© 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. 112402
    Prev
    Next
    INDEX
    Explanations

    Julies and Garrys followed by names

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

    The neuron strongly activates on personal first names (e.g. “Garry,” “Julie”).

    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
    '
    -3.42
    苹
    -2.88
    ꪹ
    -2.67
    ⭓
    -2.61
    鸨
    -2.59
    柺
    -2.55
    ？！”
    -2.52
     grün
    -2.52
     klein
    -2.42
     limitado
    -2.42
    POSITIVE LOGITS
    When
    3.78
    0
    3.47
    After
    3.36
    Since
    3.36
    It
    3.25
    Only
    3.20
    Despite
    3.16
     После
    3.09
    We
    3.06
    There
    3.06
    Activations Density 0.005%

    No Known Activations