INDEX

Explanations

The neuron strongly responds to occurrences of the phrase “do it” (and its inflected forms like “did it,” “doing it,” etc.).

do it

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

 bestehenden

-1.23

selbe

-1.20

 zawsze

-1.17

größe

-1.13

 soggiorno

-1.12

 kılıf

-1.09

能夠

-1.08

度に

-1.08

ὲν

-1.07

對於

-1.06

POSITIVE LOGITS

 with

1.43

by

1.40

س

1.23

niej

1.19

all

1.19

so

1.17

е

1.16

 without

1.15

ㄹ

1.11

 hari

1.08

Activations Density 0.031%