© 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. 58321
    Prev
    Next
    INDEX
    Explanations

    Describes things with multiple uses

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

    The neuron is triggered by occurrences of the “multi-” (or “Multi-”) prefix in hyphenated compound terms.

    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
     permitió
    -0.83
    stø
    -0.82
    viale
    -0.80
    olio
    -0.75
     ayudan
    -0.75
     peintures
    -0.75
     provocó
    -0.74
     riêng
    -0.73
    シャ
    -0.73
    zzles
    -0.73
    POSITIVE LOGITS
     function
    1.39
     purpose
    1.23
    function
    1.17
     functions
    1.05
     use
    1.00
     Function
    1.00
     purposes
    0.99
    purpose
    0.94
    Function
    0.91
     nutz
    0.90
    Activations Density 0.034%

    No Known Activations