© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9737
    Prev
    Next
    INDEX
    Explanations

    mathematical integrals, integration, sums

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

    The neuron fires on tokens that denote numeric values, variables, or mathematical terms within formulaic or code-like contexts.

    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
     Saver
    0.59
     reception
    0.56
     Reception
    0.55
    灲
    0.55
     craftsmanship
    0.53
     concession
    0.52
     transnational
    0.51
     Crafts
    0.51
     bulan
    0.50
    診
    0.50
    POSITIVE LOGITS
     integrand
    1.13
    ∫
    1.12
     интегра
    1.04
    積分
    1.02
     ∫
    1.02
    积分
    0.95
    integr
    0.95
     التكامل
    0.94
     Integr
    0.93
    
    0.93
    Activations Density 0.044%

    No Known Activations