INDEX

Explanations

frustration and irritation

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

 terrified

-1.20

 scared

-1.19

 frightened

-1.17

 shocked

-1.00

 afraid

-0.97

 horrified

-0.94

 uncomfortable

-0.91

 astonished

-0.90

 appalled

-0.89

 fearful

-0.88

POSITIVE LOGITS

 frustration

2.81

 frustrations

2.73

 Frust

2.56

 frustrated

2.42

Frust

2.36

frust

2.33

 frustrating

2.14

 frust

2.00

 frustrate

1.85

 frustra

1.73

Activations Density 0.031%