© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 10556
    Prev
    Next
    INDEX
    Explanations

    possessive "of" in multiple languages

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

    This neuron activates on nouns and noun phrases that emphasize people’s well-being, health, personal information, and impact (e.g., “health,” “well-being,” “lives,” “personal information”) in a donation appeal context.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    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
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ığı
    0.68
    :~
    0.65
    תוך
    0.63
    }|=
    0.63
    scode
    0.63
    वून
    0.63
    whose
    0.63
    〒
    0.62
    塊
    0.62
    டுவது
    0.60
    POSITIVE LOGITS
    ของผู้
    1.45
    របស់
    1.16
     của
    1.16
     каждого
    1.16
     của
    1.15
    ของ
    1.13
     degli
    1.07
     человека
    1.06
    ผู้
    1.05
    របស់អ្នក
    1.05
    Activations Density 0.542%

    No Known Activations