© 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. 128737
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on the “&” (and occasionally on “and”) when it appears in multi-part organization or proper names (e.g. “X & Y” in company or author listings).

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

    company name suffixes

    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
    .
    -1.80
    on
    -1.45
     should
    -1.32
     get
    -1.27
     Konkur
    -1.19
     from
    -1.18
     нашем
    -1.16
     new
    -1.16
     -
    -1.16
     ("
    -1.13
    POSITIVE LOGITS
     cependant
    1.55
     یا
    1.49
    vably
    1.36
     toutefois
    1.36
     MÉXICO
    1.35
     DOMINGO
    1.30
    իվ
    1.30
    dersfield
    1.29
     decidedly
    1.28
     svært
    1.27
    Activations Density 0.039%

    No Known Activations