© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8363
    Prev
    Next
    INDEX
    Explanations

    geometric dimensions

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

    The neuron is keyed to tokens representing numeric values (especially large or floating‐point 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_31_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
     B
    0.73
     T
    0.70
     ,
    0.70
     O
    0.69
     of
    0.69
     to
    0.67
     in
    0.64
     S
    0.64
     for
    0.62
     V
    0.62
    POSITIVE LOGITS
    ንዳንድ
    0.71
    ያንዳንዱ
    0.69
    блиоте
    0.66
     машиналары
    0.66
     машинасы
    0.63
    実際の
    0.62
     машиналарын
    0.59
    ሁሉም
    0.59
    боры
    0.58
    ట్స్‌మన్
    0.58
    Activations Density 0.001%

    No Known Activations