INDEX

Explanations

uses screws and drivers

The neuron fires on mentions of screw‐related tools and hardware (e.g. “screwdriver,” “driver,” “screws,” etc.).

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

 berge

-1.45

裢

-1.30

 виправи

-1.30

 militer

-1.27

 veren

-1.27

нена

-1.27

)()

-1.25

》

-1.25

 ossia

-1.24

<td>

-1.23

POSITIVE LOGITS

↵↵

1.59

1.20

afe

1.15

頃から

1.13

 also

1.11

Letting

1.11

 наш

1.09

鎏

1.08

 possibly

1.06

 położ

1.05

Activations Density 0.028%