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

    Neuronpedia

    INDEX
    Explanations

    half of memory, art, mouth problems

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    зе
    0.43
    ⊆
    0.42
     bridesmaid
    0.42
     mugs
    0.41
     McInt
    0.40
     disappointing
    0.40
    fourths
    0.40
    ②
    0.40
    eringen
    0.39
    حن
    0.39
    POSITIVE LOGITS
    பு
    0.44
    Tr
    0.42
    σή
    0.41
    s
    0.40
    tir
    0.40
    समझ
    0.39
    r
    0.39
    ี
    0.39
    யா
    0.39
     chiq
    0.39
    Act Density 8.348%

    No Known Activations