INDEX

Explanations

introduces questions or statements

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

 corre

-0.10

Tib

-0.08

peq

-0.08

 Spicer

-0.08

istrovstvÃŃ

-0.07

oa

-0.07

Eu

-0.07

icks

-0.07

inton

-0.07

 bureaucr

-0.07

POSITIVE LOGITS

 Below

0.10

 specifically

0.08

rech

0.08

asha

0.08

 lear

0.08

 Specifically

0.08

 below

0.08

0.07

Below

0.07

ehler

0.07

Activations Density 0.019%