© 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. 34-GEMMASCOPE-RES-131K
    7. 63545
    Prev
    Next
    INDEX
    Explanations

    Indonesian food and place names

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

    The neuron strongly responds to fragments of non-English or foreign-language words (e.g. loanwords, diacritic-laden tokens, and substrings of Vietnamese and Indonesian terms).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ,
    -0.76
    甚
    -0.75
    たった
    -0.73
    se
    -0.71
    lll
    -0.71
    S
    -0.71
    xlim
    -0.71
     cosy
    -0.70
     Jr
    -0.69
     Victorian
    -0.68
    POSITIVE LOGITS
     kalend
    0.94
    gantung
    0.94
    manuelle
    0.90
    ❗
    0.88
     تجربه
    0.87
    Kullan
    0.87
    isteren
    0.87
    vailability
    0.85
    kunci
    0.83
     Chartres
    0.83
    Activations Density 0.005%

    No Known Activations