INDEX

Explanations

vehicle movement directions

The neuron principally fires on numeric tokens and quantifiable specifications (e.g. gear numbers, measurements, percentages).

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

搥

-1.55

鴞

-1.52

ﮯ

-1.41

氿

-1.41

Karten

-1.41

 neces

-1.40

 technisch

-1.38

 récol

-1.38

 搞笑

-1.37



-1.37

POSITIVE LOGITS

</i>

2.16

This

1.80

this

1.76

 Since

1.59

1.57

 Unlike

1.52

The

1.52

 After

1.51

 only

1.50

</td>

1.48

Activations Density 0.021%