© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 73371
    Prev
    Next
    INDEX
    Explanations

    monotone operator, maximal operators, Minimising

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

    This neuron detects content from formal mathematical texts—particularly tokens in theorem‐style statements, definitions, and LaTeX‐like notation.

    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
     catég
    -0.77
     Media
    -0.74
     Tun
    -0.73
    Media
    -0.72
    жкой
    -0.71
     ciągu
    -0.70
    tring
    -0.70
     Hast
    -0.70
    ocus
    -0.69
    meno
    -0.69
    POSITIVE LOGITS
     selections
    0.85
     Gór
    0.78
     epic
    0.77
     selectors
    0.75
     Sonne
    0.74
     Epic
    0.74
     selection
    0.73
     inclusions
    0.71
     EPIC
    0.71
     demand
    0.70
    Activations Density 0.023%

    No Known Activations