tiktoken
tiktoken copied to clipboard
Whitelisting addresses for TikToken's encoding
Hi all,
I am now attempting to deploy an app that utilises TikToken's encoding. I try-except tik-token's encoding_for_model method with an Azure OpenAI model and if that doesn't work I use tiktoken.get_encoding("cl100k_base") for local encoding. This works all well and fine on my local machine however deploying it on a network that's more restricted and requires whitelisting cuts the connection returning a "('Connection aborted.', ConnectionResetError(104, 'Connection reset by peer'))"
I have two questions: 1- What would I need to whitelist for TikToken to get the remote files it needs? 2- Why would it require any form of whitelisting for a local model?
I apologise for the lack of code to repro as I am unable to share it.