© 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. 30881
    Prev
    Next
    INDEX
    Explanations

    personal devices and communications

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

    The neuron activates on occurrences of the word “personal” (in phrases like “personal attacks,” “personal message,” “personal non-commercial use,” etc.).

    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
     \%}$
    -0.85
    raiz
    -0.84
    べる
    -0.82
     技术
    -0.81
    жом
    -0.80
     personages
    -0.78
     парамет
    -0.78
    आ
    -0.78
    ("~/
    -0.78
    לְ
    -0.77
    POSITIVE LOGITS
     personal
    1.38
    personal
    1.27
     Personal
    1.27
    Personal
    1.10
     PERSONAL
    0.90
    ersonal
    0.85
     persoonlijke
    0.84
    Matters
    0.84
     persönliche
    0.83
    izable
    0.81
    Activations Density 0.039%

    No Known Activations