INDEX

Explanations

The neuron detects language about making requests or asking others to do favors.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 endast

0.98

 jedynie

0.77

 également

0.74

 सम्पूर्ण

0.74

 เพียง

0.71

まさに

0.69

irá

0.68

自身の

0.68

 hanya

0.67

 संपूर्ण

0.66

POSITIVE LOGITS

 stupid

1.69

 fucking

1.57

 damned

1.39

 lousy

1.38

 stuff

1.36

 damn

1.34

 crappy

1.33

 shit

1.31

 fucked

1.30

 pissed

1.29

Activations Density 1.206%