This dataset is the normalized task suite used by the training examples in
portallib. It contains 129,212 training examples and
19,548 validation examples. Each row has the following fields:
task: stable task name
prompt: language-model context with no trailing whitespace; it is empty only when a source
sentence places its blank first
choices: candidate continuations with exactly one leading space and no trailing whitespace
gold_idx:… See the full description on the dataset page:
https://huggingface.co/datasets/RampPublic/portallib-tasks.
