easy5 pts · first answer countstransformerstraining

What knowledge distillation does

A team distils a large LLM into a small one. What does that mean?

A.The large model's weights are compressed into a zip fileB.Half of the large model's layers are deleted at randomC.The small model is trained only on the large model's tokenizerD.The small student model is trained to match the large teacher's outputs

Discussion 0

Join free to post in the discussion.

What knowledge distillation does

Discussion 0

Related questions