Large Language Models (LLMs)

What is an LLM?

Large Language Models (LLMs) are reasoning engines trained on hundreds of billions of words, processing and analyzing data through trillions of connections (tokens). They come in two types:

• Base LLMs: Predict next word based on training data

  • Example: GPT-3 (original), LLaMA (base)
  • Primarily used as foundation models for further tuning
  • Input: "The capital of France is"
  • Output: "Paris the most populous city in France and serves as the country's major" (continues predicting text pattern)

• Instruction Tuned LLMs: Fine-tuned models that follow instructions and incorporate human feedback (RLHF)

  • Example: All current frontier models (GPT-4, Claude 3, Gemini, etc.)
  • Optimized for helpful, accurate responses
  • Input: "What's the capital of France?"
  • Output: "Paris is the capital of France." (direct answer to query)

Most modern LLMs, especially those available for public or commercial use, are instruction-tuned as they're more practical and safer for real-world applications.

Key Capabilities:

• API Integration
• Iterative Processing
• Summarizing
• Inferring
• Transforming
• Expanding

Important Notes:

• LLMs are reasoning engines, not knowledge stores
• Reliability comes through RAG (Retrieval-Augmented Generation)
• Should not be used as primary information sources
• Accuracy depends on clear, complete instructions

Effective Prompting Principles

1. Use delimiters (```)
2. Request structured outputs (CSV, JSON)
3. Include condition checks for responses
4. Implement few-shot prompting (provide examples)

AGI Development Levels

Level	DeepMind (Nov/2023)	OpenAI (Jul/2024)
Level 0	No AI	-
Level 1	Emerging (Equal to or somewhat better than an unskilled human)	Chatbots (AI with conversational language)
Level 2	Competent (At least 50th percentile of skilled adults)	Reasoners (Human-level problem solving)
Level 3*	Expert (At least 90th percentile of skilled adults)	Agents (Systems that can take actions)
Level 4	Virtuoso (At least 99th percentile of skilled adults)	Innovators (AI that can aid in invention)
Level 5	Superhuman (Outperforms 100% of humans)	Organizations (AI that can do the work of an organization)

Reference LLM: Claude 3.5 Sonnet

*via claude.ai It can generate code inline (like a jupyter notebook). (games, etc..) Feature: 3.5 sonnet + artifacts