INDEX

Explanations

technical code snippets

The neuron fires on capitalized technical or domain‐specific proper names—e.g. library or API identifiers, journal abbreviations, and other named entities.

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

and

-1.91

 here

-1.76

by

-1.63

-1.49

but

-1.48

-1.41

The

-1.38

}$

-1.37

on

-1.34

in

-1.31

POSITIVE LOGITS

 hamil

1.49

 directement

1.46

 dök

1.41

June

1.41

 diza

1.40

ientras

1.40

 kalori

1.35

 alcune

1.34

 فره

1.34

 diny

1.34

Activations Density 0.234%