easy5 pts · first answer countstransformerstransformer-architecture

What a decoder-only model predicts

GPT, Llama and most chat LLMs are decoder-only transformers. What are they trained to do?

A.Fill in randomly masked words using both left and right contextB.Translate an input sentence produced by a separate encoderC.Classify a whole document into one of a fixed set of labelsD.Predict the next token from all the tokens before it

Discussion 0

Join free to post in the discussion.

What a decoder-only model predicts

Discussion 0

Related questions