INDEX

Explanations

has been/become/evolved/grown

The neuron strongly activates on numeric/statistical tokens (especially decimal numbers) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ﻤ

1.43

ﻧ

1.36

ﺭ

1.34

 상당히

1.34

 최근

1.30

 Органи

1.27

 दिक्कत

1.27

 ollut

1.26

 ఇటీవల

1.25

punyai

1.24

POSITIVE LOGITS

 which

1.62

if

1.52

 when

1.52

or

1.51

1.46

1.38

of

1.31

 that

1.29

 while

1.29

 from

1.26

Activations Density 0.001%