INDEX

Explanations

`timestamp` value

The neuron activates on numeric expressions and statistical or measurement values in the text (e.g. percentages, decimal figures, counts).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Enlarg

0.84

 یعنی

0.79

 obscene

0.79

 Zionist

0.78

 homotopy

0.77

 Representations

0.77

 Linking

0.77

 inscreve

0.77

 Registration

0.75

 muon

0.75

POSITIVE LOGITS

同様

0.71

Avec

0.67

Questo

0.65

Artikel

0.64

Lorsque

0.64

Dieses

0.63

ذلك

0.63

Đối

0.63

 ресур

0.61

Dieser

0.61

Activations Density 0.000%