INDEX

Explanations

characteristic of

The neuron activates on numeric expressions (especially decimal or measurement values) 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

 also

-2.06

its

-1.98

Its

-1.66

 what

-1.51

</strong>

-1.46

因为它

-1.43

 Erwart

-1.32

 Tiefe

-1.31

junctive

-1.31

 sadece

-1.31

POSITIVE LOGITS

 this

2.41

of

1.95

 which

1.52

1.35

ですから

1.34

 данного

1.33

(*^^*)

1.31

blurRadius

1.30

enance

1.28

了

1.27

Activations Density 0.073%