model_740103356_mocov3_giant.py

Model Overview

A giant-scale implementation of the mocov3 architecture, built for generation tasks.

Architecture

• Architecture: mocov3
• Scale: giant
• Attention: sparse
• Fusion strategy: bilinear
• Task head: generation
• Activation: gelu tanh
• Normalization: groupnorm
• Initialization: xavier uniform

Training

• Optimizer: sgd
• LR scheduler: onecycle

Files

• model_740103356_mocov3_giant.py — main artifact of this repository

License

See the license field above.