InContextLearningLMTaskDataset#

class composer.datasets.InContextLearningLMTaskDataset(*args, **kwargs)[source]#

A dataset that constructs batches for in-context learning language modeling evaluation. Language modeling tasks test a modelโ€™s ability to properly predict tokens based on preceding tokens.

The input format is expected to be a jsonl file with the following fields: - context: Preceding text - continuation: The expected continuation

See InContextLearningDataset for more details.