INDEX

Explanations

This neuron is effectively inactive on these inputs and does not reliably detect or respond to any particular pattern.

using System

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

by

-1.23

 their

-1.21

。

-1.20

 those

-1.19

そして

-1.16

-1.05

！』

-1.04

 londres

-1.02

 using

-1.00

する

-1.00

POSITIVE LOGITS

ਗ਼

1.22

 intentos

1.18

 meninos

1.16

 diarios

1.15

 cristales

1.09

躥

1.09

 paquetes

1.06

 pasos

1.05

 homens

1.03

 fanáticos

1.03

Activations Density 0.000%