INDEX

Explanations

Greene and Zoe, Croatia

The neuron activates on named entities or proper nouns (e.g. personal or place names).

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

-3.78

↵↵

-3.50

-3.38

<i>

-2.98

the

-2.50

for

-2.42

</h1>

-2.41

呖

-2.41

le

-2.39

POSITIVE LOGITS

鵟

3.13

 menyambut

2.72

 dibanding

2.70

 Ganzen

2.58

Ꮈ

2.58

 selben

2.50

時間も

2.47

是我的

2.45

ейчас

2.45

 berdua

2.41

Activations Density 0.003%