Concept · Building with AI

Training

Also called: model training, trained, training data, pre-training, pretraining

Teaching an AI model by showing it a very large number of examples before anyone uses it. The first, largest round is called pre-training. Fine-tuning is extra training on a smaller, narrower set of examples.

For example: An LLM is trained on a vast amount of text once. Later, a company might fine-tune it on its own support conversations so it answers in the company's style.

Connects to 27

• LLM Concept
• Inference Concept
• RAG Concept
• Gradient descent Concept
• Loss function Concept
• Neural network Concept
• RLHF Concept
• NVIDIA Organization, tool
• Model Concept
• Fine-tuning Concept
• Parameters Concept
• GPU Concept
• Knowledge cutoff Concept
• Next-token prediction Concept
• Transformer Concept
• Parametric knowledge Concept
• Sycophancy Concept
• MLOps Concept
• Deep learning Concept
• Overfitting Concept
• Distillation Concept
• MicroGPT Tool
• Python Tool
• CUDA Tool
• CS50 AI Program
• NVIDIA Inception Program
• AI addendum Concept