INDEX

Explanations

expressions and looks

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

Not in Any Lists

No Comments

Negative Logits

ï½²

-0.09

ANGLES

-0.08

è±¡

-0.08

 ÙĤØ¯Ùħ

-0.08

jab

-0.08

ceptar

-0.08

.scalablytyped

-0.08

å¥Ĺ

-0.08

ÎµÏģÏĮ

-0.08

POSITIVE LOGITS

 expression

0.54

 expressions

0.46

 look

0.45

expression

0.42

 Expression

0.42

Expression

0.38

è¡¨æĥħ

0.35

 looks

0.35

look

0.35

-expression

0.34

Activations Density 0.086%