© 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. 129088
    Prev
    Next
    INDEX
    Explanations

    theology and theologians

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

    The main thing this neuron does is detect occurrences of the word “theology” (and its close variants).

    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
     was
    -3.36
    ?"
    -2.58
     These
    -2.52
    !"
    -2.47
     Of
    -2.38
    d
    -2.22
    致力于
    -2.17
    t
    -2.16
     Only
    -2.14
    ….
    -2.11
    POSITIVE LOGITS
     Schluß
    2.55
     attentive
    2.41
    狆
    2.33
     ¤
    2.31
    0
    2.30
    鈈
    2.22
    .
    2.19
    2.16
    ahhh
    2.14
    事も
    2.13
    Activations Density 0.004%
    No Known Activations