Model Mechanics
Today
Explore
Practice
Library
Exam
Certificate
Search
⌘K
ខ្មែរ
◐
Sign in
Today
Explore
Practice
Library
Profile
Weights, Training, and Inference
Today
1. Inside a Language Model · 1 of 4
Mark read
Prev
Next
Orient
Use four buckets: model, request, saved data, outside data
Trace why a new chat remembers Dara’s language
Predict
Classify three changed answers
Close
Practice
Related
Weights, Training, and Inference
Beginner
☆
📑
When an AI behaves differently, find what changed before you blame the model
Orient
Use four buckets: model, request, saved data, outside data
Trace why a new chat remembers Dara’s language
Predict
Classify three changed answers
Close
Practice
Related
Practice activities and scenarios are optional. Mark the lesson complete after reading, and return to practice whenever it helps.
Restoring your saved work…
07
Optional practice
🧠 Optional scenario practice
Q1
A chatbot remembers Dara’s language in a new conversation. What should you check first?
A
Inspect whether the application saved the preference and added it to the new request
B
Assume the user’s earlier message retrained the model
C
Ask the chatbot where the preference is stored and accept its answer
D
Change the response temperature and see whether the memory disappears
Q2
Which case belongs in the “model changed” bucket?
A
The application sends an old chat message again
B
A JSON-format instruction is added to the current request
C
A newly trained model version behaves differently on the same saved request
D
The application inserts a saved language preference
08
Up Next
Beginner
Tokens and Embeddings: Text Becomes Numbers
Intermediate
Attention and Context: What the Model Can Use
Intermediate
Next-Token Generation: From Logits to a Response
📝 Notes
Mark lesson complete
Next: Tokens and Embeddings: Text Becomes Numbers →