INDEX

Explanations

product series names

The neuron is tuned to spot the word “series” (typically in product or model names).

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

 necessary

-1.91

をします

-1.58

костюм

-1.56

鳏

-1.54

 mtrl

-1.51

,--

-1.48

 utilizzato

-1.48

lifter

-1.46

one

-1.45

 naves

-1.45

POSITIVE LOGITS

of

1.93

 paramInt

1.80

of

1.78

–

1.63

providedIn

1.61

—

1.60

 гр

1.57

You

1.54

 تصميم

1.52

 fermeture

1.52

Activations Density 0.032%