INDEX

Explanations

when followed by descriptive clause

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

to

-1.46

not

-1.43

};

-1.34

ными

-1.31

 paladin

-1.30

uropa

-1.29

 rita

-1.28

it

-1.27

 hierbei

-1.26

tabilidade

-1.25

POSITIVE LOGITS

 новые

1.64

 ними

1.53

któ

1.45

 asomb

1.42

أو

1.41

ślę

1.39

 них

1.38

đầm

1.38

他们

1.37

 ihren

1.34

Activations Density 0.004%

No Known Activations