© 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. 6435
    Prev
    Next
    INDEX
    Explanations

    quoted text classification instruction

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

    The neuron strongly activates on tokens that begin new segments or lines (e.g. section headings or the first word of a paragraph).

    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
    覺得
    0.82
     मोहन
    0.75
    MOS
    0.75
    但是
    0.75
    堊
    0.74
     Цвет
    0.74
     oleh
    0.73
     ys
    0.73
    以為
    0.73
    這些
    0.72
    POSITIVE LOGITS
    ’
    0.88
    ix
    0.83
    un
    0.80
    ,’
    0.79
    ів
    0.73
    ’”
    0.73
    irection
    0.72
    isso
    0.71
    aluronic
    0.70
    liquidacion
    0.70
    Activations Density 0.000%

    No Known Activations