INDEX

Explanations

codes and names

The neuron primarily responds to numeric tokens (digits and numbers).

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

 baratas

-1.18

 botellas

-1.11

rası

-1.09

魨

-1.05

கம்

-1.03

 Trouvez

-1.02

 médaille

-1.02

 bisous

-1.02

 when

-1.00

 retratos

-1.00

POSITIVE LOGITS

1.21

1.12

TODO

1.07

)})

1.05

 poteva

1.04

1.03

1.01

ains

0.99

 nachfol

0.99

 ممکن

0.98

Activations Density 0.082%