INDEX

Explanations

Real Madrid, encoder

The neuron consistently lights up on atomic named‐entities or highly domain‐specific terms (e.g. “Real” in Real Madrid, “encoder” in autoencoder, “SP2” in the cell line, “immunore–” in immunoreactivity, etc.).

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

 Then

-2.03

 such

-1.58

 three

-1.45

 sogenannten

-1.40

 Actually

-1.39

 такъ

-1.38

 tienen

-1.37

 Apparently

-1.37

 zwölf

-1.35

POSITIVE LOGITS

}$.

1.73

蹀

1.69



1.63

}}$.

1.63

$.

1.52

袮

1.52

 mentale

1.51

ㄩ

1.49

 }}"></

1.46

۔

1.45

Activations Density 0.140%