INDEX

Explanations

numbers and units

The neuron is detecting standalone numeric tokens (including measurements and counts).

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

 from

-1.41

on

-1.41

to

-1.39

but

-1.16

so

-1.13

and

-1.11

XXXXXX

-1.11

If

-1.08

-1.07

 There

-1.07

POSITIVE LOGITS

↵↵

1.71

 těchto

1.41

 überprü

1.34

 části

1.28

 această

1.28

 tých

1.26

 درخواست

1.23

 krás

1.21

Groet

1.18

Đăng

1.17

Activations Density 0.025%