© 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. 84202
    Prev
    Next
    INDEX
    Explanations

    words starting with Za, Zah, Zá

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

    The neuron is strongly triggered by the character sequence “Za,” especially at the start of names or country‐related terms.

    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
     Nagar
    -0.77
     ases
    -0.76
    tigste
    -0.73
    OrDefault
    -0.70
    complete
    -0.69
    商务
    -0.69
    lectron
    -0.69
    feiri
    -0.69
    ower
    -0.69
    tions
    -0.69
    POSITIVE LOGITS
    eringen
    0.78
    ăt
    0.74
     Za
    0.74
     jossa
    0.73
    missie
    0.71
    verifier
    0.70
     SPA
    0.69
     zoomed
    0.69
    udah
    0.68
     Marc
    0.68
    Activations Density 0.016%

    No Known Activations