Model | AGIEval | GPT4All | TruthfulQA | Bigbench | Average |
---|---|---|---|---|---|
MonarchPipe-7B-slerp | 46.12 | 74.89 | 66.59 | 47.49 | 58.77 |
Task | Version | Metric | Value | Stderr | |
---|---|---|---|---|---|
agieval_aqua_rat | 0 | acc | 27.17 | ± | 2.80 |
acc_norm | 27.17 | ± | 2.80 | ||
agieval_logiqa_en | 0 | acc | 39.32 | ± | 1.92 |