© 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. 2149
    Prev
    Next
    INDEX
    Explanations

    punctuation and sentence structure

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

    This neuron marks the first substantive content word right after each new segment (i.e. after each <start> boundary).

    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
     Букови
    0.85
     Як
    0.81
     ମ
    0.81
    クサ
    0.80
    Ტ
    0.80
     boda
    0.79
    лини
    0.78
    𐰴
    0.77
     Κα
    0.76
     buzzer
    0.75
    POSITIVE LOGITS
    );
    0.81
    ;
    0.79
    UN
    0.75
    )(
    0.72
     tav
    0.72
    dd
    0.71
    ;(
    0.70
    LL
    0.70
    ds
    0.69
    l
    0.69
    Activations Density 0.000%

    No Known Activations