INDEX

Explanations

prepositions introducing context

The neuron strongly activates on tokens that occur in the context of explicit numeric quantities or measurements (i.e. numbers and the words immediately around them).

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

 just

-1.16

:^{

-1.16

最近

-1.09

 also

-1.09

 bouteille

-1.07

加減

-1.05

so

-1.03

最近の

-1.02

not

-1.02

aneamente

-1.02

POSITIVE LOGITS

の方に

1.28

 памя

1.27

 هذه

1.25

を楽しむ

1.24

感じです

1.21

instorm

1.13

 Coloque

1.11

 Confira

1.11

久し

1.09

 baisse

1.09

Activations Density 0.088%