© 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. 67238
    Prev
    Next
    INDEX
    Explanations

    names and places

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

    The neuron detects labelled place names and administrative divisions (e.g. town, village, community) in structured info sections.

    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
     freaking
    -2.23
     estrela
    -2.09
    隂
    -2.09
     garota
    -2.09
     sonido
    -2.03
     excruciating
    -1.98
     defiant
    -1.94
    ٧
    -1.92
    熾
    -1.90
    蟥
    -1.89
    POSITIVE LOGITS
     —
    2.23
    ments
    2.19
     and
    1.97
    </em>
    1.91
    ineſs
    1.83
     länger
    1.77
     competitivo
    1.74
     Ekonomi
    1.73
    1.72
     gelb
    1.70
    Activations Density 0.002%

    No Known Activations