© 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. 128456
    Prev
    Next
    INDEX
    Explanations

    The neuron detects occurrences of the bibliographic editor abbreviation “ed” or “eds.”

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

    editors and trustees

    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
     to
    -1.64
     womit
    -1.04
     use
    -1.02
     about
    -0.95
     have
    -0.94
     know
    -0.93
     umgangssprach
    -0.92
     make
    -0.90
     obwohl
    -0.90
     they
    -0.89
    POSITIVE LOGITS
     sportif
    1.31
    Beberapa
    1.28
    Penjelasan
    1.27
    Kemudian
    1.26
     manifeste
    1.25
     hırka
    1.24
     conva
    1.23
    Namun
    1.23
     joyeux
    1.21
     réve
    1.19
    Activations Density 0.012%

    No Known Activations