INDEX

Explanations

`on` followed by `address` or `list`

This neuron activates on numeric tokens (i.e. literal numbers and numeric sequences).

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

and

-1.22

on

-1.15

 with

-1.05

 without

-1.05

at

-0.96

you

-0.96

of

-0.91

 when

-0.89

by

-0.89

 after

-0.87

POSITIVE LOGITS

 banheiro

1.04

 procedimento

1.01

ventud

0.96

cedure

0.96

 Palembang

0.96

 siyang

0.95

ඍ

0.94

욘

0.93

ូ

0.93

 tejido

0.93

Activations Density 0.002%