© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2133
    Prev
    Next
    INDEX
    Explanations

    tables, headers, and lists

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

    The neuron fires primarily on numeric tokens, i.e. occurrences of numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    formatting codes used in academic citations and bibliographic references, particularly special characters like `&###xA0;` (non-breaking spaces) and punctuation marks in citation contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    bibliographic reference formatting and web markup artifacts (e.g., citation metadata, URLs, brackets, and HTML/Unicode space entities) in academic-style entries.

    oai_token-act-pair · gpt-5Triggered 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
    tres
    0.53
    gae
    0.52
    tage
    0.52
    وامی
    0.51
     locis
    0.51
     dispos
    0.51
     modal
    0.50
    leme
    0.50
    ில்லை
    0.50
    modal
    0.49
    POSITIVE LOGITS
    {$\
    0.71
    $\
    0.69
     {\
    0.65
     文字
    0.63
    {\
    0.62
     Пусть
    0.62
     ${\
    0.61
     $|\
    0.61
    )}}\
    0.61
     俺
    0.61
    Activations Density 0.012%

    No Known Activations