INDEX

Explanations

in ascending order or Python

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

 conjunction

-0.13

noon

-0.10

uen

-0.09

 Everyday

-0.09

 regards

-0.09

owing

-0.09

-0.08

ynchronous

-0.08

daq

-0.08

_yaml

-0.08

POSITIVE LOGITS

 bulk

0.12

 reverse

0.12

 place

0.12

-place

0.12

 manner

0.11

ucc

0.11

à¸Ĺà¸°

0.11

liers

0.11

SelectionMode

0.10

 situ

0.10

Activations Density 0.120%