© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10353
    Prev
    Next
    INDEX
    Explanations

    rust code

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

    The neuron primarily detects numeric tokens (e.g. version numbers, IDs, and other digit sequences).

    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
    Canada
    0.95
     Canada
    0.88
     Mississauga
    0.88
    CANADA
    0.85
     맛
    0.83
     CANADA
    0.83
     Ontario
    0.82
     Manitoba
    0.82
    UAE
    0.82
    atchewan
    0.81
    POSITIVE LOGITS
     usize
    1.14
    usize
    0.98
     crate
    0.95
     crates
    0.95
     eprintln
    0.95
     {:?}",
    0.89
    crates
    0.84
    serde
    0.84
     serde
    0.82
     nonempty
    0.77
    Activations Density 0.013%

    No Known Activations