model_620005363_beit_xlarge.py

Model Overview

A xlarge-scale implementation of the beit architecture, built for multitask tasks.

Architecture

• Architecture: beit
• Scale: xlarge
• Attention: flash
• Fusion strategy: tensor fusion
• Task head: multitask
• Activation: gelu
• Normalization: layernorm
• Initialization: kaiming

Training

• Optimizer: sgd
• LR scheduler: constant warmup

Files

• model_620005363_beit_xlarge.py — main artifact of this repository

License

See the license field above.