© 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. 113157
    Prev
    Next
    INDEX
    Explanations

    physical objects and locations

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

    The neuron fires on mentions of physical, material things—especially the word “physical” when contrasting tangible objects or documents with digital/virtual ones.

    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
    ant
    -1.73
    ні
    -1.61
    å
    -1.57
     Betrach
    -1.55
     inmediatamente
    -1.52
    [
    -1.52
    ation
    -1.49
     flatly
    -1.48
     alguien
    -1.47
    后
    -1.46
    POSITIVE LOGITS
     delicado
    1.68
    tabria
    1.67
     パウダー
    1.67
     redondo
    1.67
     fantástico
    1.65
     comprometido
    1.64
    َب
    1.63
     bingung
    1.62
    gambarkan
    1.62
     sólido
    1.60
    Activations Density 0.017%

    No Known Activations