DecisionTransformerInterpretability icon indicating copy to clipboard operation
DecisionTransformerInterpretability copied to clipboard

Add static interpretability visualizations to wandb dashboard.

Open jbloomAus opened this issue 1 year ago • 0 comments

Add static interpretability visualizations to wandb dashboard. Seems like a cool idea I just had.

Some stuff: QK/OV circuit viz, different attribution embeddings, time embedding, L2 Norms of different components.

jbloomAus avatar May 03 '23 03:05 jbloomAus