INDEX

Explanations

Wildflower

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

<|begin_of_text|>

-0.14

ï»¿#

-0.11

 ë¸

-0.11

ç½ĳåĪĬ

-0.11

ÐĲÑĢÑħÑĸÐ²

-0.10

 Erotische

-0.10

«ĺ

-0.10

çĦ¡ãģĹãģ

-0.09

ãģĹãģªãģĦ

-0.09

 nakne

-0.09

POSITIVE LOGITS

 Korea

0.11

 Korean

0.10

rig

0.10

contents

0.09

æľ

0.09

kim

0.09

Contents

0.09

PD

0.09

onde

0.08

Activations Density 0.007%

No Known Activations