INDEX

Explanations

high energy and related concepts

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

 hardships

-0.96

 gydy

-0.96

 sized

-0.89

nofollow

-0.88

μένων

-0.88

ėj

-0.88

 hoeveel

-0.88

اهد

-0.88

ことがあります

-0.86

 inférieure

-0.85

POSITIVE LOGITS

(>

1.63

high

1.30

 high

1.27

高

1.23

(>

1.18

High

1.09

 उच्च

1.07

 High

0.99

 HIGH

0.97

dff

0.91

Activations Density 0.140%