© 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. 10072
    Prev
    Next
    INDEX
    Explanations

    multilingual text fragments

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

    The neuron is activated by numeric tokens in reference metadata—e.g. page numbers, volume/issue identifiers, year ranges, and other digit sequences in citations.

    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
    cdot
    0.68
     skating
    0.68
     cycle
    0.66
     night
    0.66
     he
    0.64
     could
    0.64
     began
    0.63
     tall
    0.63
     -
    0.63
     teeth
    0.63
    POSITIVE LOGITS
    众多
    0.75
    偈
    0.74
    细腻
    0.72
    льника
    0.71
    ιών
    0.69
     wenigen
    0.67
     mêmes
    0.66
    ให้น
    0.65
    抠
    0.64
    CaO
    0.64
    Activations Density 0.093%

    No Known Activations