© 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. 1387
    Prev
    Next
    INDEX
    Explanations

    adjective order context

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

    The neuron is picking out numeric tokens—especially numbers and decimals (including section references and measurements).

    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
     sobreviv
    0.77
     apro
    0.75
     определя
    0.71
    閒
    0.71
    户
    0.71
    stripped
    0.70
    Opt
    0.69
     intérieur
    0.69
    來自
    0.69
    Wür
    0.68
    POSITIVE LOGITS
     จำนวน
    0.84
    nak
    0.83
    ಿಂದ
    0.83
     trener
    0.77
     નં
    0.75
     modals
    0.75
    পাড়া
    0.73
    ﾓ
    0.73
     keynote
    0.73
     당연
    0.73
    Activations Density 0.000%

    No Known Activations