© 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. 10012
    Prev
    Next
    INDEX
    Explanations

    data loss prevention and documentatio

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

    The neuron specializes in spotting the tokens that occur in or adjacent to hyphens in hyphenated compound phrases.

    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
    щий
    0.75
    цию
    0.70
    setHorizontal
    0.68
    гра
    0.67
    ylmethyl
    0.67
     шей
    0.64
    ttet
    0.64
    䊀
    0.64
    зы
    0.64
    isHorizontal
    0.63
    POSITIVE LOGITS
     dulu
    0.90
     legislation
    0.88
     perusahaan
    0.87
     scandal
    0.85
     Singapore
    0.83
     pokaż
    0.83
    y
    0.82
    ’
    0.81
     กล่าว
    0.80
     società
    0.80
    Activations Density 2.641%

    No Known Activations