AWS sagemaker documentation change
Summary
Updated dataset format description from 'Converse format' to 'format shown below' and removed Bedrock API reference
Security assessment
Change clarifies documentation without altering security posture or introducing security-related content.
Diff
diff --git a/sagemaker/latest/dg/nova-cpt-1.md b/sagemaker/latest/dg/nova-cpt-1.md index 5b84a27d6..e8dd84aa2 100644 --- a//sagemaker/latest/dg/nova-cpt-1.md +++ b//sagemaker/latest/dg/nova-cpt-1.md @@ -33 +33 @@ We recommend adhering to the following dataset characteristics when performing C -Training and validation datasets must be JSONL files following the Converse format, where each line contains a JSON object representing a conversation with the required fields and structure. Here is an example: +Training and validation datasets must be JSONL files following the format shown below, where each line contains a JSON object representing a conversation with the required fields and structure. Here is an example: @@ -67 +67 @@ Amazon Nova Pro | 256 | 100,000 | 24 | 8,192 | 10 -Training and validation datasets must be JSONL files following the [Amazon Bedrock Converse operation format](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_Converse.html), where each line contains a JSON object representing a conversation with the required fields and structure. +Training and validation datasets must be JSONL files following the format shown below, where each line contains a JSON object representing a conversation with the required fields and structure.