© 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. 3097
    Prev
    Next
    INDEX
    Explanations

    information, state, full lace, Stoicism, particulate matter

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

    This neuron fires strongly on words that start a new sentence or segment — i.e. the first token following a boundary.

    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
     pérdida
    0.87
     desviación
    0.86
    髅
    0.85
     угро
    0.83
     polvo
    0.80
     stenosis
    0.80
     швид
    0.79
     étale
    0.78
     ridges
    0.77
     mortalidad
    0.77
    POSITIVE LOGITS
    A
    0.94
     Museums
    0.84
    U
    0.82
    IA
    0.81
    Presenter
    0.78
    V
    0.78
    N
    0.77
     presenter
    0.77
    C
    0.77
    FileName
    0.75
    Activations Density 0.002%

    No Known Activations