© 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. 128752
    Prev
    Next
    INDEX
    Explanations

    classifying oil drilling muds

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

    The neuron specifically fires on Italian (and closely related Romance‐language) content words—especially nouns in –zione and verbs in –are/–ere/–ire—i.e. characteristic Italian lexical suffixes.

    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
    鬭
    -2.61
    s
    -2.48
    *
    -2.39
    r
    -2.38
    At
    -2.33
    If
    -2.31
    ܇
    -2.28
    mathrm
    -2.27
    {
    -2.27
    We
    -2.22
    POSITIVE LOGITS
    潆
    2.94
    Ꮟ
    2.89
    蹀
    2.84
    骉
    2.67
    鲀
    2.63
    椟
    2.59
     possano
    2.56
     Anſ
    2.50
    も良い
    2.45
    THEY
    2.41
    Activations Density 0.003%

    No Known Activations