INDEX

Explanations

от чего-то

The neuron fires on the Russian preposition “от” (and its subword fragments) whenever it appears in the text.

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

 desses

-2.64

がたくさん

-2.52

 kostet

-2.36

啖

-2.25

不时

-2.25

 dieser

-2.23

 jabón

-2.20

 возможность

-2.17

 destes

-2.14

 brancas

-2.13

POSITIVE LOGITS

The

3.88

怱

2.45

噚

2.36

2.30

 gemeinsame

2.27

 Vordergrund

2.27

ཋ

2.25

 gewähr

2.22

Einen

2.20

And

2.19

Activations Density 0.001%