© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 35505
    Prev
    Next
    INDEX
    Explanations

    children and sons

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

    The neuron spotlights mentions of family relationships where numbers quantify children (e.g., “have two children,” “their five-year-old son”).

    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
     to
    -1.52
     it
    -1.30
     by
    -1.25
     while
    -1.24
     from
    -1.19
     for
    -1.14
     in
    -1.13
     mientras
    -1.11
     if
    -1.05
     or
    -1.03
    POSITIVE LOGITS
     children
    1.33
     passagers
    1.31
     quí
    1.28
     tissus
    1.25
     raiſ
    1.24
     vélos
    1.24
     véhicules
    1.23
    ifies
    1.23
     achats
    1.21
     porcelana
    1.20
    Activations Density 0.009%

    No Known Activations