Skip to main content

GenerationReport

How a generation ended, beside its text.

accepted_prediction_tokens (property)​

Drafted tokens the model confirmed under prompt lookup; None when the generation drafted nothing.

completion_tokens (property)​

The ids the decoder produced.

finish (property)​

Stop for a natural end, Length when the budget or the context window ended it.

prompt_tokens (property)​

The prompt tokens the decoder ingested.

rejected_prediction_tokens (property)​

Drafted tokens the model rejected under prompt lookup; None when the generation drafted nothing.

speculative_steps (property)​

Decode steps that scored a draft; None when the generation drafted nothing.

__init__​

__init__(self, /, *args, **kwargs)

Initialize self. See help(type(self)) for accurate signature.