© 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. 15289
    Prev
    Next
    INDEX
    Explanations

    The neuron activates strongly on URL domain parts (e.g. “com,” “org,” “gov”), i.e. top-level domain tokens in web addresses.

    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
     כמ
    0.75
     brooding
    0.70
     damp
    0.68
     instar
    0.67
    䀘
    0.67
     supper
    0.65
     eraser
    0.65
     diction
    0.64
     ered
    0.64
     étages
    0.64
    POSITIVE LOGITS
    VAN
    0.79
     prefeito
    0.75
    ä
    0.73
    μα
    0.71
    ガ
    0.71
    CVE
    0.70
    ਤਰ
    0.69
     Congresso
    0.69
    َا
    0.68
    타
    0.68
    Activations Density 0.000%

    No Known Activations