INDEX

Explanations

impact and its effects

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

玩意

1.76

닝

1.70

eker

1.69

iguiente

1.67

 dessus

1.67

heba

1.66

厲

1.66

 amantes

1.65

_[

1.63

חר

1.63

POSITIVE LOGITS

surgeon

1.51

 chemotherapy

1.48

operatively

1.44

AddRange

1.43

 salic

1.38

surgery

1.36

 neoplas

1.35

 radiotherapy

1.34

 nhạc

1.30

<tr>

1.30

Activations Density 0.098%