© 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. 1152
    Prev
    Next
    INDEX
    Explanations

    Course Goal

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

    This neuron strongly detects common low-content “function” words (like any, be, just, including).

    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
     tevé
    0.81
    ين
    0.77
     Plenum
    0.76
    раў
    0.76
     envisaged
    0.75
     acrylonitrile
    0.75
    垪
    0.75
     combinatorial
    0.75
    ары
    0.73
    lö
    0.71
    POSITIVE LOGITS
     електро
    0.83
    きました
    0.81
     зовніш
    0.74
    啸
    0.70
    했어요
    0.69
     особи
    0.68
     '**
    0.68
    お
    0.68
    hq
    0.67
     миру
    0.67
    Activations Density 0.000%

    No Known Activations