© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2788
    Prev
    Next
    INDEX
    Explanations

    column A

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

    The neuron strongly activates on numeric tokens (e.g. digits, counts, measurements, episode or section numbers) indicating it’s looking for 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
    the
    0.91
    pro
    0.83
    path
    0.71
    stat
    0.71
    о
    0.70
    file
    0.70
    al
    0.70
    analytics
    0.70
    ’
    0.70
    ocean
    0.69
    POSITIVE LOGITS
     तीन
    1.90
     three
    1.77
     drei
    1.75
     δύο
    1.75
     दोन
    1.73
     TWO
    1.73
     três
    1.73
     dwóch
    1.72
     two
    1.71
     zwei
    1.71
    Activations Density 1.068%

    No Known Activations