INDEX

Explanations

numbers and ids

This neuron picks out hexadecimal characters and digits within long identifier or hash-like strings.

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

 those

-1.27

 several

-1.09

 many

-1.08

 most

-1.04

or

-1.02

 both

-1.02

 some

-1.00

 with

-0.98

 very

-0.98

 relatively

-0.94

POSITIVE LOGITS

please

1.04

isei

1.01

inside

0.96

eniably

0.91

pg

0.88

ENV

0.88

requested

0.87

 contenedores

0.85

across

0.85

 SEVEN

0.85

Activations Density 0.001%