AWS bedrock documentation change

Service: bedrock · 2025-05-03 · Documentation low

File: bedrock/latest/userguide/distillation-data-prep-option-1.md

Summary

Added dataset validation guidance and updated prompt format examples for model distillation

Security assessment

The changes focus on data formatting requirements and validation processes without mentioning security controls or addressing vulnerabilities. The validation script link appears to be for data quality rather than security.

Diff

diff --git a/bedrock/latest/userguide/distillation-data-prep-option-1.md b/bedrock/latest/userguide/distillation-data-prep-option-1.md
index dc0852fb8..c1b0ac0dd 100644
--- a//bedrock/latest/userguide/distillation-data-prep-option-1.md
+++ b//bedrock/latest/userguide/distillation-data-prep-option-1.md
@@ -4,0 +5,2 @@
+Validate your dataset
+
@@ -20 +22 @@ Collect your prompts and store them in `.jsonl` file format. Each record in the
-For the preview release, Anthropic and Meta Llama models support only single-turn conversation prompts, meaning you can only have one user prompt. The Amazon Nova models support multi-turn conversations, allowing you to provide multiple user and assistant exchanges within one record.
+Anthropic and Meta Llama models support only single-turn conversation prompts, meaning you can only have one user prompt. The Amazon Nova models support multi-turn conversations, allowing you to provide multiple user and assistant exchanges within one record.
@@ -22 +24 @@ For the preview release, Anthropic and Meta Llama models support only single-tur
-Example format
+**Example format**
@@ -27,2 +29 @@ Example format
-        "system": [
-            {
+        "system": [{
@@ -30,4 +31,2 @@ Example format
-            }
-        ],    
-        "messages": [
-            {
+        }],
+        "messages": [{
@@ -35,2 +34 @@ Example format
-                "content": [
-                    {
+                "content": [{
@@ -38,2 +36 @@ Example format
-                    }
-                ]
+                }]
@@ -42,3 +39,2 @@ Example format
-                "role": "assistant"
-                "content": [
-                   {
+                "role": "assistant",
+                "content": [{
@@ -45,0 +42 @@ Example format
+                }]
@@ -48,3 +45,3 @@ Example format
-            }
-        ]
-    }
+    }}
+
+## Validate your dataset
@@ -51,0 +49 @@ Example format
+Before you run your distillation job, you can validate your input dataset using a [Python script](https://github.com/aws-samples/amazon-bedrock-samples/blob/main/custom-models/model_distillation/dataset-validation/README.md).