INDEX

Explanations

numbers and programming code

The neuron activates on numeric values—integers or decimals—wherever they appear in the text.

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

ashqai

-1.61

ApJ

-1.39

-1.38

to

-1.36

 through

-1.31

tappe

-1.30

';

-1.29

zembro

-1.29

ViewChild

-1.27

 with

-1.27

POSITIVE LOGITS

 your

1.99

 that

1.70

1.43

爱好

1.41

that

1.32

 unless

1.23

 тези

1.19

 renovables

1.19

дає

1.16

頂きました

1.16

Activations Density 0.063%