© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 21145
    Prev
    Next
    INDEX
    Explanations

    dependencies, snapshot, module

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

    The neuron fires on section‐header keywords in code comments—most strongly on “dependencies” headings.

    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
     letzte
    -1.76
     komik
    -1.65
     habile
    -1.63
     tromper
    -1.60
     abiye
    -1.59
     большие
    -1.59
     vendeur
    -1.55
     étan
    -1.55
     tunik
    -1.52
     conducteur
    -1.52
    POSITIVE LOGITS
     for
    2.20
     the
    2.16
     which
    1.92
     by
    1.86
     is
    1.86
     další
    1.80
     with
    1.77
     to
    1.72
     and
    1.71
     so
    1.69
    Activations Density 0.003%

    No Known Activations