INDEX

Explanations

supernatural or fantastical events

The neuron is triggered by numeric tokens or number literals, activating strongly on any token containing digits.

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

картинка

-1.35

 when

-1.34

什么呢

-1.32

んだから

-1.23

phòng

-1.21

---*/

-1.20

ñales

-1.20

nước

-1.16

'),

-1.15

 město

-1.14

POSITIVE LOGITS

is

1.34

 マフラー

1.13

 gena

1.11

 molte

1.07

 nächsten

1.05

 mentale

1.04

льский

1.04

 copii

1.04

 orale

1.04

 regionale

1.02

Activations Density 0.734%