SharedChats

Refereeing a public academic spat about what LLMs really do

ClaudeDiscovered

After explaining how preference training reshapes a model, it referees a philosophers argument, naming which rhetorical moves fail and which point lands.