© 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. 10113
    Prev
    Next
    INDEX
    Explanations

    single or double quotes

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

    The neuron primarily activates on numerical tokens and number‐related expressions (e.g. digits, percentages, prices, measurements).

    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
     Paise
    0.80
     $("<
    0.79
     주인공
    0.77
     Studying
    0.75
     accompagne
    0.73
     Просто
    0.71
     Ичиго
    0.71
    $};
    0.71
     Пло
    0.71
     ११
    0.70
    POSITIVE LOGITS
     vê
    0.86
    al
    0.79
     করিয়ে
    0.77
    i
    0.77
    ស
    0.77
     guru
    0.77
     nonsense
    0.77
     gere
    0.76
    データの
    0.75
    ac
    0.75
    Activations Density 0.000%

    No Known Activations