© 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. 22-GEMMASCOPE-RES-131K
    7. 35429
    Prev
    Next
    INDEX
    Explanations

    html document components

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

    The neuron strongly responds to runs of leading whitespace (indentation) before markup tags.

    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
    illerato
    -0.99
     palestra
    -0.97
     аксе
    -0.95
     lampada
    -0.93
    indeki
    -0.93
     росли
    -0.91
    diens
    -0.90
     terape
    -0.90
    文化的
    -0.89
     terapi
    -0.89
    POSITIVE LOGITS
     and
    0.97
    ズル
    0.89
     بذلك
    0.88
     "
    0.88
     atractivos
    0.86
     who
    0.84
    MM
    0.83
    æder
    0.83
     Cox
    0.82
     дают
    0.82
    Activations Density 0.007%

    No Known Activations