© 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. 34-GEMMASCOPE-RES-131K
    7. 47703
    Prev
    Next
    INDEX
    Explanations

    points and spatial location

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

    The neuron strongly responds to numeric and mathematical notation (numbers, variables, math‐mode symbols like “$”), i.e. tokens that indicate formulas or quantitative expressions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ányi
    -0.77
    头
    -0.76
    )}]{
    -0.76
     οποίο
    -0.75
    Orlando
    -0.75
    aringan
    -0.74
     تضيفلها
    -0.73
     contains
    -0.73
     addSubview
    -0.72
     Undertaker
    -0.71
    POSITIVE LOGITS
     on
    1.04
     points
    0.96
     lie
    0.85
     위치
    0.83
     O
    0.82
     equidist
    0.81
     lies
    0.81
     (−
    0.81
    iglie
    0.80
     ubicado
    0.79
    Activations Density 0.010%

    No Known Activations