INDEX

Explanations

after events or actions

The neuron strongly responds to sentence-initial temporal transition cues—words like “After,” “Last,” “weeks,” etc., that introduce a temporal frame for what follows.

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

it

-2.22

has

-1.59

It

-1.44

is

-1.41

are

-1.38

 will

-1.36

 pietre

-1.34

-1.29

 sesso

-1.27

矣

-1.27

POSITIVE LOGITS

ffichage

1.22

鵜

1.21

věd

1.20

noin

1.18

 ampliamente

1.17

Bayesian

1.16

 několika

1.16

 různých

1.15

钠

1.13

 которые

1.13

Activations Density 0.052%