© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13662
    Prev
    Next
    INDEX
    Explanations

    more than or perfectly

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

    This neuron fires on numerical quantity expressions—especially “more than [number]” or other large numeric comparisons.

    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_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
    स्‍या
    0.66
    サン
    0.61
    $+
    0.61
     lekin
    0.61
    ​,
    0.60
    ,...,
    0.59
    ກ
    0.58
    ‌,
    0.58
    osomal
    0.57
    theless
    0.57
    POSITIVE LOGITS
     suficiente
    0.87
     cukup
    0.82
     suffisamment
    0.80
    足够的
    0.80
     suficientes
    0.79
     voldoende
    0.77
     conceivably
    0.76
     đủ
    0.72
     충분
    0.72
    充分
    0.71
    Activations Density 0.123%

    No Known Activations