3 ms·
I assume the former is just the foundation model (which only predicts text) while the latter is instruction tuned.
by cubefox 4y ago
I assume the former is just the foundation model (which only predicts text) while the latter is instruction tuned.