INDEX

Explanations

contractions and names like Kaelen, Camila

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

The neuron fires strongly on HTML‐style character entities and other encoded punctuation marks (e.g. “'”, “&”, “;”).

oai_token-act-pair · o4-mini

Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 various

0.92

 historically

0.88

 geopolitical

0.84

 diverse

0.81

the

0.80

 variations

0.80

 widespread

0.79

 notable

0.79

 extensive

0.77

 widely

0.77

POSITIVE LOGITS

 もう

1.04

 тебя

1.00

🥺

0.96

łaś

0.96

可愛い

0.94

 えっと

0.93

 дочь

0.93

 Jungkook

0.91

 ហើយ

0.91

 میرے

0.90

Activations Density 0.053%

contractions and names like Kaelen, Camila

The neuron fires strongly on HTML‐style character entities and other encoded punctuation marks (e.g. “'”, “&”, “;”).

No Known Activations

contractions and names like Kaelen, Camila

The neuron fires strongly on HTML‐style character entities and other encoded punctuation marks (e.g. “'”, “&”, “;”).

No Known Activations

contractions and names like Kaelen, Camila

The neuron fires strongly on HTML‐style character entities and other encoded punctuation marks (e.g. “&apos;”, “&”, “;”).

No Known Activations

contractions and names like Kaelen, Camila

The neuron fires strongly on HTML‐style character entities and other encoded punctuation marks (e.g. “&apos;”, “&”, “;”).

No Known Activations

The neuron fires strongly on HTML‐style character entities and other encoded punctuation marks (e.g. “'”, “&”, “;”).

The neuron fires strongly on HTML‐style character entities and other encoded punctuation marks (e.g. “'”, “&”, “;”).