© 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. 7251
    Prev
    Next
    INDEX
    Explanations

    exclamations or surprise

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

    The neuron flags emphatic interjections and intensifiers (e.g. “What a…!”, “Boy…”, “O God,” “ever,” “greatly”) that signal strong emotional emphasis or exclamation.

    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
     হয়ত
    0.75
     because
    0.68
     '=
    0.68
    finex
    0.66
    스테
    0.66
     besten
    0.66
    '])[
    0.66
    ంలో
    0.65
     specials
    0.64
    很有
    0.63
    POSITIVE LOGITS
    !
    1.01
    !);
    0.98
    !
    0.96
    !;
    0.93
    !!
    0.88
    多么
    0.88
    !!!
    0.86
    !।
    0.86
    !!!!
    0.85
    !");
    0.85
    Activations Density 0.174%

    No Known Activations