INDEX

Explanations

asking questions

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

 luis

-0.84

forgettable

-0.81

Zab

-0.80

дача

-0.79

 saad

-0.79

 démarches

-0.78

Xy

-0.77

setGravity

-0.77

vimos

-0.75

 inget

-0.75

POSITIVE LOGITS

 questions

8.19

questions

6.22

Questions

5.94

 Questions

5.91

 QUESTIONS

5.06

 question

4.88

QUESTIONS

4.69

 Fragen

4.66

 preguntas

4.47

 вопросы

4.25

Activations Density 0.153%