© 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. 10-GEMMASCOPE-RES-131K
    7. 75492
    Prev
    Next
    INDEX
    Explanations

    HTML, PHP, code tags

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

    The neuron spikes on long runs of leading whitespace (indentation) 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_10/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
     Saltar
    -1.53
    ↵↵↵↵↵
    -1.36
    <h2>
    -1.33
     "
    -1.31
     an
    -1.28
     a
    -1.27
     which
    -1.27
    ↵↵↵↵
    -1.25
    ↵↵↵↵↵↵
    -1.24
     still
    -1.23
    POSITIVE LOGITS
     tunik
    1.45
     drugim
    1.38
    ngày
    1.32
    旆
    1.31
    ﹙
    1.29
    瘆
    1.29
    ancang
    1.27
     précip
    1.27
     instruk
    1.27
    Ohne
    1.27
    Activations Density 0.021%

    No Known Activations