model_145100976_flamingo_giant.py

Model Overview

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

Architecture

• Architecture: flamingo
• Scale: giant
• Attention: linear
• Fusion strategy: low rank
• Task head: matching
• Activation: gelu tanh
• Normalization: layernorm
• Initialization: xavier uniform

Training

• Optimizer: novograd
• LR scheduler: onecycle

Files

• model_145100976_flamingo_giant.py — main artifact of this repository

License

See the license field above.