© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9796
    Prev
    Next
    INDEX
    Explanations

    injection into, build a, melt and, Premium, inhaled into, Employer, monopolies, download

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

    The neuron flags formal obligation- and duty-related nouns (e.g. duty, obligation, role, power).

    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
    .
    0.96
    9
    0.95
    ique
    0.92
    ,
    0.87
     Jackson
    0.86
     Davis
    0.85
     Williams
    0.85
    0
    0.82
    4
    0.79
    ↵↵
    0.77
    POSITIVE LOGITS
     таких
    1.09
    こういう
    1.09
     মানুষদের
    1.05
    人々
    0.99
     লোকদের
    0.98
    このような
    0.97
     آن‌ها
    0.96
     такие
    0.95
    Estos
    0.95
     মানুষকে
    0.94
    Activations Density 3.828%

    No Known Activations