© 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. 77827
    Prev
    Next
    INDEX
    Explanations

    sworn in

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

    The neuron reliably fires on terms associated with formal oath‐taking or inauguration events (e.g. “inauguration,” “swear(s)/swore in,” “sworn in”).

    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
    asiswa
    -0.73
    ⸙
    -0.71
     càng
    -0.70
     isOpen
    -0.69
    amodel
    -0.69
     bouw
    -0.67
     қара
    -0.66
    ROP
    -0.64
    úmenes
    -0.64
     bake
    -0.64
    POSITIVE LOGITS
     sworn
    3.84
     swearing
    3.72
     oath
    3.69
     swore
    3.16
     oaths
    2.89
     swear
    2.88
     Oath
    2.69
     swears
    2.45
     Sw
    1.88
    Sw
    1.77
    Activations Density 0.051%

    No Known Activations