© 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. 1775
    Prev
    Next
    INDEX
    Explanations

    towards abstract/social concepts

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

    The neuron consistently activates on content words naming key concepts or outcomes—especially nouns like “risk,” “protection,” “withdrawal,” “satisfaction,” “tolerance,” etc.

    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
    𝗇
    0.71
    contador
    0.70
    compan
    0.68
    Ку
    0.67
    く
    0.65
    culpa
    0.64
    Eternal
    0.64
    나
    0.64
    있는
    0.63
    بے
    0.63
    POSITIVE LOGITS
     populations
    0.85
     biomedical
    0.75
     neuroscience
    0.73
     psychosocial
    0.72
     pharmaceuticals
    0.72
     subpopulations
    0.71
     neuropsych
    0.70
     biomed
    0.67
     pharmacological
    0.67
     subgroups
    0.66
    Activations Density 0.000%

    No Known Activations