© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6334
    Prev
    Next
    INDEX
    Explanations

    Configuration, Efficiency, unique, significant, product

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

    The neuron lights up on salient content words—especially proper names, product titles, or key-noun concepts—marking the main “about-what” terms in a passage.

    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
     sogenannte
    0.96
    lera
    0.89
    i
    0.89
    lz
    0.89
    eber
    0.88
    citing
    0.87
    cito
    0.86
    ecek
    0.83
     улицы
    0.82
    sah
    0.82
    POSITIVE LOGITS
    や
    0.93
     smoot
    0.79
    目前
    0.78
    等
    0.78
    或
    0.75
    光
    0.75
    сть
    0.74
     sàng
    0.73
    神奇
    0.73
    能
    0.72
    Activations Density 0.001%

    No Known Activations