© 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. 3580
    Prev
    Next
    INDEX
    Explanations

    SELECT, DISTINCT, sort, init

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

    The neuron flags formal instruction and disclaimer wording—terms often found in user‐obligation or legal‐notice passages (e.g. “expected,” “advance,” “contact,” “terms,” “conditions,” “ensure,” “download,” “acknowledge”).

    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
     LongNumber
    0.82
    ಿದರು
    0.77
    لای
    0.76
    养老
    0.75
     войны
    0.75
    娲
    0.75
     fudai
    0.75
     Senado
    0.75
     leprosy
    0.74
     legions
    0.72
    POSITIVE LOGITS
     fotograf
    0.95
    ↵
    0.91
     papar
    0.82
     couvert
    0.75
     o
    0.73
    ings
    0.73
     о
    0.73
     spanning
    0.71
     april
    0.71
     ook
    0.70
    Activations Density 0.002%

    No Known Activations