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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 3621
    Prev
    Next
    INDEX
    Explanations

    immoral or harmful contexts

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

    The neuron sharply activates on longer, Latinate or technical words—especially those carrying common Latin‐derived suffixes (e.g. –ed, –ion, –ize, –ice).

    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
     財布
    0.60
     đều
    0.59
    ിക്കാം
    0.56
     V
    0.51
    ઈએ
    0.50
     भी
    0.49
     husband
    0.49
     bese
    0.48
     पनि
    0.48
     sekal
    0.48
    POSITIVE LOGITS
     reformas
    0.62
    και
    0.61
     शुभार
    0.58
     మరియు
    0.57
    άρ
    0.57
    ക്ര
    0.57
    Υ
    0.56
    健全
    0.55
    ори
    0.55
    佐藤
    0.54
    Activations Density 0.158%

    No Known Activations