© 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. 5732
    Prev
    Next
    INDEX
    Explanations

    systems of roads

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

    This neuron fires most strongly on tokens that serve as document or section markers—especially numeric section/paragraph identifiers and other header‐style labels.

    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
    /
    0.65
    0.59
    ↵
    0.59
    )))
    0.55
    )
    0.55
    ,/
    0.55
    )/
    0.54
     =
    0.53
    Params
    0.52
    DataFrame
    0.52
    POSITIVE LOGITS
     devastating
    0.61
     debilitating
    0.57
     многочис
    0.56
     Obwohl
    0.56
     Помимо
    0.56
     dismantling
    0.54
     peculiarly
    0.54
     అనేక
    0.52
     exhilarating
    0.52
     sadistic
    0.52
    Activations Density 0.051%

    No Known Activations