INDEX

Explanations

extent to which

The neuron strongly activates on common English function words—especially “to,” “with,” and “which.”

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

半个

-1.77

⡞

-1.72

 întâ

-1.71

 apaixon

-1.70

 músculos

-1.70

⛦

-1.67

 disfraces

-1.66

 felizes

-1.66

 XXII

-1.65

描く

-1.63

POSITIVE LOGITS

of

2.20

 they

1.91

1.77

这些

1.73

 that

1.73

 which

1.65

:"#

1.62

1.61

how

1.59

ING

1.57

Activations Density 0.007%