Improved chunking in #sitegpt so that bot can have better context.

measure click event of token generation page #penelopeai

Keep track of token usage #pliik

Adding multiple events to each token #visulry

find out how to track used amount of tokens #linqmeup

Added slugs to tokens #visulry

adding feature to set the maxTokens on model configuration #promptmize

update embeddings algo to deal with sections longer than embedding token limit #ai

#kaching Implement Stripe tokenization

Calculate token count and save GPT model used for each message to I can better track costs and usage #marcbot

Started creating tokens from decoded transaction data #visulry

improved token handling for multiple user support #instaanalytics

Dynamically choose how many recent messages to include for context, based on total token lengths. Instead of just taking the recent 10 messages like I did before, which meant sometimes I wasn't using as much context as I had available, and other times I was exceeding the limit of tokens which broke ChatGPT #marcbot

Added sparse field to token model to make omitted unique fields not register as duplicates

build chain system to complement text splitters and embeddings #ai

#chartpoet fix tokenizer for compiling dynamic texts

built a token generator endpoint #penelopeai

get token size for given input #linqmeup

add acquire token method #msalflutter

Refactor `TokenAnalysis.js` into a Stimulus controller #looksmutable