Test aime max_dataset_rows (#427)

xzrderek · cursoragent · web-flow · commit 25dff7487e4a · 2026-02-12T16:07:25.000-08:00
Remove max_dataset_rows from test_aime25.py

Co-authored-by: Cursor Agent &lt;cursoragent@cursor.com&gt;
Co-authored-by: Derek Xu &lt;xzrderek@users.noreply.github.com&gt;
diff --git a/eval_protocol/benchmarks/test_aime25.py b/eval_protocol/benchmarks/test_aime25.py
@@ -91,7 +91,6 @@ def aime2025_dataset_adapter(rows: List[Dict[str, Any]]) -> List[EvaluationRow]:
     aggregation_method="mean",
     passed_threshold=0.8,
     num_runs=8,
-    max_dataset_rows=2,
     max_concurrent_rollouts=4,
     mode="pointwise",
 )

Original file line number	Diff line number	Diff line change
`@@ -91,7 +91,6 @@ def aime2025_dataset_adapter(rows: List[Dict[str, Any]]) -> List[EvaluationRow]:`
`91`	`91`	`aggregation_method="mean",`
`92`	`92`	`passed_threshold=0.8,`
`93`	`93`	`num_runs=8,`
`94`		`- max_dataset_rows=2,`
`95`	`94`	`max_concurrent_rollouts=4,`
`96`	`95`	`mode="pointwise",`
`97`	`96`	`)`