model_260384870_coca_nano.py

Model Overview

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

Architecture

• Architecture: coca
• Scale: nano
• Attention: standard
• Fusion strategy: gated fusion
• Task head: retrieval
• Activation: mish
• Normalization: scalenorm
• Initialization: trunc normal

Training

• Optimizer: novograd
• LR scheduler: polynomial

Files

• model_260384870_coca_nano.py — main artifact of this repository

License

See the license field above.