© 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. 85372
    Prev
    Next
    INDEX
    Explanations

    United States, Mexico, Ukraine, Israel, China

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

    The neuron strongly activates on hyphens and other dash‐style punctuation (i.e. the connectors in hyphenated compounds and name or date ranges).

    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
     at
    -2.47
     in
    -2.39
    '
    -2.27
     if
    -2.16
     when
    -2.02
     is
    -1.94
     for
    -1.92
     has
    -1.71
     or
    -1.70
     If
    -1.65
    POSITIVE LOGITS
     Schluß
    1.80
    蓠
    1.70
     Vergnügen
    1.70
    縢
    1.65
     haupts
    1.63
    wasn
    1.62
     übri
    1.61
    mogen
    1.61
     dergleichen
    1.59
     reichlich
    1.55
    Activations Density 0.070%

    No Known Activations