© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4168
    Prev
    Next
    INDEX
    Explanations

    spacing and line breaks

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

    This neuron fires on tokens specifying numerical formatting or layout parameters—e.g. font sizes, spacing measurements, page or text dimensions.

    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
     criaturas
    0.65
    有一天
    0.64
    %@",
    0.63
    今天是
    0.63
     apuestas
    0.60
     game
    0.59
     aniversario
    0.59
    jogo
    0.58
     kavram
    0.57
    i
    0.57
    POSITIVE LOGITS
     الخط
    0.71
     AcOEt
    0.68
    RIAL
    0.68
    ất
    0.64
     வள்ள
    0.64
    álás
    0.64
    conv
    0.63
     LINES
    0.63
     RMB
    0.63
    ów
    0.63
    Activations Density 0.014%

    No Known Activations