.03 | | |
|
| 254 |
+
|mmlu_high_school_geography | 0|alias | - high_school_geography | | |
|
| 255 |
+
| | |acc,none |0.86 | | |
|
| 256 |
+
| | |acc_stderr,none|0.02 | | |
|
| 257 |
+
|mmlu_high_school_government_and_politics| 0|alias | - high_school_government_and_politics| | |
|
| 258 |
+
| | |acc,none |0.90 | | |
|
| 259 |
+
| | |acc_stderr,none|0.02 | | |
|
| 260 |
+
|mmlu_high_school_macroeconomics | 0|alias | - high_school_macroeconomics | | |
|
| 261 |
+
| | |acc,none |0.75 | | |
|
| 262 |
+
| | |acc_stderr,none|0.02 | | |
|
| 263 |
+
|mmlu_high_school_mathematics | 0|alias | - high_school_mathematics | | |
|
| 264 |
+
| | |acc,none |0.43 | | |
|
| 265 |
+
| | |acc_stderr,none|0.03 | | |
|
| 266 |
+
|mmlu_high_school_microeconomics | 0|alias | - high_school_microeconomics | | |
|
| 267 |
+
| | |acc,none |0.86 | | |
|
| 268 |
+
| | |acc_stderr,none|0.02 | | |
|
| 269 |
+
|mmlu_high_school_physics | 0|alias | - high_school_physics | | |
|
| 270 |
+
| | |acc,none |0.45 | | |
|
| 271 |
+
| | |acc_stderr,none|0.04 | | |
|
| 272 |
+
|mmlu_high_school_psychology | 0|alias | - high_school_psychology | | |
|
| 273 |
+
| | |acc,none |0.87 | | |
|
| 274 |
+
| | |acc_stderr,none|0.01 | | |
|
| 275 |
+
|mmlu_high_school_statistics | 0|alias | - high_school_statistics | | |
|
| 276 |
+
| | |acc,none |0.68 | | |
|
| 277 |
+
| | |acc_stderr,none|0.03 | | |
|
| 278 |
+
|mmlu_high_school_us_history | 0|alias | - high_school_us_history | | |
|
| 279 |
+
| | |acc,none |0.85 | | |
|
| 280 |
+
| | |acc_stderr,none|0.02 | | |
|
| 281 |
+
|mmlu_high_school_world_history | 0|alias | - high_school_world_history | | |
|
| 282 |
+
| | |acc,none |0.85 | | |
|
| 283 |
+
| | |acc_stderr,none|0.02 | | |
|
| 284 |
+
|mmlu_human_aging | 0|alias | - human_aging | | |
|
| 285 |
+
| | |acc,none |0.76 | | |
|
| 286 |
+
| | |acc_stderr,none|0.03 | | |
|
| 287 |
+
|mmlu_human_sexuality | 0|alias | - human_sexuality | | |
|
| 288 |
+
| | |acc,none |0.78 | | |
|
| 289 |
+
| | |acc_stderr,none|0.04 | | |
|
| 290 |
+
|mmlu_humanities |N/A |alias | - humanities | | |
|
| 291 |
+
| | |acc,none |0.63 | | |
|
| 292 |
+
| | |acc_stderr,none|0.01 | | |
|
| 293 |
+
|mmlu_international_law | 0|alias | - international_law | | |
|
| 294 |
+
| | |acc,none |0.79 | | |
|
| 295 |
+
| | |acc_stderr,none|0.04 | | |
|
| 296 |
+
|mmlu_jurisprudence | 0|alias | - jurisprudence | | |
|
| 297 |
+
| | |acc,none |0.79 | | |
|
| 298 |
+
| | |acc_stderr,none|0.04 | | |
|
| 299 |
+
|mmlu_logical_fallacies | 0|alias | - logical_fallacies | | |
|
| 300 |
+
| | |acc,none |0.80 | | |
|
| 301 |
+
| | |acc_stderr,none|0.03 | | |
|
| 302 |
+
|mmlu_machine_learning | 0|alias | - machine_learning | | |
|
| 303 |
+
| | |acc,none |0.52 | | |
|
| 304 |
+
| | |acc_stderr,none|0.05 | | |
|
| 305 |
+
|mmlu_management | 0|alias | - management | | |
|
| 306 |
+
| | |acc,none |0.83 | | |
|
| 307 |
+
| | |acc_stderr,none|0.04 | | |
|
| 308 |
+
|mmlu_marketing | 0|alias | - marketing | | |
|
| 309 |
+
| | |acc,none |0.89 | | |
|
| 310 |
+
| | |acc_stderr,none|0.02 | | |
|
| 311 |
+
|mmlu_medical_genetics | 0|alias | - medical_genetics | | |
|
| 312 |
+
| | |acc,none |0.78 | | |
|
| 313 |
+
| | |acc_stderr,none|0.04 | | |
|
| 314 |
+
|mmlu_miscellaneous | 0|alias | - miscellaneous | | |
|
| 315 |
+
| | |acc,none |0.85 | | |
|
| 316 |
+
| | |acc_stderr,none|0.01 | | |
|
| 317 |
+
|mmlu_moral_disputes | 0|alias | - moral_disputes | | |
|
| 318 |
+
| | |acc,none |0.75 | | |
|
| 319 |
+
| | |acc_stderr,none|0.02 | | |
|
| 320 |
+
|mmlu_moral_scenarios | 0|alias | - moral_scenarios | | |
|
| 321 |
+
| | |acc,none |0.48 | | |
|
| 322 |
+
| | |acc_stderr,none|0.02 | | |
|
| 323 |
+
|mmlu_nutrition | 0|alias | - nutrition | | |
|
| 324 |
+
| | |acc,none |0.77 | | |
|
| 325 |
+
| | |acc_stderr,none|0.02 | | |
|
| 326 |
+
|mmlu_other |N/A |alias | - other | | |
|
| 327 |
+
| | |acc,none |0.75 | | |
|
| 328 |
+
| | |acc_stderr,none|0.01 | | |
|
| 329 |
+
|mmlu_philosophy | 0|alias | - philosophy | | |
|
| 330 |
+
| | |acc,none |0.78 | | |
|
| 331 |
+
| | |acc_stderr,none|0.02 | | |
|
| 332 |
+
|mmlu_prehistory | 0|alias | - prehistory | | |
|
| 333 |
+
| | |acc,none |0.77 | | |
|
| 334 |
+
| | |acc_stderr,none|0.02 | | |
|
| 335 |
+
|mmlu_professional_accounting | 0|alias | - professional_accounting | | |
|
| 336 |
+
| | |acc,none |0.57 | | |
|
| 337 |
+
| | |acc_stderr,none|0.03 | | |
|
| 338 |
+
|mmlu_professional_law | 0|alias | - professional_law | | |
|
| 339 |
+
| | |acc,none |0.50 | | |
|
| 340 |
+
| | |acc_stderr,none|0.01 | | |
|
| 341 |
+
|mmlu_professional_medicine | 0|alias | - professional_medicine | | |
|
| 342 |
+
| | |acc,none |0.71 | | |
|
| 343 |
+
| | |acc_stderr,none|0.03 | | |
|
| 344 |
+
|mmlu_professional_psychology | 0|alias | - professional_psychology | | |
|
| 345 |
+
| | |acc,none |0.73 | | |
|
| 346 |
+
| | |acc_stderr,none|0.02 | | |
|
| 347 |
+
|mmlu_public_relations | 0|alias | - public_relations | | |
|
| 348 |
+
| | |acc,none |0.76 | | |
|
| 349 |
+
| | |acc_stderr,none|0.04 | | |
|
| 350 |
+
|mmlu_security_studies | 0|alias | - security_studies | | |
|
| 351 |
+
| | |acc,none |0.78 | | |
|
| 352 |
+
| | |acc_stderr,none|0.03 | | |
|
| 353 |
+
|mmlu_social_sciences |N/A |alias | - social_sciences | | |
|
| 354 |
+
| | |acc,none |0.81 | | |
|
| 355 |
+
| | |acc_stderr,none|0.01 | | |
|
| 356 |
+
|mmlu_sociology | 0|alias | - sociology | | |
|
| 357 |
+
| | |acc,none |0.86 | | |
|
| 358 |
+
| | |acc_stderr,none|0.02 | | |
|
| 359 |
+
|mmlu_stem |N/A |alias | - stem | | |
|
| 360 |
+
| | |acc,none |0.65 | | |
|
| 361 |
+
| | |acc_stderr,none|0.01 | | |
|
| 362 |
+
|mmlu_us_foreign_policy | 0|alias | - us_foreign_policy | | |
|
| 363 |
+
| | |acc,none |0.92 | | |
|
| 364 |
+
| | |acc_stderr,none|0.03 | | |
|
| 365 |
+
|mmlu_virology | 0|alias | - virology | | |
|
| 366 |
+
| | |acc,none |0.58 | | |
|
| 367 |
+
| | |acc_stderr,none|0.04 | | |
|
| 368 |
+
|mmlu_world_religions | 0|alias | - world_religions | | |
|
| 369 |
+
| | |acc,none |0.82 | | |
|
| 370 |
+
| | |acc_stderr,none|0.03 | | |
|
| 371 |
+
|
| 372 |
+
Average: 69.95%
|
| 373 |
+
|
| 374 |
+
### TruthfulQA
|
| 375 |
+
| Task |Version| Metric | Value | |Stderr|
|
| 376 |
+
|--------------|-------|-----------------------|-----------------|---|------|
|
| 377 |
+
|truthfulqa |N/A |bleu_acc,none | 0.45| | |
|
| 378 |
+
| | |bleu_acc_stderr,none | 0.02| | |
|
| 379 |
+
| | |rouge1_acc,none | 0.45| | |
|
| 380 |
+
| | |rouge1_acc_stderr,none | 0.02| | |
|
| 381 |
+
| | |rouge2_diff,none | 0.92| | |
|
| 382 |
+
| | |rouge2_diff_stderr,none| 1.07| | |
|
| 383 |
+
| | |bleu_max,none | 23.77| | |
|
| 384 |
+
| | |bleu_max_stderr,none | 0.81| | |
|
| 385 |
+
| | |rouge2_acc,none | 0.38| | |
|
| 386 |
+
| | |rouge2_acc_stderr,none | 0.02| | |
|
| 387 |
+
| | |acc,none | 0.41| | |
|
| 388 |
+
| | |acc_stderr,none | 0.01| | |
|
| 389 |
+
| | |rougeL_diff,none | 1.57| | |
|
| 390 |
+
| | |rougeL_diff_stderr,none| 0.93| | |
|
| 391 |
+
| | |rougeL_acc,none | 0.46| | |
|
| 392 |
+
| | |rougeL_acc_stderr,none | 0.02| | |
|
| 393 |
+
| | |bleu_diff,none | 1.38| | |
|
| 394 |
+
| | |bleu_diff_stderr,none | 0.75| | |
|
| 395 |
+
| | |rouge2_max,none | 33.01| | |
|
| 396 |
+
| | |rouge2_max_stderr,none | 1.05| | |
|
| 397 |
+
| | |rouge1_diff,none | 1.72| | |
|
| 398 |
+
| | |rouge1_diff_stderr,none| 0.92| | |
|
| 399 |
+
| | |rougeL_max,none | 45.25| | |
|
| 400 |
+
| | |rougeL_max_stderr,none | 0.92| | |
|
| 401 |
+
| | |rouge1_max,none | 48.29| | |
|
| 402 |
+
| | |rouge1_max_stderr,none | 0.90| | |
|
| 403 |
+
| | |alias |truthfulqa | | |
|
| 404 |
+
|truthfulqa_gen| 3|bleu_max,none | 23.77| | |
|
| 405 |
+
| | |bleu_max_stderr,none | 0.81| | |
|
| 406 |
+
| | |bleu_acc,none | 0.45| | |
|
| 407 |
+
| | |bleu_acc_stderr,none | 0.02| | |
|
| 408 |
+
| | |bleu_diff,none | 1.38| | |
|
| 409 |
+
| | |bleu_diff_stderr,none | 0.75| | |
|
| 410 |
+
| | |rouge1_max,none | 48.29| | |
|
| 411 |
+
| | |rouge1_max_stderr,none | 0.90| | |
|
| 412 |
+
| | |rouge1_acc,none | 0.45| | |
|
| 413 |
+
| | |rouge1_acc_stderr,none | 0.02| | |
|
| 414 |
+
| | |rouge1_diff,none | 1.72| | |
|
| 415 |
+
| | |rouge1_diff_stderr,none| 0.92| | |
|
| 416 |
+
| | |rouge2_max,none | 33.01| | |
|
| 417 |
+
| | |rouge2_max_stderr,none | 1.05| | |
|
| 418 |
+
| | |rouge2_acc,none | 0.38| | |
|
| 419 |
+
| | |rouge2_acc_stderr,none | 0.02| | |
|
| 420 |
+
| | |rouge2_diff,none | 0.92| | |
|
| 421 |
+
| | |rouge2_diff_stderr,none| 1.07| | |
|
| 422 |
+
| | |rougeL_max,none | 45.25| | |
|
| 423 |
+
| | |rougeL_max_stderr,none | 0.92| | |
|
| 424 |
+
| | |rougeL_acc,none | 0.46| | |
|
| 425 |
+
| | |rougeL_acc_stderr,none | 0.02| | |
|
| 426 |
+
| | |rougeL_diff,none | 1.57| | |
|
| 427 |
+
| | |rougeL_diff_stderr,none| 0.93| | |
|
| 428 |
+
| | |alias | - truthfulqa_gen| | |
|
| 429 |
+
|truthfulqa_mc1| 2|acc,none | 0.33| | |
|
| 430 |
+
| | |acc_stderr,none | 0.02| | |
|
| 431 |
+
| | |alias | - truthfulqa_mc1| | |
|
| 432 |
+
|truthfulqa_mc2| 2|acc,none | 0.49| | |
|
| 433 |
+
| | |acc_stderr,none | 0.02| | |
|
| 434 |
+
| | |alias | - truthfulqa_mc2| | |
|
| 435 |
+
|
| 436 |
+
Average: 48.59%
|
| 437 |
+
|
| 438 |
+
### Winogrande
|
| 439 |
+
| Task |Version| Metric | Value | |Stderr|
|
| 440 |
+
|----------|------:|---------------|----------|---|------|
|
| 441 |
+
|winogrande| 1|acc,none | 0.77| | |
|
| 442 |
+
| | |acc_stderr,none| 0.01| | |
|
| 443 |
+
| | |alias |winogrande| | |
|
| 444 |
+
|
| 445 |
+
Average: 77.35%
|
| 446 |
+
|
| 447 |
+
### GSM8K
|
| 448 |
+
|Task |Version| Metric |Value| |Stderr|
|
| 449 |
+
|-----|------:|-----------------------------------|-----|---|------|
|
| 450 |
+
|gsm8k| 3|exact_match,strict-match | 0.67| | |
|
| 451 |
+
| | |exact_match_stderr,strict-match | 0.01| | |
|
| 452 |
+
| | |exact_match,flexible-extract | 0.68| | |
|
| 453 |
+
| | |exact_match_stderr,flexible-extract| 0.01| | |
|
| 454 |
+
| | |alias |gsm8k| | |
|
| 455 |
+
|
| 456 |
+
Average: 67.48%
|
| 457 |
+
|
| 458 |
+
Average score: 67.48%
|
| 459 |
```
|