Sign inSign in
ConceptAI & ML

Supervised finetuning

Training a pre-trained model on a smaller, supervised dataset to adapt it for a specific downstream task.
as Andrej Karpathy defines itIndependentThe wording closest to all 3 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.

Finetuning1Language model pretraining1Model parameters1Neural networks1Pre-training1Supervised learning1

Lecturers

3

Each in their own terms, from the easiest level.

  • Ming Ding

    Stanford University · Undergraduate

    An ordinary fine-tuning process for language models using high-quality datasets, often annotated by human experts.
    • 24:37From Large Language Models to Large Multimodal Models
  • Negar Mehr

    UC Berkeley · Research

    SFT overwrites the parameters important to each task and forgets the previous task.
    • 50:05How Can Robots Learn to Interact and Coordinate with Others?

Connections

5

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

3 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.