INDEX

Explanations

Saab 9-5

The neuron strongly activates on numeric tokens or digit sequences (e.g. model numbers, years, quantities).

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

اخ

-0.79

neling

-0.78

Nai

-0.75

ùi

-0.73

 Trafalgar

-0.73

})$}

-0.71

лайд

-0.70

بق

-0.69

 pietra

-0.69

 Kindergarten

-0.68

POSITIVE LOGITS

 obowią

0.81

 saab

0.79

のかもしれません

0.76

 बो

0.74

 スーパー

0.73

Tid

0.72

Tid

0.71

 Saab

0.69

лючение

0.69

 Pradhan

0.69

Activations Density 0.047%