INDEX

Explanations

forms of "is" or "be"

This neuron fires on the sentence’s principal content word (the main noun or verb that carries the core meaning).

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

 determinadas

-1.30

 каждой

-1.24

 внутренних

-1.23

 этих

-1.17

这一点

-1.16

 determinados

-1.16

 внутренней

-1.14

 двох

-1.12

齣

-1.11

dibujo

-1.09

POSITIVE LOGITS

the

6.50

THE

2.02

our

1.69

The

1.53

才是

1.41

the

1.30

 един

1.25



1.20

The

1.16

 gebruikt

1.15

Activations Density 0.259%