© 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. 15738
    Prev
    Next
    INDEX
    Explanations

    `never` known, `never` seen, `backing` up, `William`, `final`, `heavily`

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

    The neuron fires strongly on the first substantive word of a newly started line or paragraph, marking the start of a new block of text.

    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
    ל
    0.86
    0.75
    يب
    0.68
    ん
    0.68
    *{
    0.65
    ش
    0.64
     viel
    0.64
    但是
    0.64
     rebut
    0.64
    ور
    0.63
    POSITIVE LOGITS
    greens
    0.90
     Mikh
    0.88
     विकल्पों
    0.81
    δο
    0.79
    രോധ
    0.79
    agamanam
    0.78
     incluidos
    0.78
    །
    0.78
     Sauce
    0.77
     карты
    0.77
    Activations Density 0.000%

    No Known Activations