INDEX

Explanations

marketing and promotion

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

iras

-0.10

 prosec

-0.10

ENU

-0.09

orta

-0.09

atoi

-0.09

reon

-0.09

 inst

-0.09

 transf

-0.09

æ®

-0.08

 fundra

-0.08

POSITIVE LOGITS

 promotion

0.31

 promote

0.27

promotion

0.25

 Promotion

0.25

 promoting

0.25

 promotes

0.23

 awareness

0.23

 marketing

0.22

-prom

0.21

 promoted

0.20

Activations Density 0.117%