metadata
license: cc-by-4.0
tags:
- classification
- coca
- flash
- gelu
- kaiming-normal
- layernorm
- lion
- polynomial
- tiny
- tucker
model_266791081_coca_tiny.py
Model Overview
A tiny-scale implementation of the coca architecture, built for classification tasks.
Architecture
- Architecture: coca
- Scale: tiny
- Attention: flash
- Fusion strategy: tucker
- Task head: classification
- Activation: gelu
- Normalization: layernorm
- Initialization: kaiming normal
Training
- Optimizer: lion
- LR scheduler: polynomial
Files
model_266791081_coca_tiny.py— main artifact of this repository
License
See the license field above.