INDEX

Explanations

Senate, Joined, joined

This neuron is highly sensitive to numeric tokens (e.g. dates, years, and other standalone 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

 ucran

-0.82

谙

-0.78

/#{

-0.76

 choosing

-0.74

getCmp

-0.73

 Whatever

-0.72

 perhaps

-0.72

 foncé

-0.69

 leaving

-0.69

ět

-0.69

POSITIVE LOGITS

###

1.16

 Joined

0.95

Joined

0.92

 Senator

0.91

joined

0.90

 joined

0.90

 STATEMENTS

0.89

 Grüßen

0.89

 Disponible

0.88

ois

0.88

Activations Density 0.032%