INDEX

Explanations

key features, difference, or application

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

Framework

-0.10

odal

-0.09

 GetEnumerator

-0.09

Helpers

-0.09

Globals

-0.09

ifa

-0.08

Hints

-0.08

 Bris

-0.08

icot

-0.08

_graphics

-0.08

POSITIVE LOGITS

 features

0.20

 benefits

0.18

 characteristics

0.17

 benefit

0.17

 characteristic

0.16

 Features

0.16

 feature

0.16

 goals

0.15

key

0.15

features

0.14

Activations Density 0.106%