© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 70351
    Prev
    Next
    INDEX
    Explanations

    UN Secretary-General Ban Ki-moon

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

    The neuron fires on structured “metadata” tokens—numbers, punctuation/parentheses, uppercase identifiers, and other code‐ or markup‐style syntax elements.

    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
    justin
    -0.65
     Types
    -0.63
     ASA
    -0.63
     parapet
    -0.62
     muros
    -0.62
     types
    -0.62
     Stuck
    -0.61
    ambah
    -0.61
     кін
    -0.61
    いち
    -0.60
    POSITIVE LOGITS
    规划
    0.86
     Planner
    0.72
    xue
    0.71
    arkov
    0.71
     Ark
    0.70
    ew
    0.70
    空き
    0.69
     dc
    0.69
     qubit
    0.68
    cosity
    0.68
    Activations Density 0.081%

    No Known Activations