INDEX

Explanations

style of writing

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

 pronunciation

-0.10

anford

-0.10

 meaning

-0.09

ERIC

-0.09

ORB

-0.09

 Speaker

-0.09

mis

-0.08

 succinct

-0.08

SCO

-0.08

 Oriental

-0.08

POSITIVE LOGITS

 plain

0.17

 Plain

0.14

 convers

0.14

 style

0.13

 third

0.12

 neutral

0.12

plain

0.12

leg

0.11

 formal

0.11

 styles

0.11

Activations Density 0.233%