© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14549
    Prev
    Next
    INDEX
    Explanations

    introductions and political language

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

    This neuron consistently spikes on the very first token of each new segment (the token immediately after a <start>), effectively marking paragraph or segment boundaries.

    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
    s
    0.97
    ب
    0.90
    sama
    0.89
    ed
    0.88
    د
    0.87
    D
    0.85
     của
    0.85
    R
    0.84
    sold
    0.83
    A
    0.82
    POSITIVE LOGITS
     Politique
    0.86
     Politicians
    0.85
     zemlji
    0.79
     लड़का
    0.77
     rumores
    0.76
     преступ
    0.74
     politiques
    0.73
     asesinato
    0.71
     государство
    0.71
     remoto
    0.71
    Activations Density 0.000%

    No Known Activations