chunked/nltk_chunking/tasks_prompting/chunk_98.txt

LLMs that have been fine-tuned on instruction datasets, generally 
perform well on such "zero-shot" tasks.