INDEX

Explanations

numbers followed by standard

The neuron activates on three‐digit numeric tokens (particularly those in the “13x” series).

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

In

-3.53

-2.47

-2.39

 Vertreter

-2.27

-2.09

 betreffenden

-1.99

was

-1.97

es

-1.93

,”

-1.93

٨

-1.91

POSITIVE LOGITS

 sinyal

2.20

嚒

2.11

ቷ

1.98

1.97

mathrm

1.91

嵛

1.89

)」

1.88

࿙

1.85

 dévelo

1.81

ógica

1.77

Activations Density 0.016%