1 paper · 1 filter
Amal Alnouri, Andreas Hinterreiter, Christina Humer +2
Large language model (LLM) outputs arise from complex interactions among prompts, system instructions, model parameters, and architecture. We refer to specific configurations of th…