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

    Neuronpedia

    INDEX
    Explanations

    difference

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    duh
    0.47
    Paw
    0.43
     arbeitet
    0.43
    grund
    0.42
    Flor
    0.42
    rawler
    0.41
    deen
    0.41
    car
    0.41
    Ri
    0.41
    ür
    0.40
    POSITIVE LOGITS
     difference
    0.51
     differences
    0.50
     διαφο
    0.49
     disclosure
    0.48
     additional
    0.46
     Differ
    0.46
     orig
    0.45
     difer
    0.45
     डिफर
    0.44
    额外的
    0.44
    Act Density 0.001%

    No Known Activations