INDEX

Explanations

present perfect tense auxiliary verbs

The neuron strongly activates on numeric tokens (e.g. dates, figures, counts), i.e. it detects numbers in the text.

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

 become

-0.95

encounter

-0.83

꯱

-0.82

 بالفعل

-0.82

一楼

-0.78

creek

-0.75

iculture

-0.74

 falsi

-0.73

PROTOBUF

-0.73

iction

-0.73

POSITIVE LOGITS

has

2.30

 have

2.06

 đã

1.71

 نے

1.62

was

1.51

 telah

1.44

 heeft

1.35

 hasn

1.33

 उन्होंने

1.32

 wasn

1.30

Activations Density 0.004%