© 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. 2434
    Prev
    Next
    INDEX
    Explanations

    shadows lengthen, fled, or deepen

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

    The neuron fires strongly on substantive content words—especially nouns and verbs carrying key informational load in formal (historical or analytical) prose.

    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
     içerisinde
    0.90
     curvy
    0.85
     hitta
    0.84
     koska
    0.84
     transitioning
    0.81
     someone
    0.79
    /
    0.78
     pesky
    0.78
     debes
    0.77
     worry
    0.77
    POSITIVE LOGITS
     นับ
    0.86
    itherto
    0.82
    izdat
    0.78
    ,—
    0.77
     нередко
    0.75
     hitherto
    0.75
     속에
    0.71
    Verlag
    0.70
     뒤
    0.69
    ⓤ
    0.68
    Activations Density 0.054%

    No Known Activations