© 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. 66689
    Prev
    Next
    INDEX
    Explanations

    public, private

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

    The neuron strongly activates on runs of leading whitespace (indentation) in code lines.

    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
    توضیحات
    -1.90
     if
    -1.84
     every
    -1.82
     it
    -1.79
     while
    -1.66
     more
    -1.63
     each
    -1.62
     even
    -1.61
     their
    -1.54
     its
    -1.51
    POSITIVE LOGITS
    private
    1.38
    public
    1.34
     inconvénients
    1.30
     alltså
    1.28
    '
    1.27
    reportWebVitals
    1.23
     quí
    1.23
     stadig
    1.18
     først
    1.17
     nemlig
    1.15
    Activations Density 0.027%

    No Known Activations