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

    Neuronpedia

    INDEX
    Explanations

    comparison words and names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    真是
    1.18
    果然
    1.16
    Did
    1.14
     adjournment
    1.13
    éditeur
    1.10
    must
    1.05
     must
    1.04
     selalu
    1.04
     begle
    1.04
     Disco
    1.04
    POSITIVE LOGITS
     than
    4.81
    than
    4.27
    Than
    3.88
     Than
    3.86
     niż
    3.58
     decât
    3.54
     daripada
    3.48
     än
    3.30
     THAN
    3.28
     kuin
    3.06
    Act Density 1.015%

    No Known Activations