INDEX

Explanations

don't have a

The neuron fires on numeric tokens (integers or decimals) 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

"]));

-2.06

 Reſ

-1.89

 全新

-1.84

viembre

-1.82

ignée

-1.80

 doivent

-1.77

少し

-1.75

臜

-1.70

我们在

-1.65

蚰

-1.64

POSITIVE LOGITS

 which

2.70

 reason

2.06

1.95

 there

1.93

 this

1.91

if

1.86

 support

1.70

li

1.67

 does

1.62

 while

1.59

Activations Density 0.074%