INDEX

Explanations

alkaline substances and conditions

The neuron detects the word "alkaline" in scientific and technical contexts.

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

</i>

-2.78

.’

-2.27

’;

-2.25

蕷

-2.13

わけで

-2.05

鉕

-1.98

发现了

-1.97

’,

-1.95

 schwarze

-1.95

 confuso

-1.95

POSITIVE LOGITS

\&

2.59

2.42

ны

2.28

most

2.25

because

2.20

\%

2.19

what

2.17

going

2.05

2.03

两个

2.00

Activations Density 0.010%