© 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. 24-GEMMASCOPE-2-RES-16K
    7. 2089
    Prev
    Next
    INDEX
    Explanations

    meditation

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

    The neuron is specifically picking out passages containing quantitative or statistical content—numbers, measurements, dates, or references to numerical relations.

    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_24_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
     längst
    0.87
     valmist
    0.80
     lattices
    0.79
     aspires
    0.73
     Auctions
    0.72
     tycker
    0.71
     bidders
    0.70
    ’:
    0.67
    ക്ക്‌
    0.67
    upol
    0.66
    POSITIVE LOGITS
    ט
    0.88
    قه
    0.82
    ୦
    0.82
    일
    0.80
    يام
    0.80
    esperienza
    0.78
    Cantidad
    0.76
    鼕
    0.75
    bble
    0.75
    olique
    0.75
    Activations Density 0.000%

    No Known Activations