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

    Neuronpedia

    INDEX
    Explanations

    checking conditional negation ("if not", "if !")

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    вій
    0.46
    ﺩ
    0.45
    bian
    0.42
     thirds
    0.41
     frapp
    0.40
    cstdlib
    0.40
    AAC
    0.40
    greater
    0.39
    EXCEPTION
    0.39
     instalment
    0.38
    POSITIVE LOGITS
    ని
    0.51
    尼亚克
    0.48
    ॉप
    0.46
    te
    0.46
    व
    0.44
     미리
    0.44
    സ്
    0.42
     remarked
    0.42
     або
    0.42
     preved
    0.41
    Act Density 0.025%

    No Known Activations