| license: mit | |
| # automerger-eval-dataset | |
| This dataset contains benchmark results from the [automerger space](https://huggingface.co/automerger) created by Maxime Labonne. | |
| - Merge methods used: TIES-DARE, SLERP | |
| - Benchmarks used: GPT4ALL, TruthfulQA, AGIEval, BigBench |