© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 11024
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on large decimal numbers (e.g. conversion factors), i.e. numeric tokens with multi-digit integer parts and fractional components.

    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
     verbess
    0.73
    krieg
    0.70
    zust
    0.69
    iencia
    0.69
     भरोसा
    0.67
     보유
    0.66
    푞
    0.66
     verbessern
    0.66
     öffnen
    0.64
    .>
    0.64
    POSITIVE LOGITS
     şi
    0.76
    ळी
    0.74
    ه‌های
    0.72
     і
    0.71
    א
    0.71
    સ્ત
    0.70
     opération
    0.70
    ève
    0.67
     হিসাব
    0.65
    лите
    0.63
    Activations Density 0.000%

    No Known Activations