INDEX

Explanations

liner before, graphite drawings, Quantum Zeno

The neuron strongly responds to cosmetic and makeup–related terminology, especially product names and application instructions (e.g., liners, lashes, eyeshadows, brushes, glue).

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

 السياس

-2.17

 الحكومة

-1.71

民众

-1.58

 zamiast

-1.54

música

-1.45

 vinaigre

-1.40

之事

-1.37

 organizaciones

-1.36

 widać

-1.36

 costs

-1.33

POSITIVE LOGITS

 प्रोड

1.47

you

1.38

 Prefeitura

1.35

 سینم

1.32

 ویتامین

1.30

 consom

1.29

loài

1.28

 الشاشة

1.21

 cortado

1.21

 黑色

1.20

Activations Density 0.026%