© 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. 58623
    Prev
    Next
    INDEX
    Explanations

    door, property, sensors, applications

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

    The neuron flags mentions of mechanical lock hardware—specifically words naming lock components (e.g. door, lock, key, bolt, cylinder).

    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
     they
    -1.78
     today
    -1.47
     yesterday
    -1.30
    catore
    -1.26
     estarán
    -1.23
    esterday
    -1.21
     satisfacer
    -1.20
     ejercer
    -1.20
     again
    -1.19
     londres
    -1.19
    POSITIVE LOGITS
     other
    1.20
     двер
    1.18
    return
    1.13
     questi
    1.12
    Other
    1.12
     chapelle
    1.09
     the
    1.09
    ׳
    1.09
     projets
    1.08
     a
    1.07
    Activations Density 0.209%

    No Known Activations