model_229241357_albef_tiny.py

Model Overview

A tiny-scale implementation of the albef architecture, built for generation tasks.

Architecture

• Architecture: albef
• Scale: tiny
• Attention: sparse
• Fusion strategy: tensor fusion
• Task head: generation
• Activation: gelu tanh
• Normalization: scalenorm
• Initialization: kaiming normal

Training

• Optimizer: lamb
• LR scheduler: step

Files

• model_229241357_albef_tiny.py — main artifact of this repository

License

See the license field above.