© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 108265
    Prev
    Next
    INDEX
    Explanations

    administration and roles

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

    The neuron activates on words containing the root “administ” (e.g. administrator, administratrix, administrador), i.e. administrative‐role terms.

    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
    ঝ
    -0.77
    斥
    -0.77
     emboss
    -0.72
    isEnabled
    -0.71
    nę
    -0.71
     Aston
    -0.70
    idelity
    -0.69
    emptory
    -0.68
    まりました
    -0.68
    Abi
    -0.68
    POSITIVE LOGITS
     administration
    3.56
     Administration
    2.61
    Administration
    2.59
    administration
    2.59
     administrators
    2.47
     administrator
    2.45
     administer
    2.44
     administra
    2.44
     administ
    2.39
     admin
    2.38
    Activations Density 0.098%

    No Known Activations