metadata
license: apache-2.0
tags:
- cosine
- gated-fusion
- gelu
- generation
- layernorm
- linear
- mocov3
- orthogonal
- sgd
- small
model_716144702_mocov3_small.py
Model Overview
A small-scale implementation of the mocov3 architecture, built for generation tasks.
Architecture
- Architecture: mocov3
- Scale: small
- Attention: linear
- Fusion strategy: gated fusion
- Task head: generation
- Activation: gelu
- Normalization: layernorm
- Initialization: orthogonal
Training
- Optimizer: sgd
- LR scheduler: cosine
Files
model_716144702_mocov3_small.py— main artifact of this repository
License
See the license field above.