INDEX

Explanations

practice writings

The neuron consistently lights up on technical or academic terminology—especially words you’d see in scientific papers or research‐study descriptions.

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

-1.19

 суть

-1.11

</u>

-1.11

 vitales

-1.10

]),

-1.09

});

-1.07

("#

-1.07

 After

-1.05

 because

-1.02

()};

-1.02

POSITIVE LOGITS

萝卜

1.15

亮点

1.12

 mesmos

1.10

ビック

1.08

 órgãos

1.08

运营

1.07

 there

1.07

olah

1.05

 rester

1.04

仮

1.04

Activations Density 0.136%