INDEX

Explanations

dancer

The neuron activates on words related to dancing (e.g. “dancer,” “dancers,” “dancing”).

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

 bellissimo

-2.77

Ꮞ

-2.69

踅

-2.47

 purpoſe

-2.41

廡

-2.39

不可思议

-2.30

larda

-2.28

之为

-2.27

 Reglas

-2.20

で一番

-2.20

POSITIVE LOGITS

3.06

</td>

2.83

2.80

↵↵

2.72

</i>

2.70

</h2>

2.39

 получает

2.27

媧

2.27

主に

2.25

sibirsk

2.22

Activations Density 0.004%