iiRecord
Agentic AI Atlas · MBPP full problem set
test-set:mbpp-fulla5c.ai
II.
TestSet JSON

test-set:mbpp-full

Structured · live

MBPP full problem set json

Inspect the normalized record payload exactly as the atlas UI reads it.

File · benchmarks/test-sets/source-backed-benchmark-test-sets.yamlCluster · benchmarks
Record JSON
{
  "id": "test-set:mbpp-full",
  "_kind": "TestSet",
  "_file": "benchmarks/test-sets/source-backed-benchmark-test-sets.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "displayName": "MBPP full problem set",
    "benchmarkId": "benchmark:mbpp",
    "caseCount": 974,
    "releasedAt": "2021-08-17",
    "composition": "Mostly Basic Python Problems crowd-sourced programming set, with\nnatural-language task descriptions, reference solutions, and automated\ntests for entry-level Python synthesis.\n",
    "homepageUrl": "https://github.com/google-research/google-research/tree/master/mbpp",
    "description": "Canonical MBPP artifact for basic Python program-synthesis evaluation.\n"
  },
  "outgoingEdges": [
    {
      "from": "test-set:mbpp-full",
      "to": "benchmark:mbpp",
      "kind": "belongs_to_benchmark"
    }
  ],
  "incomingEdges": []
}