INDEX

Explanations

starts with "no,"

The neuron is detecting standalone “No” (especially as a sentence‐initial negation/discourse marker).

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

 enormous

-2.77

 burgeoning

-2.58

 There

-2.55

 sizable

-2.48

 showed

-2.45

 increases

-2.41

 HUGE

-2.38

 blatantly

-2.38

 noticeably

-2.38

 unusually

-2.34

POSITIVE LOGITS

 dando

2.22

銠

2.19

狴

2.16

having

2.16



2.09

 extravag

2.08

秾

2.06

癭

2.06

猛然

2.06

 raiſ

2.06

Activations Density 0.022%