© 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. 59465
    Prev
    Next
    INDEX
    Explanations

    mathematical set difference or division

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

    This neuron fires on code/math operator symbols, especially exponentiation (“**”) and integer‐division (“//”) tokens.

    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
     pliable
    -1.72
     The
    -1.68
     Româ
    -1.63
    ࣭
    -1.59
     tomando
    -1.58
    "})
    -1.58
     ghz
    -1.56
    旰
    -1.56
    urably
    -1.56
    扨
    -1.54
    POSITIVE LOGITS
     /
    1.85
    (?)
    1.84
    at
    1.83
    —
    1.78
    (
    1.76
    人都
    1.72
    ،
    1.71
     υπάρχ
    1.70
    ボート
    1.69
     —
    1.67
    Activations Density 0.019%

    No Known Activations