© 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. 22-GEMMASCOPE-RES-131K
    7. 72372
    Prev
    Next
    INDEX
    Explanations

    DC - DC converter

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

    The neuron selectively flags unusually rare or “information‐heavy” tokens—proper names, unique symbols or bracket characters, and infrequent words—rather than common function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     phiếu
    -1.00
    嚮
    -1.00
     místě
    -0.99
    nový
    -0.94
    日から
    -0.94
     what
    -0.92
    дры
    -0.89
     Vorstellungen
    -0.89
     überzeugt
    -0.88
     Erwartungen
    -0.87
    POSITIVE LOGITS
    Sedangkan
    1.16
     '-';
    1.15
    Jeudi
    1.02
     plafon
    1.02
     vigueur
    1.02
     enfant
    1.02
    ГЛА
    1.00
     poêle
    0.97
    ܆
    0.96
    genen
    0.96
    Activations Density 0.002%

    No Known Activations