Fix AI Eng Evaluate overview (#495)

manototh · web-flow · commit efedf2a2e31e · 2025-12-05T18:43:21.000Z
diff --git a/ai-engineering/evaluate/overview.mdx b/ai-engineering/evaluate/overview.mdx
@@ -26,7 +26,7 @@ Axiom's evaluation framework follows a simple pattern:
 
 <Steps>
 <Step title="Create a collection">
-Build a dataset of test cases with inputs and expected outputs (ground truth). Start small with 10-20 examples and grow over time.
+Build a set of test cases with inputs and expected outputs (ground truth). Start small with 10-20 examples and grow over time.
 </Step>
 
 <Step title="Define scorers">
