INDEX

Explanations

and now this

The neuron activates on present-tense content verbs (especially ones that drive a narrative or describe actions).

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

</sup>

-1.80

Questão

-1.47

 will

-1.35

 cualquier

-1.34

യും

-1.33

of

-1.26

 between

-1.24

 also

-1.23

 then

-1.22

 età

-1.20

POSITIVE LOGITS

 orchestr

1.29

on

1.22

 Polskiej

1.20

 выглядит

1.16

えました

1.11

ञ

1.11

two

1.10

Mě

1.09

Such

1.08

 portátil

1.08

Activations Density 0.048%