INDEX

Explanations

effective results

The neuron fires on words emphasizing effectiveness, problem-solving, and delivering measurable results.

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

䊐

-1.46

 especially

-1.38

 further

-1.34

 another

-1.24

of

-1.21

 including

-1.20

 particularly

-1.20

 testCase

-1.20

 over

-1.18

 such

-1.17

POSITIVE LOGITS

 eficaz

1.36

tabile

1.32

:"#

1.24

 eredmény

1.17

 noires

1.16

tingen

1.16

ји

1.15

فيذ

1.13

 skincare

1.13

已經

1.13

Activations Density 0.048%