© 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. 8366
    Prev
    Next
    INDEX
    Explanations

    numbers, N, K

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

    The neuron fires strongly on tokens naming or referring to downloadable documents (e.g. “download,” “pdf,” or publication titles), i.e. it spots mentions of PDF/document downloads.

    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
    Ссылки
    0.82
    the
    0.77
     لوگ
    0.77
     ஃப
    0.76
    ן
    0.75
     миллиар
    0.74
    ды
    0.74
     ਬ
    0.74
    sej
    0.73
     ਕ
    0.72
    POSITIVE LOGITS
     Pyrénées
    0.80
     décider
    0.77
     généraux
    0.77
     Nationality
    0.73
     Kochubei
    0.73
     género
    0.71
     vêtements
    0.71
     travaux
    0.71
     Estudios
    0.71
    ][/
    0.71
    Activations Density 0.003%

    No Known Activations