© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 15172
    Prev
    Next
    INDEX
    Explanations

    fill the space or dominate

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

    The neuron activates on present-tense verbs that describe how something spatially covers, fills, lines, or dominates an area.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    _
    -1.77
    </h2>
    -1.55
    $
    -1.55
     to
    -1.54
     "
    -1.53
    u
    -1.53
     
    -1.48
     something
    -1.41
    al
    -1.39
     two
    -1.34
    POSITIVE LOGITS
     the
    3.09
     our
    2.02
    鱨
    1.68
    鷽
    1.61
    dáms
    1.60
    ٭
    1.55
     parfü
    1.52
     estadounid
    1.51
     their
    1.48
     parfüm
    1.48
    Activations Density 0.031%

    No Known Activations