© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66513
    Prev
    Next
    INDEX
    Explanations

    places and names starting with B, L, or Mc

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

    The neuron fires on Irish place names—particularly those starting with “Bally” (e.g. Ballycar Castle, Ballymena)—i.e. geographic proper names in Ireland.

    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
     is
    -1.45
    ौन
    -1.40
     แต่
    -1.34
     самая
    -1.34
    好人
    -1.33
    龶
    -1.32
     Теперь
    -1.30
     Ahora
    -1.30
     pasaron
    -1.30
    龵
    -1.30
    POSITIVE LOGITS
    ",
    1.83
    жению
    1.72
     ‘
    1.66
    ’
    1.66
    .”
    1.54
    .’
    1.48
     сист
    1.45
    ,”
    1.45
     uska
    1.44
     (…)
    1.39
    Activations Density 0.033%

    No Known Activations