© 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. 22-GEMMASCOPE-RES-131K
    7. 16232
    Prev
    Next
    INDEX
    Explanations

    code execution contexts

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

    The neuron fires on code tokens that form chained property or index access expressions (e.g. object.property_subproperty or array[index]).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     and
    -0.92
     of
    -0.89
    mäh
    -0.85
    boru
    -0.85
     internetu
    -0.84
     konkrét
    -0.84
     генерал
    -0.84
    uksessa
    -0.81
    CURSO
    -0.81
    締切
    -0.80
    POSITIVE LOGITS
    beforeSend
    1.02
    nextProps
    0.94
    otea
    0.90
    lena
    0.89
     diamantes
    0.85
     bhp
    0.84
     урна
    0.83
     확인함
    0.83
     imprimé
    0.82
    rinted
    0.80
    Activations Density 0.010%

    No Known Activations