INDEX

Explanations

causes emotions

This neuron activates on words that denote strong emotional states or reactions (e.g. happy, nervous, cry, anger).

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

you

-1.51

 that

-1.30

行きたい

-1.30

 your

-1.27

お前

-1.25

 desmotiv

-1.20

あー

-1.19

 eventuell

-1.18

 parece

-1.17

 asust

-1.16

POSITIVE LOGITS

晚

1.41

Santé

1.33

 rubles

1.33

相同

1.31

::::::::

1.30

四

1.30

izzazione

1.27

nesse

1.27

同样

1.24

 Gladiator

1.24

Activations Density 0.024%