AI Glossary — Simple Definitions for AI Terms
Embeddings
Numeric representations of words or concepts used to measure semantic similarity.
Parameters
The learnable weights in a model that define its performance and complexity.
Transformers
A deep learning architecture using self-attention to process sequences in parallel, enabling models like GPT and BERT.
LLM (Large Language Model)
A type of AI trained on vast text data to understand and generate human-like language.
Diffusion Models
A new type of generative model used in image synthesis, based on the process of gradually adding and removing noise.
Context Window
The amount of input a model can consider at once. Larger windows mean longer conversations or documents.
Tokens
Pieces of words (subwords) used in model input/output. Costs and limits are often measured in tokens.