INDEX

Explanations

quantifiable nouns followed by 'of'

The neuron selectively responds to technical terms denoting quantitative measurements or scientific metrics (e.g. resolution, magnification, cost, density, entropy, proportion, size).

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

 misiones

-1.43

ynchronously

-1.40

)]

-1.37

 MEJOR

-1.37

.‏

-1.36

 mochilas

-1.35

weren

-1.33

luster

-1.32

 muertes

-1.31

 hace

-1.30

POSITIVE LOGITS

of

4.16

for

2.22

is

1.57

in

1.53

 could

1.41

on

1.41

can

1.38

を

1.38

may

1.34

 cannot

1.34

Activations Density 0.106%