model_148450886_flamingo_giant.py

Model Overview

A giant-scale implementation of the flamingo architecture, built for contrastive tasks.

Architecture

• Architecture: flamingo
• Scale: giant
• Attention: dilated
• Fusion strategy: bilinear
• Task head: contrastive
• Activation: gelu tanh
• Normalization: layernorm
• Initialization: kaiming

Training

• Optimizer: lamb
• LR scheduler: exponential

Files

• model_148450886_flamingo_giant.py — main artifact of this repository

License

See the license field above.