Concept
Linear probe
A linear probe is a diagnostic method for measuring how much useful information a frozen feature representation already contains. You take a pre-trained model, freeze all its weights, run your labeled data through it to extract feature vectors, and train a single linear…
The rest of “Linear probe” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→