INDEX

Explanations

want to or interested in

The neuron activates on words and phrases that express desire, interest, or an invitation (e.g., “like,” “interested,” “want,” “would you like”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 deberán

-1.05

 своих

-0.96

٢

-0.93

で購入

-0.91

 เพื่อ

-0.90

自分の

-0.88

就要

-0.86

各个

-0.86

alina

-0.85

 Verkä

-0.85

POSITIVE LOGITS

 still

1.24

we

1.09

go

1.08

 being

1.05

步伐

1.02

 μας

1.01

 considérons

0.93

Selanjutnya

0.91

бовь

0.90

lateur

0.90

Activations Density 0.014%