INDEX

Explanations

English language use

The neuron strongly activates on mentions of the word “English.”

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

}}}=

1.14

larda

1.02

 действие

0.96

્સ

0.93

 ука

0.92

्स

0.91

])),

0.90

}}}{\

0.90

}"]

0.89

اذ

0.89

POSITIVE LOGITS

 TOEFL

1.31

é

1.30

 lingua

1.27

speaking

1.25

ESL

1.20

noi

1.17

 fries

1.16

т

1.16

ന്‍

1.15

 idioma

1.15

Activations Density 0.013%