© 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. 34-GEMMASCOPE-RES-131K
    7. 2973
    Prev
    Next
    INDEX
    Explanations

    french canton administration

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

    The neuron is detecting the fixed lead‐in phrase of Wikipedia entries for administrative subdivisions (e.g. “The canton of X is an administrative division of the …”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Scott
    -0.73
    ড
    -0.73
    sets
    -0.72
    RO
    -0.71
     gearbeitet
    -0.70
    痫
    -0.69
     alapján
    -0.69
    garder
    -0.69
     Offense
    -0.69
    Ŭ
    -0.68
    POSITIVE LOGITS
     cant
    0.86
    ceğini
    0.83
     SEG
    0.76
    Jawa
    0.76
     haul
    0.75
     beh
    0.74
    Anil
    0.74
    kult
    0.74
    泛
    0.73
    cant
    0.72
    Activations Density 0.022%

    No Known Activations