Author here: Sorry for the confusing formatting on the task descriptions at the end of the paper. That "4" is the human-generated target completion, not a model generated completion. I'm not sure whether the model got that particular question correct, but from Table 3.7 that GPT-3 has 36.5% accuracy on DROP in the few-shot setting.
Many other readers were confused by this so we'll update the formatting to say "target completion" to make this more clear.
Many other readers were confused by this so we'll update the formatting to say "target completion" to make this more clear.