factual
In-context learning occurs with frozen weights
In-context learning happens after training ends and the weights are frozen, because the activations change as the model processes the prompt, so learning-like computation is not confined to the training phase.
factualpending
Speaker
Chandra SripadaEvidence Quote
“Others have likened this to more like gradient descent happening within the model. I want to emphasize, this is happening after the weights are frozen.”
Created: 11/17/2026, 6:53:48 AM
Your notes
Loading notes...