INDEX

Explanations

receiving attention or study

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

などが

-1.55

などは

-1.29

präche

-1.17

游戏介绍

-1.16

などを

-1.16

aneity

-1.12

for

-1.09

游戏特色

-1.09

earing

-1.07

 lamang

-1.04

POSITIVE LOGITS

its

1.86

 moldura

1.20

ようです

1.20

 attention

1.17

Én

1.14

Being

1.13

Another

1.13

 peculiar

1.05

Its

1.05

 careful

1.05

Activations Density 0.068%