INDEX

Explanations

explaining reasons or causes

The neuron strongly activates on pronouns and demonstrative words (e.g. this, these, it, they).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 antichi

-0.94

让

-0.91

 Depois

-0.90

masına

-0.90

ماس

-0.89

 telefonu

-0.87

 myös

-0.87

 contactez

-0.85

 חיצוני

-0.85

 châssis

-0.85

POSITIVE LOGITS

due

1.67

 because

1.56

 thanks

1.34

 unlike

1.28

 díky

1.23

 devido

1.20

 aufgrund

1.20

 بسبب

1.18

 wegen

1.17

due

1.13

Activations Density 0.085%