© 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. 9272
    Prev
    Next
    INDEX
    Explanations

    di, ke, id words

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

    The neuron fires on the standalone two‐letter token “di” (the Italian preposition “di”).

    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
    -3.56
     
    -3.36
    to
    -3.30
    In
    -3.16
    Y
    -3.09
    This
    -3.02
    re
    -2.92
    2
    -2.89
    在
    -2.89
    If
    -2.88
    POSITIVE LOGITS
    嫱
    2.81
     夜景
    2.78
    猡
    2.73
    ࿆
    2.69
    AdvEx
    2.69
    🫅
    2.63
    嫑
    2.59
    donkey
    2.55
    ㌀
    2.55
    dengan
    2.53
    Activations Density 0.002%

    No Known Activations