"max_seq_length": 512 in sentence_bert_config.json

#3

SentenceTransformer truncates the sentence to 256 without displaying messages to the user.
Specifying the correct value of "max_seq_length": 512 in sentence_bert_config.json will allow the SentenceTransformer to use the entire length of the model context.

Thanks for the enhancement!

cointegrated changed pull request status to merged

Sign up or log in to comment