© 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. 3245
    Prev
    Next
    INDEX
    Explanations

    mathematics and computer science concepts

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

    The neuron is triggered by numeric literals and numbers (including digits, numerals, and superscripted exponents) in mathematical expressions.

    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
    łóż
    -1.05
    coating
    -1.01
    ？！”
    -0.91
     construida
    -0.90
     fejl
    -0.90
    łeś
    -0.89
     cokelat
    -0.88
    larınız
    -0.87
     hivat
    -0.86
     saludo
    -0.85
    POSITIVE LOGITS
     faculté
    0.90
    Here
    0.90
     эффек
    0.85
    Procedure
    0.81
    ……
    0.81
    this
    0.81
    ici
    0.81
    転
    0.80
     collègues
    0.80
     ainda
    0.80
    Activations Density 0.010%

    No Known Activations