INDEX

Explanations

inner monologue, thoughts, core, html

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

-0.10

ee

-0.10

 Bale

-0.09

spl

-0.09

ism

-0.09

acen

-0.09

-minded

-0.09

RS

-0.09

ril

-0.09

reen

-0.09

POSITIVE LOGITS

most

0.30

halb

0.23

-inner

0.20

(inner

0.18

lijke

0.17

wear

0.16

 inner

0.16

-city

0.15

 Inner

0.15

.Inner

0.15

Activations Density 0.019%