© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 8219
    Prev
    Next
    INDEX
    Explanations

    which, który, который

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

    The neuron fires on German-language words or morphemes.

    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
    はもちろん
    0.69
    cap
    0.69
    less
    0.67
    ,
    0.67
    airement
    0.66
    ividual
    0.66
    cam
    0.66
    i
    0.66
    there
    0.65
    oresis
    0.65
    POSITIVE LOGITS
     który
    1.02
     jotka
    0.97
     که
    0.95
     который
    0.95
     które
    0.91
     whose
    0.90
     która
    0.89
     التي
    0.88
     которые
    0.87
     který
    0.87
    Activations Density 0.008%

    No Known Activations