4
Vectors and embeddings
- Token ID
- One-hot vector
- Learned embedding
Distinguish IDs, one-hot vectors, and learned embeddings.
Sign up and we'll remember how far you've come.
Look inside language models, then build with Python
Start here once you have finished beginner steps 1–3.
See beginner steps 1–3 firstWhere are you in the 9 steps?
- Beginner Use AI and check its answers
- Intermediate Look inside a language model
- Advanced Build and evaluate with Python
- Level 1 AI, machine learning, and language models
- Level 2 Write and check a prompt
- Level 3 Text, tokens, and IDs
- Level 4 Vectors and embeddings
- Level 5 Training, inference, and retrieval
- Level 6 Instructions, sampling, and verification
- Level 7 Python: run, inspect, change
- Level 8 Numbers, predictions, and activations
- Level 9 Train and evaluate a small model
Intermediate
Look inside a language model
Follow the examples at your own pace-
- 5
Training, inference, and retrieval
- Training
- Weights
- Inference
Explain what changes in training and what a model uses at inference time.
- 6
Instructions, sampling, and verification
- Instructions
- Sampling
- Verification
Compare generation settings without confusing consistency with factual accuracy.
Advanced
Build and evaluate with Python
Small steps, real code- 7
Python: run, inspect, change
- Variables
- Conditions
- Functions
Use variables, conditions, and functions in the browser.
- 8
Numbers, predictions, and activations
- Input numbers
- Activation function
- Prediction
Build small functions and explain what their outputs mean.
- 9
Train and evaluate a small model
- Split the data
- Train
- Evaluate
Separate training from evaluation and explain the limits of a score.
Where each step gets hands-on
| Step | AI Lab | Build Your Own | Code lab |
|---|---|---|---|
| 4. Vectors and embeddings | Yes | No | No |
| 5. Training, inference, and retrieval | No | Yes | No |
| 6. Instructions, sampling, and verification | No | Yes | No |
| 7. Python: run, inspect, change | No | No | Yes |
| 8. Numbers, predictions, and activations | No | No | Yes |
| 9. Train and evaluate a small model | No | No | Yes |