INDEX

Explanations

copyrights and licenses

The neuron primarily activates on numeric tokens (e.g. standalone numbers like years or counts).

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

 Trotter

-0.74

לוח

-0.73

WISH

-0.69

🗜

-0.69

 THER

-0.68

acl

-0.68

ATTENTION

-0.68

ראי

-0.67

Ethio

-0.67

<0xC9>

-0.66

POSITIVE LOGITS

kod

0.78

 बिंदु

0.68

Dru

0.67

 abuso

0.67

ey

0.67

麵包

0.67

どり

0.66

 Rude

0.66

Внеш

0.65

 moved

0.64

Activations Density 0.079%