INDEX

Explanations

from virtual

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

arde

-0.10

ups

-0.09

 Unsigned

-0.09

orp

-0.09

onus

-0.08

mai

-0.08

 forthcoming

-0.08

.bb

-0.08

UpInside

-0.08

 intermitt

-0.08

POSITIVE LOGITS

 mind

0.10

 extensive

0.10

 intensive

0.10

 emin

0.09

 fancy

0.09

styl

0.09

 extreme

0.09

TimeString

0.09

 complex

0.09

Activations Density 0.134%