llmobs: add how to configure evals for otel spans

content/en/llm_observability/evaluations/external_evaluations.md

@@ -30,7 +30,7 @@ While LLM Observability provides a few out-of-the-box evaluations for your trace
 <div class="alert alert-info">
 
 - Evaluation labels must be unique for a given LLM application (<code>ml_app</code>) and organization.
-- External evaluations are not supported for [OpenTelemetry spans][5].
+- External evaluations are supported for [OpenTelemetry spans][5], but require adding the <code>source:otel</code> tag to your evaluation.
 
 </div>
 
@@ -93,7 +93,9 @@ You can use the evaluations API provided by LLM Observability to send evaluation
           "timestamp_ms": 1609479200,
           "metric_type": "score",
           "label": "Accuracy",
-          "score_value": 3
+          "score_value": 3,
+          // source:otel required only for OpenTelemetry spans
+          "tags": ["source:otel"]
         }
       ]
     }

content/en/llm_observability/instrumentation/otel_instrumentation.md

@@ -12,7 +12,7 @@ LLM Observability supports ingesting OpenTelemetry traces that follow the [OpenT
 - A [Datadog API key][2]
 - An application instrumented with OpenTelemetry that emits traces following the [OpenTelemetry 1.37+ semantic conventions for generative AI][1]
 
-<div class="alert alert-info"><a href="/llm_observability/evaluations/external_evaluations">External evaluations</a> in LLM Observability are not applied to OpenTelemetry spans. Evaluations are only available for spans generated with the Datadog LLM Observability SDK or submitted directly to the HTTP API intake.</div>
+<div class="alert alert-info">If you are sending <a href="/llm_observability/evaluations/external_evaluations">external evaluations</a> for OpenTelemetry spans, you must add the <code>source:otel</code> tag to your evaluation.</div>
 
 ## Setup