© 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. 34-GEMMASCOPE-RES-131K
    7. 58352
    Prev
    Next
    INDEX
    Explanations

    names starting with certain syllables

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

    The neuron strongly activates on capitalized name fragments and acronyms, i.e. tokens that are parts of proper names or named entities.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    zący
    -1.04
     himself
    -1.00
     brother
    -0.90
    基準
    -0.88
     uncle
    -0.84
    inius
    -0.82
    ūnas
    -0.82
    utus
    -0.81
     spokesman
    -0.80
    ロケ
    -0.80
    POSITIVE LOGITS
     herself
    1.13
    odora
    1.08
    trud
    1.03
     Thị
    1.00
     Kaur
    1.00
     caixas
    0.93
    ronique
    0.93
     nzuri
    0.93
    ώρα
    0.92
    mogen
    0.91
    Activations Density 0.093%

    No Known Activations