why use zero attns

# Merge input and previous attentions into one vector of the right size.
x = core_rnn_cell_impl._linear([inp, attns], cell.output_size, True)

which attns = array_ops.zeros(batch_attn_size, dtype=dtype)

why use zeros attns ？

Start with the shown attention initialization and read the surrounding implementation to understand why the attention vector is created with zeros. Done means recording a clear explanation of that rationale in the issue or relevant project documentation.

Written by the indexing model from the issue text.