© 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. 6015
    Prev
    Next
    INDEX
    Explanations

    1: followed by units

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

    The neuron primarily detects numeric tokens (stand-alone numbers, decimals, and measurements).

    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
     hasattr
    0.63
     ২৮
    0.61
     中
    0.59
     ૨
    0.59
     ۱۹
    0.58
     १२
    0.55
     ۱۴
    0.55
     위
    0.55
     달리
    0.54
     accelerator
    0.54
    POSITIVE LOGITS
     apiece
    0.89
     dollars
    0.86
     shillings
    0.84
     ounces
    0.79
     rubles
    0.77
    dollars
    0.76
     dólares
    0.75
     cents
    0.73
     bucks
    0.73
     dígitos
    0.73
    Activations Density 0.247%

    No Known Activations