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

    Neuronpedia

    INDEX
    Explanations

    rounding and truncation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     وهم
    0.39
    อม
    0.37
    бло
    0.36
     Inputs
    0.36
    潞
    0.36
    諰
    0.36
    入力
    0.36
    输入
    0.35
     expression
    0.35
     einge
    0.35
    POSITIVE LOGITS
     trunc
    0.71
     truncation
    0.69
     truncate
    0.60
    Trunc
    0.59
    trunc
    0.57
     towards
    0.53
     Nearest
    0.52
    Rounding
    0.52
     toward
    0.51
     Toward
    0.49
    Act Density 0.011%

    No Known Activations