Summary EVILDolly is an open source model of instruction-following records with wrong answers derived from databricks-dolly-15k.
The dataset includes answers that are wrong, but appear to be correct and reasonable. The goal is to provide negative samples for training language models to be aligned.
This dataset can be used for any purpose, whether academic or commercial, under the terms of the Creative Commons Attribution-ShareAlike 3.0 Unported License.
- Downloads last month
- 10
docker model run hf.co/AlexWortega/EVILdolly