© 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. 31-GEMMASCOPE-2-RES-16K
    7. 12820
    Prev
    Next
    INDEX
    Explanations

    mathematics and algebra

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

    This neuron detects references to mathematics (e.g. “math,” “mathematician,” “Mathematical”).

    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
     मिशन
    0.73
     ड्राइविंग
    0.73
     Empresa
    0.66
    蓓
    0.63
    鲵
    0.63
     регули
    0.60
     Mojo
    0.60
     일정
    0.60
    瘪
    0.59
     Photoshop
    0.59
    POSITIVE LOGITS
     mathematicians
    2.44
     mathematical
    2.25
     Mathematical
    2.17
     mathematics
    2.16
    Mathematical
    2.16
     математи
    2.14
    数学
    2.11
     mathematician
    2.11
     Mathematic
    2.09
     matemat
    2.03
    Activations Density 0.254%

    No Known Activations