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

    Neuronpedia

    INDEX
    Explanations

    numbers and calculations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ¾
    0.67
     ½
    0.59
     ০৯
    0.53
     (.
    0.50
     Half
    0.46
    ¾
    0.46
     half
    0.45
     pół
    0.44
    ⅝
    0.44
     ৭৫
    0.43
    POSITIVE LOGITS
     sunk
    0.37
    尊
    0.36
    تقد
    0.36
    lla
    0.34
     sinking
    0.34
    mniej
    0.33
    issner
    0.33
     vars
    0.32
     እንኳን
    0.32
    0.32
    Act Density 0.020%

    No Known Activations