INDEX

Explanations

quick development and control

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

Neh

-0.10

 Simone

-0.09

quirer

-0.09

~-

-0.09

eger

-0.09

ĳľ

-0.09

PM

-0.09

pu

-0.09

HEX

-0.09

POSITIVE LOGITS

 midd

0.12

 bishop

0.12

 Ston

0.11

awns

0.11

 Exchange

0.11

ro

0.11

 Engines

0.10

sac

0.10

 Fritz

0.10

 pawn

0.10

Activations Density 0.010%