model_323037297_tiny_transformer_xlarge.py

Model Overview

A xlarge-scale implementation of the tiny transformer architecture, built for classification tasks.

Architecture

• Architecture: tiny transformer
• Scale: xlarge
• Attention: sliding window
• Fusion strategy: gated fusion
• Task head: classification
• Activation: gelu
• Normalization: groupnorm
• Initialization: xavier uniform

Training

• Optimizer: adam
• LR scheduler: step

Files

• model_323037297_tiny_transformer_xlarge.py — main artifact of this repository

License

See the license field above.