© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92030
    Prev
    Next
    INDEX
    Explanations

    C++ keywords

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

    The neuron fires on runs of (leading) whitespace—i.e. it detects code indentation.

    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
     kurie
    -0.80
     their
    -0.77
     όμως
    -0.76
    厭
    -0.76
    他們的
    -0.74
     Thomas
    -0.73
    ReadKey
    -0.73
     stojí
    -0.73
     like
    -0.73
     will
    -0.73
    POSITIVE LOGITS
    auto
    5.19
     auto
    4.72
    自动
    2.91
     авто
    2.86
     Auto
    2.80
     AUTO
    2.78
    自動
    2.77
    Auto
    2.70
     autos
    2.55
    AUTO
    2.50
    Activations Density 0.010%

    No Known Activations