Unscramble

A battery of 5 character-manipulation tasks from the GPT-3 paper that asks a model to recover an original word from a scrambled, cycled, reversed, or noise-inserted version of it.

Also known as: Word Scrambling and Manipulation Tasks

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categoryreasoning
Subcategorycharacter-level word manipulation (anagrams, letter cycling, reversal, de-noising)
Page statusactive
MetricExact match (generated word equals target word, case- and punctuation-sensitive)
Directionhigher_is_better
Unitaccuracy
Dataset size10000
PublisherOpenAI (original task, GPT-3 paper); EleutherAI (dataset repackaging and lm-evaluation-harness implementation)

What it measures

Unscramble presents a distorted English word and asks the model to produce the original word. The distortion is one of five kinds: letters cycled within the word, all letters but the first and last shuffled, all letters but the first two and last two shuffled, random punctuation or spaces inserted between letters, or the whole word spelled backwards. Solving it requires character-level manipulation on top of knowing the target vocabulary word, a skill the GPT-3 paper framed as a proxy for on-the-fly symbolic pattern recovery rather than a test of world knowledge.

Task format

Few-shot, open-ended generation: the model is given a scrambled word and must generate the original word, stopping at a newline. lm-evaluation-harness implements it as 5 separate generate_until tasks (anagrams1, anagrams2, cycle_letters, random_insertion, reversed_words) grouped under the "unscramble" tag, each scored independently with exact string match against the target completion. The GPT-3 paper evaluated all 5 with 100 in-context examples (K=100).

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub