© 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. 2887
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on words naming business sectors or industries (e.g. “retail,” “e-commerce,” “industry,” “sector”).

    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
    ਮੀ
    0.64
     earphones
    0.63
     파일을
    0.62
     জ
    0.59
    设备的
    0.59
     पी
    0.58
    ädig
    0.58
     گوشی
    0.58
    ოველ
    0.58
    噸
    0.58
    POSITIVE LOGITS
     paradigm
    0.83
     realm
    0.79
    领域
    0.79
     ámbito
    0.72
     triad
    0.71
     system
    0.71
     pillar
    0.69
     arena
    0.69
     continuum
    0.68
     domain
    0.68
    Activations Density 0.340%

    No Known Activations