← Questions
Pack fine-tuning samples into fixed-length sequences
0
I was asked this
No reports yet. Sign in to say where you were asked it.
For supervised fine-tuning, pack many tokenized samples of different lengths into sequences of a fixed maximum length so little space is wasted. Return the packed sequences with the boundaries and attention or position information needed so samples do not attend to one another. Discuss the packing heuristic and samples longer than the limit. Only the title of this report was public.
No official answer yet. Write yours on the right and run it.
Nothing published yet.
Sign in to submit your code; every attempt is kept here.
Write and test your solution herePython in your browser, the question's tests level by level, every submission kept for you.Sign in to start coding