INDEX

Explanations

months in dates

The neuron fires on date components—especially month names (or abbreviations) and their accompanying day numbers.

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

 October

-1.16

 September

-1.12

 November

-1.09

 December

-1.05

 February

-0.99

 several

-0.96

 Tenggara

-0.96

 most

-0.93

 primavera

-0.92

 sexto

-0.92

POSITIVE LOGITS

 arrivare

0.93

 thingy

0.92

aż

0.91

 toco

0.89

MouseAdapter

0.85

২

0.85

 مكان

0.85

hed

0.83

mun

0.82

̣i

0.82

Activations Density 0.123%