© 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. 22-GEMMASCOPE-RES-131K
    7. 25443
    Prev
    Next
    INDEX
    Explanations

    found this

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

    The neuron spotlights verbs that denote searching or discovering (e.g. “found,” “looking”).

    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
     for
    -1.04
     into
    -0.96
     to
    -0.93
     senza
    -0.92
    seeing
    -0.91
    star
    -0.90
    till
    -0.89
    achi
    -0.87
    cf
    -0.87
    lo
    -0.86
    POSITIVE LOGITS
    kvm
    1.16
    Ингредиенты
    1.12
     vuestro
    1.07
     encender
    1.06
     arreglar
    1.06
    romyal
    1.05
    mittags
    1.03
     blí
    1.02
     anywhere
    1.01
     desbloquear
    1.00
    Activations Density 0.154%

    No Known Activations