INDEX

Explanations

shoe size fitting

The neuron activates on numeric tokens—particularly measurements and other numbers (e.g. “170,” “116.5,” “99,” 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

 Mejora

-1.32

 Евгений

-1.21

 amélior

-1.15

 réjou

-1.12

 mérit

-1.11

его

-1.10

 reconoci

-1.10

 étend

-1.10

 rédu

-1.10

 Puedes

-1.10

POSITIVE LOGITS

 this

1.85

 этой

1.20

 dieser

1.07

 umani

1.06

toxicity

1.05

 вклад

1.05

[{

1.03

 questa

1.03

 questo

1.02

 piccoli

1.01

Activations Density 0.018%