© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 13739
    Prev
    Next
    INDEX
    Explanations

    Py array, Corresponding Table, technological surveillance

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

    The neuron fires strongly on the first “real” content word of a new instruction or paragraph – in other words, it flags section or step beginnings.

    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
     vết
    0.82
     vr
    0.77
     AS
    0.71
    ('--
    0.71
    gé
    0.68
     décret
    0.68
     thèse
    0.68
     এস
    0.67
     시행
    0.67
     \'
    0.66
    POSITIVE LOGITS
    率为
    0.88
    4
    0.86
    的时间
    0.84
    തിരെ
    0.82
    PLICATIONS
    0.81
    ،
    0.79
    สำหรับการ
    0.77
    的水
    0.77
    5
    0.77
    ンド
    0.74
    Activations Density 0.001%

    No Known Activations