INDEX

Explanations

pressing buttons and keys

The neuron fires on mentions of physically triggering interface controls (e.g. “press,” “pressed,” “hit,” or clicking buttons).

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

we

-1.30

 have

-1.27

</h3>

-1.26

見えます

-1.25

tických

-1.22

FONTE

-1.20

过来了

-1.19

しかし

-1.18

is

-1.13

จาก

-1.11

POSITIVE LOGITS

 aktivi

1.22

ごめん

1.21

噐

1.20

madeus

1.19

饪

1.19

 agrega

1.18

関連記事

1.18

えっ

1.16

nein

1.15

ἄ

1.12

Activations Density 0.025%