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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 16078
    Prev
    Next
    INDEX
    Explanations

    instructions for ai roles

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

    The neuron fires on formal descriptions of a tragic event or investigation, especially around location (“district,” “where”) and its victims (“lives,” “affected”).

    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
     famously
    0.83
    ’
    0.73
     your
    0.73
    鈥
    0.73
    r
    0.72
    **
    0.71
     arguably
    0.71
    K
    0.70
     burgeoning
    0.68
     bolstered
    0.66
    POSITIVE LOGITS
     kegiatan
    0.92
     činn
    0.90
     деятельности
    0.89
    を変更
    0.82
     atividades
    0.81
     actividades
    0.80
     діяльності
    0.79
     செய்தால்
    0.78
     aktivnosti
    0.78
     hereunder
    0.77
    Activations Density 0.509%

    No Known Activations