© 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. 5031
    Prev
    Next
    INDEX
    Explanations

    explains why

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

    The neuron strongly activates on proper‐noun tokens and special UI/linkage symbols (e.g. brand or site names like “ForexTime,” “LinkedIn,” location labels like “China,” and navigation markers like “>>” or “›”).

    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
    ను
    0.81
    noon
    0.79
    {
    0.73
    rische
    0.73
    presentation
    0.72
    となる
    0.71
    becomes
    0.70
    சால
    0.69
    பட
    0.68
    stitches
    0.68
    POSITIVE LOGITS
    ุ
    0.80
     Архиви
    0.71
     Secondly
    0.71
     Послед
    0.71
    ික
    0.70
     木
    0.70
    鱿
    0.69
     oczy
    0.68
    팜
    0.68
    USA
    0.68
    Activations Density 0.001%

    No Known Activations