INDEX

Explanations

heroic deeds and accomplishments

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

 bilden

-0.87

аа

-0.84

 geboren

-0.83

蔵

-0.78

RemoveField

-0.77

styleUrls

-0.75

 nació

-0.73

兵

-0.71

RIL

-0.71

МО

-0.70

POSITIVE LOGITS

 achievements

5.88

 achievement

5.47

 accomplishments

5.31

 accomplishment

4.84

achievement

4.06

 Achievements

4.00

 successes

3.94

Achievements

3.86

 Achievement

3.67

 success

3.53

Activations Density 0.082%