© 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. 11602
    Prev
    Next
    INDEX
    Explanations

    Bootstrap, Layer, function, health, policies

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

    The neuron flags occurrences of numeric data—words or tokens associated with statistics, measurements, or figures.

    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
    ти
    0.95
    ري
    0.88
    त
    0.85
    ли
    0.81
    ла
    0.79
    cích
    0.77
    ро
    0.77
    ਆਂ
    0.77
     عنا
    0.75
    х
    0.74
    POSITIVE LOGITS
    isan
    0.86
     aussi
    0.79
     encour
    0.78
    ጅም
    0.76
     ankaŭ
    0.75
    inton
    0.75
    anch
    0.74
     également
    0.74
    anju
    0.74
     ook
    0.73
    Activations Density 0.000%

    No Known Activations