INDEX

Explanations

mentions of frustration

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Biv

-0.91

 ImGui

-0.72

awtextra

-0.66

 Fait

-0.63

 Suit

-0.63

therly

-0.61

Entrée

-0.59

 /\.(

-0.59

enschappelijke

-0.59

TPC

-0.59

POSITIVE LOGITS

 frustration

1.49

 frustrations

1.43

 frustrated

1.30

frust

1.30

Frust

1.25

 frustrating

1.24

 Frust

1.23

 frustra

1.20

 frust

1.19

 frustrate

1.16

Activations Density 0.005%