INDEX

Explanations

be + past participle/adjective

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

aeda

-0.11

ylon

-0.10

gger

-0.09

ÑģÑı

-0.09

 Ting

-0.09

 concaten

-0.09

ardu

-0.09

 segreg

-0.09

 interpol

-0.09

ilyn

-0.08

POSITIVE LOGITS

 subjected

0.14

 further

0.12

 kept

0.12

 broken

0.11

 left

0.10

 sent

0.10

 made

0.10

 brought

0.10

 nurs

0.09

Activations Density 0.093%