INDEX

Explanations

implanted medical devices

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

 Transmission

-0.11

 UNESCO

-0.10

fra

-0.09

eren

-0.09

 Paula

-0.09

wer

-0.09

Jub

-0.09

aad

-0.09

 extr

-0.09

/ext

-0.09

POSITIVE LOGITS

 Class

0.11

FCC

0.11

 immunity

0.11

EMC

0.11

CEPT

0.11

icom

0.10

_exempt

0.10

 radi

0.10

fcc

0.10

inee

0.09

Activations Density 0.015%