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

    Neuronpedia

    INDEX
    Explanations

    confidently fabricated information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    描写
    0.66
     داری
    0.66
    ণিজ্য
    0.65
    ጾ
    0.65
     Window
    0.65
    submenu
    0.64
     Dom
    0.64
    型
    0.63
     Interrupt
    0.63
     programming
    0.63
    POSITIVE LOGITS
     falsehood
    2.08
     disinformation
    1.89
     debunk
    1.78
     hoax
    1.75
     credibility
    1.74
     거짓
    1.72
     skepticism
    1.72
     disbelief
    1.72
     misinformation
    1.72
     myths
    1.68
    Act Density 0.378%

    No Known Activations