INDEX

Explanations

make hair look bouncy

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

10,000 prompts, 128 tokens each

Dataset (Dashboard)

lmsys/lmsys-chat-1m

Embeds

IFrame

Link

Negative Logits

 lighten

-0.11

Hed

-0.09

ckill

-0.09

ilk

-0.09

 Blasio

-0.09

 stiff

-0.09

deg

-0.09

exc

-0.09

iska

-0.09

 tavs

-0.09

POSITIVE LOGITS

 flattering

0.15

fl

0.15

 illusion

0.14

 shape

0.13

 slim

0.12

lines

0.11

 Shape

0.11

 lines

0.11

 optical

0.11

 illusions

0.11

Activations Density 0.066%

No Known Activations