AI Dictionary › AI Fundamentals

Training Data

Dati di addestramento (Training Data)

Training data is the set of examples an AI model is trained on. For large language models this means enormous text corpora, books, articles, web pages, code, from which the model learns the statistical patterns of human language.

Definition

The quality and composition of training data determine what the model knows and how it behaves: a model trained mostly on English text will be weaker in Italian; a dataset containing prejudice will pass it on to the model (bias). Even the knowledge cutoff date depends on the data: the model doesn't know about events after its training.

For professional use, this has a practical implication: recent, internal or confidential information isn't in the training data and must be supplied in the prompt (context) or through techniques like RAG.

Related terms

• Machine Learning
• AI Bias
• Fine-tuning
• RAG (Retrieval Augmented Generation)
• Context (in prompting)

More in AI Fundamentals

• LLM (Large Language Model)
• Token
• Context Window
• Temperature
• AI Hallucination
• Fine-tuning
• RAG (Retrieval Augmented Generation)
• Embedding

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

📱 Download the Android app (beta) iOS coming soon

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.