© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6828
    Prev
    Next
    INDEX
    Explanations

    large numbers like billions and millions

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

    The neuron detects large numeric-quantity expressions (e.g. “thousands,” “hundreds,” “tens of millions”).

    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
     Miscellaneous
    0.68
    som
    0.65
     மொத்தம்
    0.62
     ಸ್ವಲ್ಪ
    0.61
     aproximadamente
    0.60
    +](=
    0.60
     মোটামুটি
    0.60
     gewisse
    0.59
     Somewhat
    0.59
    uento
    0.59
    POSITIVE LOGITS
     thousands
    1.94
     hundreds
    1.84
     millions
    1.78
    Thousands
    1.77
    Hundreds
    1.76
    thousands
    1.70
     Thousands
    1.62
     tens
    1.59
     decenas
    1.57
     Hundreds
    1.56
    Activations Density 0.174%

    No Known Activations