INDEX

Explanations

validation and verification

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

袍

2.02

了

1.96

marker

1.89

 брига

1.85

 définie

1.83

pä

1.82

ণু

1.81

mitt

1.81

ături

1.80

1.78

POSITIVE LOGITS

![

1.78

𝗇

1.60

![

1.57

!\

1.54

נה

1.46

[/

1.39

textit

1.38

ார

1.35

ാരണ

1.33

 আশ্চর্যজনক

1.33

Activations Density 0.001%