INDEX

Explanations

introducing an explanation

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

ﾙ

3.27

3.21

궬

2.90

\"",

2.82

FileDialog

2.74

">\

2.65

allclasses

2.64

2.63

valueChanged

2.61

```

2.61

POSITIVE LOGITS

tion

2.70

ب

2.58

san

2.51

sy

2.47

ائ

2.41

nich

2.34

tene

2.27

دع

2.25

tır

2.22

nas

2.22

Activations Density 0.014%