© 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. 34-GEMMASCOPE-RES-131K
    7. 11926
    Prev
    Next
    INDEX
    Explanations

    programming language keywords and identifiers

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

    The neuron is primarily detecting semicolon tokens, i.e. statement terminators 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_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
     política
    -0.81
     الوطن
    -0.77
     hilo
    -0.77
    ndice
    -0.75
    こんにちは
    -0.74
     édition
    -0.73
     torno
    -0.73
     like
    -0.71
    quire
    -0.71
     apples
    -0.71
    POSITIVE LOGITS
    constexpr
    1.16
     these
    0.94
    these
    0.93
    static
    0.92
     uusi
    0.90
     kimia
    0.89
     static
    0.88
     konstant
    0.88
     således
    0.87
    const
    0.84
    Activations Density 0.007%

    No Known Activations