model_238759927_albef_base.py

Model Overview

A base-scale implementation of the albef architecture, built for retrieval tasks.

Architecture

• Architecture: albef
• Scale: base
• Attention: flash
• Fusion strategy: co attention
• Task head: retrieval
• Activation: gelu tanh
• Normalization: layernorm
• Initialization: orthogonal

Training

• Optimizer: adam
• LR scheduler: cosine

Files

• model_238759927_albef_base.py — main artifact of this repository

License

See the license field above.