INDEX

Explanations

determining properties and characteristics

The neuron strongly responds to academic/scientific nominalizations that denote quantitative methods or measurements (e.g. reconstruction, estimation, determination, configuration).

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

 выбран

-1.59

 مثل

-1.38

ごちそう

-1.35

 faisant

-1.34

you

-1.28

 neumáticos

-1.26

éditeur

-1.25

 venant

-1.25

 conocimientos

-1.24

↵↵↵

-1.23

POSITIVE LOGITS

of

1.92

 with

1.87

 through

1.76

1.71

 when

1.66

for

1.60

 whose

1.55

 without

1.55

and

1.52

 more

1.49

Activations Density 0.324%