INDEX

Explanations

direction and movement

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

 downward

-0.15

 backward

-0.12

quil

-0.11

 dictionaryWith

-0.10

 downwards

-0.09

 ëĴ

-0.09

aras

-0.09

 å»

-0.09

 Verde

-0.09

 bottoms

-0.09

POSITIVE LOGITS

 toward

0.34

 towards

0.34

 hacia

0.25

åĲĳ

0.23

 Towards

0.22

Tow

0.21

Towards

0.21

owards

0.19

oward

0.18

 away

0.18

Activations Density 0.109%