The readme.md says that this model has a context size of 32768 but the tokenizer_config.json shows "model_max_length": 131072.
· Sign up or log in to comment
