© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92316
    Prev
    Next
    INDEX
    Explanations

    Wilkes-Barre and NOW,

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

    This neuron is sharply tuned to spot the place‐name token “Wilkes” (as it appears in “Wilkes‐Barre”).

    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
     li
    -0.73
    昏
    -0.68
     TT
    -0.65
    眈
    -0.65
    zugs
    -0.64
    fot
    -0.64
     달
    -0.64
     gleich
    -0.64
    ELDS
    -0.63
     лежа
    -0.63
    POSITIVE LOGITS
    补充
    0.74
     Klinik
    0.68
    hooter
    0.66
    бам
    0.65
    Shuffle
    0.65
    addError
    0.63
    érature
    0.63
    Obsah
    0.63
     설정
    0.62
     業務
    0.62
    Activations Density 0.064%

    No Known Activations