General AI

Models, training and inference

Reference only. Project progress and credentials come from assessed work.

A model contains learned numerical patterns. Training adjusts those patterns using data and an objective. Inference runs the trained model to produce a result.

For language models

  • Text is divided into tokens, which may be words, word pieces or punctuation.
  • The context window is the amount of tokenised information available for one generation.
  • A large language model predicts a likely next token repeatedly.
  • Parameters are learned values inside the model, not stored company documents.

Three kinds of adaptation

  1. Prompt and context: change what the model receives for this request.
  2. Retrieval: find external information and add relevant parts to the context.
  3. Fine-tuning: further train a model to influence behaviour or task performance.

These approaches solve different problems. Fine-tuning is not a reliable replacement for retrieving current facts.

Models, training and inference | Mariete Academy