Large Language Model (LLM)
A neural network with billions of parameters trained on massive text datasets to understand and generate human language, forming the foundation of modern AI assistants.
How They Work
LLMs are typically decoder-only transformers trained via next-token prediction on trillions of tokens. This simple objective, at sufficient scale, produces models that can reason, code, translate, and perform creative tasks.
Key Properties
Scaling produces emergent capabilities (abilities that appear at certain scales). In-context learning (learning from examples in the prompt). Instruction following (through alignment training). The field is rapidly evolving with new techniques, architectures, and capabilities.