© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128640
    Prev
    Next
    INDEX
    Explanations

    The neuron detects backslash characters (i.e. LaTeX/TeX command initiators) in mathematical formula markup.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    mathematical symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     (“
    -1.90
    締
    -1.89
    は
    -1.76
    があり
    -1.75
     保証
    -1.73
    で
    -1.73
     hunde
    -1.73
     visuelle
    -1.72
    ’,
    -1.71
    が
    -1.71
    POSITIVE LOGITS
    튿
    2.42
     acogedor
    2.14
     \
    2.11
     acoged
    2.06
    苤
    2.05
     impresionantes
    2.02
     maravillosas
    1.98
     alrededores
    1.98
     gorgeous
    1.98
    ところです
    1.89
    Activations Density 0.022%

    No Known Activations