© 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. 13952
    Prev
    Next
    INDEX
    Explanations

    ships disappearing hull first

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

    The neuron fires on structured metadata or heading‐style snippets—things like category paths, course codes, PDF/form reference lines, or other formal label blocks.

    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
     И
    0.81
    ’
    0.79
    '
    0.77
    Па
    0.73
    Read
    0.68
     така
    0.68
    И
    0.68
     talented
    0.68
    Path
    0.68
    IN
    0.66
    POSITIVE LOGITS
    oxid
    0.83
     höheren
    0.80
    न्दु
    0.79
    媟
    0.79
     orbifold
    0.78
    má
    0.77
    ikam
    0.77
     sır
    0.77
    드의
    0.77
    ე
    0.77
    Activations Density 0.000%

    No Known Activations