learnGPT - team 2

week 1

test_2.py? (HeadVer1)

taking the past into account - why?

• Compare: GPTVer1-generated completion vs GPTVer2-generated completion
• What difference do you notice?
• Why is there a difference?

test_3.py? (HeadVer2)

vectorizing for loops - why?

• How is HeadVer2 logically the same as HeadVer1?
• Why is HeadVer2 faster than HeadVer1?

test_4.py? (HeadVer3)

taking the past into account with masking & normalization - how?

• How is HeadVer3 logically the same as HeadVer1?
• Why mask wei with -inf? Why not 0?

week 2

...

Contributors