INDEX

Explanations

future outcomes

The neuron detects occurrences of the modal “will” introducing a verb—that is, future‐tense “will [verb]” 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

lenir

-1.33

ürzen

-1.28

keiten

-1.26

棹

-1.24

áticas

-1.24

дение

-1.23



-1.22

 fontes

-1.20

逦

-1.20

ttino

-1.18

POSITIVE LOGITS

1.35

 após

1.35

be

1.26

 Would

1.23

--

1.16

</sup>

1.15

 продукта

1.14

–

1.14

 khá

1.10

—

1.10

Activations Density 0.098%