INDEX

Explanations

one of several

The neuron detects “one of the…” constructions that quantify someone or something as part of a group (e.g. “one of the first,” “one of the most,” “one of the leading”).

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

 each

-1.41

its

-1.40

it

-1.37

 itself

-1.34

as

-1.18

 their

-1.07

 during

-1.05

by

-1.04

 there

-1.04

 which

-1.03

POSITIVE LOGITS

 several

1.84

 flera

1.71

several

1.61

 kilku

1.55

few

1.48

 tylko

1.47

 only

1.42

 three

1.41

only

1.40

 několika

1.40

Activations Density 0.052%