model_074483456_perceiver_nano.py

Model Overview

A nano-scale implementation of the perceiver architecture, built for matching tasks.

Architecture

• Architecture: perceiver
• Scale: nano
• Attention: standard
• Fusion strategy: concat mlp
• Task head: matching
• Activation: gelu tanh
• Normalization: scalenorm
• Initialization: trunc normal

Training

• Optimizer: lamb
• LR scheduler: constant warmup

Files

• model_074483456_perceiver_nano.py — main artifact of this repository

License

See the license field above.