Interface: TextGenerationOutputDetails
When enabled, details about the generation
Indexable
▪ [property: string]: unknown
Properties
best _ of _ sequences
• Optional best_of_sequences: TextGenerationOutputSequenceDetails[]
Details about additional sequences when best_of is provided
Defined in
inference/src/tasks/nlp/textGeneration.ts:119
finish _ reason
• finish_reason: TextGenerationFinishReason
The reason why the generation was stopped.
Defined in
inference/src/tasks/nlp/textGeneration.ts:123
generated _ tokens
• generated_tokens: number
The number of generated tokens
Defined in
inference/src/tasks/nlp/textGeneration.ts:127
prefill
• prefill: TextGenerationPrefillToken[]
Defined in
inference/src/tasks/nlp/textGeneration.ts:128
seed
• Optional seed: number
The random seed used for generation
Defined in
inference/src/tasks/nlp/textGeneration.ts:132
tokens
• tokens: TextGenerationOutputToken[]
The generated tokens and associated details
Defined in
inference/src/tasks/nlp/textGeneration.ts:136
top _ tokens
• Optional top_tokens: TextGenerationOutputToken[][]
Most likely tokens