© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 28932
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and computer code

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

    This neuron detects mathematical power‐ or exponentiation notation (the “**” operator and its surrounding arithmetic context).

    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
    ebly
    -1.69
    蘗
    -1.46
    了不少
    -1.41
    unno
    -1.38
    /*
    -1.37
    min
    -1.35
    vare
    -1.34
    ¨
    -1.31
     angegeben
    -1.31
    section
    -1.31
    POSITIVE LOGITS
     would
    1.88
     will
    1.64
    dịch
    1.52
    hamburg
    1.44
     could
    1.42
    bản
    1.41
     it
    1.40
    bình
    1.38
    colori
    1.35
    짹
    1.34
    Activations Density 0.083%

    No Known Activations