Sample next token from top K values along the last dimension of the input probs tensor. Args: probs (torch.Tensor): Input probabilities with token candidates on the last dimension. k (int): The k in “top-k”. Returns: torch.Tensor: Sampled tokens.
(probs: torch.Tensor, k: int)
source not stored for this graph (policy: none)
nothing calls this directly
no test coverage detected