© 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. 3890
    Prev
    Next
    INDEX
    Explanations

    mathematical formulas

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

    This neuron activates on numeric literals (especially decimal or large numeric constants) in code snippets and formulas.

    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
     smrti
    0.61
    থিল
    0.56
     ezzel
    0.54
    mez
    0.54
    hez
    0.54
    henyl
    0.53
     nondegenerate
    0.53
    껐
    0.53
     Compagnie
    0.52
     assass
    0.52
    POSITIVE LOGITS
     বেলু
    0.59
     검색
    0.59
     дав
    0.55
     дъл
    0.53
     форум
    0.53
     suficientes
    0.52
     +
    0.52
    ఋ
    0.52
     بلوچستان
    0.51
     ห
    0.51
    Activations Density 0.001%

    No Known Activations