© 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. 12077
    Prev
    Next
    INDEX
    Explanations

    prior delay

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

    The neuron primarily responds to numeric tokens showing decimal‐point values (floating‐point numbers).

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

    the word "for" when it appears before time durations or periods.

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

    phrases that mark time or duration, such as temporal transitions, dates, and multi-year spans.

    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
    ಿಕೆ
    0.82
    $};
    0.79
    icist
    0.79
     conmigo
    0.76
    秧
    0.76
     রোধ
    0.75
    язы
    0.75
     islas
    0.75
    ようになった
    0.72
     flujo
    0.72
    POSITIVE LOGITS
    于
    0.72
     à
    0.67
    crate
    0.65
    à
    0.65
    微
    0.65
    謂
    0.65
    RELAND
    0.64
    С
    0.64
    dub
    0.64
    版本
    0.63
    Activations Density 0.000%

    No Known Activations