INDEX

Explanations

duration up to a point

The neuron activates on numerical tokens (digits or number-related words).

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

of

-1.90

-1.61

思った

-1.59

 dieser

-1.49

</h2>

-1.47

”。

-1.46

tne

-1.46

ふう

-1.45

 первые

-1.38

POSITIVE LOGITS

 until

2.20

 jusqu

2.14

HostException

1.99

 Until

1.85

 என்று

1.77

]]);

1.75

ја

1.75

 extent

1.74

 untill

1.68

 indiví

1.67

Activations Density 0.071%