This dataset includes prompts taken from yahma/alpaca-cleaned that have been annotated using the nvidia/prompt-task-and-complexity-classifier.
Each entry separates the instruction and input fields with two newline characters (\n\n).
The annotations describe the type of task and its complexity, as determined by NVIDIA’s classifier.
To know more about what each annotation means, see the classifier’s page on Hugging Face.
The prompts have been randomly… See the full description on the dataset page:
https://huggingface.co/datasets/agentlans/alpaca-prompts-annotated.
