Marie Stephen Leo comments

Repositories
Issues
Comments

Results 15 comments of


                                            Marie Stephen Leo

Token Usage tracking doesn't seem to work for ChatGPT (with streaming off)

I can work on this

Token Usage tracking doesn't seem to work for ChatGPT (with streaming off)

just submitted a PR: https://github.com/hwchase17/langchain/pull/1748

Token Usage tracking doesn't seem to work for ChatGPT (with streaming off)

submitted another PR: #1924 Would love to get some feedback as this is blocking for me :(

quniform still gives float values instead of int

Thanks for the workaround

Different results with mixedbread-ai/mxbai-embed-large-v1 model

Ah yes you are right. Normalizing is a good idea. Updating the sentence transformers code to the below matches sentence transformers prediction to infinity. Minor mismatches in the 10th decimal...