© 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. 22-GEMMASCOPE-RES-131K
    4. 67864
    Prev
    Next
    INDEX
    Explanations

    invitations to join, participate, or attend

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

    The neuron fires on invitation or request phrases—that is, tokens in segments where someone is being asked, invited, or offered to do something.

    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
     trenes
    -0.97
     and
    -0.96
     glisser
    -0.96
     hábit
    -0.95
     tromper
    -0.94
     manguera
    -0.94
     např
    -0.93
    Trying
    -0.93
     relying
    -0.92
    wirkungen
    -0.90
    POSITIVE LOGITS
     because
    1.28
     for
    1.20
    dienne
    1.11
    Historio
    1.09
     some
    1.02
     idé
    1.02
    ニューアル
    1.02
     an
    1.02
    遽
    0.96
    CRIPTION
    0.96
    Activations Density 0.045%

    No Known Activations