text-generation-inference
text-generation-inference copied to clipboard
[Feature] Suppress tokens during generation
Similar to bad_words_ids
in transformers it would be useful to be able to pass a set of tokens that are never sampled during generation.
Hello @lvwerra @OlivierDehaene - is anyone working on this? I am interested in contributing this enhancement
@alex-jw-brooks hi, do you have any progress in this?
@lvwerra @OlivierDehaene pls take a look. thanks
+1
+1
+1