INDEX

Explanations

numbered lists

The neuron fires on numeric sequences with decimal points (e.g. scores, rankings, or measurements).

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

 গ্রন্থে

0.83

 গ্রন্

0.79

<start_of_turn>

0.79

 könne

0.78

 própri

0.77

kében

0.77

 vielfält

0.77

 codebase

0.77

审美

0.77

<unused19>

0.76

POSITIVE LOGITS

1.30

1.28

1.25

1.24

1.23

1.18

1.16

1.15

1.10

1.09

Activations Density 0.038%