INDEX

Explanations

numbers and scores

The neuron activates on numerical statistic tokens—especially point and score values in sports 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

its

-1.49

ܙ

-1.38

乚

-1.33

 fijne

-1.27

 uniquement

-1.27

 fijn

-1.26

 gezel

-1.25

 unieke

-1.23

他们是

-1.22

甃

-1.22

POSITIVE LOGITS

minecraftforge

1.21

'';

1.18

 provide

1.18

 動画

1.16

 assist

1.15

 Menschheit

1.13

ன்று

1.13

 Sünde

1.12

ванні

1.12

our

1.12

Activations Density 0.040%