INDEX

Explanations

professor of

The neuron is essentially detecting standalone numeric tokens (i.e. numbers) 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

↵

-1.64

it

-1.45

 myriad

-1.44

swire

-1.41

iamo

-1.38

─

-1.35

 любых

-1.35

tocks

-1.33

 (!__

-1.32

－

-1.32

POSITIVE LOGITS

who

3.47

 which

2.23

2.22

 from

2.19

 whose

2.11

 whom

1.90

of

1.77

one

1.63

an

1.60

 který

1.59

Activations Density 0.107%