© 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. 10-GEMMASCOPE-RES-131K
    7. 22268
    Prev
    Next
    INDEX
    Explanations

    last names of people

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

    This neuron selectively activates on capitalized personal names (particularly surnames) in sports write-ups.

    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
     rédig
    -1.44
    țiuni
    -1.39
     nový
    -1.39
     payé
    -1.38
    iyle
    -1.36
     páns
    -1.35
     dámské
    -1.35
     vertellen
    -1.34
     kertoo
    -1.34
     váš
    -1.30
    POSITIVE LOGITS
     other
    1.96
     we
    1.54
     другими
    1.44
     others
    1.43
     anderen
    1.35
     altri
    1.33
     our
    1.32
    为大家
    1.32
     who
    1.29
     outros
    1.26
    Activations Density 0.062%

    No Known Activations