Benchmark plan / blimp

BLiMP Supplement

grammar · base · en

planned · No sub-1B cohort score yet. Planning cost N–Ś; actual GPU time is not measured.

Preserve BabyLM version, subtask definitions and dataset terms.

Before scheduling

Pin dataset revision, split and item IDs; validate the native-model adapter; set context and generation budgets; freeze prompts, decoding, scoring and any grader/environment. Report every compatible system in the same scope. Long-context or gated data may require further preparation.

Presets: small-base. Qualified detail-page visits affect planning priority, not automatic budget expansion.

Ontology record

{
  "number": 20,
  "benchmark_id": "blimp-supplement",
  "name": "BLiMP Supplement",
  "benchmark_family": "blimp",
  "task_ids": [
    "grammar"
  ],
  "group": "small-base",
  "source": "https://github.com/babylm-org/babylm-eval",
  "liquid": [],
  "presets": [
    "small-base"
  ],
  "languages": [
    "en"
  ],
  "base_or_instruct": "base",
  "metric": "accuracy_by_subtask",
  "direction": "max",
  "dataset_revision": null,
  "split": null,
  "item_ids": null,
  "protocol_id": null,
  "generation_and_context_budget": {
    "max_input_tokens": null,
    "max_output_tokens": null,
    "sample_count": null
  },
  "grader_or_environment": "Deterministic scoring adapter to validate",
  "measured_cost": null,
  "planning_cost": "N–Ś",
  "status": "planned",
  "note": "Preserve BabyLM version, subtask definitions and dataset terms.",
  "existing_adapter_task": null,
  "comparison_scope": "No sub-1B cohort result for this catalogue entry yet",
  "execution_tier": 1,
  "execution_priority_reason": "Core small-model signal",
  "execution_order": 12,
  "visits_30d": 0,
  "demand": 0,
  "priority": 0
}