© 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. 10-GEMMASCOPE-RES-131K
    7. 127837
    Prev
    Next
    INDEX
    Explanations

    equations with variables

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

    The neuron fires most strongly on digits and numeric expressions (numbers and arithmetic symbols).

    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
     to
    -1.71
     raso
    -1.45
     antik
    -1.43
    Zdrav
    -1.39
     which
    -1.36
     döner
    -1.35
     magasiner
    -1.30
     assister
    -1.30
    eronau
    -1.25
     ū
    -1.25
    POSITIVE LOGITS
     Artículos
    1.54
     Accesorios
    1.45
    </u>
    1.43
     Tarjeta
    1.41
    л
    1.27
    вания
    1.26
    zám
    1.25
     et
    1.23
     T
    1.23
    TB
    1.22
    Activations Density 0.666%

    No Known Activations