model_262074163_coca_nano.py

Model Overview

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

Architecture

• Architecture: coca
• Scale: nano
• Attention: flash
• Fusion strategy: gated fusion
• Task head: classification
• Activation: relu
• Normalization: groupnorm
• Initialization: trunc normal

Training

• Optimizer: novograd
• LR scheduler: linear warmup

Files

• model_262074163_coca_nano.py — main artifact of this repository

License

See the license field above.