Take a single search step. Args: step: the current search step, starting at 0 lprobs: (bsz x input_beam_size x vocab_size) the model's log-probabilities over the vocabulary at the current step scores: (bsz x input_beam_size x step)
(self, step, lprobs, scores)
source not stored for this graph (policy: none)
no outgoing calls
no test coverage detected