© 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. 99885
    Prev
    Next
    INDEX
    Explanations

    function definitions

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

    This neuron activates on function, method, or procedure names (i.e. identifier tokens at the head of a definition).

    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
     dîner
    -1.63
     charmant
    -1.59
     bisous
    -1.55
    .
    -1.52
     huid
    -1.51
     magnifiques
    -1.48
     devront
    -1.46
     soucis
    -1.46
     exasper
    -1.45
     Akhir
    -1.44
    POSITIVE LOGITS
     it
    2.16
     when
    2.05
     its
    1.96
     can
    1.93
     not
    1.73
     is
    1.73
     or
    1.70
     neuen
    1.53
     cuando
    1.49
     doesn
    1.48
    Activations Density 0.009%

    No Known Activations