© 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. 128304
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on typographic markers (em‐dashes “–” and colons) that denote section or feature headings in the text.

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

    your hand, you feel

    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
    tedy
    -1.45
    ším
    -1.41
    США
    -1.28
     ImportError
    -1.23
    آب
    -1.22
     nodig
    -1.16
    کری
    -1.16
     sombras
    -1.15
     what
    -1.14
     then
    -1.14
    POSITIVE LOGITS
     This
    1.20
     extremely
    1.16
    no
    1.16
     incredibly
    1.16
    this
    1.15
    看似
    1.12
     you
    1.12
    while
    1.11
     exceedingly
    1.11
    у
    1.09
    Activations Density 0.002%

    No Known Activations