3 ms·
When OP wrote about LLMs "thinking" he implied that they have an internal conceptual self-reflecting state. Which they don't, they *are* merely next token predi
by otabdeveloper4 6mo ago
When OP wrote about LLMs "thinking" he implied that they have an internal conceptual self-reflecting state. Which they don't, they *are* merely next token predicting statistical machines.