© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 128784
    Prev
    Next
    INDEX
    Explanations

    can or could followed by verbs

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

    The neuron activates on modal auxiliaries expressing ability or possibility (e.g. “can,” “could”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     kesi
    -1.90
    kembangkan
    -1.71
    Merken
    -1.66
    の魅力
    -1.63
     miliki
    -1.62
    Papier
    -1.61
    ^{*}
    -1.60
    ⽹
    -1.60
    CELONA
    -1.55
    AutoModerator
    -1.55
    POSITIVE LOGITS
    an
    2.16
    as
    2.08
    st
    1.94
    x
    1.77
    g
    1.75
    is
    1.73
    i
    1.70
    q
    1.66
    t
    1.64
    first
    1.63
    Activations Density 0.014%

    No Known Activations