INDEX

Explanations

bring

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ResumeLayout

-0.90

Autoritní

-0.90

featureID

-0.89

+#+#

-0.87

principalColumn

-0.85

DllImport

-0.81

HasForeignKey

-0.81

 ModelExpression

-0.80

 MessageBoxIcon

-0.80

 snippetHide

-0.79

POSITIVE LOGITS

 about

0.65

about

0.62

back

0.53

 back

0.45

んでも

0.43

 ABOUT

0.42

pert

0.41

it

0.40

Back

0.40

これで

0.40

Activations Density 0.024%