AI glossary

A

Training Alignment Making AI systems pursue the goals their developers and users intend, and behave helpfully, honestly and safely.

B

Training Backpropagation The algorithm that computes how much each weight contributed to an error, so gradient descent knows how to adjust it.

C

Training Compute The processing power used to train or run AI models, usually measured in GPU hours or floating-point operations. Training Constitutional AI Anthropic's method of training models to follow a written set of principles, using AI feedback guided by those principles.

D

Training Distillation Training a smaller 'student' model to imitate a larger 'teacher' model, keeping much of its ability at a fraction of the cost.

F

Training Fine-tuning Further training of an existing model on a smaller, specific dataset to specialise it for a task, style or domain.

G

Training GPU A graphics processing unit; the parallel processor that trains and runs most AI models. Training Gradient descent The optimisation method that nudges a model's parameters, step by step, in the direction that most reduces its loss.

I

Training Instruction tuning Fine-tuning a pre-trained model on examples of instructions and good responses, so it follows requests instead of just continuing text.

L

Training LoRA (low-rank adaptation) A cheap fine-tuning technique that trains a small set of extra weights instead of updating the whole model. Training Loss function The formula that measures how wrong a model's predictions are during training; training tries to minimise it.

M

Training Model size How large a model is, usually stated in parameters (for example 8B or 120B), which roughly predicts its capability and running cost.

O

Training Overfitting When a model memorises its training examples instead of learning general patterns, so it performs worse on new data.

P

Training Pre-training The first, largest stage of training, where a model learns language and world knowledge by predicting text across a huge dataset.

Q

Training Quantization Storing a model's weights with fewer bits (for example 4 or 8 instead of 16) to cut memory use and speed it up, with a small loss in quality.

R

Training Reinforcement learning Training by trial and reward: the model tries actions or answers and is updated toward those that score well. Training RLHF (reinforcement learning from human feedback) A training method where people rank model responses and the model is optimised to produce the kind of answers they prefer.

S

Training Scaling laws Empirical findings that model performance improves predictably as you increase parameters, data and compute. Training Synthetic data Training data generated by AI models rather than collected from people or the world.

T

Training TPU Tensor processing unit; Google's custom AI accelerator, used to train and serve Gemini.