INDEX

Explanations

introduce explanations

The neuron detects common discourse‐marker phrases (e.g. “long story short,” “as far as my understanding goes,” “truth be told,” “the fact that,” etc.) that introduce summaries or commentary.

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

and

-1.92

to

-1.57

];

-1.55

in

-1.44

for

-1.41

 einverstanden

-1.38

 dergleichen

-1.34

安い

-1.34

参考に

-1.33

關注

-1.33

POSITIVE LOGITS

2.16

 there

2.05

it

1.60

Év

1.59

 they

1.55

 impecable

1.53

 Sollte

1.52

 alkoh

1.48

 ofrecerte

1.47

過ぎる

1.46

Activations Density 0.066%