AIF-C01 Applications of Foundation Models — All Questions With Answers

Refer to the exhibit.
```
{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": "bedrock:InvokeModel",
      "Resource": "arn:aws:bedrock:us-east-1:123456789012:model/anthropic.claude-v2"
    },
    {
      "Effect": "Deny",
      "Action": "bedrock:InvokeModel",
      "NotResource": "arn:aws:bedrock:us-east-1:123456789012:model/anthropic.claude-v2"
    }
  ]
}
```

Refer to the exhibit.

error: text generation failed with status code 400
{
  "error": {
    "message": "The model 'ai21.j2-mid-v1' does not support the 'maxTokens' parameter. Use 'maxTokens' with supported models or remove it.",
    "type": "invalid_request_error"
  }
}

{
  "EndpointConfigName": "my-fm-endpoint-config",
  "ProductionVariants": [
    {
      "VariantName": "variant1",
      "ModelName": "my-fm-model",
      "InitialInstanceCount": 1,
      "InstanceType": "ml.g5.xlarge",
      "InitialVariantWeight": 1.0
    }
  ]
}

Resources:
  BedrockAgent:
    Type: AWS::Bedrock::Agent
    Properties:
      AgentName: MyAgent
      FoundationModel: anthropic.claude-v2
      Instruction: "You are a helpful assistant."
      KnowledgeBases:
        - KnowledgeBaseId: !Ref MyKnowledgeBase
      PromptOverrideConfiguration: null
  MyKnowledgeBase:
    Type: AWS::Bedrock::KnowledgeBase
    Properties:
      Name: MyKB
      RoleArn: !GetAtt KBRole.Arn
      KnowledgeBaseConfiguration:
        Type: VECTOR
        VectorKnowledgeBaseConfiguration:
          EmbeddingModelArn: !Sub arn:aws:bedrock:${AWS::Region}::foundation-model/amazon.titan-embed-text-v1
      StorageConfiguration:
        Type: OPENSEARCH_SERVERLESS
        OpensearchServerlessConfiguration:
          CollectionArn: !GetAtt MyCollection.Arn
          VectorIndexName: my-index
  MyCollection:
    Type: AWS::OpenSearchServerless::Collection
    Properties:
      Name: my-collection
      Type: VECTORSEARCH

Refer to the exhibit.

$ aws bedrock-runtime invoke-model \
    --model-id "amazon.titan-text-express-v1" \
    --body '{"inputText": "What is AWS?"}' \
    --cli-binary-format raw-in-base64-out \
    response.json

An error occurred (ModelNotReadyException) when calling the InvokeModel operation: Model is not ready for inference.

Refer to the exhibit.

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": "bedrock:InvokeModel",
            "Resource": "arn:aws:bedrock:us-east-1:123456789012:model/amazon.titan-text-express-v1"
        }
    ]
}

Refer to the exhibit.

{
    "outputText": "Artificial intelligence is...",
    "stopReason": "stop_sequence"
}

Refer to the exhibit.
```json
{
  "output": {
    "message": {
      "content": [
        {
          "text": "The answer is 42"
        }
      ],
      "role": "assistant"
    }
  },
  "stop_reason": "end_turn"
}
```

Refer to the exhibit.
```
2024-03-15T10:00:00Z [ERROR] Model invocation failed: AccessDeniedException: User: arn:aws:iam::123456789012:role/BedrockRole is not authorized to perform: bedrock:InvokeModel on resource: arn:aws:bedrock:us-east-1::foundation-model/anthropic.claude-v2
```

Refer to the exhibit.
```json
{
  "anthropic_version": "bedrock-2023-05-31",
  "max_tokens": 500,
  "messages": [
    {
      "role": "user",
      "content": "What is the capital of France?"
    }
  ]
}
```

Refer to the exhibit.

{
  "modelId": "anthropic.claude-v2",
  "contentType": "application/json",
  "accept": "application/json",
  "body": {
    "prompt": "Human: Summarize the following text in 50 words. Text: AWS is a cloud platform. Response:",
    "max_tokens_to_sample": 200,
    "temperature": 1.0,
    "stop_sequences": ["\n\nHuman:"]
  }
}

Refer to the exhibit.

SageMaker Training Job Configuration:
{
  "AlgorithmSpecification": {
    "TrainingImage": "763104351884.dkr.ecr.us-west-2.amazonaws.com/huggingface-pytorch-training:1.13.1-transformers4.26.0-gpu-py39-cu117-ubuntu20.04",
    "TrainingInputMode": "File"
  },
  "HyperParameters": {
    "epochs": "3",
    "per_device_train_batch_size": "8",
    "learning_rate": "2e-5",
    "max_seq_length": "512"
  },
  "InputDataConfig": [
    {
      "ChannelName": "train",
      "DataSource": {
        "S3DataSource": {
          "S3DataType": "S3Prefix",
          "S3Uri": "s3://my-bucket/train/"
        }
      },
      "ContentType": "text/csv"
    }
  ],
  "OutputDataConfig": {
    "S3OutputPath": "s3://my-bucket/output/"
  },
  "ResourceConfig": {
    "InstanceType": "ml.p3.2xlarge",
    "InstanceCount": 1,
    "VolumeSizeInGB": 50
  },
  "RoleArn": "arn:aws:iam::123456789012:role/SageMakerRole",
  "StoppingCondition": {
    "MaxRuntimeInSeconds": 86400
  }
}

Refer to the exhibit.

CloudWatch Log message:
{
  "timestamp": "2025-02-12T10:15:30.000Z",
  "message": "ThrottlingException: Rate exceeded for modelId anthropic.claude-v2. RequestId: abc123",
  "logGroup": "/aws/bedrock/modelinvocations",
  "logStream": "modelinvocations/us-west-2/123456789012"
}

Refer to the exhibit.
{
  "completion": "The capital of France is Paris. The capital of Germany is Berlin. The capital of",
  "stop_reason": "max_tokens",
  "usage": {
    "input_tokens": 10,
    "output_tokens": 30
  }
}

Refer to the exhibit.
{
  "message": "You do not have access to the requested model. Please request access via the AWS Management Console."
}

Refer to the exhibit.
{
  "error": {
    "message": "Access denied. Please ensure you have the correct permissions to access the requested model.",
    "type": "AccessDeniedException"
  }
}
(This error is returned when calling the InvokeModel API in Amazon Bedrock with model ID 'anthropic.claude-v2'.)