© 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. 6476
    Prev
    Next
    INDEX
    Explanations

    ECP Connection Definition & Usage

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

    The neuron fires strongly on words that serve as section headings or labels—i.e. tokens marking major topics, categories, or documentation headings.

    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
     latter
    0.68
    odio
    0.63
    {},
    0.61
    ៕
    0.59
    odynam
    0.58
    first
    0.56
    抱着
    0.56
    initializing
    0.56
    ｍ
    0.56
    inafter
    0.56
    POSITIVE LOGITS
     There
    2.52
     This
    2.17
     It
    2.03
     These
    2.02
     They
    1.99
     Although
    1.93
     However
    1.91
     The
    1.88
     When
    1.84
     Those
    1.83
    Activations Density 1.357%

    No Known Activations