INDEX

Explanations

future states and causes

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

Negative Logits

 revolution

-0.11

 Revolution

-0.09

bab

-0.09

Ã´t

-0.09

vern

-0.09

-0.08

 revolutions

-0.08

 selfish

-0.08

 Secret

-0.08

 Ying

-0.08

POSITIVE LOGITS

 events

0.21

 curve

0.20

 Murphy

0.18

 things

0.18

events

0.18

-events

0.17

 plans

0.17

 circumstances

0.16

curve

0.16

 Events

0.16

Activations Density 0.085%

No Known Activations