Skip to content
The Chat Archive
Human-written demonstration (OpenAssistant volunteers)· OpenAssistant (human-written) OpenAssistant Conversations (oasst2)

what is the difference between under-fitting and over-fitting

In OpenAssistant data, “assistant” turns were written by human volunteers as demonstrations. They are not model outputs.
  1. User

    what is the difference between under-fitting and over-fitting

  2. Assistant

    Assuming you are referring to data science, overfitting is the opposite of underfitting, occurring when the model has been overtrained or when it contains too much complexity, resulting in high error rates on test data. Overfitting a model is more common than underfitting one, and underfitting typically occurs in an effort to avoid overfitting through a process called “early stopping.”

  3. User

    How can you tell when a model is overfit?

  4. Assistant

    You would be able to tell when a model is overfit when it is has 'memorized' the training data. A symptom of an overfit model is if it performs great on the training data but cannot perform well on data that it has never seen before.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.