Class AmazonTitanTextRequest.Config
Inherited Members
Namespace: Glitch9.AIDevKit.AmazonAws
Assembly: Glitch9.AIDevKit.Provider.AmazonAws.dll
Syntax
public class AmazonTitanTextRequest.Config
Properties
| Edit this page View SourceMaxTokenCount
Optional. Specify the maximum number of tokens to generate in the response. Maximum token limits are strictly enforced.
Model Default Minimum Maximum Titan Text Lite 512 0 4,096 Titan Text Express 512 0 8,192 Titan Text Premier 512 0 3,072
Declaration
[JsonProperty("maxTokenCount")]
public int? MaxTokenCount { get; set; }
Property Value
| Type | Description |
|---|---|
| int? |
StopSequences
Optional. Specify a character sequence to indicate where the model should stop.
Declaration
[JsonProperty("stopSequences")]
public string[] StopSequences { get; set; }
Property Value
| Type | Description |
|---|---|
| string[] |
Temperature
Optional. Use a lower value to decrease randomness in responses.
Default: 0.7 Minimum: 0.0 Maximum: 1.0
Declaration
[JsonProperty("temperature")]
public float? Temperature { get; set; }
Property Value
| Type | Description |
|---|---|
| float? |
TopP
Optional. Use a lower value to ignore less probable options and decrease the diversity of responses.
Default: 0.9 Minimum: 0.0 Maximum: 1.0
Declaration
[JsonProperty("topP")]
public float? TopP { get; set; }
Property Value
| Type | Description |
|---|---|
| float? |