We view Large Language Models as stochastic language layers in a network, where the learnable parameters are the natural language prompts at each layer. We stack two such layers, feeding the output of one layer to the next. We call the stacked architecture a Deep Language Network - DLN
MIT License
87
stars
12
forks
source link
Allows loading templates from custom directory #19
Allows specifying a custom directory from where templates are loaded. Otherwise, users are limited to the templates specified by DLN.
It also includes 'gpt-35-turbo' to available models for azure compatibility.