metadata
license: cc-by-4.0
tags:
- approx-gelu
- batchnorm
- bilinear
- generation
- huge
- kaiming
- perceiver
- polynomial
- sgd
- standard
model_102597058_perceiver_huge.py
Model Overview
A huge-scale implementation of the perceiver architecture, built for generation tasks.
Architecture
- Architecture: perceiver
- Scale: huge
- Attention: standard
- Fusion strategy: bilinear
- Task head: generation
- Activation: approx gelu
- Normalization: batchnorm
- Initialization: kaiming
Training
- Optimizer: sgd
- LR scheduler: polynomial
Files
model_102597058_perceiver_huge.py— main artifact of this repository
License
See the license field above.