INDEX

Explanations

speaking, language

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 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

 GenerationType

-0.80

/**

-0.68

openzeppelin

-0.68

-0.67

 FetchType

-0.65

Tikang

-0.63

 CascadeType

-0.62



-0.61

Efq

-0.60

IntoConstraints

-0.60

POSITIVE LOGITS

 speak

1.13

 speaks

0.98

 Speak

0.96

speak

0.94

Speak

0.92

 Language

0.88

 language

0.87

 speaking

0.85

 spoke

0.80

Language

0.77

Activations Density 0.000%