Comment on Can LLM be trained like training Go AI? Such as alphagoComments−Jax_Hax2yThere are models trained via RNN, but LLMs usually use Transformer architecture with a bit of human feedback on top which sort of uses reinforcement learning like AlphaGo
Comments
There are models trained via RNN, but LLMs usually use Transformer architecture with a bit of human feedback on top which sort of uses reinforcement learning like AlphaGo