© 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. 34432
    Prev
    Next
    INDEX
    Explanations

    news agency and personal names

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

    This neuron spotlights news‐story metadata such as wire-service attributions (e.g. Reuters, AP) and accompanying dates or numerical dateline elements.

    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
     temen
    -1.00
     Chávez
    -0.96
    能够
    -0.96
     valla
    -0.96
     aucune
    -0.95
    Amigos
    -0.93
     sokak
    -0.92
    Libro
    -0.91
     rejuvenation
    -0.91
    PRODUCTS
    -0.90
    POSITIVE LOGITS
    ՝
    1.14
     also
    1.08
    lables
    1.05
    ην
    1.03
    moveToNext
    1.02
     by
    0.99
     voltaje
    0.99
    pinMode
    0.96
    Ỹ
    0.96
    packer
    0.95
    Activations Density 0.006%

    No Known Activations