Sign inSign in
ConceptAI & ML

Supervised fine-tuning

Training stage where the model learns how to behave in a certain way using examples of the harness, tools, and ideal responses.
as Lee Robinson defines itStanford UniversityThe wording closest to all 7 definitions
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Opens up

Concepts that name this one as a prerequisite.

Lecturers

4

Each in their own terms, from the easiest level.

  • Andrej Karpathy

    Independent · Undergraduate

    Supervised fine-tuning is the stage where a base model is trained on a high-quality dataset of prompt-response pairs written by contractors to learn assistant behavior.
    • 1:52:31Let's Build GPT: From Scratch, in Code, Spelled Out
    • 11:37State of GPT: Pretraining, SFT, Reward Modeling & RLHF Pipeline
    • 3:59:56Let's Reproduce GPT-2 (124M): Distributed PyTorch, FlashAttention & FineWeb
    • 2:24:00Deep Dive into Large Language Models: Pretraining to RLHF
  • Tristan Buckmaster

    New York University · Graduate

    A training process where models imitate curated responses including reasoning traces written by other models.
    • 43:41A 'Deep Blue-Kasparov Moment for Mathematics': Euler Equations and AI
  • Nathan Lambert

    Stanford University · Graduate

    Training a model to learn task-specific capabilities, usually via autoregressive LM loss.
    • 10:14Aligning Open Language Models

Lectures that assume it

6

Connections

23

The lectures that define it or assume it, and the concepts next to it.

4 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.