© 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. 63783
    Prev
    Next
    INDEX
    Explanations

    program, hereby, microchipping, Ainsley, Q

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

    The neuron fires on numeric tokens—especially section or standard numbers (e.g. the “2 3 0” in “ISA 230 Documentation”).

    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
    ясь
    -0.92
    tention
    -0.82
     comprimidos
    -0.77
    recur
    -0.72
    getDay
    -0.72
    lapses
    -0.71
    rowData
    -0.71
     addObject
    -0.71
    testData
    -0.70
    macar
    -0.70
    POSITIVE LOGITS
     nerfed
    0.78
     biological
    0.77
    budget
    0.76
    𝑢
    0.75
    UNE
    0.73
    Budget
    0.73
    部
    0.72
    onomy
    0.72
     étranger
    0.71
    ROI
    0.71
    Activations Density 0.009%

    No Known Activations