model_284114063_coca_xlarge.py
Model Overview
A xlarge-scale implementation of the coca architecture, built for retrieval tasks.
Architecture
- Architecture: coca
- Scale: xlarge
- Attention: linear
- Fusion strategy: concat mlp
- Task head: retrieval
- Activation: relu
- Normalization: groupnorm
- Initialization: orthogonal
Training
- Optimizer: lion
- LR scheduler: step
Files
model_284114063_coca_xlarge.py — main artifact of this repository
License
See the license field above.
