INDEX

Explanations

codes, IDs, and citations

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 their

-1.95

for

-1.80

 also

-1.77

it

-1.76

any

-1.66

 from

-1.66

 նաև

-1.61

 Even

-1.56

 твоя

-1.50

 while

-1.49

POSITIVE LOGITS

1.81

.,

1.68

",

1.63

Instead

1.57

',

1.55

1.52

 тех

1.52

1.51

 intric

1.47

忘了

1.45

Activations Density 0.006%