© 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. 12067
    Prev
    Next
    INDEX
    Explanations

    October dates

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

    The neuron activates on numeric tokens—especially decimal‐formatted numbers—such as citation page numbers and retrieval timestamps.

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

    dates in numeric format, particularly those formatted with slashes or following words like "on" or "expires".

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

    numbers with decimal points, especially those appearing in citation or metadata lines such as retrieval/archival details.

    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
     компетен
    0.60
     கரு
    0.58
     деятельность
    0.58
    จำ
    0.58
    狙
    0.57
    uits
    0.57
    튿
    0.57
     bifur
    0.56
     resultó
    0.56
    IActionResult
    0.56
    POSITIVE LOGITS
    0.70
     الأمريكي
    0.57
     ٢
    0.56
     Apr
    0.55
     ١
    0.54
     बढ़ी
    0.54
     '/':
    0.53
     ۱۴
    0.52
     Sementara
    0.51
     ១
    0.51
    Activations Density 0.064%

    No Known Activations