© 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. 6342
    Prev
    Next
    INDEX
    Explanations

    mathematical notation and operators

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

    The neuron fires on numeric tokens—especially decimal and measurement values—i.e. it picks out numbers (including floats) in the text.

    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
     factors
    0.66
     asymptotic
    0.64
     ±
    0.62
    factors
    0.57
     factor
    0.57
    ±
    0.56
    factor
    0.55
     عوامل
    0.55
     conjugates
    0.55
     factored
    0.55
    POSITIVE LOGITS
    -(-
    0.85
    ((-
    0.76
     सूट
    0.75
    --(
    0.71
    }-(
    0.70
     (($
    0.70
     ((
    0.67
    {(-
    0.66
    。(
    0.66
     [(
    0.65
    Activations Density 0.013%

    No Known Activations