© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43772
    Prev
    Next
    INDEX
    Explanations

    category followed by specific publication

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

    The neuron fires on capitalized names of media outlets, websites, or brands (e.g., “Golf Channel,” “Cyclingnews,” “Travel Channel.com,” “Law360,” etc.).

    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
    jeuner
    -1.76
    ulitis
    -1.57
     you
    -1.51
     снова
    -1.45
    Вам
    -1.41
     cocinar
    -1.38
    fiées
    -1.38
    Ƀ
    -1.36
     pintadas
    -1.36
     gratuitas
    -1.34
    POSITIVE LOGITS
     of
    2.05
     a
    1.78
    
    1.54
    mogen
    1.54
    皷
    1.53
    ⎔
    1.48
     jajaja
    1.46
     desselben
    1.43
    eeeeeeee
    1.41
     that
    1.41
    Activations Density 0.355%

    No Known Activations