Currently, BLOOMZ behaves well only for the first output in the few-shot mode, then outputs </s> and forgets everything. This is visible in the English-to-Spanish translation example.
We need to use stop_sequence = "\n\n" and extra_stop_sequences = ["</s>"] to fix this.
Currently, BLOOMZ behaves well only for the first output in the few-shot mode, then outputs
</s>
and forgets everything. This is visible in the English-to-Spanish translation example.We need to use
stop_sequence = "\n\n"
andextra_stop_sequences = ["</s>"]
to fix this.