© 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. 67024
    Prev
    Next
    INDEX
    Explanations

    units of measurement

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

    The neuron reliably fires on numeric values (including decimals and percentages) presented inside parentheses or brackets.

    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
    seemed
    -1.54
    tried
    -1.52
     intérêts
    -1.51
     piaci
    -1.45
    seems
    -1.43
    僕が
    -1.41
     worden
    -1.41
     huiles
    -1.41
     ingrédients
    -1.39
    新しく
    -1.38
    POSITIVE LOGITS
     of
    1.52
     September
    1.43
    ongono
    1.38
     otros
    1.37
     in
    1.35
    禾
    1.34
    ,
    1.34
     June
    1.34
     July
    1.31
     weeks
    1.30
    Activations Density 0.010%

    No Known Activations