© 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. 24-GEMMASCOPE-2-RES-16K
    7. 15003
    Prev
    Next
    INDEX
    Explanations

    biographical dates and names

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

    The neuron selectively activates on numeric tokens—especially years, dates, and other numerical data.

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

    birth years in biographical text, particularly four-digit years in the format "19XX" when following parentheses and the word "born".

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

    biographical parenthetical date information—especially birth/death years and birth details—following people’s names.

    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_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
     जिसने
    0.58
     خاندان
    0.57
    及ひ
    0.56
     Seniors
    0.55
    很多人
    0.55
     Officers
    0.55
     Witnesses
    0.54
     Singers
    0.54
     officers
    0.54
     Menschen
    0.53
    POSITIVE LOGITS
    ocat
    0.59
    etition
    0.59
    valuate
    0.57
     завтра
    0.57
    ванию
    0.56
    ineuse
    0.56
    vanje
    0.55
    iksa
    0.54
     sicuro
    0.54
    ?)
    0.54
    Activations Density 0.012%

    No Known Activations