© 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. 71624
    Prev
    Next
    INDEX
    Explanations

    product names and model numbers

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

    The neuron is highly sensitive to specific technical terms, acronyms, and product or brand names (e.g. ATM, equatorial mount, GoTo, Eliquid, Nitaku, Bachmann).

    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
    你现在
    -1.19
    當時
    -1.17
     Thursday
    -1.16
     those
    -1.16
     Geduld
    -1.15
     somebody
    -1.14
    новниш
    -1.13
     Certainly
    -1.11
    ホームページ
    -1.11
    Where
    -1.09
    POSITIVE LOGITS
    ципа
    1.07
     Сол
    1.07
     Krakowie
    1.06
    reviewed
    1.05
     OGSÅ
    1.04
     ſo
    1.03
    dienne
    1.02
    кова
    1.02
     Liste
    1.02
     én
    1.01
    Activations Density 0.029%

    No Known Activations