Resources
Model History and Background
- GPT-2 Release Blog Post - OpenAI’s original GPT-2 announcement
- GPT-3 Paper - “Language Models are Few-Shot Learners”
- InstructGPT Paper - “Training language models to follow instructions with human feedback”
- Meta’s Llama Models - Official Llama model family page
Chat Templates and Model Formats
- Hugging Face Chat Templates Guide - Comprehensive guide to chat templates
OpenAI API
- OpenAI Platform - Create an account and get API keys
- OpenAI API Documentation - Official API reference
- Chat Completions Guide - How to use the chat completions endpoint
- Structured Outputs Guide - Guide to structured outputs
OpenRouter
- OpenRouter Home Page - Unified API for hundreds of models
- OpenRouter Documentation - API docs and model listings
- OpenRouter Models - Browse available models and pricing
Server-Sent Events (SSE) and Streaming
- MDN: Server-Sent Events - Technical overview of SSE
- EventSource API - Browser API for SSE
- OpenAI Streaming Guide - How to implement streaming with OpenAI API
Gradio
- Gradio Home Page
- Gradio Documentation - Official documentation
- Gradio ChatInterface - Chat interface component documentation
- Gradio Guides - Tutorials and examples
- Hugging Face Spaces - Platform for deploying Gradio apps
RLHF and Model Training
- RLHF Explainer - Hugging Face’s comprehensive guide to RLHF
- Reinforcement Learning from Human Feedback Paper - Original RLHF research
Context Windows and Token Management
- Understanding Context Windows - Anthropic’s post on long context
- Token Counting Best Practices - OpenAI cookbook example
- Tiktoken - Token counting library for estimating costs