© 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. 71435
    Prev
    Next
    INDEX
    Explanations

    bank or bank(s)

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

    The neuron detects occurrences of the word “bank” (including its plural form).

    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
     and
    -2.02
     However
    -1.80
     Unfortunately
    -1.61
     well
    -1.59
    رر
    -1.55
     When
    -1.55
     that
    -1.54
     There
    -1.52
     Despite
    -1.50
     Compared
    -1.48
    POSITIVE LOGITS
    锖
    1.70
     دانلود
    1.64
    ۴
    1.59
    ۳
    1.58
     untill
    1.57
     پیشن
    1.55
     diffusi
    1.54
     نوج
    1.53
     وزارة
    1.53
     品牌
    1.53
    Activations Density 0.028%

    No Known Activations