metadata
license: mit
tags:
- adamw
- albef
- contrastive
- grouped-query
- kaiming
- relu
- scalenorm
- step
- tensor-fusion
- tiny
model_231544602_albef_tiny.py
Model Overview
A tiny-scale implementation of the albef architecture, built for contrastive tasks.
Architecture
- Architecture: albef
- Scale: tiny
- Attention: grouped query
- Fusion strategy: tensor fusion
- Task head: contrastive
- Activation: relu
- Normalization: scalenorm
- Initialization: kaiming
Training
- Optimizer: adamw
- LR scheduler: step
Files
model_231544602_albef_tiny.py— main artifact of this repository
License
See the license field above.