© 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. 5612
    Prev
    Next
    INDEX
    Explanations

    content:

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

    The neuron fires on numeric tokens that are part of tennis match set scores (e.g. the games‐won numbers in “6-2,” “7-6,” 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
    ्स
    0.88
    ין
    0.84
    ח
    0.81
    마
    0.81
    اس
    0.80
    dah
    0.79
    nueva
    0.79
    ುದು
    0.78
    ل
    0.78
    אם
    0.77
    POSITIVE LOGITS
     Teich
    0.71
     Features
    0.70
     COMPET
    0.69
     Abbreviations
    0.69
     Kein
    0.69
    EK
    0.67
     Estab
    0.66
     Listings
    0.66
     Annals
    0.65
     கும்ப
    0.65
    Activations Density 0.001%

    No Known Activations