© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 71568
    Prev
    Next
    INDEX
    Explanations

    computing sums

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

    The neuron fires whenever it sees the word “sum” (e.g. as a variable or function name in code).

    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
    ?”
    -2.92
     pomi
    -2.33
    ).”
    -2.30
     producenta
    -2.13
     listop
    -2.11
     całym
    -2.05
     paździer
    -2.03
     โดย
    -1.98
    ”:
    -1.95
     pokud
    -1.93
    POSITIVE LOGITS
    '
    2.48
    /
    2.38
    \
    2.31
    ?
    2.20
    2
    2.13
    _{
    2.05
    =
    2.05
    {
    1.95
    过年
    1.88
    ung
    1.85
    Activations Density 0.006%

    No Known Activations