metadata
license: cc-by-4.0
tags:
- adam
- batchnorm
- classification
- coca
- flash
- gelu
- low-rank
- step
- tiny
- trunc-normal
model_267007225_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: low rank
- Task head: classification
- Activation: gelu
- Normalization: batchnorm
- Initialization: trunc normal
Training
- Optimizer: adam
- LR scheduler: step
Files
model_267007225_coca_tiny.py— main artifact of this repository
License
See the license field above.