© 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. 130327
    Prev
    Next
    INDEX
    Explanations

    The neuron responds to isolated punctuation or other non-alphanumeric symbols (e.g. “?”, “<”, “|”, “~”, “&”, etc.).

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

    symbols and operators

    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
    Should
    -1.54
    Here
    -1.53
    What
    -1.53
    Need
    -1.44
    Our
    -1.40
    Really
    -1.38
    About
    -1.34
    おわりに
    -1.33
    You
    -1.33
    влении
    -1.31
    POSITIVE LOGITS
     as
    1.52
    ledad
    1.43
    内外
    1.37
     that
    1.32
     iedereen
    1.28
    oczes
    1.27
     daire
    1.27
     participado
    1.27
    creens
    1.18
     sebagian
    1.18
    Activations Density 0.056%

    No Known Activations