© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8965
    Prev
    Next
    INDEX
    Explanations

    ASCII art separators

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

    This neuron strongly activates on sequences of punctuation marks and quotation‐style characters (e.g., strings of quotes, periods, ellipses, and other nonalphanumeric symbols).

    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_31_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
     $^{
    0.44
     ्
    0.43
     Acad
    0.43
    нь
    0.43
     അദ്ദേ
    0.43
    谀
    0.42
    ప
    0.42
     dd
    0.41
     thèse
    0.41
    ்த்த
    0.41
    POSITIVE LOGITS
    ••
    0.60
    •
    0.59
    ■
    0.57
    ">•
    0.57
    ・
    0.52
    Encrypt
    0.51
    を含む
    0.50
    ♦
    0.49
    ###
    0.49
    ••••
    0.48
    Activations Density 0.022%

    No Known Activations