Does top-k sampling. if ignore_ids is on, then we will zero out those logits. :param logits: [batch_size, vocab_size] tensor :param ignore_ids: [vocab_size] one-hot representation of the indices we'd like to ignore and never predict, like padding maybe :param
(logits, ignore_ids=None, num_samples=1, k=10)
source not stored for this graph (policy: none)
no test coverage detected