© 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. 73720
    Prev
    Next
    INDEX
    Explanations

    mathematical operations found in texts

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

    This neuron activates on the “**” exponentiation operator and the digits of the exponent, i.e. instances of exponent syntax.

    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
    rdag
    -0.78
    路由
    -0.77
     καθ
    -0.77
    msgSender
    -0.77
     ресурсы
    -0.76
     Route
    -0.75
     Schloß
    -0.73
    ならない
    -0.73
    mitives
    -0.72
     inverse
    -0.72
    POSITIVE LOGITS
    Доброго
    0.96
     DLA
    0.93
    袄
    0.90
    Empereur
    0.88
     rengi
    0.87
     zweiten
    0.86
     Люди
    0.83
    ssa
    0.83
    dragen
    0.83
     tredje
    0.82
    Activations Density 0.054%

    No Known Activations