INDEX

Explanations

The neuron fires on terms associated with formal political and diplomatic processes (e.g. negotiations, peace process, resistance movement).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

그리고

0.38

 そして

0.37

そして

0.36

之际

0.36

 그리고

0.35

 ہے۔

0.34

 しかし

0.33

 illetve

0.32

 ہوا۔

0.32

。

0.31

POSITIVE LOGITS

 που

1.31

 whose

1.28

 التي

1.19

 that

1.16

 که

1.13

 الذي

1.09

që

1.07

ที่มี

1.06

 which

1.03

 cuya

1.02

Activations Density 4.199%