© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 8671
    Prev
    Next
    INDEX
    Explanations

    The neuron is specialized for spotting the “e-” hyphenated prefix (as in e-commerce, e-mail, e-health, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ar
    1.63
    یر
    1.58
    at
    1.56
    y
    1.56
    u
    1.54
    ли
    1.49
    lardan
    1.42
    ي
    1.41
    al
    1.38
    le
    1.36
    POSITIVE LOGITS
     thorac
    0.95
    ୍
    0.94
    ৯
    0.92
     conscience
    0.89
     funkt
    0.88
    ভাবে
    0.86
     resten
    0.86
     counterfe
    0.85
    써
    0.85
    лки
    0.84
    Activations Density 0.359%

    No Known Activations