aLSTM(depth-2)+RecurrentPolicy (WT2)
University of ManchesterAlan Turing InstituteLanguage modeling
Developed by University of Manchester,Alan Turing Institute in 2018, aLSTM(depth-2)+RecurrentPolicy (WT2) is a language modeling model with 32000000.0 parameters.
About aLSTM(depth-2)+RecurrentPolicy (WT2)
Standard neural network architectures are non-linear only by virtue of a simple element-wise activation function, making them both brittle and excessively large. In this paper, we consider methods for making the feed-forward layer more flexible while
Details
- Provider
- University of Manchester,Alan Turing Institute
- Task
- Language modeling
- Parameters
- 32000000.0
- Released
- 2018-05-22
- Open weights
- No