© 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. 13467
    Prev
    Next
    INDEX
    Explanations

    language names and technical terms

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

    This neuron primarily detects numeric values, especially quantitative measurements (e.g., dimensions, heights, dates, and other number-based data).

    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
    이고
    0.68
     거고
    0.64
    이며
    0.62
     있고
    0.61
    }$,
    0.58
    ँग
    0.57
    했고
    0.55
     δυνα
    0.53
     آقای
    0.53
     hight
    0.52
    POSITIVE LOGITS
    ]]]]
    0.65
    "/></
    0.58
    ','')
    0.55
     Tất
    0.55
     XNUMX
    0.54
    جميع
    0.54
    Xem
    0.53
     উভয়ে
    0.52
     모두
    0.51
    TargetFramework
    0.51
    Activations Density 0.010%

    No Known Activations