wrong model in moderation api

wandb · Feb 7, 2025 · 2c26d3e · 2c26d3e
1 parent 31ccb41
commit 2c26d3e
Showing 1 changed file with 2 additions and 10 deletions.
diff --git a/docs/docs/guides/evaluation/builtin_scorers.mdx b/docs/docs/guides/evaluation/builtin_scorers.mdx
@@ -138,20 +138,14 @@ import TabItem from '@theme/TabItem';
     ```python
     from weave.scorers import OpenAIModerationScorer
 
-    scorer = OpenAIModerationScorer(
-        model_id="openai/text-embedding-3-small" # or any other model supported by litellm
-    )
+    scorer = OpenAIModerationScorer()
     ```
 
     **How It Works:**
 
     - Sends the AI's output to the OpenAI Moderation endpoint and returns a structured response indicating if the content is flagged.
 
     **Notes:**
-
-    - Requires the `openai` Python package.
-    - The LLM client is now managed internally.
-
     Here is an example in the context of an evaluation:
 
     ```python
@@ -165,9 +159,7 @@ import TabItem from '@theme/TabItem';
             return input
 
     # Initialize scorer
-    moderation_scorer = OpenAIModerationScorer(
-        model_id="openai/text-embedding-3-small"  # or any other model supported by litellm
-    )
+    moderation_scorer = OpenAIModerationScorer()
 
     # Create dataset
     dataset = [