Large Language Models - Topics

What are Large Language Models (LLM)? GPT vs o-series: when reasoning mode earns its cost Why ChatGPT answers the same prompt differently each time How GPT models handle instructions Few-shot examples for consistent Claude output Claude Opus for complex reasoning and multi-step problems How large is Claude's context window in practice? What fits inside Claude's context window in practice Claude Opus context limits in practice Prompting mistakes that waste Claude's long context Why Claude Sonnet is the default choice for most work Handling long documents with the Claude API context window Why agent evals are harder than model evals Post-training stages explained end to end Common mistakes when feeding ChatGPT a long document Why assistants struggle with long multi-step requests Using ChatGPT without letting it make things up When fine-tuning beats prompting Few-shot examples: how many is too many? Tool descriptions matter more than the model Chain-of-thought prompting with Claude: when it helps Why does ChatGPT sometimes make things up? What is retrieval-augmented generation in a chatbot? Why do chatbots hallucinate wrong answers? Short-term vs long-term memory in an AI agent How a large language model predicts the next token What an LLM context window really limits How Claude Mythos differs from the main Claude models Handling long context windows in the Anthropic API Anthropic API pricing: how input and output tokens are billed Cutting hallucinations in a production LLM app Why large AI models hallucinate at all