INDEX

Explanations

names starting with abbreviations

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|>

-0.10

enco

-0.10

ëį°ìĿ´íĬ¸

-0.09

¦æĥħ

-0.09

ÐĲÑĢÑħÑĸÐ²

-0.09

łéĻ¤

-0.09

'gc

-0.09

Å¡ak

-0.09

$MESS

-0.09

âĸı

-0.09

POSITIVE LOGITS

0.09

\-

0.08

 Brilliant

0.08

 Mour

0.08

 doub

0.07

oria

0.07

Ket

0.07

ained

0.07

nb

0.07

lain

0.07

Activations Density 0.105%

No Known Activations