INDEX

Explanations

floor plans and tents

The neuron detects quantitative information, i.e. numbers and their surrounding measurement or numeric-specification context.

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

<td>

-1.80

理石

-1.72

 strictement

-1.71

 vuitton

-1.69

<bos>

-1.63

 lesquelles

-1.62

 regularmente

-1.59

𝖢

-1.59

Quais

-1.59

Concepto

-1.55

POSITIVE LOGITS

in

2.28

 that

2.08

if

2.05

 или

1.91

are

1.87

 will

1.85

1.84

of

1.82

1.80

at

1.71

Activations Density 0.019%