3 ms·
The top entry is based on a Transformer architecture. It's basically an LLM, but trained on the input instead.
by jonas21 2y ago
The top entry is based on a Transformer architecture. It's basically an LLM, but trained on the input instead.
loading thread…