© 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. 13865
    Prev
    Next
    INDEX
    Explanations

    mathematical operators and programming code

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

    The neuron activates on numeric literals (constants) 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
     have
    -2.22
     They
    -2.03
     However
    -1.92
     they
    -1.86
     These
    -1.76
     Despite
    -1.76
     Although
    -1.75
     Instead
    -1.74
     Though
    -1.72
     Following
    -1.72
    POSITIVE LOGITS
     \&
    1.73
    biór
    1.63
    '
    1.54
     ساعد
    1.52
    pontos
    1.52
    ühner
    1.45
    等の
    1.43
    jugador
    1.42
     beträ
    1.41
     gorro
    1.41
    Activations Density 0.017%

    No Known Activations