© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 99436
    Prev
    Next
    INDEX
    Explanations

    Arab population minorities, Shia

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

    The neuron fires strongly on named national, ethnic, religious, and place identifiers—especially Israeli, Palestinian, Arab/Muslim, Jewish labels and associated Middle Eastern toponyms—i.e. it detects references to Middle Eastern identity groups and locations.

    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
    quarie
    -0.81
     Settle
    -0.81
    流し
    -0.79
    quilla
    -0.75
     Félix
    -0.73
     Amalia
    -0.73
    echos
    -0.71
     Afghans
    -0.71
    adah
    -0.71
    Zona
    -0.71
    POSITIVE LOGITS
     Arab
    1.09
     minority
    0.92
    Arab
    0.91
     Dru
    0.90
     Minority
    0.86
    Inside
    0.83
    xfc
    0.83
     Inside
    0.83
    ALE
    0.82
     внутри
    0.81
    Activations Density 0.013%

    No Known Activations