INDEX

Explanations

past participle endings

This neuron activates on Russian words formed with the “-нн-” participle/adjective ending (e.g. past participles or adjectives built with the ‑нн- suffix).

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

-1.33

are

-1.25

 perché

-1.24

 porque

-1.24

 flera

-1.20

-1.16

-1.13

 bardak

-1.13

 следу

-1.09

击败

-1.09

POSITIVE LOGITS

by

1.65

 from

1.36

 throughout

1.31

in

1.30

到这里

1.20

我们

1.13

我們

1.10

 during

1.06

此时

1.02

at

1.01

Activations Density 0.011%