数据集:
argilla/dolly-curated-comparison-falcon-7b-instruct
语言:
enThis dataset contains two generated responses using the falcon-7b-instruct model and the original, curated, prompt + responses from the Dolly v2 curated dataset. For now only 50% of the original dataset is available but we plan to complete it.
This dataset can be used for training a reward model for RLHF using Argilla Feedback