INDEX

Explanations

про used with verbs or in Russian requests

This neuron responds to the three‐letter sequence “pro” at the start of words.

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

 porque

-2.64

媄

-2.56

-2.52

ſelf

-2.48

of

-2.34

 BECAUSE

-2.30

 they

-2.30

を見つけ

-2.28

 berühmten

-2.27

POSITIVE LOGITS

}$

2.78

買

2.47

 nemesis

2.45

倌

2.44

 بهداشتی

2.42

’

2.41

 ($('#

2.39

鬭

2.39

/////////

2.36

 зді

2.36

Activations Density 0.001%