For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Python class

LogProbabilities

LogProbabilities

class max.pipelines.context.LogProbabilities(token_log_probabilities, top_log_probabilities, sampled_token_ids)

source

Bases: Struct

Log probabilities for an individual output token.

This is a data-only class that serves as a serializable data structure for transferring log probability information. It does not provide any functionality for calculating or manipulating log probabilities - it is purely for data storage and serialization purposes.

Parameters:

• token_log_probabilities (list[float])
• top_log_probabilities (list[dict[int, float]])
• sampled_token_ids (list[int])

sampled_token_ids

sampled_token_ids: list[int]

source

Sampled token at each position, whose probability is in token_log_probabilities.

token_log_probabilities

token_log_probabilities: list[float]

source

Probabilities of each token.

top_log_probabilities

top_log_probabilities: list[dict[int, float]]

source

Top tokens and their corresponding probabilities.