INDEX

Explanations

leave alone / stop bothering

np_acts-logits-general · gemini-2.5-flash-lite

The neuron detects occurrences of the word “alone,” especially when used in phrases like “leave … alone.”

oai_token-act-pair · o4-mini Triggered by @jyhe0408

phrases that tell others to leave someone or something alone, be left on their own, or back off.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

the phrase "leave [something/someone] alone" or variations of it.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ajouter

-1.68

 verminder

-1.59

-1.52

看到了

-1.46

 rafra

-1.45

裥

-1.44

 nahil

-1.42

巂

-1.40

狳

-1.40

 desapar

-1.40

POSITIVE LOGITS

涎

1.49

大半

1.48

↵

1.45

if

1.43

海南

1.39

Эти

1.37

Snapshots

1.37

星空

1.36

 ними

1.36

for

1.36

Activations Density 0.005%

leave alone / stop bothering

The neuron detects occurrences of the word “alone,” especially when used in phrases like “leave … alone.”

phrases that tell others to leave someone or something alone, be left on their own, or back off.

the phrase "leave [something/someone] alone" or variations of it.

No Comments

No Known Activations

leave alone / stop bothering

The neuron detects occurrences of the word “alone,” especially when used in phrases like “leave … alone.”

phrases that tell others to leave someone or something alone, be left on their own, or back off.

the phrase "leave [something/someone] alone" or variations of it.

No Comments

No Known Activations