INDEX

Explanations

used to verbs

The neuron detects strong “action” verbs that describe producing, generating, or creating something in technical or procedural contexts.

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

its

-1.52

的確

-1.48

しくは

-1.27

-1.25

ないし

-1.25

たって

-1.23

 entsprechenden

-1.23

が必要です

-1.22

 üblichen

-1.22

 notwendigen

-1.21

POSITIVE LOGITS

by

2.05

viele

1.63



1.50

1.42

潆

1.41



1.41

 sebag

1.40

)&

1.38

䊊

1.38

drei

1.38

Activations Density 0.127%