大規模言語 model (LLM)
GPT (generative pre-trained transformer)
transformer
neuron
$ y=\varphi\left(\sum_{i=1}^m w_i x_i+b\right)
入力$ x_1,\dots,x_m
重み附け$ w_1,\dots,x_m
bias 項$ b
出力$ y
活性化函數 (activation function。傳達函數 (transfer function))
非線形函數
$ y=H\left(\sum_{i=1}^m w_i x_i-h\right)
閾値$ h
注意機構 (attention mechanism)
擴散 model (diffusion model)
semconv
OpenLLMetry
Export for AI