© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 82682
    Prev
    Next
    INDEX
    Explanations

    insulinoma, simulans, ultraluminous

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

    The neuron strongly activates on any token containing the character sequence “ul.”

    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
    eland
    -0.83
     Witnesses
    -0.75
    ごはん
    -0.75
    立派
    -0.75
    izante
    -0.75
    ไตล์
    -0.75
     jiwa
    -0.73
    filetype
    -0.69
     Mankind
    -0.68
    каун
    -0.68
    POSITIVE LOGITS
     inclination
    0.89
    드
    0.85
    livan
    0.75
    bula
    0.73
    ارتمان
    0.72
    лич
    0.71
    ड
    0.71
    如何
    0.71
     regular
    0.70
    ski
    0.68
    Activations Density 0.069%

    No Known Activations