INDEX

Explanations

time durations after events

The neuron principally detects mentions of time spans or durations (e.g. “year,” “after,” numeric time intervals).

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

 może

-1.75

and

-1.72

wth

-1.63

føre

-1.63

 diferite

-1.61

 poate

-1.55

 possa

-1.54

 може

-1.53

っていますが

-1.52

遑

-1.50

POSITIVE LOGITS

2.11

it

1.84

for

1.65

 other

1.53

this

1.52

just

1.51

.......

1.43

await

1.42

there

1.41

 miesięcy

1.41

Activations Density 0.023%