© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92815
    Prev
    Next
    INDEX
    Explanations

    kur followed by various endings

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

    The neuron consistently activates on tokens containing the letter sequence “kur” (e.g. “Kur,” “Kurth,” “kur”).

    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
     certific
    -0.82
    melten
    -0.74
    服务的
    -0.74
    鸪
    -0.73
    iencing
    -0.72
     Interrupt
    -0.72
     naming
    -0.71
    質
    -0.70
    naming
    -0.70
    ciare
    -0.70
    POSITIVE LOGITS
     Kur
    1.14
     kur
    1.13
    Kur
    0.93
     القر
    0.78
     KUR
    0.77
     кур
    0.76
     Kurt
    0.74
    BOOL
    0.73
    kur
    0.73
    Sae
    0.71
    Activations Density 0.008%

    No Known Activations