© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 111234
    Prev
    Next
    INDEX
    Explanations

    detecting ampersand connective

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

    Looking at the activations, the neuron consistently activates strongly (ranging from ~156 to 198) on the ampersand symbol "&" when it appears between two related items or concepts in a list or pairing. **The neuron detects the ampersand symbol

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron responds to occurrences of the ampersand character (“&”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     has
    -2.09
     And
    -1.96
     you
    -1.79
     any
    -1.75
     to
    -1.73
     Not
    -1.73
     и
    -1.72
     about
    -1.66
     they
    -1.64
     That
    -1.56
    POSITIVE LOGITS
    闐
    1.73
    注意事
    1.71
     geniales
    1.69
     ANOTHER
    1.66
      (
    1.66
     poils
    1.64
     oeil
    1.63
     köln
    1.63
     pittores
    1.62
     baumwolle
    1.61
    Activations Density 0.014%

    No Known Activations