© 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. 34795
    Prev
    Next
    INDEX
    Explanations

    sewing or business names

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

    The neuron fires strongly on numbers (dates, measurements, statistics, addresses) and the punctuation that typically surrounds them.

    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
     will
    -1.59
     should
    -1.54
     if
    -1.45
     would
    -1.45
     may
    -1.38
     your
    -1.29
     provide
    -1.26
     suggested
    -1.20
     in
    -1.20
     useful
    -1.17
    POSITIVE LOGITS
     مقالات
    1.38
     satisfactor
    1.36
     xadrez
    1.31
     unicórnio
    1.26
    ໋
    1.23
     tuesday
    1.22
    “(
    1.22
     WANT
    1.16
    vielen
    1.14
    according
    1.13
    Activations Density 0.053%

    No Known Activations