© 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. 80716
    Prev
    Next
    INDEX
    Explanations

    names like Gayle, Dele, Cele

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

    The neuron activates on character sequences containing “le.”

    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
     stade
    -0.80
    iexcl
    -0.77
     AppCompatTheme
    -0.76
    JUD
    -0.75
    Ų
    -0.75
     GES
    -0.75
     وانت
    -0.73
    тельным
    -0.71
    court
    -0.71
     ссыл
    -0.71
    POSITIVE LOGITS
    れる
    0.71
     Luxury
    0.71
     przysz
    0.71
    erweise
    0.70
     colorida
    0.69
    épa
    0.68
    咐
    0.68
    mei
    0.66
    休息
    0.66
     inherently
    0.66
    Activations Density 0.015%

    No Known Activations