INDEX

Explanations

clock related to counter

The neuron activates on words that indicate time or the passage of time (e.g., minutes, stroke, time running out).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 മണ്ണി

0.63

吭

0.54

Sham

0.53

 कच्च

0.51

糞

0.51

 যোগ্য

0.50

 സംസ്

0.50

 Issled

0.50

槛

0.50

^{[\

0.49

POSITIVE LOGITS

 clock

2.88

clock

2.69

 clocks

2.69

Clock

2.63

 Clock

2.61

時計

2.53

 reloj

2.31

CLOCK

2.22

 timepiece

2.14

 clocked

2.13

Activations Density 0.431%