© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 2878
    Prev
    Next
    INDEX
    Explanations

    places and businesses

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

    The neuron fires strongly on title‐case or headline words—particularly proper nouns in headings (e.g. company names, publication titles, product or location names).

    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
     then
    -1.16
    兩人
    -1.02
    のおすすめ
    -1.00
     disparition
    -0.99
    Paket
    -0.99
     seminario
    -0.97
    Размеры
    -0.94
    beging
    -0.93
    estatus
    -0.93
    两者
    -0.91
    POSITIVE LOGITS
     &
    1.20
     is
    1.11
     препратки
    1.02
    queur
    1.00
    ={{
    0.99
    =}
    0.99
    :],
    0.98
    ∫
    0.96
    ('\
    0.96
     \%}$
    0.96
    Activations Density 0.078%

    No Known Activations