metadata
license: mit
tags:
- adam
- generation
- layernorm
- linear-warmup
- multi-query
- poolformer
- relu
- tiny
- tucker
- xavier-uniform
model_416625896_poolformer_tiny.py
Model Overview
A tiny-scale implementation of the poolformer architecture, built for generation tasks.
Architecture
- Architecture: poolformer
- Scale: tiny
- Attention: multi query
- Fusion strategy: tucker
- Task head: generation
- Activation: relu
- Normalization: layernorm
- Initialization: xavier uniform
Training
- Optimizer: adam
- LR scheduler: linear warmup
Files
model_416625896_poolformer_tiny.py— main artifact of this repository
License
See the license field above.