© 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. 22-GEMMASCOPE-RES-131K
    7. 70638
    Prev
    Next
    INDEX
    Explanations

    public, private, virtual, static

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

    The neuron is looking for C++ operator‐overload definitions (i.e. occurrences of the “operator” keyword in code).

    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
    same
    -1.02
     protagonista
    -0.88
     same
    -0.87
    Ȟ
    -0.85
     efekty
    -0.85
    Podob
    -0.85
     gitara
    -0.84
     intelig
    -0.82
    と思いました
    -0.82
    tidas
    -0.81
    POSITIVE LOGITS
     will
    1.20
     static
    1.05
    public
    0.89
     alltså
    0.88
     wollen
    0.86
    private
    0.85
    static
    0.85
    吁
    0.84
     abstract
    0.84
     void
    0.82
    Activations Density 0.047%

    No Known Activations