INDEX

Explanations

historical suffering caused by actions

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

 dynam

-0.11

 Brill

-0.10

HQ

-0.10

 Gree

-0.09

hq

-0.09

enha

-0.09

HQ

-0.09

 mole

-0.09

 trait

-0.09

POSITIVE LOGITS

 millions

0.14

Naz

0.14

 Holocaust

0.14

Sho

0.13

 exter

0.13

hol

0.13

Hol

0.13

Hit

0.12

gen

0.12

 Einsatz

0.11

Activations Density 0.039%