INDEX

Explanations

explaining with \"to say\"

The neuron activates on numeric tokens—digits or numbers (including decimals) appearing in the text.

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

-1.61

-1.60

 blandt

-1.55

的

-1.47

 after

-1.41

'].'

-1.41

 ponieważ

-1.41

-1.39

还真

-1.38

POSITIVE LOGITS

 alliés

1.59

matically

1.57

ingang

1.54

 espectador

1.53

Préparation

1.52



1.52

できました

1.51

意思是

1.49

стрі

1.47

ответ

1.45

Activations Density 0.850%