model_275209502_coca_base.py

Model Overview

A base-scale implementation of the coca architecture, built for classification tasks.

Architecture

• Architecture: coca
• Scale: base
• Attention: sparse
• Fusion strategy: cross attention
• Task head: classification
• Activation: gelu
• Normalization: scalenorm
• Initialization: orthogonal

Training

• Optimizer: lion
• LR scheduler: linear warmup

Files

• model_275209502_coca_base.py — main artifact of this repository

License

See the license field above.