© 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. 13950
    Prev
    Next
    INDEX
    Explanations

    scripting and commands

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

    The neuron fires on section‐header–style tokens—things like document labels, headings, dates, and other metadata markers (e.g. “font,” “Wednesday,” “JAN /”).

    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
     toll
    0.78
     tolls
    0.78
    數量
    0.75
     oli
    0.75
     highlights
    0.73
    șa
    0.73
    點
    0.73
    
    0.73
     helps
    0.71
    ᄈ
    0.71
    POSITIVE LOGITS
    MAGNET
    0.78
    те
    0.75
    REIB
    0.74
    вате
    0.74
    WINDOW
    0.73
    теле
    0.71
     receptacles
    0.71
    TRAN
    0.70
    デイ
    0.69
    OT
    0.69
    Activations Density 0.001%

    No Known Activations