© 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. 5304
    Prev
    Next
    INDEX
    Explanations

    adding text to names

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

    This neuron detects numeric tokens—especially decimal numbers representing amounts, measurements, or input values.

    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
    azuje
    0.58
    ningen
    0.57
     опубликован
    0.56
    oribus
    0.56
    зки
    0.54
    andte
    0.54
     ات
    0.52
     الشرق
    0.52
     pengambilan
    0.52
    eluarkan
    0.52
    POSITIVE LOGITS
     names
    1.09
     gewüns
    1.09
     desired
    1.07
     имена
    1.05
    desired
    1.02
    姓名
    1.02
    用户名
    1.02
     원하는
    0.99
     이름을
    0.98
     название
    0.94
    Activations Density 0.391%

    No Known Activations