© 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. 6825
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on nouns denoting ongoing physical processes or actions (especially verbal nouns/gerunds like “pampering,” “fondling,” “ferment,” “labour”).

    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
    鬯
    0.94
     الدراسي
    0.91
     immortality
    0.87
     względu
    0.87
    گنڈ
    0.87
     ค่ะ
    0.85
    鳧
    0.85
    ルの
    0.84
     ……..
    0.84
     อ่ะ
    0.83
    POSITIVE LOGITS
    ’
    0.92
    ও
    0.83
    (|
    0.75
    ERS
    0.71
     이에
    0.71
    া
    0.70
    аб
    0.68
    SignedIn
    0.66
    ’”
    0.66
    россий
    0.64
    Activations Density 0.000%

    No Known Activations