© 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. 13843
    Prev
    Next
    INDEX
    Explanations

    a niche

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

    The neuron strongly activates on numeric tokens—especially decimal‐formatted numbers and page or volume references as found in academic citations.

    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
    crever
    0.84
     matières
    0.83
    嫘
    0.80
     variétés
    0.78
     EGFR
    0.77
     Zariski
    0.75
    示す
    0.75
    جون
    0.75
     démon
    0.75
     algebras
    0.74
    POSITIVE LOGITS
    го
    0.73
    ্স
    0.72
    𝑠
    0.72
    in
    0.71
    *,
    0.71
     zuletzt
    0.71
     weren
    0.70
    ̦
    0.69
    est
    0.69
    АО
    0.68
    Activations Density 0.002%

    No Known Activations