INDEX

Explanations

duration of time

The neuron detects references to historical duration or lineage—phrases that indicate something dates back or has been handed down over many years.

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

 February

-0.93

Eleven

-0.88

ִּ

-0.86

 October

-0.83

dollar

-0.82

 July

-0.82

 January

-0.81

 Month

-0.81

 Veteran

-0.80

month

-0.79

POSITIVE LOGITS

 centuries

4.44

 thousands

4.44

 hundreds

4.09

 thousand

3.38

 millennia

3.31

thousands

3.25

hundreds

3.06

 hundred

2.98

Thousands

2.98

 Thousands

2.80

Activations Density 0.109%