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

    Neuronpedia

    INDEX
    Explanations

    .py file operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     landen
    0.44
     precipitation
    0.43
     snowman
    0.40
     secretory
    0.40
     KIR
    0.40
     settling
    0.39
     kosong
    0.39
     ghi
    0.39
     articulation
    0.39
     secretion
    0.38
    POSITIVE LOGITS
     وعند
    0.47
    然后在
    0.45
    读取
    0.42
    可以使用
    0.42
    awak
    0.40
     späteren
    0.40
    accessed
    0.40
    access
    0.40
    consulté
    0.40
    зже
    0.40
    Act Density 0.104%

    No Known Activations