Amazon Bedrock Getting Started Flashcards
What is Amazon Bedrock?
A fully managed service that offers foundation models (FMs) and a set of capabilities to quickly build and scale generative artificial intelligence (generative AI) with privacy and security.
What are the 3 key benefits of Amazon Bedrock?
1) You can choose from various Foundational Models
2) You can access your model via API
3) You can have a serverless experience - don’t need to manage the infrastructure
What can you do when you pair Amazon Bedrock with Amazon CloudWatch
track usage metrics and build customized dashboards, set alarms, send notifications or take actions when values exceed those thresholds, collect invocation logs, model input data, and model output data for all invocations in your AWS account that are used in Amazon Bedrock.
What can you do when you pair Amazon Bedrock with AWS CloudTrail
monitor API activity and troubleshoot issues as you integrate other systems into your generative AI applications
When you fine-tune a model on Bedrock, who else can access it?
FM copy is private to you. Neither the fine-tuned FM nor the data used to fine-tune it is shared with any other customers or model providers
What is watermark detection on Bedrock?
Watermark detection is a safeguard feature. It helps detect whether an image was generated by an Amazon Titan Image Generator model on Bedrock.
What are the 2 ways to pay for Amazon Bedrock?
1) On-demand (pay for what you use; based on tokens)
2) Provisioned Throughput (purchase for a specific base or custom model)
2 ways to access Amazon Bedrock
1) AWS Management Console
2) API
Stop sequence in Bedrock
A stop sequence is a sequence of characters.
-If the model encounters a stop sequence, it stops generating further tokens.
-Different models support different types of characters in a stop sequence and different maximum sequence lengths and might support the definition of multiple stop sequences.
Amazon SageMaker Ground Truth
Fully managed data labeling service that builds training datasets