INDEX

Explanations

pirate or informal greetings

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

Gee

-0.11

 GIVEN

-0.09

berman

-0.09

!=(

-0.09

gee

-0.09

aha

-0.08

wee

-0.08

ocop

-0.08

arend

-0.08

boom

-0.08

POSITIVE LOGITS

dat

0.12

uv

0.11

 Accent

0.11

listen

0.11

Dat

0.10

mae

0.10

 casual

0.10

 listen

0.10

dat

0.10

urr

0.10

Activations Density 0.100%