© 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. 890
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on floating-point numeric tokens (i.e. numbers with decimal points).

    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
    /
    0.64
    c
    0.63
     ไม่
    0.63
    when
    0.62
    k
    0.62
    人に
    0.61
    เมื่อ
    0.60
    x
    0.59
    '
    0.58
    F
    0.58
    POSITIVE LOGITS
     several
    3.02
    several
    2.66
     нескольких
    2.48
     plusieurs
    2.47
     mehrere
    2.47
     چندین
    2.39
     two
    2.38
     няколко
    2.34
     varios
    2.31
     flera
    2.31
    Activations Density 1.488%

    No Known Activations