INDEX

Explanations

Theoretical approach or framework

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ل

2.91

oğlu

2.42

िक

2.41

್

2.41

ो

2.38

ной

2.34

لل

2.30

ী

2.25

2.20

৯

2.19

POSITIVE LOGITS

纷

1.73

 Breath

1.58

 itulah

1.55

 الرئيسي

1.48

 pictured

1.45

 وعند

1.45

্লাহ

1.43

ς

1.40

 dotycz

1.39

 Monat

1.38

Activations Density 0.004%