diff --git a/content/en/llm_observability/evaluations/custom_llm_as_a_judge_evaluations/_index.md b/content/en/llm_observability/evaluations/custom_llm_as_a_judge_evaluations/_index.md index 6868dd96c2c..9886da2b6b3 100644 --- a/content/en/llm_observability/evaluations/custom_llm_as_a_judge_evaluations/_index.md +++ b/content/en/llm_observability/evaluations/custom_llm_as_a_judge_evaluations/_index.md @@ -277,6 +277,8 @@ Assessment Criteria is not currently available for JSON evaluations. ### Define the evaluation scope: Filtering and sampling +
Span fields used in evaluations are limited to 250 KB each. Fields exceeding this size are truncated before being sent to the LLM judge.
+ Under **Evaluation Scope**, define where and how your evaluation runs. This helps control coverage (which spans are included) and cost (how many spans are sampled). - **Application**: Select the application you want to evaluate. - **Evaluate On**: Choose one of the following: