© 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. 68499
    Prev
    Next
    INDEX
    Explanations

    structural separators and code elements

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

    This neuron activates on rows of ASCII-table divider characters (e.g. lines of dashes or pipes) that mark table borders.

    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
     also
    -1.09
     another
    -1.00
     however
    -0.97
     Although
    -0.95
    ूरत
    -0.94
    However
    -0.94
     However
    -0.90
    鬢
    -0.89
     Also
    -0.88
    』。
    -0.88
    POSITIVE LOGITS
     yılında
    1.03
     none
    1.00
    ęcie
    0.99
    none
    0.92
     None
    0.91
     napisał
    0.91
     представителей
    0.90
    !
    0.90
     putative
    0.90
     senza
    0.89
    Activations Density 0.008%

    No Known Activations