INDEX

Explanations

the word "conflict"

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

StructEnd

-0.72

['$

-0.71

Ceb

-0.68

OGND

-0.66

 Darius

-0.65

Darius

-0.65

Edm

-0.63

sej

-0.62

Emile

-0.61

eves

-0.59

POSITIVE LOGITS

 Conflicts

1.14

FLICT

1.08

 conflict

1.05

 Conflict

1.05

 conflicts

1.03

conflict

0.96

Conflicts

0.89

CONFLICT

0.88

 conflicted

0.83

 chi̍t

0.82

Activations Density 0.006%