© 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. 5007
    Prev
    Next
    INDEX
    Explanations

    The neuron responds most strongly to tokens that are names, trademarks, or numeric/measurement terms—that is, proper nouns and numbers.

    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
    setIcon
    0.47
    返回值
    0.44
    、
    0.43
    addKill
    0.43
    दर्शक
    0.42
    దని
    0.42
    ICIENCY
    0.41
    致
    0.41
     завер
    0.40
     சேர்த்து
    0.40
    POSITIVE LOGITS
     êtres
    0.60
     hommes
    0.57
     accidentes
    0.56
    hood
    0.54
    TZ
    0.54
    oodles
    0.54
    T
    0.54
     accountants
    0.53
    ায়
    0.52
     athletes
    0.52
    Activations Density 24.188%

    No Known Activations