© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6810
    Prev
    Next
    INDEX
    Explanations

    dash it all, flash cards, cash

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

    This neuron detects capitalized proper names (e.g. people, places, or organizations).

    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_31_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
     মাধ্য
    0.70
    蜱
    0.67
    ening
    0.62
    ened
    0.61
    ška
    0.57
    zy
    0.57
    enu
    0.57
    ⵖ
    0.56
    дцать
    0.56
    entz
    0.55
    POSITIVE LOGITS
    ׂ
    1.03
    ׁ
    0.78
    تباه
    0.68
    hhhh
    0.63
    terre
    0.62
    warz
    0.62
    ield
    0.61
    hhh
    0.61
    hhhhhhhh
    0.60
    اشة
    0.58
    Activations Density 0.417%

    No Known Activations