© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 13589
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly flags prominent topic words—especially proper names, brands, products, and key service terms (e.g. “exhibition,” “Dubai,” “rental,” “contractors”)—essentially highlighting domain-specific or headline-style keywords.

    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
     vendita
    0.74
    ราคา
    0.69
     prezzo
    0.68
    t
    0.66
     usato
    0.66
    を購入
    0.63
     komb
    0.61
     preço
    0.61
     fixe
    0.60
     vyroben
    0.60
    POSITIVE LOGITS
    Oil
    0.66
     NSUTF
    0.61
     判断
    0.58
    และการ
    0.57
     qDebug
    0.57
    🛫
    0.57
    ্বেও
    0.56
    ♉
    0.55
    🤰
    0.55
     एवं
    0.55
    Activations Density 3.462%

    No Known Activations