© 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. 14491
    Prev
    Next
    INDEX
    Explanations

    quantity estimates

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

    The neuron picks out numeric expressions—counts, amounts, years, and other statistical figures—in the text.

    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_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
    ifestyle
    0.64
    strap
    0.60
     সীমিত
    0.59
    ennies
    0.58
    inuous
    0.57
    azem
    0.57
    ąz
    0.57
    啕
    0.55
     кише
    0.55
    ాన్ని
    0.55
    POSITIVE LOGITS
    約
    0.83
    合計
    0.82
     eight
    0.80
     approximately
    0.79
     eighty
    0.79
     કુલ
    0.79
     Insgesamt
    0.79
     மொத்தம்
    0.79
     nearly
    0.78
    合计
    0.78
    Activations Density 0.137%

    No Known Activations