© 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. 63566
    Prev
    Next
    INDEX
    Explanations

    prefix "ме", "поме", "спе", "meta", "per"

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

    The neuron is triggered by subword tokens beginning with the letters “me” (e.g. Me-, mé-, πε-, μετα-), i.e. the “me” prefix.

    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
    seventh
    -0.80
     ninth
    -0.79
     ventre
    -0.78
     SERIES
    -0.77
     enge
    -0.76
     seconde
    -0.75
    servez
    -0.75
     serien
    -0.75
     Poste
    -0.73
     Anglia
    -0.73
    POSITIVE LOGITS
     schönes
    0.79
    umfang
    0.77
    ITAGE
    0.73
    東大
    0.72
     Abschluss
    0.71
    介
    0.71
     kiedyś
    0.71
     turnovers
    0.69
     vermittelt
    0.69
     umfasst
    0.69
    Activations Density 0.021%

    No Known Activations