INDEX

Explanations

performance events and performers

The neuron fires strongly on tokens naming musical performance metadata—especially credit phrases like “Performed by,” “Arranged by,” “Performance,” and similar.

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

 wtedy

-1.61

for

-1.53

símbolo

-1.46

晅

-1.44

 fiquei

-1.44

 episódio

-1.40

lequin

-1.34

 クリーム

-1.34

 coincident

-1.34

':[

-1.32

POSITIVE LOGITS

 asma

1.54

ALTH

1.42

LEAR

1.42

 alſo

1.33

😊

1.32

灕

1.31

骜

1.30

 enceinte

1.29

 hilly

1.27

 mitu

1.27

Activations Density 0.039%