INDEX

Explanations

short film or story

The neuron consistently lights up on tokens referring to short films and related film‐length descriptors (e.g. “short,” “film,” “shot,” “subject,” “story,” etc.).

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

 method

-1.30

 frankfurt

-1.28

と

-1.27

袮

-1.23

ceği

-1.23

there

-1.22

-1.20

logías

-1.20

 maquina

-1.19

ۉ

-1.19

POSITIVE LOGITS

1.48

1.29

 überhaupt

1.27

れました

1.27

Ⓦ

1.24

Iss

1.24

 that

1.24

balar

1.23

 those

1.23

簠

1.23

Activations Density 0.029%