INDEX

Explanations

becoming more X

The neuron activates on the comparative intensifier “more,” especially in repeated or “more and more” 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

-1.94

舅

-1.63

istically

-1.54

 még

-1.52

 Still

-1.51

начально

-1.49

 ſou

-1.47

 Certainly

-1.47

 still

-1.45

ness

-1.45

POSITIVE LOGITS

1.51

}${

1.40

 جيدة

1.36

</td>

1.31

1.30

 jako

1.30

Namara

1.29

1.27

這一

1.27

",

1.26

Activations Density 0.016%