INDEX

Explanations

impact of or on

The neuron activates on mentions of something’s “impact,” i.e. words indicating the effect or consequence of an action or factor.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ור

1.44

ときの

1.20

коль

1.16

шње

1.16

ك

1.15

Το

1.13

 vrchol

1.11

וה

1.10

ク

1.10

ুদ্ধে

1.08

POSITIVE LOGITS

 Auswirkungen

1.23

影响

1.17

的变化

1.15

 impacto

1.12

ן

1.10

ेल

1.09

 samping

1.08

mathrm

1.05

 auront

1.02

Bä

1.00

Activations Density 0.048%