model_244504518_albef_large.py

Model Overview

A large-scale implementation of the albef architecture, built for multitask tasks.

Architecture

• Architecture: albef
• Scale: large
• Attention: multi query
• Fusion strategy: cross attention
• Task head: multitask
• Activation: gelu tanh
• Normalization: layernorm
• Initialization: trunc normal

Training

• Optimizer: adamw
• LR scheduler: onecycle

Files

• model_244504518_albef_large.py — main artifact of this repository

License

See the license field above.