INDEX

Explanations

project and projects

This neuron specifically detects mentions of “project” (and its plural form) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 řekl

-1.63

῞

-1.55

)=\

-1.45

 jurí

-1.45

isn

-1.41

mô

-1.40

}$,

-1.39

 lakini

-1.38

 aumentando

-1.38

)$,

-1.38

POSITIVE LOGITS

Also

1.73

cenzo

1.71

PAIGN

1.66

剛剛

1.64

もちろん

1.61

We

1.59

待って

1.58

requent

1.56

Currently

1.55

プロジェクト

1.55

Activations Density 0.033%