INDEX

Explanations

for followed by pronoun or noun

The neuron is keyed to the concessive “too … for …” pattern, firing on the “for” (and its following pronouns) in “too X for Y” constructions.

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

</em>

-1.82

is

-1.80

</h2>

-1.70

 eſſ

-1.70

])

-1.64

 fhall

-1.55

kt

-1.48

</h3>

-1.46

 unſ

-1.45

</strong>

-1.44

POSITIVE LOGITS

也得

1.90

 your

1.88

 what

1.79

',

1.69

all

1.66

一瞬

1.59

 demander

1.55

媜

1.55

ㄩ

1.55

Activations Density 0.010%