Скачать или смотреть Gen AI - Getting Started - Step 2: Under The Hood - Token Limits, Workarounds, and Planning

In this second episode, we introduce some new terminology that will help us understand what's happening under the hood.

We learn about Attention, & Token Limits, and how to work around them.

Finally, we look at using Chat GPT for planning projects interactively.

Episode Resources:
https://chat.openai.com/
https://platform.openai.com/tokenizer
https://help.openai.com/en/articles/4...
https://platform.openai.com/docs/mode...

Video Chapters:
00:00:07 - Intro and Recap
00:00:55 -Prompt Engineering
00:02:22 - LLM Charlie
00:03:36 - Conversation Limits (Forgetfulness)
00:04:16 - The Cheat Sheet Analogy
00:05:00 - Autocomplete vs. Attention
00:07:12 - Humans have Limited Attention, Too.
00:08:35 - Workarounds & The Power of Summaries
00:12:04 - What are Tokens?
00:15:52 - Other LLM Factors (Parameters)
00:21:00 - Recap on Finding Advertised LLM Details
00:21:54 - Prompt Engineering: Interactive Planning
00:25:39 - Refining a Plan Conversationally
00:29:11 - Wrap Up and GPT-Pro Teaser

Thank you for Watching!

Digital Heresy on Substack: https://www.digitalheresy.com/
Digital Heresy on Spotify: open.spotify.com/show/60g7lzt9jRCDt7sMawXITE?si=10524dcb2b13421c