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

    Neuronpedia

    INDEX
    Explanations

    I. sections

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     voilà
    0.45
    Ainsi
    0.42
     nowe
    0.41
     matchups
    0.40
    有趣的
    0.38
    raseña
    0.36
     trajectories
    0.36
     liné
    0.36
     assembler
    0.36
    釹
    0.36
    POSITIVE LOGITS
    首先
    0.61
     Firstly
    0.59
    Firstly
    0.59
     divided
    0.58
     首先
    0.58
    分为
    0.54
     Divided
    0.54
    First
    0.53
     primero
    0.53
     먼저
    0.52
    Act Density 0.149%

    No Known Activations