© 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. 11153
    Prev
    Next
    INDEX
    Explanations

    numbers and structure

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

    The neuron responds almost exclusively to numeric literals containing a decimal point.

    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
     HREF
    0.72
    袂
    0.68
     sepen
    0.68
    طيع
    0.67
     menik
    0.65
     menyatakan
    0.65
    agenda
    0.64
     Artinya
    0.64
     donnant
    0.63
     చెబు
    0.63
    POSITIVE LOGITS
     pharmacological
    0.75
     motility
    0.71
     trespassing
    0.71
     geotechnical
    0.70
    лимпий
    0.68
    diaz
    0.68
     水
    0.68
     প্রাণী
    0.67
     snoring
    0.67
     पजल
    0.67
    Activations Density 0.000%

    No Known Activations