INDEX

Explanations

pronoun relations

The neuron fires on formal, abstract nouns or generalizing technical terms (e.g. “such,” “devices,” “state,” “type,” “direction,” “mortality,” “disease”) typical of patent- or academic-style descriptions.

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

 marchio

-1.40

 camadas

-1.32

 paziente

-1.30

 Reſ

-1.23

 wię

-1.21

łóż

-1.20

 campionato

-1.19

 molta

-1.18

に乗って

-1.18

Either

-1.17

POSITIVE LOGITS

its

1.57

</strong>

1.33

are

1.20

</b>

1.17

her

1.12

 standen

1.10

„

1.10

 тоже

1.07

<i>

1.05

“

1.05

Activations Density 0.229%