INDEX

Explanations

time followed by m

The neuron activates on time‐of‐day expressions, specifically hours marked with “a.m.” or “p.m.”.

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

踯

-1.99

⎔

-1.94

 cuja

-1.94

髫

-1.93

 famed

-1.88

糴

-1.87

 divertidos

-1.83

 accél

-1.80



-1.79

 semelh

-1.77

POSITIVE LOGITS

 with

3.25

and

3.02

in

2.88

at

2.81

by

2.78

 from

2.72

as

2.27

had

2.25

an

2.13

out

2.13

Activations Density 0.002%