Instruction Tuning
Generative AI & LLMsChapter 4: Choosing the LLMTerm 75 of 253
Fine-tuning on instruction→response pairs so the model follows directions.
At Sunrise School
Gopal asksWhy does Bit follow “reply in 3 bullet points”?
Bit answersInstruction tuning: training on thousands of instruction-and-answer pairs, so I do what you ask instead of just continuing the text.
In the real worldTeaching it to answer, not just continue text.
Watch the short
Short coming soonSeason 5 · Day 16Follow @haigopala
Short coming soonSeason 5 · Day 16Follow @haigopala.telugu
Short coming soonSeason 5 · Day 16Follow @haigopala.hindi
Quick check
Quick check. Does quality usually improve with more data and compute? Yes or no?
Yes! That's the scaling law.
Taught together with
Practise it
Try this term and all 253 others in the free Model Training Arena.
Open the Arena →Follow Bit daily: YouTube English · తెలుగు · हिंदी | Instagram English · తెలుగు · हिंदी