© 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. 128061
    Prev
    Next
    INDEX
    Explanations

    power and relations

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

    The neuron is primarily detecting the concept “power,” i.e. occurrences of the word “power” (and closely related forms like “powerful”).

    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
     '
    -2.05
    '
    -1.98
    "));
    -1.77
    ⾞
    -1.73
    賸
    -1.70
     provenant
    -1.66
    a
    -1.65
     frequentemente
    -1.65
    孓
    -1.65
    ."
    -1.63
    POSITIVE LOGITS
     WHAT
    1.95
     It
    1.88
     SOME
    1.85
     einer
    1.65
     THE
    1.60
     Conversely
    1.55
    </h4>
    1.55
     noch
    1.50
    </sup>
    1.49
     ONE
    1.49
    Activations Density 0.021%

    No Known Activations