Transformer (Adaptive Input Embeddings) WT103
Facebook AI ResearchLanguage modelingOpen weights
Transformer (Adaptive Input Embeddings) WT103 is a language modeling model from Facebook AI Research released in 2018 with 247000000.0 parameters.
About Transformer (Adaptive Input Embeddings) WT103
We introduce adaptive input representations for neural language modeling which extend the adaptive softmax of Grave et al. (2017) to input representations of variable capacity. There are several choices on how to factorize the input and output layers
Details
- Provider
- Facebook AI Research
- Task
- Language modeling
- Parameters
- 247000000.0
- Released
- 2018-09-28
- Open weights
- Yes