model_608159809_beit_small.py

Model Overview

A small-scale implementation of the beit architecture, built for generation tasks.

Architecture

• Architecture: beit
• Scale: small
• Attention: dilated
• Fusion strategy: gated fusion
• Task head: generation
• Activation: gelu tanh
• Normalization: batchnorm
• Initialization: orthogonal

Training

• Optimizer: adafactor
• LR scheduler: constant warmup

Files

• model_608159809_beit_small.py — main artifact of this repository

License

See the license field above.