model_208611746_clip_large.py

Model Overview

A large-scale implementation of the clip architecture, built for contrastive tasks.

Architecture

• Architecture: clip
• Scale: large
• Attention: dilated
• Fusion strategy: concat mlp
• Task head: contrastive
• Activation: swish
• Normalization: rmsnorm
• Initialization: trunc normal

Training

• Optimizer: novograd
• LR scheduler: cosine

Files

• model_208611746_clip_large.py — main artifact of this repository

License

See the license field above.