INDEX

Explanations

combined followed by diverse terms

The neuron activates on occurrences of the word “combined.”

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

 berühmten

-1.69

-1.68

-1.61

комплек

-1.58

</u>

-1.57

 الأخيرة

-1.56

 octubre

-1.52

 Entonces

-1.52

 Septiembre

-1.48

for

-1.43

POSITIVE LOGITS

でしたが

1.76

 voyez

1.71

uttgart

1.70

 recevez

1.63

quelize

1.62

 trouvez

1.59

 και

1.54

きましたが

1.52

 využí

1.48

 supported

1.47

Activations Density 0.042%