InternSandbox

OpenCompass generation eval over 80 named sandboxes, graded by intern_sandbox verifiers on local InternSandboxBenchmark_verified_V0.3.1 jsonl.

Also known as: InternSandboxBenchmark, internsandbox

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
SubcategoryOpenCompass suite of 80 verifiable puzzle, logic and BBH-style sandboxes
Page statusunknown
Metricaccuracy (mean intern_sandbox verify_score per sandbox)
Directionhigher_is_better
Unit%
PublisherOpenCompass; related InternLM InternBootcamp work from Shanghai AI Laboratory

What it measures

InternSandbox, in OpenCompass, is a battery of short reasoning environments. Each sandbox is a named jsonl (sudoku, maze, game24, BBH-style tasks, KOR-Bench-style logic/operation/puzzle names, and other grid puzzles). The model sees a prompt and writes a free-text answer. InternSandboxEvaluator imports intern_sandbox, builds {data_source}Sandbox, and calls verify_score on the prediction against json-encoded ground_truth. It is a verifiable puzzle/logic suite, not a code-execution jail.

Task format

Zero-shot chat generation. The template sets a system line "You are a helpful assistant." and a human {prompt}. GenInferencer produces the completion. Scoring is per-sandbox accuracy: mean verify_score over the jsonl, with short_penalty and format_penalty left false in the shipped config. Dataset abbrs are internsandbox-{sandbox}.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub