INDEX

Explanations

last release, album, or event

This neuron strongly activates on mentions of something being a “last” or “final” occurrence—e.g. a last album, last competition, or final release.

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

 commencent

-1.16

 costado

-1.09

 premiers

-1.03

 budu

-1.02

 DIFFERENT

-1.02

mittelt

-0.98

ंभ

-0.97

 первые

-0.96

 शुरु

-0.96

BEC

-0.96

POSITIVE LOGITS

 before

2.47

 after

1.51

for

1.40

 Before

1.38

 until

1.37

 sebelum

1.33

on

1.30

 remaining

1.30

 will

1.21

of

1.20

Activations Density 0.039%