© 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. 10908
    Prev
    Next
    INDEX
    Explanations

    LLMs

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

    The neuron fires strongly on proper names and product or brand terms—i.e. capitalized entity names (e.g. “Safflower,” “Mews,” “Wiking,” “Costa,” “Zovirax”) within the text.

    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
    에서도
    0.55
    ’
    0.52
    PED
    0.51
    ๆ
    0.51
    '
    0.50
     발생
    0.50
    olm
    0.49
    fuels
    0.49
    /
    0.49
    ।
    0.49
    POSITIVE LOGITS
     Tidak
    0.69
     waktu
    0.63
     Jehova
    0.62
     tidak
    0.61
     Recherche
    0.61
     Fiverr
    0.60
     jarang
    0.59
     advertisers
    0.59
     Literatura
    0.57
     Durchschnitt
    0.57
    Activations Density 0.018%

    No Known Activations