Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    directory where saved

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Tested
    0.41
     preexisting
    0.40
    ifiée
    0.38
    ходили
    0.38
    allery
    0.37
     ነገ
    0.37
     condiment
    0.36
    liques
    0.36
    glichen
    0.36
     எடு
    0.36
    POSITIVE LOGITS
    刚才
    0.68
     saving
    0.64
     Save
    0.64
     saved
    0.63
     Saving
    0.58
     保存
    0.57
     save
    0.55
     saves
    0.53
     Saved
    0.52
    剛剛
    0.52
    Act Density 0.001%

    No Known Activations